Iej 04 2009

doi:10.1111/j.1365-2591.2008.01537.
REVIEW
The mysterious appearance of enterococci in lled root canals
M. Zehnder1 & B. Guggenheim2

Department of Preventive Dentistry, Periodontology, and Cariology and 2Institute of Oral Biology, University of Zu rich Center of Dental Medicine, Zu rich, Switzerland
1
Abstract
Zehnder M, Guggenheim B. The mysterious appearance of
enterococci in lled root canals. International Endodontic Journal, 42, 277287, 2009.
In this narrative review, the potential reasons for the high occurrence of enterococci in lled root canals are explored. The pulpless root canal appears to be a habitat for these bacteria, particularly for Enterococcus faecalis. However, re-surveying the literature in caries research, it can be concluded that, contrary to earlier belief, enterococci are rare if ever found at the advancing front of dentinal lesions. The same is the case for true primary endodontic infections, but some uncertainty remains, because the coronal seal and the history of teeth harbouring enterococci have rarely been accurately investigated. Furthermore, from longitudinal studies with a known infection at the initiation of treatment, which was carried out under controlled
asepsis, it is questionable whether enterococci are as difcult to eliminate from the canal system as is commonly held. A more likely explanation for the high occurrence of enteroccci in lled root canals is that they enter after treatment, but from which source? The intriguing nding in this context is that enterococci do not appear to be colonizers of the oral cavity. They are merely transient oral bacteria, unless there is a predilection site such as the unsealed necrotic or lled root canal. The origin of this infection is most likely food. Using the example of enterococci in lled root canals, this paper highlights the possible importance of transient microorganisms in the oral cavity and changes in a microenvironment that can create favourable conditions for infection. Keywords: enterococci, Enterococcus faecalis, food, review, root canal.
Received 1 September 2008; accepted 20 November 2008
Introduction
The search term enterococcus and root canal yielded 353 related articles from the PubMed database on May 29, 2008, when the work on this manuscript was initiated. This indicates a high interest in this bacterial genus, especially the species Enterococcus faecalis. However, other enterococcal species such as Enterococcus
Correspondence: Dr Matthias Zehnder, Department of Preventive Dentistry, Periodontology, and Cariology, University of Zu rich Center for Dental Medicine, Plattenstrasse 11, CH 8032 Zu rich, Switzerland (Tel.: +41 44 632 8610; fax: +41 44 634 4308; e-mail: matthias.zehnder@zzmk.uzh.ch).
faecium, Enterococcus casseliavus and Enterococcus dur` re 1975, ans have also been identied in root canals (Meja Ferrari et al. 2005). When talking to clinicians, there appears to be a general consensus that enterococci are hardy inhabitants of the necrotic root canal system, which are more difcult to eliminate than other taxa and are likely to survive chemomechanical root canal treatment (Stuart et al. 2006). However, this assumption may be wrong, or at least not entirely correct. Whilst there are some excellent reviews on microbiological aspects of enterococci and their elimination via antimicrobials (Portenier et al. 2003), there appears to be a lack of knowledge when it comes to the question why enterococci are likely to be found in lled root canal
2009 International Endodontic Journal
International Endodontic Journal, 42, 277287, 2009
277
The case of the Enterococcus Zehnder & Guggenheim
systems. The necrotic or improperly lled root canal system appears to be a habitat for enterococci, especially E. feacalis. Contrary to most other species, E. faecalis can survive on its own and appears to tolerate the ecological stress in the root canal niche better than most other taxa, which, in turn, may prot from its presence (Fabricius et al. 1982a). There can be no doubt that in those regions of the world where such analyses have been performed, enterococci appear to be common colonizers of lled root canals (Molander et al. 1998, Peciuliene et al. 2000, Hancock et al. 2001, Pinheiro et al. 2003, Rocas et al. 2008). Enterococci have been identied in a considerable proportion of teeth with persisting periapical lesions that had technically adequate (Sundqvist et al. 1998) as well as in counterparts that had insufcient root llings (Peciuliene et al. 2000). They appear to occur frequently in root lled teeth both in regions where calcium hydroxide is commonly used as an interim dressing (Molander et al. 1998) and in countries where this topical antiseptic is usually not applied (Hancock et al. 2001). In one study, the occurrence of viable enterococci in root lled teeth with apical periodontitis was as high as 64% of the culture-positive cases (Peciuliene et al. 2001). In theory, there are two possibilities to explain this nding: (i) enterococci are primary colonizers of the root canal system as it becomes necrotic and then survive endodontic treatment including root lling better than other taxa; or (ii) enterococci are opportunistic coronal invaders of the improperly sealed necrotic or lled root canal system and hence enter this system during or after treatment. The aim of this review is to investigate which of the above possibilities is more likely to explain the high occurrence of enterococci recovered from lled root canals.
Is it really difcult to eliminate enterococci from the root canal?

The eradication of enterococci from the root canal system and its tolerance to most antimicrobials is not the primary topic of this review. As indicated above, excellent reports on this issue have been published (Portenier et al. 2006). It should be realized, however, that the only study design that can conclusively assess the resistance of enterococci to chemo-mechanical root canal treatment is a longitudinal study in humans with a known infection at the initiation of the treatment and controlled asepsis during all the steps that follow. Ex vivo studies using models with infected dentinal tubules are certainly helpful for comparing the effectiveness of certain antimicrobials, but on the other hand may be
misleading when it comes to estimating the ability of enterococci to survive chemo-mechanical root canal treatment in situ. Enterococci are found in lled root canals regardless of the antimicrobials that were used during treatment (Hancock et al. 2001). Furthermore, the ability of enterococci to enter dentinal tubules observed in bovine teeth is probably clinically irrelevant, because tubules of the apical root dentine in the adult human tooth are sclerotic (Vasiliadis et al. 1983, Mjo r et al. 2001). Sclerotic dentine is not invaded by microorganisms (Shovelton 1964). Infected ramications of the root canal system that are in direct contact to the periodontium are clinically more important (Nair et al. 2005), yet the microorganisms therein have not been identied. Clinical studies with samples that were sent in from private practitioners are prone to another type of systematic error, namely, samples collected by individuals that are not primarily involved in microbiological research are likely not to represent the true root canal microbiota (Bender et al. 1964, Morse 1971). Last but not least, studies that found enterococci in lled root canals with persisting apical lesions cannot elucidate whether the target bacteria have survived treatment, entered during treatment, or even accessed the canal space after the root lling procedure. In terms of properly controlled longitudinal studies, the classical work on enterococci in root canals was reported by Engstro m (1964) who investigated the culturable microorganisms from root canals of 223 teeth that were either root lled or had a necrotic pulp. In this material, E. faecalis (no other enterococci were identied) was found in 20 teeth, i.e. in 9% of the cases. The canals were then treated chemomechanically using an iodophor or a quartenary ammonium compound in aqueous solution in combination with Dakins solution (0.5% NaOCl buffered with bicarbonate) to irrigate the canals. Between visits, 5% iodinepotassium iodide was sealed into the canal system. The rst antibacterial treatment failed in 13/20 cases infected with E. faecalis, compared to 45/114 cases infected with other taxa. Whilst Engstro m stated that this difference was not statistically signicant based on a non-disclosed test, it actually is (Fishers exact test, twotailed, P = 0.049). On the other hand, the cases with enterococcal infection were relatively few, and it would thus not be justied to make clinical conclusions based on these observations. Moreover, the true clinical outcome (i.e. the healing rate) was not assessed in Engstro ms study. In a more recent study (Ferrari et al. 2005), enterococci were identied in 6/25 (24%) of
278
Zehnder & Guggenheim The case of the Enterococcus
single-rooted teeth with intact pulp chamber and apical periodontitis. The teeth were instrumented and rinsed with 0.5% NaOCl and then EDTA. At the end of the rst treatment, no canals harboured culturable enterococci, whilst 5/25 canals still contained other species. The canal was then left empty and sealed with a zinc oxide cement for 7 days. At the initiation of the second visit, 14/25 canals harboured enterococci including E. faecalis, E. faecium and E. casseliavus. The authors explained the occurrence of enterococci at the second visit in canals that did not show these bacteria at the beginning with an initial number of enterococci that was below the detection limit. However, despite the unusually high occurrence of enterococci in 24% of the primarily infected cases reported by these authors, there were too few observations to allow any general conclusions from this material. In all other longitudinal studies on root canal disinfection published so far, too few cases contained enterococci to allow any statistical analysis (see below: Enterococci in primary root canal infections). Studies with a controlled infection in monkey teeth draw a slightly different picture regarding the survival of enterococci in root canals. When canals containing necrotic pulp tissue were autogenously infected with oral microorganisms and then sealed, enterococci, if present, were recovered in similar absolute numbers during an experimental period of 1060 days, but were gradually outnumbered by strict anaerobes (Fabricius et al. 1982b). In a later study using a ve-strain combination including E. faecalis, the latter could be re-isolated from 24/24 monkey root canals 812 months after closure (Mo ller et al. 2004). Enterococcus faecalis appeared to also survive chemomechanical treatment using 1% NaOCl and 10% H2O2 slightly better than the anerobes in the ve-strain combination; however, only in 3/24 cases E. faecalis was the only culturable strain, compared to 14/24 root canals, in which E. faecalis survived treatment together with other taxa. Nevertheless, in 21/24 root canals inoculated with the ve-strain combination one or more strains survived, compared to 99/160 inoculated with the four-strain combination (Fishers exact test, twotailed, P = 0.02). Furthermore, the presence of E. faecalis in the original inoculum made it more likely that some bacteria remained viable after the root lling procedure in the long term (Fabricius et al. 2006). In summary, it may indeed be so that once a root canal is infected with enterococci, proper disinfection may be harder to achieve. However, the relative importance of enterococci in this context appears to
be over-estimated, and the high occurrence of enterococci in lled root canals cannot be explained based simply on a higher resistance to antimicrobials of this compared with other species. So the question remains as to when enterococci are most likely to enter the root canal. Furthermore, the source of this infection is of interest. The rest of the review will thus focus on the possible entry of enterococci during the necrotizing process of the pulp, the root lling process and the restorative phase.
Enterococci in dentinal caries

There appears to be a consensus that the primary causes of pulpal necrosis and the subsequent occurrence of apical periodontitis are dental caries and its sequelae. The progressive infection of dentine eventually leads to microabscesses in the pulp and tissue breakdown mediated by proteolytic enzymes (Langeland 1987, Gusman et al. 2002, McLachlan et al. 2004). It has been known for some time that, as a carious lesion progresses into the dentine close to the pulp, the microbial inltrate therein resembles the one in primary root canal infections (Edwardsson 1974). This was recently reconrmed in a study using contemporary molecular biology methods (Martin et al. 2002). In this context, it would be of interest to know whether enterococci are present in dentinal caries and consequently would be amongst the early invaders of the necrotizing pulp space. In 1933, Wohlfeil speculated that enterococci could cause dental caries based on the observation that they occurred more frequently around carious teeth and in individuals with bad oral hygiene than in orally healthy patients (Wohlfeil 1933). However, at that time, proper methods for the identication of oral streptococci (enterococci were included in the streptococci) were not available (Isenberg et al. 1970). A prominent example for this is the fact that Streptococcus mutans, although rst detected and associated with caries in 1924 (Clarke 1924), could not be properly discriminated from other oral streptococci until the 1960s (Carlsson 1967, Guggenheim 1968). Enterococci were traditionally identied by their characteristic morphology, Gram staining, haemolysis, catalase test and their ability to grow in methylene blue milk or on eosin-methylene blue agar (Sherman 1938, Isaacs & Scouller 1948). However, these methods do not differentiate between enterococci and oral streptococci (Guggenheim 1968). Moreover, non-enterococcal streptococci bearing Lanceelds group-D antigen, such as Streptococcus bovis, have even
279
more similarities with enterococci than other oral streptococci when identied with phenotypic identication methods (Facklam 1973). The genus Enterococcus was proposed only in 1984 based on genomic differences between Streptococcus faecalis and Streptococcus faecium and other streptococci bearing the Lanceeld group-D antigen (Schleifer & Klipper-Baelz 1984). Streptococcus bovis is found in the oral cavity (Crawford & Russell 1983) and is more cariogenic in the gnotobiotic rat than E. faecalis (Drucker & Green 1981, Willcox et al. 1990). The rst experimental study that found an association between enterococci and dental caries was undertaken in the cotton rat (Wakeman et al. 1948). However, the identication tests used were far from sufcient to properly identify enterococci (Guggenheim 1968, Facklam 1973). Burnett & Scherp (1951) published a study on the culturable microorganisms from the advancing front of dentinal lesions. They isolated aciduric streptococci, which they presumed to be enterococci because they were non-haemolytic, catalase-negative and grew in methylene-blue milk. Furthermore, these bacteria grew on 40% bile agar and at pH 9.6. The latter, but not the former feature appears to make it likely that they were correctly identied as enterococci and not S. mutans-like strains (Edwardsson 1968). However, the production of extracellular polysaccharides in 5% sucrose agar was not tested, which would have provided a clearer differentiation (Guggenheim 1968). Based on the Burnett and Scherps study, the rst dental caries in a gnotobiotic rat model was produced by a S. faecalis-like strain in combination with a proteolytic bacillus (Orland et al. 1955). Later, when proper biochemical identication tools for enterococci were available, Gold et al. (1975) showed that indeed some of the enterococcus species isolated from human oral cavities were able to cause caries in germ-free rats. However, they also observed that in the oral cavity of conventionally maintained rats, these strains could not survive and were not detectable 2 weeks after inoculation. Hence, interest was lost to study caries induction by enterococci any further in conventionally maintained rats. Later studies conrmed the low cariogenic potential of different E. faecalis strains under laboratory conditions (Drucker & Green 1981, Chestnutt et al. 1994). A further bias in early studies on caries close to the pulp was the fact that proper anaerobic techniques for the recovery of strict anaerobes were not available before 1969 (Aranki et al. 1969). Consequently, as was shown later (Hoshino 1985), the relative number of
facultative anaerobes was over-estimated in early caries studies. Interestingly, none of the researchers who controlled contamination of caries samples from the tooth surface and used modern culture and identication techniques found enterococci in carious dentine close to the pulp (Hahn et al. 1991, Hoshino et al. 1992, Martin et al. 2002). In a well-controlled study of the microorganisms related to early ssure caries in naval recruits, cultures from 48 carious ssures and 20 healthy control ssures yielded only 4 E. faecalis-positive samples from carious and 0 from non-carious sites, as compared to 48 and 17 positive cultures for S. mutans, respectively (Meiers et al. 1982). Even in positive samples, the counts of E. faecalis were three orders of magnitude below those of S. mutans. In conclusion, early studies erroneously linked enterococcci with caries, because this genus was known from other elds of microbiology, was easy to culture, but could not be properly differentiated from oral streptococci that were later identied in carious lesions and dental plaque. Given the evidence we have today it is rather unlikely that enterococci occur at the forefront of carious lesions.
Enterococci in primary root canal infections

Even if not present in caries, enterococci may theoretically still enter the necrotizing pulp space at an early stage. Not all primary endodontic infections are caused by caries (Abbott 2004). Many teeth contain cracks (Ratcliff et al. 2001), and thus it is conceivable that enterococci found in primary root canal infections entered via that route rather than dentinal tubules in carious teeth. Furthermore, inadequate coronal restorations with open margins have been linked to the occurrence of apical periodontitis (Kirkevang et al. 2007). However, the issue of studying primary root canal infections is difcult, especially when trying to determine how and when the microorganisms that eventually lead to pulpal breakdown entered the endododontic system. One of the most prominent problems in this context is again contamination from saliva or plaque from the outer tooth surface (Mo ller 1966). Because root canal infections are usually polymicrobial (Sundqvist 1994), and the most common invaders of the endodontium can be found in other sites of the oral cavity, it is impossible in the laboratory to discern between the microorganisms that were actually present in the root canal at the time of sampling and contaminants. As already highlighted by Engstro m,
280
many of the teeth that harbour enterococci in the root canal system also show positive enterococcal growth on outer tooth surfaces (Engstro m 1964). False positive results are even more likely when PCR is used to detect specic DNA sequences of microorganisms suspected to be present in the root canal (Nair 2007). Hence, a meticulous sampling technique including disinfection of the tooth and the access cavity with the respective sterility checks from both sites is a prerequisite to yield meaningful results. Whilst such protocols have been validated and published for both culture and molecular methods (Mo ller 1966, Ng et al. 2003), relatively few studies have complied with these guidelines. In studies on the culturable microbiota of primary endodontic infections with proper sterility checks of the access cavity, enterococci have usually been found in a rather low proportion of infected canals or not at all (for review, see Portenier et al. 2003). This is in contrast to the high frequency of enterococci encountered in lled root canal systems associated with treatment failure (Molander et al. 1998, Sundqvist et al. 1998, Peciuliene et al. 2000, Hancock et al. 2001). On the other hand, studies employing DNA-DNA hybridization or PCR techniques (without checking the access cavity) found a somewhat higher occurrence of E. faecalis in primary root canal infections as compared to those investigations, which were performed using culture techniques (Siqueira et al. 2002, Pirani et al. 2008). Nevertheless, the occurrence of E. faecalis in a PCR assay was still signicantly lower in teeth containing necrotic pulps compared with root lled counterparts with apical lesions (Pirani et al. 2008). Contrary to these ndings, a relatively high occurence of E. faecalis in primary root canal infections has been reported when nested PCR was used to identify this taxon and was compared with a conventional culture technique: E. faecalis was identied in 82% vs. 4%, respectively (Gomes et al. 2006). The authors concluded that E. faecalis could be present in numbers below the detection level for culturing or that they could be in a viable but non-culturable state. On the other hand, the authors also conceded that 49/50 of their sampled teeth had coronal leakage. Consequently, proper decontamination of the access for PCR was a difcult, if not impossible task, and enterococcal DNA could thus have originated from sources outside the root canal. Furthermore, nested PCR is notoriously difcult to perform; the paper describes only sequencing a single band from the gels this would afford little condence in the band identication. Moreover, it is well-known that E. faecalis is one of the easiest species to culture,
and it does not enter a viable but non-culturable state (Bogosian et al. 1998). In summary, as already suspected from the low or non-existing occurrence of enterococci in caries, members of the genus Enterococcus probably exist relatively rarely in primary root canal infections. To the best of our knowledge, only one study so far has specically targeted the difference in the microbiota recovered from necrotic root canals between teeth with an exposed and counterparts with a non-exposed pulp space. However, in that investigation, cracks were not taken into consideration. Furthermore, the incidence of E. faecalis was merely 0/45 vs. 2/43 in pulps from teeth that had a visible contact to the oral cavity compared to those that did not, respectively (Chu et al. 2005). These numbers are again too low to allow any conclusions. The low occurrence of enterococci in primary root canal infections makes one possible pathway for the colonization of necrotic pulp tissue that cannot be terminally excluded unlikely: anachoresis, i.e. the transfection of microorganisms via the blood stream (Tunnicliff & Hammond 1937). Animal experiments with high numbers of microorganisms injected in the blood stream have shown that a colonization of necrotizing pulp tissue is possible (Gier & Mitchell 1968, Tziagas 1989). In the case of enterococci, it is conceivable that these bacteria could enter the bloodstream from the large intestine and then enter necrotic areas of a tooth with terminal pulpitis. However, as indicated above, the low occurrence of enterococci in primary root canal infections and the low likelihood of any bacterium to colonize necrotic pulp tissue via the blood stream make the pathway of direct oral entry more likely.
Enterococcal invasion of the root canal during or after treatment

Few data exist in the literature to support or contradict the theory that enterococci could enter the root canal system during or even after endodontic treatment. It has been surmised from longitudinal studies that culture reversals with the sudden occurrence of enterococci after the initial treatment session could be due to leakage through the temporary lling (Sjo gren et al. 1991, Sundqvist et al. 1998). However, again these observations were too infrequent to allow generalization. The only study so far that has addressed the correlation between the clinical occurrence of enterococci and other enteric bacteria and the root canal seal was based on samples that were sent in by private practitioners, accompanied by a questionnaire regard-
281
n ing the treatment steps that had been performed (Sire et al. 1997). As it turned out, there was a signicant positive correlation between the occurrence of the target species and the number of visits, as well as leaving the canal unsealed between treatment sessions. Studies on the occurrence of enterococci in root lled teeth (Engstro m 1964, Molander et al. 1998, Kaufman et al. 2005) also suggest that these bacteria could have entered the canal system after the root lling procedure. Enterococci are able to induce and maintain an apical lesion as monoinfectants (Fabricius et al. 1982a, Ferrari et al. 2005). On the other hand, they have been found more frequently in lled canals without a radiographic lesion compared with counterparts with a lesion (Kaufman et al. 2005). It appears rather unlikely that enterococci from a primary infection survived treatment and root lling procedures only in the coronal aspect of the canal and not in the apical portion (which would, with a high degree of certainty, result in a lesion). Hence, it may be assumed from everything that we know at this point that enterococci in lled root canals without apical rarefaction are likely to have entered after the root was lled. In this context, it should be stated that almost one general shortcoming in endodontic articles is the lack of information regarding the restoration and the history of the teeth under investigation. Depending on the coronal restoration, lled root canal systems may invariably have been exposed to the oral cavity at one point during treatment. This is especially the case with teeth that receive an indirect restoration which undergo a phase of temporization. However, clinical studies that investigated microbial leakage around temporary llings or crowns are few, and the involved microorganisms have not been identied (Beach et al. 1996). Taken together, the little evidence currently available points in the direction that enterococci enter the root canal system at some time after the root canal treatment has been initiated. The source of infection for pulpless root canals appears to be the oral cavity with its currently more than 700 identied bacterial species or phylotypes. However, whether the oral cavity is a habitat for enterococci is the next question.
Are enterococci colonizers of the oral cavity?

This question may appear somewhat stupid, because common sense would dictate that, of course, if enterococci can maintain in necrotic root canals, why should they not be present at other oral sites? However,
microorganisms that can enter a specic niche in the oral cavity that is not present in all individuals (in our case the unsealed pulpless root canal) need not necessarily be consistent inhabitants of adjacent sites. The issue that is often overlooked in this context is infections caused by transient oral microorganisms that have recently gained some interest in connection with the survival of probiotics in the oral cavity (Meurman 2005). It should be realized that humans live in an environment surrounded by a complex microbiota, continuously inhaling and ingesting microorganisms that are unable to colonize epithelial and tooth surfaces permanently (Pamer 2007). The healthy gastrointestinal mucosal surfaces are inhabited by microbial populations that, in aggregate, are called the commensal ora (outdated term) or microbiota (Ley et al. 2006). The local composition of the commensal microbiota is site-specic and can vary considerably. The main habitat of enterococci, as the name indicates, is the gastrointestinal tract. In humans, E. faecalis and E. faecium are the predominant species, whilst E. faecium is the predominant species in poultry and pigs. However, enterococci are not part of the typical commensal microbiota of the oral cavity, that has recently been dened using culture-independent techniques (Aas et al. 2005). Sampled sites included tongue dorsum, lateral sides of tongue, buccal epithelium, hard palate, soft palate, supragingival plaque of tooth surfaces, subgingival plaque, maxillary anterior vestibule and tonsils. Interestingly, enterococci were not found in any of the sites of the ve individuals that were screened. Instead, most sites were covered with protective microorganisms such as Streptococcus mitis and Streptococcus salivarius, but the bacterial richness was striking, with well over 100 predominant bacterial taxa representing six phyla that were identied, almost half of which had never been cultured. As reviewed by Engstro m, early studies that specifically screened the oral cavity for the presence of viable enterococci were inconsistent (Engstro m 1964). The most thorough of these early studies was probably the one carried out by Williams et al. (1950). The authors were looking for enterococci and yeasts in saliva based on an earlier nding in their related dental clinic that these microorganisms were frequently recovered from root canals that had been dressed with a penicillin/ streptomycin mix (Grossman & Steward 1949). Saliva samples from 206 individuals were collected and analysed for enterococci. From most donors, saliva was collected more than once. Overall, 45 persons
282
(21.8%) had enterococci (mostly E. faecalis) in their saliva at least once. However, the carriage was not consistent, meaning that on one experimental day an individual tested positive, on another day the same individual tested negative. This correlates rather well with the above-mentioned observation in rats with a normal oral microbiota, namely, the nding that enterococci that were introduced into the oral cavity did not maintain over time (Gold et al. 1975). In a recent study, the prevalence, phenotype and genotype of oral enterococci was studied (Sedgley et al. 2004). Enterococci were detected in oral rinse samples from 11/100 patients receiving endodontic treatment and 1/ 100 dental students with no history of endodontic treatment (P = 0.0027). All enterococcal isolates were identied as E. faecalis. The data from that study could be interpreted in the direction that the occurrence of E. faecalis could have something to do with oral hygiene, as dental students should technically have a better oral hygiene than endodontic patients. This, however, was not assessed. In an even more recent study no viable entrococci were found in any of the oral rinse samples from 50 dental students with good oral hygiene and few llings (Razavi et al. 2007). Taken together, it could be stated that the most common enterococcal species found in the oral cavity occasionally is E. faecalis. This is consistent with the predominance of this species in the root canal. However, enterococci are rarely found in the oral cavity of healthy subjects with a good oral hygiene, and are not consistently recovered from the oral cavity of average patients. How can this be? The most logical explanation for this is that enterococci are just transient bacteria in the oral cavity. And the most likely source of enterococci that can be found in the oral cavity at times is food.
Enterococci in food
Enterococcus faecalis is an indicator for contamination of food, water and inanimate objects from human sources (Reuter 1992). On the other hand, enterococci are ubiquitous in food products for raw consumption (Franz et al. 2003). Only few health problems have been reported related to this fact (Kayser 2003). Enterococci are even used as veterinary food supplements. Since February 2004, nine different strains of E. faecium are authorized as additives in feeding stuff in the European Union (Foulquie Moreno et al. 2006). Furthermore, E. faecium SF 68 (Cylactin, Hoffmann-La Roche, Basel, Switzerland) is a probiotic used in humans, which was
shown to be effective against different types of diarrhoea (Foulquie Moreno et al. 2006). The resistance of enterococci to high temperatures and their ability to grow under a wide array of conditions is the reason why they can be found in food not only manufactured from raw materials but also in heat-treated merchandise. Milk products, meat products for raw consumption, vegetables and olives commonly contain enterococci (Franz et al. 2003). Enterococcus faecalis is mostly found in milk products such as cheese and in fermented sausages for raw consumption, but can also commonly be isolated from various other food products such as minced beef, minced pork and sh/crustacea (Klein 2003, Peters et al. 2003). The presence of enterococci in food is not necessarily always unwanted, as they can add to the specic taste of, for instance, Mediterranean cheeses. Enterococci have even been used as starter cultures to ferment olives and cheese (Giraffa 2002). The possibility that enterococci could cause a transient oral infection has only been identied recently (Razavi et al. 2007). In a study with healthy volunteers, the clearance of E. faecalis from a highly contaminated cheese (Brie de Meaux containing a mean of 4.8 104 E. faecalis colony-forming units g)1) was investigated. The volunteers refrained from eating or drinking between cheese ingestion and the collection of an oral rinse sample 1, 10, or 100 min after cheese ingestion. Between experimental days, the volunteers maintained their normal oral hygiene and diet habits for 1 week. One minute after ingestion, a median of 5480 E. faecalis colony-forming units was recovered. Bacterial counts were reduced after 10 min, dropped after 100 min to levels that were signicantly (P < 0.005) different from 1 and 10-min scores and were below the detection limit after 1 week. These ndings suggest that enterococci do adhere to oral tissues of healthy subjects but fail to grow when competing with the normal oral microbiota and are thus gradually eliminated. However, the rate of clearance of the food-derived enterococci from the oral cavity was such that infection of predilection sites cannot be excluded. This adds weight to the hypothesis that enterococcal root canal infections could be foodborne. In a follow-up study, the possibility that E. faecalis from food could enter the endodontium was explored in an articial oral environment in a mastication apparatus (Kampfer et al. 2007). It was shown that viable enterococci could indeed leak through a calcium sulphate-based temporary lling material, if its thickness was below 4 mm.
283
Concluding remarks and call for future studies

Enterococci, especially E. faecalis strains, appear to adapt to the habitat of a treated root canal better than other taxa. From the information available, it may be concluded that these bacteria are not amongst the early invaders of the necrotizing root canal system. They may enter the root canal at any point in time during or after treatment if the coronal seal is inadequate. Their source is most likely food. In individuals with an adequate level of oral hygiene, they do not colonize the oral cavity. However, they may enter the unsealed root canal system, where they nd a habitat that allows their growth and survival. Future studies should be directed at several issues. First, virulence factors that favour the occurrence of enterococci in lled root canals should be identied. This work has already started and has yielded some interesting results (Sedgley 2007). Gelatinase production is one of the virulence factors that may be associated with the survival of E. faecalis in lled root canals. The E. faecalis strains producing gelatinase were termed S. faecalis var. liquefaciens in early studies and were frequently recovered from root canal systems (Guthof 1953, Winkler & van Amerongen ` re 1975). Interestingly 1959, Engstro m 1964, Meja and perhaps obviously, gelatinase production is also a factor that promotes the presence of enterococci in fermented food products (Franz et al. 2003). The origin of enterococci in root canals should be further identied by comparing the scheme of highly preserved genes between clones found in root canals and counterparts from food products, preferably from a similar area (Ruiz-Garbajosa et al. 2006). On a less sophisticated level, contamination of different enterococcal species in typical regional foods could be compared with the recovery of these species from root canals in a given country or area. Last but not least, it has been known for a long time that the healthy microbiota of the oral cavity can defend against potential pathogens (Deyloff & Sanders 1980). It would be interesting to identify the mechanisms preventing the colonization of the oral cavity by enterococci.
References
Aas JA, Paster BJ, Stokes LN, Olsen I, Dewhirst FE (2005) Dening the normal bacterial ora of the oral cavity. Journal of Clinical Microbiology 43, 572132. Abbott PV (2004) Assessing restored teeth with pulp and periapical diseases for the presence of cracks, caries and marginal breakdown. Australian Dental Journal 49, 339. Aranki A, Syed SA, Kenney EB, Freter R (1969) Isolation of anaerobic bacteria from human gingiva and mouse cecum by means of a simplied glove box procedure. Applied Microbiology 17, 56876. Beach CW, Calhoun JC, Bramwell JD, Hutter JW, Miller GA (1996) Clinical evaluation of bacterial leakage of endodontic temporary lling materials. Journal of Endodontics 22, 459 62. Bender IB, Seltzer S, Turkenkopf S (1964) To culture or not to culture? Oral Surgery Oral Medicine Oral Pathology 18, 527 40. Bogosian G, Morris PJ, ONeil JP (1998) A mixed culture recovery method indicates that enteric bacteria do not enter the viable but nonculturable state. Applied Environmental Microbiology 64, 173642. Burnett GW, Scherp HW (1951) Bacteriologic studies of the advancing dentinal lesion. Journal of Dental Research 30, 76677. Carlsson J (1967) A medium for isolation of Streptococcus mutans. Archives of Oral Biology 12, 16578. Chestnutt IG, MacFarlane TW, Stephen KW (1994) An in vitro investigation of the cariogenic potential of oral streptococci. Archives of Oral Biology 39, 58993. Chu FC, Tsang CS, Chow TW, Samaranayake LP (2005) Identication of cultivable microorganisms from primary endodontic infections with exposed and unexposed pulp space. Journal of Endodontics 31, 4249. Clarke JK (1924) On the bacterial factor in the aetiology of caries. British Journal of Experimental Pathology 5, 141 7. Crawford I, Russell C (1983) Streptococci isolated from the bloodstream and gingival crevice of man. Journal of Medical Microbiology 16, 2639. Deyloff JL, Sanders CC (1980) Inhibition of Streptococcus mutans by human plaque ora. Journal of Dental Research 59, 19539. Drucker DB, Green RM (1981) The relative cariogenicity of different streptococci in the gnotobiotic WAG/RIJ rat. Archives of Oral Biology 26, 599606. Edwardsson S (1968) Characteristics of caries-inducing human streptococci resembling Streptococcus mutans. Archives of Oral Biology 13, 63746. Edwardsson S (1974) Bacteriological studies on deep areas of carious dentine. Odontologisk Revy Suppl 32, 1143. Engstro m B (1964) The signicance of Enterococci in root canal treatment. Odontologisk Revy 15, 87106.
Acknowledgements
The authors thank Tuomas Waltimo and Monika Marending for their thoughtful review of this manuscript.
284
JR (1982a) Inuence n G, Holm SE, Mo Fabricius L, Dahle ller A of combinations of oral bacteria on periapical tissues of monkeys. Scandinavian Journal of Dental Research 90, 2006. JR (1982b) hman AE, Mo n G, O Fabricius L, Dahle ller A Predominant indigenous oral bacteria isolated from infected root canals after varied times of closure. Scandinavian Journal of Dental Research 90, 13444. JR n G, Sundqvist G, Happonen RP, Mo Fabricius L, Dahle ller A (2006) Inuence of residual bacteria on periapical tissue healing after chemomechanical treatment and root lling of experimentally infected monkey teeth. European Journal of Oral Sciences 114, 27885. Facklam RR (1973) Comparison of several laboratory media for presumptive identication of enterococci and group D streptococci. Applied Microbiology 26, 13845. Ferrari PH, Cai S, Bombana AC (2005) Effect of endodontic procedures on enterococci, enteric bacteria and yeasts in primary endodontic infections. International Endodontic Journal 38, 37280. Foulquie Moreno MR, Sarantinopoulos P, Tsakalidou E, De Vuyst L (2006) The role and application of enterococci in food and health. International Journal of Food Microbiology 106, 124. Franz CM, Stiles ME, Schleifer KH, Holzapfel WH (2003) Enterococci in foods a conundrum for food safety. International Journal of Food Microbiology 88, 10522. Gier RE, Mitchell DF (1968) Anachoretic effect of pulpitis. Journal of Dental Research 47, 56470. Giraffa G (2002) Enterococci from foods. FEMS Microbiology Reviews 26, 16371. Gold OG, Jordan HV, van Houte J (1975) The prevalence of enterococci in the human mouth and their pathogenicity in animal models. Archives of Oral Biology 20, 4737. Gomes BP, Pinheiro ET, Sousa EL et al. (2006) Enterococcus faecalis in dental root canals detected by culture and by polymerase chain reaction analysis. Oral Surgery Oral Medicine Oral Pathology Oral Radiology and Endodontics 102, 24753. Grossman LI, Steward GG (1949) An effective penicillin streptomycin suspension for endodontic treatment. Oral Surgery Oral Medicine Oral Pathology 2, 3748. Guggenheim B (1968) Streptococci of dental plaques. Caries Research 2, 14763. Gusman H, Santana R, Zehnder M (2002) Matrix metalloproteinase levels and gelatinolytic activity in clinically healthy and inamed human dental pulps. European Journal of Oral Sciences 110, 3537. Guthof O (1953) Zahnbakteriologische Fragen im Zusammenhang mit Wurzelbehandlung und Herdinfektion. Zahna rztliche Rundschau 62, 1903. Hahn CL, Falkler WAJ, Minah GE (1991) Microbiological studies of carious dentine from human teeth with irreversible pulpitis. Archives of Oral Biology 36, 14753. Hancock HHr, Sigurdsson A, Trope M, Moiseiwitsch J (2001) Bacteria isolated after unsuccessful endodontic treatment in
a North American population. Oral Surgery Oral Medicine Oral Pathology Oral Radiology and Endodontics 91, 57986. Hoshino E (1985) Predominant obligate anaerobes in human carious dentin. Journal of Dental Research 64, 11958. Hoshino E, Ando N, Sato M, Kota K (1992) Bacterial invasion of non-exposed dental pulp. International Endodontic Journal 25, 25. Isaacs A, Scouller JM (1948) Catalase production by Grampositive cocci: a simple test for differentiating enterococci from micrococci. Journal of Pathology and Bacteriology 60, 1356. Isenberg HD, Goldberg D, Sampson J (1970) Laboratory studies with a selective Enterococcus medium. Applied Microbiology 20, 4336. Kampfer J, Go hring TN, Attin T, Zehnder M (2007) Leakage of food-borne Enterococcus faecalis through temporary llings in a simulated oral environment. International Endodontic Journal 40, 4717. ngberg L, Barry J, Fouad AF (2005) Kaufman B, Spa Enterococcus spp. in endodontically treated teeth with and without periradicular lesions. Journal of Endodontics 31, 8516. Kayser FH (2003) Safety aspects of enterococci from the medical point of view. International Journal of Food Microbiology 88, 25562. Kirkevang LL, Vaeth M, Ho rsted-Bindslev P, Bahrami G, Wenzel A (2007) Risk factors for developing apical periodontitis in a general population. International Endodontic Journal 40, 2909. Klein G (2003) Taxonomy, ecology and antibiotic resistance of enterococci from food and the gastro-intestinal tract. International Journal of Food Microbiology 88, 12331. Langeland K (1987) Tissue response to dental caries. Endodontics and Dental Traumatology 3, 14971. Ley RE, Peterson DA, Gordon JI (2006) Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell 124, 83748. Martin FE, Nadkarni MA, Jacques NA, Hunter N (2002) Quantitative microbiological study of human carious dentine by culture and real-time PCR: association of anaerobes with histopathological changes in chronic pulpitis. Journal of Clinical Microbiology 40, 1698704. McLachlan JL, Sloan AJ, Smith AJ, Landini G, Cooper PR (2004) S100 and cytokine expression in caries. Infection and Immunity 72, 41028. Meiers JC, Wirthlin MR, Shklair IL (1982) A microbiological analysis of human early carious and non-carious ssures. Journal of Dental Research 61, 4604. ` re B (1975) Streptococcus faecalis and Streptococcus faecium Meja in infected dental root canals at lling and their susceptibility to azidocillin and some comparable antibiotics. Odontologisk Revy 26, 193204. Meurman JH (2005) Probiotics: do they have a role in oral medicine and dentistry? European Journal of Oral Sciences 113, 18896.
285
Mjo r IA, Smith MR, Ferrari M, Mannocci F (2001) The structure of dentine in the apical region of human teeth. International Endodontic Journal 34, 34653. n G, Kvist T (1998) Microbiological Molander A, Reit C, Dahle status of root-lled teeth with apical periodontitis. International Endodontic Journal 31, 17. JR (1966) Microbiological examination of root canals Mo ller A and periapical tissues of human teeth. Methodological studies. Odontologisk Tidskrift 74 (Suppl.), 1380. JR, Fabricius L, Dahle n G, Sundqvist G, Happonen RP Mo ller A (2004) Apical periodontitis development and bacterial response to endodontic treatment. Experimental root canal infections in monkeys with selected bacterial strains. European Journal of Oral Sciences 112, 20715. Morse DR (1971) The endodontic culture technique. Factors leading to false positive and negative results. New York State Dental Journal 37, 8894. Nair PNR (2007) Abusing technology? Culture-difcult microbes and microbial remnants Oral Surgery Oral Medicine Oral Pathology Oral Radiology and Endodontics 104, 56970. Nair PNR, Henry S, Cano V, Vera J (2005) Microbial status of apical root canal system of human mandibular rst molars with primary apical periodontitis after one-visit endodontic treatment. Oral Surgery Oral Medicine Oral Pathology Oral Radiology and Endodontics 99, 23152. Ng YL, Spratt D, Sriskantharajah S, Gulabivala K (2003) Evaluation of protocols for eld decontamination before bacterial sampling of root canals for contemporary microbiology techniques. Journal of Endodontics 29, 317 20. Orland FJ, Blayney JR, Harrison RW et al. (1955) Experimental caries in germfree rats inoculated with enterococci. Journal of the American Dental Association 50, 25972. Pamer EG (2007) Immune responses to commensal and environmental microbes. Nature Immunology 8, 11738. Peciuliene V, Balciuniene I, Eriksen HM, Haapasalo M (2000) Isolation of Enterococcus faecalis in previously root-lled canals in a Lithuanian population. Journal of Endodontics 26, 5935. Peciuliene V, Reynaud AH, Balciuniene I, Haapasalo M (2001) Isolation of yeasts and enteric bacteria in root-lled teeth with chronic apical periodontitis. International Endodontic Journal 34, 42934. Peters J, Mac K, Wichmann-Schauer H, Klein G, Ellerbroek L (2003) Species distribution and antibiotic resistance patterns of enterococci isolated from food of animal origin in Germany. International Journal of Food Microbiology 88, 3114. Pinheiro ET, Gomes BP, Ferraz CC, Teixeira FB, Zaia AA, Souza Filho FJ (2003) Evaluation of root canal microorganisms isolated from teeth with endodontic failure and their antimicrobial susceptibility. Oral Microbiology and Immunology 18, 1003. Pirani C, Bertacci A, Cavrini F et al. (2008) Recovery of Enterococcus faecalis in root canal lumen of patients with
primary and secondary endodontic lesions. New Microbiology 31, 23540. Portenier I, Waltimo TMT, Haapasalo M (2003) Enterococcus faecalis the root canal survivor and star in posttreatment diesease. Endodontic Topics 6, 13560. Portenier I, Waltimo T, Orstavik D, Haapasalo M (2006) Killing of Enterococcus faecalis by MTAD and chlorhexidine digluconate with or without cetrimide in the presence or absence of dentine powder or BSA. Journal of Endodontics 32, 13841. Ratcliff S, Becker IM, Quinn L (2001) Type and incidence of cracks in posterior teeth. Journal of Prosthetic Dentistry 86, 16872. Razavi A, Gmu r R, Imfeld T, Zehnder M (2007) Recovery of Enterococcus faecalis from cheese in the oral cavity of healthy subjects. Oral Microbiology and Immunology 22, 24851. Reuter G (1992) Culture media for enterococci and group D-streptococci. International Journal of Food Microbiology 17, 10111. Rocas IN, Hu lsmann M, Siqueira JFJ (2008) Microorganisms in root canal-treated teeth from a German population. Journal of Endodontics 34, 92631. Ruiz-Garbajosa P, Bonten MJ, Robinson DA et al. (2006) Multilocus sequence typing scheme for Enterococcus faecalis reveals hospital-adapted genetic complexes in a background of high rates of recombination. Journal of Clinical Microbiology 44, 22208. Schleifer KH, Klipper-Baelz R (1984) Transfer of Streptococcus faecalis and Streptococcus faecium to the genus Enterococcus nom.rev. as Enterococcus faecalis comb.nov. and Enterococcus faecium comb.nov. International Journal of Systematic Bacteriology 34, 314. Sedgley CM (2007) The inuence of root canal sealer on extended intracanal survival of Enterococcus faecalis with and without gelatinase production ability in obturated root canals. Journal of Endodontics 33, 5616. Sedgley CM, Lennan SL, Clewell DB (2004) Prevalence, phenotype and genotype of oral enterococci. Oral Microbiology and Immunology 19, 95101. Sherman JM (1938) The enterococci and related streptococci. Journal of Bacteriology 35, 8193. Shovelton DS (1964) The presence and distribution of microorganisms within non-vital teeth. British Dental Journal 117, 1017. Siqueira JFJ, Rocas IN, Souto R, de Uzeda M, Colombo AP (2002) Actinomyces species, streptococci, and Enterococcus faecalis in primary root canal infections. Journal of Endodontics 28, 16872. n EK, Haapasalo MP, Ranta K, Salmi P, Kerosuo EN Sire (1997) Microbiological ndings and clinical treatment procedures in endodontic cases selected for microbiological investigation. International Endodontic Journal 30, 915. ngberg L, Sundqvist G (1991) The Sjo gren U, Figdor D, Spa antimicrobial effect of calcium hydroxide as a short-term
286
intracanal dressing. International Endodontic Journal 24, 11925. Stuart CH, Schwartz SA, Beeson TJ, Owatz CB (2006) Enterococcus faecalis: its role in root canal treatment failure and current concepts in retreatment. Journal of Endodontics 32, 938. Sundqvist G (1994) Taxonomy, ecology, and pathogenicity of the root canal ora. Oral Surgery Oral Medicine Oral Pathology 78, 52230. Sundqvist G, Figdor D, Persson S, Sjo gren U (1998) Microbiologic analysis of teeth with failed endodontic treatment and the outcome of conservative re-treatment. Oral Surgery Oral Medicine Oral Pathology Oral Radiology and Endodontics 85, 8693. Tunnicliff R, Hammond C (1937) Presence of bacteria in pulps of intact teeth. Journal of the American Dental Association & Dental Cosmos 24, 16636. Tziagas D (1989) Experimental bacterial anachoresis in dog dental pulps capped with calcium hydroxide. Journal of Endodontics 15, 5915.
Vasiliadis L, Darling AI, Levers BG (1983) The amount and distribution of sclerotic human root dentine. Archives of Oral Biology 28, 6459. Wakeman EJ, Smith JK, Zepplin M, Sarles WB, Philipps PH (1948) Microorganisms associated with dental caries in the cotton rat. Journal of Dental Research 27, 48992. Willcox MD, Drucker DB, Green RM (1990) In vivo dental plaque-forming ability and cariogenicity of the bacterium Streptococcus bovis in gnotobiotic rats. Archives of Oral Biology 35, 1636. Williams NB, Forbes MA, Blau E, Eickenberg CF (1950) A study of the simultaneous occurrence of Enterococci, Lactobacilli, and yeasts in saliva from human beings. Journal of Dental Research 29, 56370. Winkler KC, van Amerongen J (1959) Bacteriologic results from 4000 root canal cultures. Oral Surgery Oral Medicine Oral Pathology 12, 85775. Wohlfeil T (1933) Die Enterokokken als Krankheitserreger. Deutsche Medizinische Wochenschrift 28, 1112.
287
doi:10.1111/j.1365-2591.2008.01540.x
REVIEW
The properties and applications of chlorhexidine in endodontics
Z. Mohammadi1,2 & P. V. Abbott3

Department of Endodontics, School of Dentistry, Hamedan University of Medical Sciences, Hamedan, Iran; 2Iranian Centre for Endodontic Research (ICER); and 3School of Dentistry, University of Western Australia, Perth, WA, Australia
1
Abstract
Mohammadi Z, Abbott PV. The properties and applications
of chlorhexidine in endodontics. International Endodontic Journal, 42, 288302, 2009.
Microorganisms and their by-products are considered to be the major cause of pulp and periradicular pathosis. Hence, a major objective in root canal treatment is to disinfect the entire root canal system, which requires that all contents of the root canal system be eliminated as possible sources of infection. This goal may be accomplished using mechanical instrumentation and chemical irrigation, in conjunction with medication of the root canal system between treatment sessions. To reduce or eliminate bacteria, various irrigation solutions have been advocated. Chlorhexidine is a cationic molecule, which can be used during treatment. It has a wide range of antimicrobial activity. Its cationic structure provides a unique property named substantivity. The purpose of this paper is to review the structure and mechanism of
action of CHX, its antibacterial and antifungal activity, its effect on biolm, its substantivity (residual antibacterial activity), its tissue solvent ability, its interaction with calcium hydroxide and sodium hypochlorite, its anticollagenolytic activity, its effect on coronal and apical leakage of bacteria, its toxicity and allergenicity and the modulating effect of dentine and root canal components on its antimicrobial activity. A Medline search was performed from 1981 to the end of March 2008 and was limited to English-language papers. The keywords searched on Medline were chlorhexidine AND endodontics, chlorhexidine AND root canal therapy, chlorhexidine AND substantivity and chlorhexidine AND toxicity. The reference lists of each article were manually checked for additional articles of relevance. Keywords: chlorhexidine, medicaments, substantivity. endodontics, irrigants,
Received 1 May 2008; accepted 18 November 2008
Introduction
The major causative role of microorganisms in the pathogenesis of pulp and periapical diseases has clearly been demonstrated (Kakehashi et al. 1965, Mo ller et al. 1981, Sundqvist 1992). The elimination of microorganisms from infected root canal systems is a
Correspondence: Dr Zahed Mohammadi, Department of Endodontics, Hamedan Dental School, Shahid Fahmideh Street, Hamedan, Iran (Tel.: +98 918 8729690; fax: +98 351 6250344; e-mail: mohammadi_zahed@yahoo.com).
complicated task involving the use of various instrumentation techniques, irrigation regimens and intracanal medicaments. Mechanical instrumentation alone does not result in a bacteria-free root canal system and when the complex anatomy of the root canal system (Hess 1925) is considered, this is not surprising. On the other hand, ex vivo and clinical evidence has shown that mechanical instrumentation leaves signicant portions of the root canal walls untouched (Peters et al. 2001) and complete elimination of bacteria by instrumentation alone is unlikely to occur (Bystro m & Sundqvist 1981). It is assumed, but not demonstrated,
288
Mohammadi & Abbott Chlorhexidine in endodontics
that any pulp tissue left in the root canals can serve as a nutrient source for any remaining microorganisms. Furthermore, tissue remnants also impede the antimicrobial effects of root canal irrigants and medicaments. Therefore, some form of irrigation and disinfection is necessary to remove residual tissue and to kill microorganisms. Chemical treatment of the root canal can be arbitrarily divided into irrigants, canal rinses and inter-appointment medicaments. Chlorhexidine (CHX) is used widely as an endodontic irrigant and medicament, but there has not been an adequate review of the literature regarding CHX. Hence, the purpose of this paper is to review different aspects of CHX of relevance to endodontics. The literature review was performed using a Medline electronic search. The search was performed from 1981 to the end of March 2008 and was limited to English-language papers. The keywords searched on Medline were chlorhexidine AND endodontics, chlorhexidine AND root canal therapy, chlorhexidine AND substantivity and chlorhexidine AND toxicity. The reference lists of each article were manually checked for additional articles of relevance.
Antibacterial activity
Delany et al. (1982) evaluated 0.2% CHX-gluconate in infected root canals. Bacteriologic samples were obtained before, during, immediately after and 24 h after instrumentation, irrigation and medication either with CHX-gluconate or with sterile saline. There was a highly signicant reduction in the number of microorganisms in the CHX-treated specimens after instrumentation and irrigation. Basson & Tait (2001) compared the ex vivo effectiveness of calcium hydroxide [Ca(OH)2], iodine potassium iodide (IKI) and a CHX solution in disinfecting root canal systems that were infected with Actinomyces israelii. The root canals were exposed to either IKI, calcium hydroxide or 2% CHX for periods of 3, 7 and 60 days. CHX was the only disinfectant that was able to eliminate A. israelii from all samples at all time periods whilst 25% of the specimens treated with IKI and 50% of the specimens treated with Ca(OH)2 still had viable A. israelii after et al. (2003) evaluated the antibactreatment. Onc ag terial properties of 5.25% sodium hypochlorite (NaOCl), 2% CHX and 0.2% CHX plus 0.2% cetrimide [Cetrexidin (GABA Vebas, San Giuliano Milanese, Italy)] after 5 min and 48 h in extracted human teeth after the canals had been infected by Enterococcus faecalis. The 2% CHX and Cetrexidin were signicantly more effective against E. faecalis than the 5.25% NaOCl at both time periods. Two studies (Gomes et al. 2001, Vianna et al. 2004) have investigated the ex vivo antimicrobial activity against endodontic pathogens of three concentrations (0.2%, 1% and 2%) of two forms of CHX (gel and liquid) and compared them with ve concentrations of NaOCl (0.5%, 1%, 2.5%, 4% and 5.25%). Both the 2% gel and 2% liquid formulations of CHX eliminated Staphylococcus aureus and Candida albicans within 15 s, whereas the gel formulation killed E. faecalis within 1 min. All of the tested irrigants eliminated Porphyromonas endodontalis, Porphyromonas gingivalis and Prevotella intermedia within 15 s. The time required for 1.0% and 2.0% CHX liquid to eliminate all microorganisms was the same as the time required for 5.25% NaOCl. These studies conrm that the antimicrobial action is related to the type, concentration and presentation form of the irrigants as well as the microbial susceptibility to the formulation used. Zamany et al. (2003) examined the effects of adding a 2% CHX rinse to the conventional treatment protocol. Their results showed that cultivable bacteria were retrieved at the conclusion of the rst visit in one of the CHX cases, whereas seven of the 12 control cases
Structure and mechanism of action

Chlorhexidine is a synthetic cationic bis-guanide that consists of two symmetric 4-cholorophenyl rings and two biguanide groups, connected by a central hexamethylene chain (Greenstein et al. 1986). CHX is a positively charged hydrophobic and lipophilic molecule that interacts with phospholipids and lipopolysaccharides on the cell membrane of bacteria and then enters the cell through some type of active or passive transport mechanism (Athanassiadis et al. 2007). Its efcacy is because of the interaction of the positive charge of the molecule and the negatively charged phosphate groups on microbial cell walls (Gomes et al. 2003a,b), thereby altering the cells osmotic equilibrium. This increases the permeability of the cell wall, which allows the CHX molecule to penetrate into the bacteria. CHX is a base and is stable as a salt. The most common oral preparation, CHX gluconate, is watersoluble and at physiologic pH, it readily dissociates and releases the positively charged CHX component (Greenstein et al. 1986). At low concentration (0.2%), low molecular weight substances, specically potassium and phosphorous, will leak out of the cell. On the other hand, at higher concentration (2%), CHX is bactericidal as precipitation of the cytoplasmic contents occurs, which results in cell death (Gomes et al. 2003a).
289
Chlorhexidine in endodontics Mohammadi & Abbott
without CHX showed growth; this difference was statistically signicant. Siqueira et al. (2007) compared the effectiveness of 2.5% NaOCl and 0.12% CHX as irrigants in reducing the cultivable bacteria in infected root canal systems of teeth with apical periodontitis. They found that the two solutions had comparable effects in eliminating bacteria and they suggested that both could be used as irrigants. In a randomized clinical trial, Manzur et al. (2007) assessed the antibacterial efcacy of intracanal medication with Ca(OH)2, 2% CHX gel and a combination of both [Ca(OH)2/CHX] in teeth with chronic apical periodontitis. Bacteriological samples were obtained from the operative eld and the root canals before and after instrumentation in the rst treatment session. Further samples were taken from the canals at the commencement of the second appointment 1 week later. They concluded that the antibacterial efcacies of Ca(OH)2, CHX and a mixture of Ca(OH)2/CHX were comparable. Zerella et al. (2005) investigated the effect of a slurry of Ca(OH)2 mixed in aqueous 2% CHX versus aqueous Ca(OH)2 alone on the disinfection of the root canal system of root lled teeth that required root canal re-treatment because the canals had become infected again. Twelve (30%) of the 40 samples were positive for bacteria before root lling. The control medication disinfected 12 (60%) of 20 teeth including two of four teeth that had been originally diagnosed with enterococci. The experimental medication resulted in disinfection of 16 of 20 (80%) teeth at the beginning of the third appointment. None of the teeth originally containing enterococci showed remaining growth. They concluded that a mixture of 2% CHX and a Ca(OH)2 slurry is as efcacious as aqueous Ca(OH)2 on the disinfection of infected root lled teeth. Ercan et al. (2004) evaluated the antibacterial activity of 2% CHX and 5.25% NaOCl in infected root canals of incisors and premolars. They concluded that both CHX and NaOCl were effective irrigants for reducing the number of microorganisms in teeth with a necrotic pulp, periapical pathosis or both. Tanomaru et al. (2003) evaluated the effect of biomechanical preparation with 5% NaOCl, 2% CHX and physiological saline irrigating solutions and Ca(OH)2 dressing in the root canals of dogs teeth that contained bacterial endotoxin. They found that biomechanical preparation with the irrigating solutions did not inactivate the endotoxin, but the calcium hydroxide intracanal dressing did inactivate the effects induced by the endotoxin in vivo.
Another interesting topic is the additive effect of CHX and hydrogen peroxide. Heling & Chandler (1998) studied the antimicrobial effect of irrigant combinations within dentinal tubules ex vivo against E. faecalis and found that a specic combination of 3% hydrogen peroxide (H2O2) and CHX was superior in its antibacterial activity in dentine compared with other regimens, such as CHX alone and NaOCl. Steinberg et al. (1999) challenged E. faecalis suspensions in trypticase soy broth (a culture medium rich in peptides) with various combinations of CHX and H2O2. The experiments demonstrated that the combination of the two substances totally killed E. faecalis at concentrations much lower than that required for each component alone. According to that study, the bactericidal effect of CHX is derived from its ability to denature the bacterial cell wall whilst forming pores in the membrane, whereas H2O2 is effective against intracellular organelles, such as DNA. Although the exact synergistic mechanism of CHX and H2O2 is not known, it can be postulated that the exposure of bacteria to CHX leads to a more permeable cell wall that the H2O2 can easily penetrate and hence damage the intracellular organelles (Steinberg et al. 1999). Shabahang et al. (2008) evaluated the antibacterial efcacy of the substitution of CHX for doxycycline in MTAD against a strain of E. faecalis ex vivo. Findings showed that the presence of doxycycline in the concentration included in the MTAD formulation was effective in eliminating E. faecalis. Furthermore, the addition of 0.2% CHX did not adversely affect the antibacterial action of doxycycline. On the other hand, the substitution of 0.2% CHX did not allow the same disinfection efcacy on E. faecalis as MTAD. On the whole, although studies comparing the antibacterial effect of CHX and NaOCl have produced somewhat conicting results, it seems that when used in identical concentrations, their antibacterial effects ex vivo (in infected dentine) and in vivo (in the root canal system) are similar.
Antifungal activity
Fungi (or yeasts) constitute a small proportion of the usual oral microbiota with Candida species being the most common of the fungi present in both healthy (30 45%) and medically compromised (95%) individuals (Siqueira & Sen 2004). Fungi have occasionally been found in infected root canals that have not had any previous endodontic treatment, but they are more common in lled root canals in teeth that have become infected some time after treatment or in those that have
290
not responded to endodontic treatment (23). Overall, the occurrence of fungi reported in infected root canals varies between 1% and 17% (Waltimo et al. 2004). Fungi may be involved in cases of persistent and secondary infections associated with recalcitrant periradicular lesions and therefore the spectrum of antimicrobial activity of endodontic medicaments and irrigants should include these organisms. Thus, medicaments that have antifungal effectiveness may assist in the successful management of persistent or secondary endodontic infections caused by fungi (Siqueira & Sen 2004, Waltimo et al. 2004). To try and improve antisepsis in single-appointment endodontic treatment regimes, it has been suggested to irrigate and/or soak the root canals with either CHX or iodine-IKI solutions following irrigation with NaOCl. Aqueous CHX solution has a wide spectrum of antimicrobial activity at low concentrations and is especially effective against C. albicans. Furthermore, it binds to surrounding tissues and can then be released again slowly over extended periods of time, a phenomenon known as substantivity. Interestingly, it appears that CHX can efciently inhibit the initial adherence and perhaps further accumulation and biolm formation of fungi and other microorganisms. A recent clinical study has shown that canals that received a nal rinse with a 2% CHX solution were signicantly more often free of cultivable microorganisms than controls irrigated with NaOCl alone (Siqueira & Sen 2004, Waltimo et al. 2004). Sen et al. (1999) evaluated the antifungal properties of 0.12% CHX, 1% NaOCl and 5% NaOCl against Candida albicans using a cylindrical dentine tube model. They reported that C. albicans was more resistant to these irrigant solutions when the smear layer was present than when it was absent. When smear layer was absent, the NaOCl started to display antifungal activity after 30 min. Waltimo et al. (1999) evaluated the susceptibility of seven strains of C. albicans to four disinfectants, namely IKI, CHX-acetate (0.5%), NaOCl (5% and 0.5%) and Ca(OH)2. Each solution was tested individually as well as in pairs using all possible pairs of these four disinfectants. All C. albicans strains tested showed similar susceptibility to these medicaments. They were highly resistant to Ca(OH)2, but the NaOCl and IKI killed all cells within 30 s and the CHX-acetate showed complete killing after 5 min. Combinations of disinfectants were either equally or less effective than the more effective component of the pair tested. Siqueira et al. (2003) evaluated the effectiveness of four intracanal medications in disinfecting root dentine in bovine teeth experimentally infected with C. albicans.
Infected dentine cylinders were exposed to four different medications: Ca(OH)2/glycerin; Ca(OH)2/0.12% CHX; monoparachlorophenol/glycCa(OH)2/camphorated erin and 0.12% CHX/zinc oxide. The specimens treated with the Ca(OH)2/camphorated monoparachlorophenol/glycerin paste or with the CHX/zinc oxide paste were completely disinfected after 1 h of exposure whilst the Ca(OH)2/glycerin paste consistently eliminated the C. albicans after 7 days of exposure. Calcium hydroxide mixed with CHX was ineffective in disinfecting dentine even after 1 week. In another study, Siqueira et al. (2001) investigated the antifungal activity of several medicaments against C. albicans, Candida glabrata, Candida guilliermondii, Candida parapsilosis and Saccharomyces cerevisiae. Calcium hydroxide mixed with CPMC/glycerin as a paste showed the most pronounced antifungal effects. Calcium hydroxide in glycerin, Ca(OH)2 with CHX and CHX in detergent had less antifungal activity. Ferguson et al. (2002) sought to determine the in vitro susceptibility of C. albicans to various irrigants and medicaments. The minimum inhibitory concentrations of NaOCl, hydrogen peroxide, CHX-digluconate and aqueous Ca(OH)2 were determined. Their results revealed that NaOCl, hydrogen peroxide and CHX-digluconate were effective against C. albicans even when signicantly diluted. However, aqueous Ca(OH)2 had no anti-fungal activity. Taken together, it can be concluded that CHX is an effective antifungal agent, but its efcacy is signicantly less than NaOCl.
CHX and biolms

The term biolm was introduced to designate the thinlayered condensations of microbes that may occur on various surface structures in nature. Free-oating bacteria existing in an aqueous environment, the socalled planktonic form of microorganisms, are a prerequisite for biolm formation (Bowden & Hamilton 1998). Biolms may thus become established on any organic or inorganic surface substrate where planktonic microorganisms prevail in a water-based solution. In dental contexts, a well-known and extensively studied biolm structure is established during the attachment of bacteria to teeth to form dental plaque. Here, bacteria free in saliva (planktonic organisms) serve as the primary source of organisms for the organization of this specic biolm (Bowden & Hamilton 1998). In endodontics, the biolm concept was initially discussed mainly within the framework of bacteria on the root tips of teeth with necrotic and
291
infected pulps or pulpless and infected root canal systems. Such bacterial aggregations have been thought to be the cause of therapy-resistant apical periodontitis (Tronstad & Sunde 2003). Although not described in as much detail, bacterial condensations (that is, biolms) on the walls of infected root canals have been observed. Antimicrobial agents have often been developed and optimized for their activity against fast growing, dispersed populations containing a single microorganism. However, microbial communities grown in biolms are remarkably difcult to eradicate with antimicrobial agents and microorganisms in mature biolms can be notoriously resistant for reasons that have yet to be adequately explained (Bowden & Hamilton 1998). There are reports showing that microorganisms grown in biolms could be twofold to 1000-fold more resistant than the corresponding planktonic form of the same organisms (Svensater & Bergenholtz 2004). Spratt et al. (2001) have evaluated the effectiveness of 2.25% NaOCl, 0.2% CHX, 10% povidone iodine, 5 ppm colloidal silver and phosphate buffered solution (PBS) as a control against monoculture biolms of ve root canal isolates including P. intermedia, Peptostreptococcus miros, Streptococcus intermedius, Fusobacterium nucleatum and E. faecalis. They reported that NaOCl was the most effective antimicrobial agent followed by the iodine solution. Clegg et al. (2006) evaluated the ex vivo effectiveness against apical dentine biolms of three concentrations of NaOCl (6%, 3% and 1%), 2% CHX and a commercially available mixture of a tetracycline, an acid and a detergent known as BioPure MTAD (Dentsply, Tulsa Dental, Tulsa, OK, USA). They reported that the 6% NaOCl and 3% NaOCl were capable of disrupting and removing the biolm, but the 1% NaOCl and the MTAD were capable of disrupting the biolm, but did not eliminate the bacteria. The 2% CHX was not capable of disrupting the biolm. Viable bacteria could not be cultured from specimens exposed to 6% NaOCl, 2% CHX or 1% NaOCl followed by BioPure MTAD. Dunavant et al. (2006) evaluated the efcacy of 6% NaOCl, 1% NaOCl, Smear Clear (SybronEndo, Orange, CA, USA), 2% CHX, REDTA (Roths International Ltd, Chicago, IL, USA) and BioPure MTAD against E. faecalis biolms using a novel laboratory testing system. Biolms grown in a ow cell system were submerged in test irrigants for either 1 or 5 min. There was a signicant relationship between the test agent and the percentage kill of the bacteria in the biolm. No signicant relationship between time and percentage
kill was found. The percentage kills of the bacteria were: 6% NaOCl (>99.99%), 1% NaOCl (99.78%), Smear Clear (78.06%), 2% CHX (60.49%), REDTA (26.99%) and BioPure MTAD (16.08%). There was a signicant difference between NaOCl (both concentrations of 1% and 6%) and all other agents. Therefore, both 1% NaOCl and 6% NaOCl were more efcient in eliminating E. faecalis biolm than the other solutions tested. In another study, Lima et al. (2001) assessed the effectiveness of CHX-based or antibiotic-based (clindamycin and metronidazole) medications in eliminating 1- and 3-day E. faecalis biolms. Each biolm-containing membrane was thoroughly covered with 1 mL of the test medications and incubated for 1 day at 37 C. The treated biolms were then aseptically transferred to vials containing a neutralizing agent in saline solution and vortexed. Suspensions were diluted 10-fold, seeded onto Mitis salivarius agar plates and the colony-forming units counted after 48 h of incubation. There were signicant differences between the formulations tested. The association of clindamycin with metronidazole signicantly reduced the number of cells in the 1-day biolms. However, of all medications tested, only 2% CHX-containing medications were able to thoroughly eliminate most of both the 1-day and 3-day E. faecalis biolms. On the whole, it seems that although CHX is somewhat effective against bacterial biolms, NaOCl is the only irrigation solution with the capability of disrupting the biolms.
Substantivity
Chlorhexidine as well as tetracyclines have a unique feature in that dentine medicated with it acquires antimicrobial substantivity (Khademi et al. 2006). The positively charged ions released by CHX can adsorb into dentine and prevent microbial colonization on the dentine surface for some time beyond the actual the period of time of application of the medicament (Athanassiadis et al. 2007). The antimicrobial substantivity of CHX has been assessed in several periodontal and endodontic studies. In an in vivo periodontal study, Stabholz et al. (1993) evaluated the substantivity of the human root surface after in situ subgingival irrigation with tetracycline HCL and CHX. They found that the substantivity of tetracycline at 50 mg mL)1 was signicantly greater than that of CHX for 12 days and greater than saline for 16 days. In a laboratory study, White et al. (1997) evaluated the antimicrobial substantivity of a 2% CHX solution as an endodontic irrigant. Findings showed that the
292
substantivity lasted 72 h. In an in vivo study, Leonardo et al. (1999) evaluated the antimicrobial substantivity of 2% CHX used as a root canal irrigating solution in teeth with pulp necrosis and radiographically visible chronic periapical lesions. They reported that the CHX prevented microbial activity with residual effects in the root canal system for up to 48 h after application. However, some other studies have reported that the substantivity of CHX can last for longer periods of time. Khademi et al. (2006) found that 5 min application of 2% CHX solution induced substantivity for up to 4 weeks. Rosenthal et al. (2004) evaluated the substantivity of 2% CHX solution within the root canal system after 10 min of application and they reported that the CHX was retained in the root canal dentine in antimicrobially effective amounts for up to 12 weeks. Antimicrobial substantivity depends on the number of CHX molecules available to interact with the dentine. Therefore, medicating the canal with a more concentrated CHX preparation should result in increased resistance to microbial colonization. The antibacterial substantivity of three concentrations of CHX solution (4%, 2% and 0.2%) after 5 min of application has been evaluated. Results revealed a direct relationship between the concentration of CHX and its substantivity (Mohammadi et al. 2008). On the contrary, Lin et al. (2003a) attributed the substantivity of CHX to its ability to adsorb on to the dentine during the rst hour. They stated that it is only after the saturation point is reached after the rst hour that the antimicrobial capability of CHX increases with time. Furthermore, Komorowski et al. (2000) revealed that 5 min application of CHX did not induce substantivity and that the dentine should be treated with CHX for 7 days. Taken together, it seems that residual antimicrobial activity of CHX in the root canal system remains for up to 12 weeks.
Modulating effect of dentine on CHX

The root canal milieu is a complex mixture of a variety of organic and inorganic compounds. Hydroxyapatite, the main component of dentine, is the major representative of inorganic components present. In addition, inammatory exudate, entering the apical root canal in purulent infections, is rich in proteins, such as albumin. The relative importance of the various organic and inorganic compounds in the inactivation of root canal disinfectants have been studied restrictively (Haapasalo et al. 2000). Difculties in designing experiments that will give reliable and comparable data have been some of the greatest challenges for researchers for many
years. Haapasalo et al. (2000) introduced a new dentine powder model for studying the inhibitory effect of dentine on various root canal irrigants and medicaments. They investigated the modulating effect of dentine on the antibacterial activity of saturated CA(OH)2 solution, 1% NaOCl, 0.5% and 0.05% CHX acetate and 2/4% and 0.2/0.4% IKI. Dentine powder had an inhibitory effect on all medicaments tested. The effect was dependent on the concentration of the medicament as well as on the length of time the medicament was pre-incubated with the dentine powder before adding the bacteria. Similarly, 0.2/ 0.4% IKI lost its effect after pre-incubation for 1 h with dentine before adding the bacteria. The effect of 0.05% CHX and 1% NaOCl on E. faecalis was reduced, but not totally eliminated by the presence of dentine. No inhibition could be measured when full strength solutions of CHX and IKI were used in killing E. faecalis. Portenier et al. (2001) evaluated the inhibition of the antibacterial effect of saturated Ca(OH)2 solution, CHXacetate and IKI by dentine, hydroxylapatite (HA) and bovine serum albumin (BSA). Calcium hydroxide was totally inactivated by the presence of 28 mg of dentine powder or BSA. CHX (0.05%) was strongly inhibited by BSA and slowed down by dentine. However, HA had little or no inhibitory effect on CHX. The antibacterial effect of 0.2/0.4% IKI on E. faecalis was totally inhibited by dentine (28 mg), but was practically unaffected by HA or BSA. A stepwise reduction of dentine from 28 to 2.8 mg 150 lL)1 was followed by a similar reduction of the inhibition of the antibacterial activity of CHX. IKI was not inhibited at all with dentine amounts <28 mg. However, the effect of saturated calcium hydroxide solution was totally eliminated by dentine at all four concentrations tested. It could be assumed that inhibition by dentine of the antibacterial activity of Ca(OH)2, CHX and IKI occurs by different mechanisms (Portenier et al. 2001). Surprisingly, Ca(OH)2 was sensitive to the inhibitory effect of all three materials tested. The inhibition of Ca(OH)2 by dentine and by the other compounds is, of course, dependent on their quantitative relationships (Portenier et al. 2001). One major mechanism for resistance of survival of E. faecalis in the root canal lled with Ca(OH)2 may be the buffering effect of dentine against the pH rise. Inorganic HA had little or no inhibitory activity against CHX as compared with dentine, whereas BSA was the strongest inhibitor of CHX, with more than 10% of E. faecalis cells still viable after 24 h of incubation with the medicament. This indicates that periapical inammatory exudate entering the root canal is a greater threat to the activity
293
of CHX than the dentine walls. Dentine powder totally eliminated the antibacterial effect of IKI; whereas the major component of dentine, HA did not affect the antibacterial effect of IKI, nor did BSA. In addition, it is generally known that blood rapidly inactivates the antibacterial activity of iodine compounds (Portenier et al. 2001). In another study, Portenier et al. (2002) assessed the antibacterial activity of CHX and IKI on E. faecalis in the presence of dentine, dentine matrix, dentine pre-treated by ethylenediamine tetraacetic acid (EDTA) and citric acid, collagen and heat-killed cells of E. faecalis and Candida albicans. Dentine matrix and heat-killed microbial cells were the most effective inhibitors of CHX, whereas dentine pre-treated by citric acid or EDTA showed only slight inhibition. Dentine and skin collagen showed some inhibition at 1 h, but not after 24 h. IKI was effectively inhibited by dentine, dentine matrix and heat-killed microbial cells. Skin collagen and dentine pre-treated by EDTA or by citric acid showed little or no inhibitory effect on IKI. The inhibitory effect of dentine and BSA on the antibacterial activity of CHX and MTAD was assessed in another study (Portenier et al. 2006). The presence of dentine or BSA caused a marked delay in the killing of E. faecalis by both medicaments. The inhibitory effect of BSA on the antibacterial activity of CHX and NaOCl has been conrmed by Sassone et al. (2008). Taken together, it seems that dentine, dentine components (HA and collagen), killed microorganisms and inammatory exudate in the root canal system all reduce or inhibit the antibacterial activity of medicaments and irrigants. On the whole, it seems that dentine, dentine components (HA and collagen), killed microorganisms and inammatory exudates in the root canal system reduce the antibacterial activity of CHX.
the frequency of shaking, the amount of organic matter in relation to the amount of irrigant in the system and the surface area of tissue that was available for contact with the irrigant. Okino et al. (2004) evaluated the tissue dissolving ability of 0.5%, 1.0% and 2.5% NaOCl, 2% aqueous solution of CHX-digluconate, 2% CHX digluconate gel and distilled water as the control. Bovine pulp fragments were weighed and placed in contact with 20 mL of each tested substance in a centrifuge at 150 rpm until total dissolution. Dissolution speed was calculated by dividing the pulp weight by the dissolution time. Distilled water and both solutions of CHX did not dissolve the pulp tissue within 6 h. The mean dissolution speeds for 0.5%, 1.0% and 2.5% NaOCl solutions were 0.31, 0.43 and 0.55 mg min)1, respectively. In another study, Naenni et al. (2004) assessed the necrotic tissue dissolution capacity of 1% (w/v) NaOCl, 10% CHX, 3% and 30% hydrogen peroxide, 10% peracetic acid, 5% dichloroisocyanurate (NaDCC) and 10% citric acid. Standardized necrotic tissue samples obtained from pig palates were incubated in these solutions and their weight loss was measured over time. None of the test solutions except NaOCl had any substantial tissue dissolution capacity. It was concluded that this might be important when considering the use of irrigants other than NaOCl. On the whole, one of the major disadvantages of CHX is that it has no tissue solvent activity.
CHX and Ca(OH)2

Chlorhexidine is a cationic biguanide whose optimal antimicrobial activity is achieved within a pH range of 5.57.0 (Athanassiadis et al. 2007). Therefore, it is likely that alkalinizing the pH by adding Ca(OH)2 to CHX will lead to precipitation of the CHX molecules and thereby decreases its effectiveness. It has been demonstrated that the alkalinity of Ca(OH)2 when mixed with CHX remained unchanged. Therefore, the usefulness of mixing Ca(OH)2 with CHX still remains unclear and controversial (Athanassiadis et al. 2007). When used as an intracanal medicament, CHX was more effective than Ca(OH)2 in eliminating E. faecalis from inside dentinal tubules (Athanassiadis et al. 2007). In a study by Almyroudi et al. (2002), all of the CHX formulations used, including a CHX/ Ca(OH)2 50:50 mix, were efcient in eliminating E. faecalis from the dentinal tubules with a 1% CHX gel working slightly better than the other preparations. These ndings were corroborated by Gomes et al. (2006) in bovine dentine and Schafer & Bossmann (2005) in
Tissue-solvent effects of CHX

Several studies have been conducted in the search for an irrigant that meets the four major desirable properties for root canal irrigants namely: antimicrobial activity, nontoxicity to the periapical tissues, water solubility and the capacity to dissolve organic matter. Therefore, an ideal irrigant should dissolve the organic matter inside the root canal system. Grossman & Meiman (1941) demonstrated the importance of the solvent ability of an endodontic irrigant and emphasized that the elimination of pulp tissue from the root canal was important for the ultimate success of root canal treatment. Moorer & Wesselink (2003) showed that tissue dissolution was dependent on three factors:
294
human dentine where 2% CHX gel had greater activity against E. faecalis, followed by CHX/ Ca(OH)2 and then Ca(OH)2 used alone. In a study using agar diffusion, Haenni et al. (2003) could not demonstrate any additive antibacterial effect by mixing Ca(OH)2 powder with 0.5% CHX and they showed that the CHX had a reduced antibacterial action. However, Ca(OH)2 did not lose its antibacterial properties in such a mixture. This may be because of the deprotonation of CHX at a pH >10, which reduces its solubility and alters its interaction with bacterial surfaces as a result of the altered charge of the molecule. In an in vitro study using human teeth, Ercan et al. (2006) showed 2% CHX gel was the most effective agent against E. faecalis inside dentinal tubules, followed by a Ca(OH)2/ 2% CHX mix, whilst Ca(OH)2 alone was totally ineffective, even after 30 days. The 2% CHX gel was also signicantly more effective than the Ca(OH)2/2% CHX mix against C. albicans at 7 days, although there was no signicant difference at 15 and 30 days. Ca(OH)2 alone was completely ineffective against C. albicans. In another in vivo study using primary teeth, a 1% CHX-gluconate gel, both with and without Ca(OH)2, was more effective against E. faecalis than CH alone over a 48-h period (Oncag et al. 2006). Schafer & Bossmann (2005) reported that 2% CHXgluconate was signicantly more effective against E. faecalis than Ca(OH)2 used alone or a mixture of the two. This was also conrmed by Lin et al. (2003b) although in a study by Evans et al. (2003) using bovine dentine, 2% CHX with Ca(OH)2 was shown to be more effective than Ca(OH)2 in water. In an animal study, Lindskog et al. (1998) reported that teeth dressed with CHX for 4 weeks had reduced inammatory reactions in the periodontium (both apically and marginally) and less root resorption. Waltimo et al. (1999) reported that 0.5% CHX-acetate was more effective at killing C. albicans than saturated Ca(OH)2, whilst Ca(OH)2 combined with CHX was more effective than Ca(OH)2 used alone. The high pH of Ca(OH)2 was unaffected when combined with CHX in this study. Taken together, it seems that the usefulness of mixing Ca(OH)2 with CHX remains unclear and controversial.
CHX and coronal penetration of bacteria

Because of its antimicrobial substantivity, it seems that CHX preparations delay entry of bacteria through the coronal portion of the tooth into the root canal system. In a laboratory study, Gomes et al. (2003b) investigated the time required for recontamination of the root
canal system of teeth with coronal restorations medicated with either calcium hydroxide, 2% CHX gel or with a combination of both. The canals without a coronal restoration, but medicated with CHX, showed recontamination after an average time of 3.7 days; the group with Ca(OH)2 after 1.8 days and the group with CHX+Ca(OH)2 after 2.6 days. The canals medicated with CHX and restored with IRM showed recontamination within 13.5 days; the group with Ca(OH)2+IRM after 17.2 days and the group with CHX+ Ca(OH)2+IRM after 11.9 days. The group with no medication, but restored with IRM, showed recontamination after an average time of 8.7 days. There were statistically signicant differences between the groups (P < 0.05). All groups without a coronal restoration were recontaminated signicantly more quickly than those restored with IRM, except those teeth that had a restoration, but no medicament. The groups with intracanal medication and a coronal restoration were not signicantly different from each other. Vivacqua-Gomes et al. (2002) assessed ex vivo coronal dye penetration of extracted human teeth after root canal treatment using 1% NaOCl, 1% NaOCl + 17% EDTA, 2% CHX gel, 2% CHX gel + 1% NaOCl and distilled water. After root canal lling, the teeth were incubated at 37 C for 10 days followed by 10 days immersion in human saliva and an additional 10 days in India ink. The teeth were cleared and the maximum depth of dye penetration was determined digitally in millimetres. Results revealed that the least dye penetration occurred with 1% NaOCl + 17% EDTA and 2% CHX gel. NaOCl, distilled water and 2% CHX gel + 1% NaOCl had more dye penetration with a signicant difference compared with NaOCl + 17% EDTA and 2% CHX gel and compared with one another. Other studies have shown that viscous irrigants, including those containing CHX gluconate, were less soluble substances and they can leave residues on the root canal surfaces, which may affect the root canal lling. Lambrianidis et al. (2006) investigated the efciency of removing Ca(OH)2/CHX gel, Ca(OH)2/ CHX solution and Ca(OH)2/saline pastes with the use of instrumentation and irrigation with NaOCl and EDTA solutions. None of the techniques used in this study removed the inter-appointment root canal medicaments effectively (Lambrianidis et al. 2006). Overall, Ca(OH)2/CHX (gel) paste was associated with signicantly larger amount of residue, whereas the Ca(OH)2/ CHX mixture was associated with less residue than the other two medicaments. When all these studies are considered it appears as although CHX used as an
295
intracanal medicament and/or irrigant may delay recontamination of the root canal system via the coronal route because of its substantivity. Overall, because of its substantivity, CHX as an intracanal medicament/irrigant delays recontamination of the root canal system via the coronal route.
CHX and apical uid penetration

Marley et al. (2001) assessed the effect of 0.12% CHX-gluconate as an endodontic irrigating solution on the apical seal of root lled canals using three different cements (Roths 801, AH26 and Sealapex). At 90 and 180 days after root lling, apical uid penetration was measured using the uid ltration method. The results showed no signicant differences related to the irrigants at both the 90- and 180-day observation periods. Furthermore, the same group reported that at long-term periods (270 and 360 days), CHX-gluconate irrigant did not adversely affect the apical penetration of uid with the different root canal cements (Ferguson et al. 2003). Wuerch et al. (2004) investigated the effect of CHX gel and Ca(OH)2 on the apical seal of root canal llings. They reported that 2% CHX gel and Ca(OH)2 paste did not adversely affect the apical seal. These ndings were also conrmed by Engel et al. (2005). Overall, it seems that medication and/or irrigation with CHX does not adversely affect the ability of root llings to prevent uid penetration into the root canal system through the apical foramen.
Interaction between CHX and NaOCl

A suggested clinical protocol by Zehnder (2006) for treating the dentine before root canal lling consists of irrigation with NaOCl to dissolve the organic components, irrigation with EDTA to eliminate the smear layer and irrigation with CHX to increase the anti-microbial spectrum of activity and to impart substantivity. Although such a combination of irrigants may enhance the overall antimicrobial effectiveness (Kuruvilla & Kamath 1998), the possible chemical interactions amongst the irrigants have to be considered. Some studies have reported the occurrence of colour change and precipitation when NaOCl and CHX are combined (Vivacqua-Gomes et al. 2002, Zehnder 2006, Basrani et al. 2007). Furthermore, concern has been raised that the colour change may have some clinical relevance because of staining and that the precipitate might interfere with the seal of the root lling (Vivacqua-
Gomes et al. 2002). The formation of a precipitate could be explained by the acidbase reaction that occurs when NaOCl and CHX are mixed together. CHX, a dicationic acid has the ability to donate protons whilst NaOCl is alkaline and can accept protons from the dicationic acid. This proton exchange results in the formation of a neutral and insoluble substance, referred to as the precipitate (Basrani et al. 2007). Basrani et al. (2007) evaluated the chemical nature of this precipitate and reported that there was an immediate reaction when 2% CHX was combined with NaOCl, even at the low concentration (0.023%). Increasing of the concentration of NaOCl to 0.19% (the sixth dilution in their series) resulted in the formation of a precipitate, which consisted mainly of para-chloroaniline (PCA). This occurred through a substitution of the guanidine group in the CHX molecule. They found that the amount of PCA directly increased with the increasing concentration of NaOCl. PCA has been shown to be toxic with short-term exposure of humans to PCA resulting in cyanosis, which is a manifestation of methemoglobin formation. In another study, Bui et al. (2008) evaluated the effect of irrigating root canals with a combination of NaOCl and CHX on root dentine and dentinal tubules by using the environmental scanning electron microscope (FEI Quanta 200, Hillsboro, OR, USA) and a computer program (photoshop cs2, Adobe Systems, San Jose, CA, USA). Their ndings indicated that there were no signicant differences in the amount of debris remaining between the negative control group and the experimental groups although there were signicantly fewer patent tubules in the experimental groups when compared with the negative control group. They concluded that the NaOCl/CHX precipitate tends to occlude the dentinal tubules and suggested that until this precipitate is studied further, caution should be exercised when irrigating with both NaOCl and CHX. Taken together, the combination of NaOCl and CHX causes colour changes and formation of a neutral and insoluble precipitate, which may interfere with the seal of the root lling. Copious amounts of CHX irrigant should be administered to prevent discolouring of the tooth by this precipitate. Alternatively, the canal can be dried using paper points before the nal CHX rinse (Zehnder 2006).
CHX and dentine bonding (anticollagenolytic activity)

During the last two decades, chemical and technical advances have contributed to increases in resindentine
296
bond strength. However, the premature loss of bond strength is one of the problems that still affects adhesive restorations (Mjo r et al. 2000) and markedly reduces their durability (Carrilho et al. 2005b, De Munck et al. 2005, Frankenberger et al. 2005). The loss of bond strength has been attributed mainly to the degradation of the hybrid layer at the dentine-adhesive interface. Numerous publications have demonstrated this lack of bond stability (Wang & Spencer 2003, 2005, Yiu et al. 2004, Carrilho et al. 2005a). The notion that deterioration of dentine collagen brils contributes to the mechanism responsible for bond degradation has been reported (Hashimoto et al. 2003, Pashley et al. 2004). In this context, it has been speculated that a decreasing concentration gradient of resin monomer diffusion within the acid-etched dentine and a subsequent resin elution from hydrolytically unstable polymeric hydrogels within the hybrid layers (Wang & Spencer 2003) leaves the collagen brils unprotected and vulnerable to degradation by endogenous metalloproteinases (MMPs). The MMPs are a group of 23 mammalian enzymes capable of degrading all extracellular matrix components. Human dentin contains at least collagenase (MMP-8), gelatinases MMP-2 and -9 and enamelysin MMP-20 (Martin-De Las Heras et al. 2000, Sulkala et al. 2002, 2007, Mazzoni et al. 2006). Dentine collagenolytic and gelatinolytic activities (Pashley et al. 2004) can be suppressed by protease inhibitors (Pashley et al. 2004), indicating that MMP inhibition could be benecial in the preservation of hybrid layers. This was demonstrated in an in vivo study, in which the application of CHX, known to have a broad-spectrum MMP-inhibitory effect (Gendron et al. 1999), signicantly improved the integrity of the hybrid layer in a 6-month clinical trial (Hebling et al. 2005). Carrilho et al. (2007a) evaluated the effect of CHX on the resindentine bond stability ex vivo. Results showed that with CHX, signicantly better preservation of bond strength was observed after 6 months and protease inhibitors in the storage medium had no effect. Failure analysis showed signicantly less failure in the hybrid layer with CHX, compared with controls after 6 months. Furthermore, they evaluated the effect of CHX on the preservation of the hybrid layer in vivo. Findings showed that bond strength remained stable in the CHX-treated specimens, whilst bond strength decreased signicantly in control teeth. Resin-inltrated dentine in CHX-treated specimens exhibited normal structural integrity of the collagen network. Conversely, progressive disintegration of the brillar network was identied in control specimens. They
concluded that auto-degradation of collagen matrices can occur in resin-inltrated dentine, but may be prevented by the application of a synthetic protease inhibitor, such as CHX (Carrilho et al. 2007b). On the whole, because of its broad-spectrum MMP-inhibitory effect, CHX can signicantly improve the resindentine bond stability.
Cytotoxicity of CHX
Results from a study of the cytotoxic effect of chlorehexidine on canine embryonic broblasts and Staphylococcus aureus showed that bactericidal concentrations of CHX were lethal to canine embryonic broblasts whilst noncytotoxic concentrations allowed signicant bacterial survival (Sanchez et al. 1988). In a study by Tatnall et al. (1990), the cytotoxic effects of CHX, hydrogen peroxide and NaOCl were examined on cultured human broblasts, basal keratinocytes and a transformed keratinocyte line (SVK 14 cells). At concentrations recommended for wound cleansing, all agents produced 100% killing of all cell types. Comparison of the ED50 concentration for each agent on all cell types produced a ranking order of toxicity showing CHX to be the least toxic antiseptic agent. Results from a laboratory study on the toxicity of CHX to human gingival cells showed that the toxic potency of CHX is dependent on the length of exposure and the composition of the exposure medium (Babich et al. 1995). The addition of foetal bovine serum, albumin, lecithin and heat-killed Escherichia coli reduced the cytotoxicity of CHX, presumably because of the binding of the cationic CHX to the negatively charged chemical moieties/ sites of these components/bacteria (Babich et al. 1995). These ndings suggest that similar reactions within a root canal may reduce the potential of a cytotoxic reaction in the periapical tissues (Boyce et al. 1995). Boyce et al. (1995) found CHX (0.05%) uniformly toxic to both cultured human cells and microorganisms. Agarwal et al. (1997) found that CHX rapidly disrupts the cell membrane of both crevicular and peripheral blood neutrophils at concentrations above 0.005% within 5 min, indicating that its inhibitory effect on neutrophil function is mostly because of its lytic properties. Yesilsoy et al. (1995) assessed the short-term toxic effects of CHX in the subcutaneous tissue of guinea pigs and found moderate inammation present after 2 days, followed by a foreign-body granuloma formation at 2 weeks. Ribeiro et al. (2005) evaluated the genotoxicity
297
(potential damage to DNA) of formocresol, paramonochlorophenol, calcium hydroxide and CHX against Chinese hamster ovary cells. Results showed that none of the mentioned agents contributed to DNA damage. Taken together, in the clinically used concentrations, the biocompatibility of CHX is acceptable. The potentially toxic interactions between CHX and NaOCl were considered previously.
Allergic reactions to CHX

Although sensitivity to CHX is rare, contact dermatitis is a common adverse reaction to CHX (Krautheim et al. 2004). Apart from this, CHX may have a number of rare side effects, such as desquamative gingivitis, discolouration of the teeth and tongue or dysgeusia (distorted taste). Contact with conjunctiva can cause permanent damage and accidental contact with the tympanum can cause ototoxicity (Dukes 1992). Various allergic reactions to CHX have been described. Contact sensitivity to CHX was rst reported by Calnan (1962). Today, CHX is known to elicit allergic contact dermatitis, including connubial contact dermatitis, generally after prolonged and repeated application (Krautheim et al. 2004). It can also cause contact urticaria, photosensitivity, xed drug eruption and occupational asthma. People at particular risk of contact allergy (apart from medical and dental staff) are patients with leg ulcers and leg eczema (Krautheim et al. 2004). Overall, contact sensitivity to CHX seems to be rare as some large studies have shown a sensitization rate of about 2% (Osmundsen 1982, Bechgaard et al. 1985, Nomura et al. 1989). Even rarer are reports of immediate anaphylactic reactions because of CHX. Ohtoshi et al. (1986) demonstrated IgE antibodies in the sera of patients with anaphylaxis to CHX. Application of CHX to intact skin can cause immediate allergic reactions, such as urticaria, Quinckes edema or dyspnea and very rarely severe anaphylactic reactions (Torricelli & Wu thrich 1996, Snellman & Rantanen 1999). These reports of reactions to CHX indicate that practitioners should always be aware of this potential risk of using CHX. On the whole, although sensitivity to CHX is rare, this complication should be kept in mind during CHX application.
3. The effect of CHX on microbial biolms is signicantly less than that of NaOCl. 4. CHX has antibacterial substantivity in dentine for up to 12 weeks. 5. Dentine, dentine components (HA and collagen), killed microorganisms and inammatory exudate in the root canal system may reduce or inhibit the antibacterial activity of CHX. 6. CHX has little to no ability to dissolve organic tissues. 7. Mixing CHX with Ca(OH)2 may enhance its antimicrobial activity. 8. Medication and/or irrigation with CHX may delay the contamination of root lled teeth by bacteria entering through the coronal restoration/tooth interface. 9. Medication and/or irrigation with CHX will not adversely affect the penetration of uid through the root lled apical foramen. 10. Combination of NaOCl and CHX causes colour changes and formation a precipitate, which may interfere with the seal of the root lling. 11. CHX can signicantly improve the integrity of the hybrid layer andresindentine bond stability. 12. The biocompatibility of CHX is acceptable. 13. In rare cases, CHX may cause allergic reactions.
References
Agarwal S, Piesco NP, Peterson De et al. (1997) Effects of sanguinarium, chlorhexidine and tetracycline on neutrophil viability and functions in vitro. Journal of Periodontal Research 32, 33544. Almyroudi A, Mackenzie D, McHugh S, Saunders WP (2002) The effectiveness of various disinfectants used as endodontic intracanal medications: an in vitro study. Journal of Endodontics 28, 1637. Athanassiadis B, Abbott PV, Walsh LJ (2007) The use of calcium hydroxide, antibiotics and biocides as antimicrobial medicaments in endodontics. Australian Dental Journal 52(Suppl), S6482. Babich H, Wurzburger BJ, Rubin YL, Sinensky MC, Blau L (1995) An in vitro study on the cytotoxicity of chlorhexidine digluconate to human gingival cells. Cell Biology and Toxicology 11, 7988. Basrani B, Manek S, Sodhi RN, Fillery E, Manzur A (2007) Interaction between sodium hypochlorite and chlorhexidine gluconate. Journal of Endodontics 33, 9669. Basson NJ, Tait CM (2001) Effectiveness of three root canal medicaments to eliminate Actinomyces israelii from infected dentinal tubules in vitro. South African Dental Journal 56, 499501.
Conclusions
1. CHX has a wide range of activity against both Gram positive and Gram negative bacteria. 2. CHX is an effective antifungal agent especially against C. albicans.
298
Bechgaard E, Ploug E, Hjorth N (1985) Contact sensitivity to chlorhexidine. Contact Dermatitis 13, 535. Bowden GH, Hamilton IR. (1998) Survival of oral bacteria. Critical Reviews in Oral Biology and Medicine 9, 5484. Boyce ST, Warden GD, Holder IA (1995) Cytotoxicity testing of topical antimicrobial agents on human keratinocytes and broblasts for cultured skin grafts. Burn Care Rehabilitation 16, 97103. Bui TB, Baumgartner JC, Mitchell JC (2008) Evaluation of the interaction between sodium hypochlorite and chlorhexidine gluconate and its effect on root dentin. Journal of Endodontics 34, 1815. Bystro m A, Sundqvist G (1981) Bacteriologic evaluation of the efcacy of mechanical root canal instrumentation in endodontic therapy. Scandinavian Journal of Dental Research 89, 3218. Calnan CD. (1962) Contact dermatitis from drugs. Proceedings of Royal Society of Medicine 55, 3942. Carrilho MR, Carvalho RM, Tay FR, Yiu C, Pashley DH (2005a) Durability of resindentin bonds related to water and oil storage. American Journal of Dentistry 18, 3159. Carrilho MR, Tay FR, Pashley DH, Tja derhane L, Carvalho RM (2005b) Mechanical stability of resindentin bond components. Dental Materials 21, 23241. Carrilho MR, Carvalho RM, de Goes MF et al. (2007a) Chlorhexidine preserves dentine bond in vitro. Journal of Dental Research 86, 904. Carrilho MR, Geraldeli S, Tay F et al. (2007b) In vivo preservation of the hybrid layer by chlorhexidine. Journal of Dental Research 86, 52933. Clegg MS, Vertucci FJ, Walker C, Belanger M, Britto LR (2006) The effect of exposure to irrigant solutions on apical dentine biolms in vitro. Journal of Endodontics 32, 4347. De Munck J, Van Landuyt K, Peumans M et al. (2005) A critical review of the durability of adhesion to tooth tissue: methods and results. Journal of Dental Research 84, 11832. Delany GM, Patterson SS, Miller CH, Newton CW (1982) The effect of chlorhexidine gluconate irrigation on the root canal ora of freshly extracted necrotic teeth. Oral Surgery, Oral Medicine and Oral Pathology 53, 51823. Dukes M N (1992) Meylers Side Effects of Drugs: An Encyclopedia of Adverse Reactions and Interactions. Amsterdam: Elsevier, pp. 5769. Dunavant TR, Regan JD, Glickman GN, Solomon ES, Honeyman AL (2006) Comparative evaluation of endodontic irrigants against Enterococcus faecalis biolms. Journal of Endodontics 32, 52731. Engel GT, Goodell GG, McClanahan SB (2005) Sealer penetration and apical microleakage in smear-free dentine after a nal rinse with either 70% isopropyl alcohol or Peridex. Journal of Endodontics 31, 6203. Ercan E, Ozekinci T, Atakul F, Gu l K (2004) Antibacterial activity of 2% chlorhexidine gluconate and 5.25% sodium hypochlorite in infected root canal: an in vivo study. Journal of Endodontics 30, 847.
Ercan E, Dalli M, Du lgergil CT (2006) In vitro assessment of the effectiveness of chlorhexidine gel and calcium hydroxide paste with chlorhexidine against Enterococcus faecalis and Candida albicans. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 102, e2731. Evans MD, Baumgartner JC, Khemaleelakul SU, Xia T (2003) Efcacy of calcium hydroxide: chlorhexidine paste as an intracanal medication in bovine dentine. Journal of Endodontics 29, 3389. Ferguson JW, Hatton JF, Gillespie MJ (2002) Effectiveness of intracanal irrigants and medications against the yeast Candida albicans. Journal of Endodontics 28, 6871. Ferguson DB, Marley JT, Hartwell GR. (2003) The effect of chlorhexidine gluconate as an endodontic irrigant on the apical seal: long-term results. Journal of Endodontics 29, 914. Frankenberger R, Pashley DH, Reich SM, Lohbauer U, Petschelt A, Tay FR (2005) Characterisation of resin dentine interfaces by compressive cyclic loading. Biomaterials 26, 204352. Gendron R, Grenier D, Sorsa T, Mayrand D (1999) Inhibition of the activities of matrix metalloproteinases 2, 8, and 9 by chlorhexidine. Clinical and Diagnostic Laboratory Immunology 6, 4379. Gomes BP, Ferraz CC, Vianna ME, Berber VB, Teixeira FB, Souza-Filho FJ (2001) In vitro antimicrobial activity of several concentrations of sodium hypochlorite and chlorhexidine gluconate in the elimination of Enterococcus faecalis. International Endodontic Journal 34, 4248. Gomes BPFA, Souza SFC, Ferraz CCR et al. (2003a) Effectiveness of 2% chlorhexidine gel and calcium hydroxide against Enterococcus faecalis in bovine root dentine in vitro. International Endodontic Journal 36, 26775. Gomes BP, Sato E, Ferraz CC, Teixeira FB, Zaia AA, SouzaFilho FJ (2003b) Evaluation of time required for recontamination of coronally sealed canals medicated with calcium hydroxide and chlorhexidine. International Endodontic Journal 36, 6049. Gomes BP, Vianna ME, Sena NT, Zaia AA, Ferraz CC, de Souza Filho FJ (2006) In vitro evaluation of the antimicrobial activity of calcium hydroxide combined with chlorhexidine gel used as intracanal medicament. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 102, 54450. Greenstein G, Berman C, Jafn R (1986) Chlorhexidine: an adjunct to periodontal therapy. Journal of Periodontology 57, 3706. Grossman LI, Meiman BW (1941) Solution of pulp tissue by chemical agents. Journal of the American Dental Association 28, 2235. Haapasalo HK, Siren EK, Waltimo TM, Orstavik D, Haapasalo MP (2000) Inactivation of local root canal medicaments by dentine: an in vitro study. International Endodontic Journal 33, 12631. Haenni S, Schmidlin PR, Mueller B, Sener B, Zehnder M (2003) Chemical and antimicrobial properties of calcium
299
hydroxide mixed with irrigating solutions. International Endodontic Journal 36, 1005. Hashimoto M, Ohno H, Sano H, Kaga M, Oguchi H (2003) In vitro degradation of resindentin bonds analyzed by microtensile bond test, scanning and transmission electron microscopy. Biomaterials 24, 3795803. Hebling J, Pashley DH, Tja derhane L, Tay FR (2005) Chlorhexidine arrests subclinical degradation of dentin hybrid layers in vivo. Journal of Dental Research 84, 7416. Heling I, Chandler NP (1998) Antimicrobial effect of irrigant combinations within dentinal tubules. International Endodontic Journal 31, 814. Hess W (1925) Anatomy of Root Canals in the Teeth of the Permanent Dentition. New York: William Wood & Co. Kakehashi S, Stanley HR, Fitzgerald RJ (1965) The effects of surgical exposure of dental pulps in germ free and conventional laboratory rats. Oral Surgery, Oral Medicine and Oral Pathology 18, 3408. Khademi AA, Mohammadi Z, Havaee A (2006) Evaluation of the antibacterial substantivity of several intra-canal agents. Australian Endodontic Journal 32, 1125. Komorowski R, Grad H, Wu XY, Friedman S (2000) Antimicrobial substantivity of chlorhexidine-treated bovine root dentine. Journal of Endodontics 26, 3157. Krautheim AB, German THM, Bircher AJ (2004) Chlorhexidine anaphylaxis: case report and review of the literature. Contact Dermatitis 50, 1136. Kuruvilla JR, Kamath MP (1998) Antimicrobial activity of 2.5% sodium hypochlorite and 0.2% chlorhexidine gluconate separately and combined, as endodontic irrigants. Journal of Endodontics 24, 4726. Lambrianidis T, Kosti E, Boutsioukis C, Mazinis M (2006) Removal efcacy of various calcium hydroxide/chlorhexidine medicaments from the root canal. International Endodontic Journal 39, 5561. Leonardo MR, Tanomaru-Filho M, Silva LAB, Nelson-Filho P, Bonifacio KC, Ito IY (1999) In vivo antimicrobial activity of 2% chlorhexidine used as a root canal irrigation solution. Jouenal of Endodontics 25, 16771. Lima KC, Fava LR, Siqueira JF Jr (2001) Susceptibilities of Enterococcus faecalis biolms to some antimicrobial medications. Journal of Endodontics 27, 6169. Lin S, Zuckerman O, Weiss EI, Fuss Z (2003a) Antibacterial efcacy of a new chlorhexidine slow-releasing device to disinfect dentinal tubules. Journal of Endodontics 29, 4168. Lin YH, Mickel AK, Chogle S (2003b) Effectiveness of selected materials against Enterococcus faecalis: part 3. The antibacterial effect of calcium hydroxide and chlorhexidine on Enterococcus faecalis. Journal of Endodontics 29, 5656. Lindskog S, Pierce AM, Blomlo f L (1998) Chlorhexidine as a root canal medicament for treating inammatory lesions in the periodontal space. Endodontics and Dental Traumatology 14, 18690. lez AM, Pozos A, Silva-Herzog D, Friedman S Manzur A, Gonza (2007) Bacterial quantication in teeth with apical peri-
odontitis related to instrumentation and different intracanal medications: a randomized clinical trial. Journal of Endodontics 33, 1148. Marley JT, Ferguson DB, Hartwell GR (2001) Effects of chlorhexidine gluconate as an endodontic irrigant on the apical seal: short-term results. Journal of Endodontics 27, 7758. Martin-De Las Heras S, Valenzuela A, Overall CM (2000) The matrix metalloproteinase gelatinase A in human dentine. Archives of Oral Biology 45, 75765. Mazzoni A, Pashley DH, Nishitani Y et al. (2006) Reactivation of quenched endogenous proteolytic activities in phosphoric acid-etched dentine by etch-and-rinse adhesives. Biomaterials 27, 44706. Mjo r IA, Moorhead JE, Dahl JE (2000) Reasons for replacement of restorations in permanent teeth in general dental practice. International Dental Journal 50, 3616. Mohammadi Z, Khademi AA, Davari AR. (2008) Evaluation of the antibacterial substantivity of three concentrations of chlorhexidine in bovine root dentine. Iranian Endodontic Journal 2, 11325. hman AE, Heyden G (1981) Mo ller AJ, Fabricius L, Dahlen G, O Inuence on periapical tissues of indigenous oral bacteria and necrotic pulp tissue in monkeys. Scandinavian Journal of Dental Research 89, 47584. Moorer WR, Wesselink PR (2003) Root canal treatment, intra-canal disinfectants and bacterial culture: past and present. Nederlands Tijdschrift voor Tandheelkunde 110, 178 80. Naenni N, Thoma K, Zehnder M (2004) Soft tissue dissolution capacity of currently used and potential endodontic irrigants. Journal of Endodontics 30, 7857. Nomura M, Okano M, Okada N et al. (1989) Four cases with anaphylaxis induced by chlorhexidine. Archives of Dermatology 125, 502. Ohtoshi T, Yamauchi N, Tadokoro K et al. (1986) Ig E antibody-mediated shock reaction caused by topical application of chlorhexidine. Clinical Allergy 16, 15561. Okino LA, Siqueira EL, Santos M, Bombana AC, Figueiredo JA (2004) Dissolution of pulp tissue by aqueous solution of chlorhexidine digluconate and chlorhexidine digluconate gel. International Endodontic Journal 37, 3841. Oncag O, Gogulu D, Uzel A (2006) Efcacy of various intracanal medicaments against Enterococcus faecalis in primary teeth: an in vivo study. Journal of Clinical Pediatric Dentistry 30, 2338. O, Hos lu S, Zekiog lu O, Eronat C, Onc ag go r M, Hilmiog lu D (2003) Comparison of antibacterial and toxic Burhanog effects of various root canal irrigants. International Endodontic Journal 36, 42332. Osmundsen P E (1982) Contact dermatitis to chlorhexidine. Contact Dermatitis 8, 813. Pashley DH, Tay FR, Yiu C et al. (2004) Collagen degradation by host-derived enzymes during aging. Journal of Dental Research 83, 21621.
300
Peters OA, Laib A, Gohring TN, Barbakow F (2001) Changes in root canal geometry after preparation assessed by high resolution computed tomography. Journal of Endodontics 27, 16. Portenier I, Haapasalo H, Rye A, Waltimo T, Orstavik D, Haapasalo M (2001) Inactivation of root canal medicaments by dentine, hydroxylapatite and bovine serum albumin. International Endodontic Journal 34, 1848. Portenier I, Haapasalo H, Orstavik D, Yamauchi M, Haapasalo M (2002) Inactivation of the antibacterial activity of iodine potassium iodide and chlorhexidine digluconate against Enterococcus faecalis by dentine, dentine matrix, type-I collagen, and heat-killed microbial whole cells. Journal of Endodontics 28, 6347. Portenier I, Waltimo T, Orstavik D, Haapasalo M (2006) Killing of Enterococcus faecalis by MTAD and chlorhexidine digluconate with or without cetrimide in the presence or absence of dentine powder or BSA. Journal of Endodontics 32, 13841. Ribeiro DA, Scolastici C, de Almeida PLA, Marques PLA, Marques MEA, Salvadori MF (2005) Genotoxicity of antimicrobial endodontic compounds by single cell gel (comet) assay in Chinese hamster ovary (CHO) cells. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 99, 63740. Rosenthal S, Spangberg L, Safavi KE (2004) Chlorhexidine substantivity in root canal dentine. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 98, 48892. Sanchez IR, Nusbaum KE, Swaim SF, Hale AS, Henderson RA, McGuire JA (1988) Chlorhexidine diacetate and povidoneiodine cutotoxicity to canine embryonic broblasts and Staphylococcus aureus. Veterinary Surgery 17, 1825. Sassone LM, Fidel RAS, Murad CF, Fidel SR, Hirata R (2008) Antimicrobial activity of sodium hypochlorite and chlorhexidine by two different tests. Australian Endodontic Journal 34, 1924. Schafer E, Bossmann K (2005) Antimicrobial efcacy of chlorhexidine and two calcium hydroxide formulations against Enterococcus faecalis. Journal of Endodontics 31, 536. Sen BH, Safavi KE, Spangberg LS (1999) Antifungal effects of sodium hypochlorite and chlorhexidine in root canals. Journal of Endodontics 25, 2358. Shabahang S, Aslanyan J, Torabinejad M (2008) The substitution of chlorhexidine for doxycycline in MTAD: the antibacterial efcary against a strain of Enterococcus faecalis. Journal of Endodontics 34, 28890. Siqueira JF Jr, Sen BH (2004) Fungi in endodontic infections. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 97, 63241. Siqueira JF Jr, Rocas IN, Magalhaes FA, de Uzeda M (2001) Antifungal effects of endodontic medicaments. Australian Endodontic Journal 27, 1124. Siqueira JF Jr, Rocas IN, Lopes HP, Magalhaes FA, de Uzeda M (2003) Elimination of Candida albicans infection of the
radicular dentine by intracanal medications. Journal of Endodontics 29, 5014. Siqueira JF Jr, Rocas IN, Paiva SS, Guimaraes-Pinto T, Magalhaes KM, Lima KC (2007) Bacteriologic investigation of the effects of sodium hypochlorite and chlorhexidine during the endodontic treatment of teeth with apical periodontitis. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 104, 12230. Snellman E, Rantanen T (1999) Severe anaphylaxis after a chlorhexidine bath. Journal of the American Academy of Dermatology 40, 7712. Spratt DA, Pratten J, Wilson M, Gulabivala K (2001) An in vitro evaluation of the antimicrobial efcacy of irrigants on biolms of root canal isolates. International Endodontic Journal 34, 3007. Stabholz A, Kettering J, Aprecio R, Zimmerman G, Baker PJ, Wikesjo UM (1993) Retention of antimicrobial activity by human root surfaces after in situ subgingival irrigation with tetracycline HCl or chlorhexidine. Journal of Periodontology 64, 13741. Steinberg D, Heling I, Daniel I, Ginsburg I (1999) Antibacterial synergistic effect of chlorhexidine and hydrogen peroxide against Streptococcus sobrinus, Streptococcus faecalis and Staphylococcus aureus. Journal of Oral Rehabilitation 26, 1516. Sulkala M, Larmas M, Sorsa T, Salo T, Tja derhane L (2002) The localization of matrix metalloproteinase-20 (MMP-20, enamelysin) in mature human teeth. Journal of Dental Research 81, 6037. Sulkala M, Tervahartiala T, Sorsa T, Larmas M, Salo T, Tja derhane L (2007) Matrix metalloproteinase-8 (MMP-8) is the major collagenase in human dentin. Archives of Oral Biology 52, 1217. Sundqvist G (1992) Ecology of the root canal ora. Journal of Endodontics 18, 42730. Svensater G, Bergenholtz G (2004) Biolms in endodontic infections. Endodontic Topics 9, 2736. Tanomaru JM, Leonardo MR, Tanomaru Filho M, Bonetti Filho I, Silva LA (2003) Effect of different irrigation solutions and calcium hydroxide on bacterial LPS. International Endodontic Journal 36, 7339. Tatnall FM, Leigh IM, Gibson JR (1990) Comparative study of antiseptic toxicity on basal keratinocytes, transformed human keratinocytes and broblasts. Skin Pharmacology 3, 15763. Torricelli R, Wu thrich B (1996) Life-threatening anaphylactic shock due to skin application of chlorhexidine. Clinical and Experimental Allergy 26, 1123. Tronstad L, Sunde PT (2003) The evolving new understanding of endodontic infections. Endodontic Topics 6, 5777. Vianna ME, Gomes BP, Berber VB, Zaia AA, Ferraz CC, de Souza-Filho FJ (2004) In vitro evaluation of the antimicrobial activity of chlorhexidine and sodium hypochlorite. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 97, 7984.
301
Vivacqua-Gomes N, Ferraz CC, Gomes BP, Zaia AA, Teixeira FB, Souza-Filho FJ (2002) Inuence of irrigants on the coronal microleakage of laterally condensed gutta-percha root llings. International Endodontic Journal 35, 7915. Waltimo TM, rstavik D, Siren EK, Haapasalo MP (1999) In vitro susceptibility of Candida albicans to four disinfectants and their combinations. International Endodontic Journal 32, 4219. Waltimo TM, Haapasalo M, Zehnder M, Meyer J (2004) Clinical aspects related to endodontic yeast infections. Endodontic Topics 9, 6678. Wang Y, Spencer P (2003) Hybridization efciency of the adhesive/dentine interface with wet bonding. Journal of Dental Research 82, 1415. Wang Y, Spencer P (2005) Continuing etching of an all-in-one adhesive in wet dentine tubules. Journal of Dental Research 84, 3504. White RR, Hays GL, Janer LR (1997) Residual antimicrobial activity after canal irrigation with chlorhexidine. Journal of Endodontics 23, 22931. Wuerch RM, Apicella MJ, Mines P, Yancich PJ, Pashley DH (2004) Effect of 2% chlorhexidine gel as an intracanal
medication on the apical seal of the root-canal system. Journal of Endodontics 30, 78891. Yesilsoy C, Whitaker E, Cleveland D, Phillips E, Trope M (1995) Antimicrobial and toxic effects of established and potential root canal irrigants. Journal of Endodontics 21, 5135. Yiu CK, King NM, Pashley DH et al. (2004) Effect of resin hydrophilicity and water storage on resin strength. Biomaterials 25, 578996. Zamany A, Safavi K, Spangberg LS (2003) The effect of chlorhexidine as an endodontic disinfectant. Oral Surgery, Oral Medicine Oral Pathology Oral Radiology Endodontics 96, 57881. Zehnder M (2006) Root canal irrigants. Journal of Endodontics 32, 38998. ngberg LS (2005) Effectiveness of a Zerella JA, Fouad AF, Spa calcium hydroxide and chlorhexidine digluconate mixture as disinfectant during retreatment of failed endodontic cases. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 100, 75661.
302
doi:10.1111/j.1365-2591.2008.01509.x
A laboratory assessment of coronal bacterial leakage in root canals lled with new and conventional sealers
A. U. Eldeniz1,2 & D. rstavik3

1 Nordic Institute of Dental Materials, Haslum, Norway; 2School of Dentistry, Selcuk University, Konya, Turkey; and 3Institute of Clinical Dentistry, University of Oslo, Oslo, Norway
Abstract
Eldeniz AU, rstavik D. A laboratory assessment of coronal
bacterial leakage in root canals lled with new and conventional sealers. International Endodontic Journal, 42, 303312, 2009.
Aim To evaluate the resistance to ex vivo bacterial leakage over a 40-day period of root canal llings with ve new root canal sealers: RC Sealer, Epiphany, EndoREZ, GuttaFlow and Acroseal, compared with Apexit, AH Plus and RoekoSeal. Methodology One hundred and forty-four single rooted human teeth were divided randomly into eight test (n = 15) and two control groups (n = 12). The root canals were lled using a single cone technique with gutta-percha except in the Epiphany and EndoREZ groups. These were lled with Resilon and resin-coated gutta-percha, respectively. The gutta-percha surface of the GuttaFlow group was coated with an experimental primer prior to lling. Positive controls were lled with gutta-percha without sealer and tested with bacteria, whereas negative controls were sealed with wax to test the seal between the chambers. Filled roots were
incorporated in a split chamber model system using Streptococcus mutans as a microbial marker. Leakage was assessed for turbidity of the broth in the lower chamber every day for 40 days. Survival analysis was performed using the KaplanMeier product limit method and event times were compared using the Log-rank test (a = 0.05). Results Epiphany, GuttaFlow with test primer and Apexit prevented leakage signicantly better than AH Plus, RC Sealer, RoekoSeal, EndoREZ and Acroseal (P < 0.05). None of the specimens in the AH Plus, RC Sealer, RoekoSeal and EndoREZ groups resisted bacterial penetration for 40 days. Conclusion The new sealers, Epiphany and GuttaFlow with primer, along with Apexit, showed better resistance to bacterial penetration than the other new or traditional sealers tested. Keywords: bacterial leakage, methacrylate, new sealers, polycaprolactone, silicone.
Received 30 May 2007; accepted 4 November 2008
Introduction
The aim of a root lling is to create a bacteria-tight seal, thus minimizing the risks of infection or reinfection of the root canal system (Siqueira et al. 1999) and preventing periradicular pathosis. No available material and/or technique produce a complete seal of the
Correspondence: Ayce Unverdi Eldeniz, Faculty of Dentistry, Department of Endodontics, Selcuk University, Konya, Turkey (Tel.: 90 332 223 12 57; fax: 90 332 241 00 62; e-mail: ayce71@hotmail.com; aunverdi@selcuk.edu.tr).
entire root canal system. Therefore, root canal lling materials should be developed that possess an improved capacity to prevent bacterial ingress in the long term. Recently, biodegradable aliphatic polyester incorporated with polycaprolactone (Jia & Alpert 2003) has been introduced as a root canal lling material which performs like gutta-percha. It contains dimethacrylate resins (Jia & Alpert 2003), and it can couple to a variety of dentine adhesives and resin-type sealers, including Epiphany (Pentron Clinical Technologies, Wallingford, CT, USA), RealSeal (SybronEndo, Orange, CA, USA) and Next (Heraeus-Kulzer, Armonk, NY,
303
Bacterial leakage of new sealers Eldeniz & rstavik
USA). Another resin-based root canal sealing material is EndoREZ (Ultradent, South Jordan, Utah, USA), which is a hydrophilic, urethane-dimethacrylate-based resin sealer. It is recommended for use with the same companys resin-coated gutta-percha points for bonding between sealer and point. The methyl methacrylate/tributylborane (MMA/ TBB) resin-based RC Sealer (Test sealer-Sun Medical, Moriyama, Shiga, Japan) is modied from resin cements C & B Metabond (Parkell, Farmingdale, NY, USA) and Super Bond C & B (Sun Medical, Moriyama, Shiga, Japan) by Imai & Komabayashi (2003). The major problems with the original formulations, including a short working time, low radiopacity and difculty in the removal of the resin from the root canal have been solved to some extent by substituting the polymer component with poly (methyl-methacrylate) (PMMA) (Imai & Komabayashi 2003). This material also contains partially oxidized tri-n-butyl borane as a catalyst and 4-methacryloxyethyl trimellitate anhydride/ methyl methacrylate (4-META/MMA). Many endodontic sealers containing calcium hydroxide are available. These sealers present leakage values comparable to other types of sealers commonly used in endodontics (Chailertvanitkul et al. 1996, 1997, Haikel et al. 1999, Xu et al. 2005). The advantages of the presence of calcium hydroxide in the composition of this type of sealers have been shown by many investigators (Sonat et al. 1990, Eldeniz et al. 2007a). Acroseal is a new epoxy-based sealer with calcium hydroxide. Promising clinical and laboratory data have been reported for the silicone-based sealer RoekoSeal (Wu et al. 2002, Huumonen et al. 2003, Al-Awadhi et al. 2004). Another silicone-based material, GuttaFlow, has been developed; this sealer also contains nanosilver and gutta-percha particles. The manufacturer claims a better seal and good adaptability because of the increased owability of GuttaFlow and because of the slight expansion of this material on setting (ElAyouti et al. 2005). A wide variety of test methods have been used to assess the seal of endodontic materials (Wu & Wesselink 1993), including: methylene blue (Dummer et al. 1993, Roggendorf et al. 2007), india ink (Baumgardner et al. 1995), uid ltration (Wu et al. 1995, Brackett et al. 2006, Stratton et al. 2006), radioisotopes (Rhome et al. 1981), electrochemical circuits (Jacobson & von Fraunhofer 1976), saliva (Torabinejad et al. 1995), lipopolysaccharide (Bouillaguet et al. 2004), endotoxin (Trope et al. 1995, Alves et al.
` et al. 2002) and bacteria (Torabinejad 1998, Carratu ` et al. 2002). et al. 1990, Timpawat et al. 2001, Carratu Diffusion of dyes or other media into an obturated canal space when a tooth is suspended or submerged is still a problematic issue (Trope et al. 1995). Thus, due to inadequacies associated with these types of testing methods and as a result of the nonexistence of a universally accepted model, bacterial leakage studies might be meaningful and clinically relevant. Therefore, the purpose of this ex vivo study was to assess the penetration of S. mutans through coronally unsealed, lled root canals and the effectiveness of ve new sealers (Epiphany, EndoREZ, RC Sealer, Acroseal and GuttaFlow) to resist bacterial leakage compared with conventional sealers (AH Plus, Apexit and RoekoSeal).
Materials and methods

A total of 144 single-rooted human teeth with fully developed apices were used. Data about age, gender or reason for extraction were not available. The teeth were stored in 1% NaOCl solution until use. Bone, calculus or soft tissues on the roots were removed with scalpel blades, with care not to damage the root surface. Before the experiment, the teeth were rinsed thoroughly under running tap water for 20 h. The crowns of all specimens were removed with a diamond saw (Accutom, Struers, Copenhagen, Denmark) with water coolant and the coronal surfaces of the roots were sectioned perpendicular to the long axis of the root. Three milimetres of the root apices were similarly removed. In an attempt to standardize the length of canal involved in each experimental group, the length of all roots was measured from the coronal surface to the cut apex, and root specimens ranging 1116 mm were distributed equally to the groups. Apical patency was ensured throughout instrumentation. Each root canal was coronally enlarged with Largo Peeso Reamers (Dentsply Maillefer, Ballaigues, Switzerland) to ISO size 090 or 110. Due to the presence of equally distributed oval shaped root canals in experimental groups (20% oval-shaped canals), further preparation with ISO size 110 Largo Peeso Reamer was completed in order to obtain a round root canal shape. Hand K-les (to ISO size 090 or 110) were also used to nish the enlargement and achieve better adaptation of core materials to the root segments except the EndoREZ group (see below). A total of 10 mL of distilled water, applied with a syringe and a 26 G needle (Terumo Europe N.V., Leuven, Belgium) was
304
Eldeniz & rstavik Bacterial leakage of new sealers
used for irrigation. Organic and inorganic debris including the smear layer were removed by treatment in an ultrasonic bath (Finn Sonic m03/m, Lahti, Finland) in 5% NaOCl followed by 17% EDTA, for 3 min each, and the specimens were rinsed with distilled water for 3 min to remove remnants of these solutions and autoclaved in vials containing distilled water for 20 min at 121 C. The root canals were dried ne/Whaledent, with sterile paper points (Roeko, Colte Langenau, Germany) before lling. The experimental groups had 15 and the control groups had 12 root segments, respectively. Group 1: The roots were lled with gutta-percha and AH Plus sealer (De Trey/Dentsply, Konstanz, Germany) using a single cone technique. AH Plus sealer was applied to the canal with a lentulo spiral ller (Dentsply, Maillefer) and an approriate size guttapercha cone was coated with AH Plus sealer and placed into the root canal with tweezer until fully seated. Group 2: A single cone of Resilon with Epiphany sealer (Pentron, Wallingford, CT, USA) was used. Epiphany primer was applied to the root canals for 20 s, the excess was removed with sterile paper points, and the root canals were lled as in Group 1 using Epiphany sealer and an appropriate size of Resilon point (ISO 090 or 110). Group 3: EndoREZ sealer (Ultradent, South Jordan, Utah, USA) and the manufacturers special type of resin-coated gutta-percha, 0.06 tapered with a size of 35 EndoREZ points (0.06 Taper Assorted, Lot no. 4226C-A, Ultradent Products Inc.,Ko ln-Porz, Germany) were used to ll this group. Because of the lack of ISO 090 and 110 sizes of this special type guttapercha, Prole 0.06 instruments (Prole, Dentsply Maillefer, Ballaigues, Switzerland) were used in the enlargement of these root canals to ISO size 30 with a crown-down technique to achieve adequate adaptation of coated tapered gutta percha to the dentine walls. The teeth were otherwise treated as in Group 1. Group 4: RC sealer (Sun Medical, Moriyama, Shiga, Japan) in combination with conventional gutta-percha points was used. The Green activator from the sealer kit was applied to the root canal walls with paper points for 10 s, and the root canals were rinsed with sterile distilled water and dried. Five drops of monomer were dispensed into the chilled dispensing dish and one drop of catalyst S was added to the dispensed monomer to activate the liquid. Then one small cup of the polymer powder from the RC Sealer kit was added to the activated liquid and stirred lightly for 510 s to form a slurry, which was applied to the root canals as in Group 1.
Group 5: These roots were lled with gutta-percha and Apexit sealer (Ivoclar-Vivadent, Liechtenstein) using a single cone technique similar to Group 1. Group 6: These roots were lled with gutta-percha s, and Acroseal sealer (Septodont, Saint-Maur-des-Fosse Cedex, France) using a single cone technique similar to Group 1. Group 7: These roots were lled with gutta-percha ` ne/Whaledent, Langenau, and RoekoSeal sealer (Colte Germany) using a single cone technique similar to Group 1. Group 8: The root segments in this group were lled ` ne/Whaledent, Langenau, with GuttaFlow sealer (Colte Germany), applying a test primer (Guttapercha Primer ` ne/Whaledent, Langenau, H, Lot S17848-104, Colte Germany) provided by the manufacturer for better adhesion of GuttaFlow to gutta-percha. GuttaFlow sealer was mixed according to the manufacturers instruction and applied to the canal with a lentulo spiral ller. After applying primer to the gutta-percha surface, the guttapercha cone was placed into the root canal. Group 9: The teeth in this positive control group were lled with a single cone of gutta-percha but without any sealer. Group 10: The surfaces of the roots as well as the canal orices coronally in this negative control group were completely covered by sticky wax. The canals were lled with a single cone of gutta-percha without any sealer. During all procedures throughout the experiment the teeth were kept moist by holding the roots in gauze moistened with sterile distilled water. Excess core material and sealer, coronally and apically, was carefully removed with sterile scalpels, and the root surfaces were wiped with gauze and ethanol to remove excess sealer. All groups were stored in an incubator and allowed to set for 14 days at 37 C and 100% humidity. A modication of the microbial leakage model consisting of an upper chamber and a lower chamber as described by Torabinejad et al. (1990) was used. The upper chamber consisted of a Corning 15-mL polycarbonate centrifuge tube (Corning Inc., Corning, NY, USA) with a small hole prepared at the bottom to receive the root-end. The tooth was inserted into the tube and gently pushed through the opening until approximately one-half of it protruded through the tube. The space between the tube and the tooth was then sealed with sticky wax (Sticky Wax, Kerr Corporation, Orange, CA, USA). Approximately 4 mm of root remained in the upper chamber.
305
Testing with S. mutans

Streptococcus mutans, strain ATCC 10449, was adapted to and maintained on trypticase soy broth (TSB; Oxoid Ltd, Basingstoke, UK) with 2 mg mL)1 streptomycin (Sigma-Aldrich Chemie GmbH, Steinheim, Germany) and used as a test organism. The tip of the centrifuge tube with the tooth attached was introduced into and sealed to the neck of a at-bottomed, 20-mL, transparent scintillation vial. The tip of the root was mounted to reach approximately 2 mm into a reservoir of 10 mL sterile TSB with 2 mg mL)1 streptomycin in the lower chamber. To the upper chamber 2 mL of an overnight culture of resistant S. mutans in TSB with 2 mg mL)1 streptomycin was added (Fig. 1a). Every second day 1.9 mL of broth was removed from the upper chamber
and replaced with fresh broth. The centrifuge tube cap was replaced to prevent evaporation and contamination. The mount was stored in an aerobic incubator at 37(1) C and any changes in opacity of the broth in the apical chamber checked daily for 40 days. Bacteria penetrating along the root lling were detected by turbidity observed in the lower chamber (Fig. 1b). The time taken for this to occur was recorded as an indicator of complete root canal contamination. When this occurred, the seal was broken, and the nature and purity of the organism growing in the lower chamber conrmed by Gram stain, cultural morphology and streptomycin resistance.
Data analysis
Using the nonparametric KaplanMeier analysis, survival curves were constructed illustrating leaking specimens over time, and the median time of leakage in days was estimated for all groups. Specimens that did not leak over the 40 experimental days were computed with an event time of 40 days as censored variables. Bacterial leakage was statistically compared amongst the groups using the log-rank (Mantel Cox) test, with the alpha type error set at 0.05.
(a)
(b)
Results
All positive control teeth exhibited bacterial leakage rapidly and consistently within 2448 h, whereas the lower chamber of negative control teeth remained uncontaminated throughout the experiment. All samples which were taken from the bottom chambers after the occurence of turbidity showed the presence of S. mutans only. The number of leaking samples and the mean day of leakage per group are presented in Table 1. The resistance of all the tested sealers to the penetration of the bacteria was better when compared with Group 9 in which no sealer was used (P < 0.05). The percentage of specimens without leakage was highest for the Epiphany group (73.33%). Many specimens in the Apexit (66.67%) and GuttaFlow (60%) groups also resisted bacterial penetration up to 40 days. There was no signicant difference between the Epiphany, GuttaFlow and Apexit groups (P > 0.05). None of the specimens from the AH Plus, RC Sealer, RoekoSeal and EndoREZ groups resisted bacterial penetration for 40 days. Statistically no difference was found between these groups and Acroseal group (P > 0.05). KaplanMeier survival
Figure 1 (a) Apparatus set-up demonstrating fresh broth in lower chamber. (b) Evident turbidity of broth in lower chamber after Streptococcus mutans penetration through the specimen.
306
Table 1 Number of leaking specimens and the median time of leakage in days
Test Sealers AH Plus Epiphany EndoREZ RC Sealer Apexit Acroseal RoekoSeal GuttaFlow Positive Control Negative Control n 15 15 15 15 15 15 15 15 12 12 P 15/15 4/15 15/15 15/15 5/15 11/15 15/15 6/15 12/12 0/12 p 100 26.7 100 100 33.3 73.3 100 40 100 0 m 4 40 5 2 40 10 5 40 0 40+ D a b a a b a a b
n, number of specimens P, proportions of leaking specimens p, percentage of leaking specimens m, median time of leakage in days D, Log-rank test (P < 0.05): experimental groups with different letter are signicantly different from each other.
probabilities for all the test groups are presented in Fig. 2.
Discussion
This study conrmed the ndings of other studies that used different techniques for assessment of coronal leakage and demonstrated that leakage occurs after the loss of coronal seal in lled root canals to different tic et al. 2002, Monextents (Barthel et al. 1999, Mile ticelli et al. 2007). Removal of the smear layer may improve the resistance of lled canals to bacterial challenge from a coronal direction (Behrend et al.
1996). This could be because of the relatively weak bond of the smear layer to the underlying dentine, approximately 5 MPa (Taylor et al. 1997), which may be insufcient to withstand the shrinkage associated with the curing of resins, and the smear layer may be pulled away from the dentine and provide an avenue for microleakage (Shipper & Trope 2004). The combination of 17% EDTA and 5.25% NaOCl is a generally preferred and effective method in removing the smear layer from the canal walls and dentinal tubules (Oksan et al. 1993). NaOCl is a strong oxidizing agent and may cause problems when used as the last irrigant. It leaves behind an oxygen-rich layer on the dentine surface, which results in reduced bond strengths (Lai et al. 2001, Erdemir et al. 2004) by inhibiting the polymerization of resins (Rueggeberg & Margeson 1990), and increased microleakage (Yiu et al. 2002, Stratton et al. 2006). Therefore, it has been proposed to use NaOCl rst, followed by EDTA for removal of the smear layer after the instrumentation, and then distilled water as a nal rinse in order to minimize the compromising effect of NaOCl on primer/resin-sealer polymerization (Lai et al. 2001), and to achieve better adhesion of the sealers by permitting penetration of sealers into dentine tubules (Behrend et al. 1996, Eldeniz et al. 2005), a procedure which was followed in the present study. Although the validity of laboratory leakage tests has been criticized (Wu et al. 1993, Al-Ghamdi & Wennberg 1994), the use of bacteria as markers in an ex vivo model was introduced to overcome some of the limitations of dye leakage studies (Torabinejad et al.
Survival functions
Proportion of roots resisting bacterial leakage
1.0
0.8
0.6
AH Plus EndoREZ RC Sealer Epiphany Acroseal Apexit Roeko Seal Gutta flow
0.4
0.2
0.0
Figure 2 KaplanMeier survival curves
10
20
30
40
for all sealer groups.
Experimental period (day)
307
1990, Michailesco et al. 1996, Barrieshi et al. 1997, Malone & Donnelly 1997). The model used in this study was patterned after that designed by Torabinejad et al. (1990) and has been modied and used by several other researchers for coronal leakage studies (Shipper & Trope 2004, Shipper et al. 2004). Streptomycin-resistant bacteria and a medium containing streptomycin sulphate were used to eliminate false-positive results. The apical 3 mm of the roots were removed to eliminate the variations induced by the apical pulp ramications (Vertucci 1984). The strain of S. mutans used as a bacterial marker in this study is a nonmotile coccus that moves by Brownian movement (Wu et al. 1993). It is a facultative anaerobic organism with a size of 0.51.2 lm (Behrend et al. 1996). It was used because facultative bacteria are predominant in infections of previously treated canals (Molander et al. 1998); Streptoccocus species are often found in endodontic infections (Sundqvist 1994); they penetrate easily along root tic et al. 2002), and S. mutans is canal llings (Mile convenient and practical to use for the purpose. The number of microorganisms that caused turbidity in the lower chamber was not measured as the purpose was only to test if S. mutans was capable of penetrating through the root lled speciman. KaplanMeier survival analyses were chosen for statistical analyses. They allow visualization of the event-time patterns of all materials under investigation. Event-times were compared using the log rank test. This approach facilitates differentiation between early and late failures during the 40 days of observation period (Zehnder et al. 2007). Most sealers have both antibacterial and cytotoxic effects, and these properties may limit the ingress of bacteria. The materials physical properties, such as adhesion, adaptability and degradation, may also be important for their resistance to bacterial penetration (Timpawat et al. 2001). This study found that bacterial leakage occurred in a small proportion of specimens in root canals lled with the polycaprolactone-based Resilon and Epiphany sealer after applying dentine primer. This result is in agreement with the results of Shipper et al. (2004) and could be attributable to many factors. Pre-treatment of dentine before lling with a self-etching Epiphany primer may prevent shrinkage of the resin sealer away from the dentine wall and the bonding of Resilon to resin type sealers (Tay et al. 2005a) may create a monoblock composed of Resilon lling material, resin sealant, bonding agent and dentine (Teixeira et al.
2004). As cavity-conguration factors (C-factors: the ratio of the bonded to the unbonded surface area) are extremely high in root canals (Tay et al. 2005b), sometimes this primer-containing Epiphany lling technique may not always sufce to form monoblocks within the root canals (Skidmore et al. 2006). Epiphany sealer is made to auto-polymerize in 45 min at room temperature in order to improve the chance for the relief of shrinkage stress via resin ow (Tay et al. 2005b). As the manufacturers instructions about light-curing of dual cure Epiphany sealer might cancel out the benets derived from a sealer that is designed for very slow auto-curing dynamics and in order to not add one more variable amongst the test groups, photo polymerization was not applied to this group. The antibacterial activity of the Epiphany sealer may also contribute to resistance to the bacterial penetration (Eldeniz & rstavik 2007c). Some of the Epiphany specimens showed leakage in this group (26.67%) within 40 days, which is more than found previously (Shipper et al. 2004). This could be attributable to a higher lm thickness and greater polymerization shrinkage as a result of single cone lling technique used in this study: it has previously been demonstrated that when the thickness of the adhesive is increased, the volumetric shrinkage is increased, which results in an increase in shrinkage stress (Tay et al. 2005b). Moreover, it has also been shown that Resilon core exhibits extensive surface thinning and weight loss after incubation with bacterial and salivary enzymes (Tay et al. 2005c). Biodegradation of Resilon may thus contribute to the leakage of bacteria at the sealerResilon interface. As a result of increasing interest in the use of methacrylate resin-based sealers in endodontics (Eldeniz et al. 2005, Sevimay & Kalayci 2005, Tay et al. 2005d) in order to obtain chemical union between the polyisoprene component of gutta-percha and methacylate resins, another strategy was introduced by coating conventional gutta-percha cones with resins (Haschke 2004). A special resin is created rst and grafted to gutta-percha producing a resin coat that is bondable to a methacylate-based resin sealer (Grubbs et al. 2000, Haschke 2004). The resin-coated gutta-percha cone is recommended for use as a single master cone with the EndoREZ sealer. It has been demonstrated that although long resin tags are formed with the thin hybrid layer of dentine, gaps may form along the sealer-dentine interface (Sevimay & Kalayci 2005) as a result of sealer tags being pulled out of the tubules during polymerization shrinkage (Pashley et al. 1995,
308
Bergmans et al. 2005) and leakage could not be prevented in canals lled with this technique (Tay et al. 2005d). In the present study, even if the inner surface area of the root canals were less when compared with the other experimental groups due to the increased taper of the instruments used during the preparation of these root segments, all the specimens in this group showed bacterial leakage within 13 days. This could be due to polymerization shrinkage of the methacrylate-based sealer (Schwartz & Fransman 2005) following the increased sealer thickness as a result of using a single cone technique (De-Deus et al. 2006); the high C-factors of the root canals (Tay et al. 2005b); the incomplete removal of the smear layer in isolated areas of the root canal; and also debonding of resin sealer from resin coated gutta-percha due to the lack of an oxygen inhibition layer which is necessary for optimal coupling of methacrylate-based resins (Ruyter 1981). After the application of the coating, this inhibition layer is removed to avoid sticking of the gutta-percha cones during storage (Haschke 2004). This may have resulted in weak bonds between the resin-coated gutta-percha and the resin sealer (Tay et al. 2005d). Another reason for the rapid bacterial leakage in this sealer group could be that no or very little antibacterial activity has been demonstrated for EndoREZ (Sipert et al. 2005, Eldeniz et al. 2006). Although the RC Sealer has satisfactory physical properties (lm thickness, ow, radiopacity) according to the ISO standards 68761984 and 2001, it demonstrates higher lm thickness values than the other sealers in the present study (Eldeniz & rstavik 2005). Previous studies also demonstrated that it is slightly toxic to human gingival broblasts (Eldeniz et al. 2007b) and it shows similar apical leakage values as AH Plus and Rocanal 2 (Cobankara et al. 2006). In the present study, all the specimens lled with this sealer leaked within eight days. Again, this material also has very limited antibacterial activity (Erdemir et al. 2003). The epoxy-resin-based AH Plus sealer was chosen as a reference (Brackett et al. 2006). All the specimens lled with AH Plus sealer leaked within 13 days. This could be because of shrinkage of this epoxy type sealer during setting (rstavik et al. 2001), and/or as a result of diminished antibacterial activity of this sealer some time after setting (Pizzo et al. 2006) with no inhibitory effect on S. mutans (Kaplan et al. 1999). The results are in aggreement with the ndings of Yu cel et al. (2006) and Timpawat et al. (2001), who showed that early bacterial penetration and increased penetration after 14 days for AH Plus sealer respectively, but they are in
contrast of the ndings of other researchers (De-Deus et al. 2006) who demonstrated that a greater lm thickness did not negatively inuence the sealing property and good performance of AH Plus sealer to polymicrobial leakage. The relatively poor bonding between gutta-percha and the sealer (Sevimay & Kalayci 2005) may also contribute to bacterial leakage in the present study. The high pH of calcium hydroxide-containing sealers is known to inhibit growth of bacteria. Whilst calcium hydroxide released by Apexit may be partly neutralized by other compounds in the formula and thus limit its antibacterial effect (Kaplan et al. 1999, Timpawat et al. 2001), this sealer demonstrated relatively high resistance to bacterial penetration in the present study. This could be attributable to the highly hydrophobic, zinc stearate content of this sealer that might prevent water ingress and solubility (Eldeniz et al. 2007a), and it could be because of toxic substances in the Apexit effective against S. mutans (Eldeniz et al. 2007b). The recently introduced calcium hydroxide-based Acroseal sealer contains diglycidyl ether of bisphenol A and methenamine, which are known epoxy compounds and also found in the formula of AH 26 and Sealer 26. Most of the specimens in this group leaked within 28 days (73.33%). This could be because of polymerization contraction of the sealer as a consequence of epoxy component, and to the presence of voids in this type of sealer (Mutal & Gani 2005); voids might occur as a result of formaldehyde release during setting and through ionization of calcium hydroxide (Eldeniz et al. 2007a). All RoekoSeal lled specimens showed leakage within 9 days. These results are in contrast with the favourable leakage results reported by previous researchers (Bouillaguet et al. 2004, Roggendorf et al. 2007). This could be attributable to different leakage evaluation methods and/or tracers used. The present results conrmed the ndings reported by Economides et al. (2005) that RoekoSeal in combination with gutta percha did not yield better sealing ability than other sealers tested. Whitworth & Baco (2005) suggested that the use of RoekoSeal sealer only as backlls provided leak-proof coronal seals. This may be because the sealer-gutta percha interface is eliminated (Brackett et al. 2006). The GuttaFlow sealer, when used together with its test primer, showed comparable leakage values to the Resilon-Epiphany and Apexit sealer groups. The primer may be important for this promising result; also, the moderate expansion of the GuttaFlow sealer (ElAyouti et al. 2005, Monticelli et al. 2007) may be benecial.
309
Conclusion
Within the limitations of this laboratory study Epiphany, GuttaFlow with test primer, and Apexit sealers resisted bacterial penetration for a longer period of time than EndoREZ, RC Sealer, AH Plus, Acroseal and RoekoSeal.
Acknowledgements
The nancial support of the Research Council of Norway and Nordic Institute of Dental Materials (NIOM) is gratefully acknowledged. The authors would also like to thank to the Scientic Research Projects Coordination Center of Selcuk University, Konya/ Turkey, for partial support of this project (Project no: 07701032). The authors wish to thank Mrs. Inger Sophie Dragland and Mrs. Anne Wesman for technical assistance, Professor Leiv Sandvik for assisting with statistical analyses, and the companies Ultradent, Sun ` ne/Whaledent for generMedical, Septodont, and Colte ously providing materials used in this study.
References
Al-Awadhi S, Spears R, Gutmann JL, Opperman LA (2004) Cultured primary osteoblast viability and apoptosis in the presence of root canal sealers. Journal of Endodontics 30, 52733. Al-Ghamdi A, Wennberg A (1994) Testing of sealing ability of endodontic lling materials. Endodontics & Dental Traumatology 10, 24955. Alves J, Walton R, Drake D (1998) Coronal leakage: endotoxin penetration from mixed bacterial communities through obturated, post-prepared root canals. Journal of Endodontics 24, 58791. Barrieshi KM, Walton RE, Johnson WT, Drake DR (1997) Coronal leakage of mixed anaerobic bacteria after obturation and post space preparation. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology 84, 3104. Barthel CR, Moshonov J, Shuping G, rstavik D (1999) Bacterial leakage versus dye leakage in obturated root canals. International Endodontic Journal 32, 3705. Baumgardner KR, Taylor J, Walton R (1995) Canal adaptation and coronal leakage: lateral condensation compared to Thermal. Journal of the American Dental Association 126, 3516. Behrend GD, Cutler CW, Gutmann JL (1996) An in vitro study of smear layer removal and microbial leakage along root-canal llings. International Endodontic Journal 29, 99 107. Bergmans L, Moisiadis P, De Munck J, Van Meerbeek B, Lambrechts P (2005) Effect of polymerization shrinkage on
the sealing capacity of resin llers for endodontics use. The Journal of Adhesive Dentistry 7, 3219. Bouillaguet S, Wataha JC, Lockwood PE, Galgano C, Golay A, Krejci I (2004) Cytotoxicity and sealing properties of four classes of endodontic sealers evaluated by succinic dehydrogenase activity and confocal laser scanning microscopy. European Journal of Oral Sciences 112, 1827. Brackett MG, Martin R, Sword J et al. (2006) Comparison of seal after obturation techniques using a polydimethylsiloxane-based root canal sealer. Journal of Endodontics 32, 118890. ` P, Amato M, Riccitiello F, Rengo S (2002) Evaluation Carratu of leakage of bacteria and endotoxins in teeth treated endodontically by different techniques. Journal of Endodontics 28, 2725. Chailertvanitkul P, Saunders WP, Mackenzie D (1996) An assessment of microbial coronal leakage in teeth root lled with gutta-percha and three different sealers. International Endodontic Journal 29, 38792. Chailertvanitkul P, Saunders WP, Mackenzie D (1997) Coronal leakage of obturated root canals after long-term storage using a polymicrobial marker. Journal of Endodontics 23, 6103. Cobankara FK, Orucoglu H, Sengun A, Belli S (2006) The quantitative evaluation of apical sealing of four endodontic sealers. Journal of Endodontics 32, 6668. De-Deus G, Coutinho-Filho T, Reis C, Murad C, Paciornik S (2006) Polymicrobial leakage of four root canal sealers at two different thicknesses. Journal of Endodontics 32, 998 1001. Dummer PMH, Kelly T, Meghji A, Sheikh I, Vanitchai JT (1993) An in vitro study of the quality of root llings in teeth obturated by lateral condensation of gutta-percha or Thermal obturators. International Endodontic Journal 26, 99105. Economides N, Panagiotis B, Kolokouris I, Gogos C, Kokorikos I (2005) Comparative study of the sealing ability of a polydimethylsiloxane-based root canal sealer. Brazilian Dental Journal 16, 1458. ElAyouti A, Achleithner C, Lo st C, Weiger R (2005) Homogeneity and adaptation of a new gutta-percha paste to root canal walls. Journal of Endodontics 31, 68790. Eldeniz AU, rstavik D (2005) Physical properties of newly developed root canal sealers. International Endodontic Journal 38, 9148. Abstract no: 50. Eldeniz AU, Erdemir A, Belli S (2005) Shear bond strength of three resin based sealers to dentin with and without the smear layer. Journal of Endodontics 31, 2936. Eldeniz AU, Erdemir A, Hadimli HH, Belli S, Erganis O (2006) Assessment of antibacterial activity of EndoREZ. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology 102, 11926. Eldeniz AU, Erdemir A, Kurtoglu F, Esener T (2007a) Evaluation of pH and calcium ion release of Acroseal sealer in comparison with Apexit and Sealapex sealers. Oral
310
Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology 103, e8691. Eldeniz AU, Mustafa K, rstavik D, Dahl JE (2007b) Cytotoxicity of new resin-, calcium hydroxide- and silicone-based root canal sealers on broblasts derived from human gingiva and L929 cell lines. International Endodontic Journal 40, 32937. Eldeniz AU, rstavik D (2007c) Antibacterial activity of new root canal lling materials against Enterococcus faecalis. Abstract no: R1.29. International Endodontic Journal 40, 988. Erdemir A, Unverdi Eldeniz A, Hadimli HH, Belli S, Erganis O (2003) In vitro antibacterial activity of a new resin-based sealer. Journal of Dental Research 82, Dental Materials B-343. (Special Issue B): Abstract no: 2660. Erdemir A, Ari H, Gungunes H, Belli S (2004) Effect of medications for root canal treatment on bonding to root canal dentin. Journal of Endodontics 30, 1136. Grubbs RB, Broz ME, Dean JM, Bates FS (2000) Selectively epoxidized polyisoprene-polybutadiene block copolymers. Macromolecules 33, 230810. Haikel Y, Wittenmeyer W, Bateman G, Bentaleb A, Allemann C (1999) A new method for the quantitative analysis of endodontic microleakage. Journal of Endodontics 25, 1727. Haschke E (2004) Adhesive Endodontic Cones and Related Methods. United States Patent Application 20040202986. US Patent & Trademark Ofce. Huumonen S, Lenander-Lumikari M, Sigurdsson A, rstavik D (2003) Healing of apical periodontitis after endodontic treatment: a comparison between a silicone-based and a zinc oxide-eugenol-based sealer. International Endodontic Journal 36, 296301. Imai Y, Komabayashi T (2003) Properties of a new injectable type of root canal lling resin with adhesiveness to dentin. Journal of Endodontics 29, 203. Jacobson SM, von Fraunhofer JA (1976) The investigation of microleakage in root canal therapy. An electrochemical technique. Oral Surgery, Oral Medicine, and Oral Pathology 42, 81723. Jia WT, Alpert B (2003) Root Canal Filling Material. United States Patent Application 20030113686. US Patent and Trademark Ofce. Kaplan AE, Picca M, Gonzalez MI, Macchi RL, Molgatini SL (1999) Antimicrobial effect of six endodontic sealers: an in vitro evaluation. Endodontics & Dental Traumatology 15, 42 5. Lai SC, Mak YF, Cheung GS et al. (2001) Reversal of compromised bonding to oxidized etched dentin. Journal of Dental Research 80, 191924. Malone KH, Donnelly JC (1997) An in vitro evaluation of coronal microleakage in obturated root canals without coronal restorations. Journal of Endodontics 23, 358. Michailesco PM, Valcarcel J, Grieve AR, Levallois B, Lerner D (1996) Bacterial leakage in endodontics: an improved method for quantication. Journal of Endodontics 22, 5359.
tic I, Prpic -Mehic G, Mars Mile ic an T et al. (2002) Bacterial and fungal microleakage of AH 26 and AH Plus root canal sealers. International Endodontic Journal 35, 42832. Molander A, Reit C, Dahlen G, Kvist T (1998) Microbiological status of root-lled teeth with apical periodontitis. International Endodontic Journal 31, 17. Monticelli F, Sadek FT, Schuster GS et al. (2007) Efcacy of two contemporary single-cone lling techniques in preventing bacterial leakage. Journal of Endodontics 33, 3103. Mutal L, Gani O (2005) Presence of pores and vacuoles in set endodontic sealers. International Endodontic Journal 38, 6906. Oksan T, Aktener BO, Sen BH, Tezel H (1993) The penetration of root canal sealers into dentinal tubules. A scanning electron microscopic study. International Endodontic Journal 26, 3015. rstavik D, Nordahl I, Tibballs JE (2001) Dimensional change following setting of root canal sealer materials. Dental Materials 17, 5129. Pashley DH, Ciucchi B, Sano H, Carvalho RM, Russell CM (1995) Bond strength versus dentine structure: a modelling approach. Archives of Oral Biology 40, 110918. Pizzo G, Giammanco GM, Cumbo E, Nicolosi G, Gallina G (2006) In vitro antibacterial activity of endodontic sealers. Journal of Dentistry 34, 3540. Rhome BH, Solomon EA, Rabinowitz JL (1981) Isotopic evaluation of the sealing properties of lateral condensation, vertical condensation, and Hydron. Journal of Endodontics 7, 45861. Roggendorf MJ, Ebert J, Petschelt A, Frankenberger R (2007) Inuence of moisture on the apical seal of root canal llings with ve different types of sealer. Journal of Endodontics 33, 313. Rueggeberg FA, Margeson DH (1990) The effect of oxygen inhibition on an unlled/lled composite system. Journal of Dental Research 69, 16528. Ruyter IE (1981) Unpolymerized surface layers on sealants. Acta Odontologica Scandinavica 39, 2732. Schwartz RS, Fransman R (2005) Adhesive dentistry and endodontics: materials, clinical strategies and procedures for restoration of access cavities: a review. Journal of Endodontics 31, 15165. Sevimay S, Kalayci A (2005) Evaluation of apical sealing ability and adaptation to dentine of two resin-based sealers. Journal of Oral Rehabilitation 32, 10510. Shipper G, Trope M (2004) In vitro microbial leakage of endodontically treated teeth using new and standard obturation techniques. Journal of Endodontics 30, 1548. Shipper G, rstavik D, Teixeira FB, Trope M (2004) An evaluation of microbial leakage in roots lled with a thermoplastic synthetic polymer-based root canal lling material (Resilon). Journal of Endodontics 30, 3427. Sipert CR, Hussne RP, Nishiyama CK, Torres SA (2005) In vitro antimicrobial activity of Fill Canal, Sealapex, Mineral
311
Trioxide Aggregate, Portland cement and EndoRez. International Endodontic Journal 38, 53943. Siqueira JF, Roc as IN, Lopes HP, de Uzeda M (1999) Coronal leakage of two root canal sealers containing calcium hydroxide after exposure to human saliva. Journal of Endodontics 25, 146. Skidmore LJ, Berzins DW, Bahcall JK (2006) An in vitro comparison of the intraradicular dentin bond strength of Resilon and gutta-percha. Journal of Endodontics 32, 9636. Sonat B, Dalat D, Gu nhan O (1990) Periapical tissue reaction to root llings with Sealapex. International Endodontic Journal 23, 4652. Stratton RK, Apicella MJ, Mines P (2006) A uid ltration comparison of gutta-percha versus Resilon, a new soft resin endodontic obturation system. Journal of Endodontics 32, 6425. Sundqvist G (1994) Taxonomy, ecology, and pathogenicity of the root canal ora (review). Oral Surgery, Oral Medicine, and Oral Pathology 78, 52230. Tay FR, Loushine RJ, Weller RN et al. (2005a) Ultrastructural evaluation of the apical seal in roots lled with a polycaprolactone-based root canal lling material. Journal of Endodontics 31, 51419. Tay FR, Loushine RJ, Lambrechts P, Weller RN, Pashley DH (2005b) Geometric factors affecting dentin bonding in root canals: a theoretical modeling approach. Journal of Endodontics 31, 5849. Tay FR, Pashley DH, Yiu CK et al. (2005c) Susceptibility of a polycaprolactone-based root canal lling material to degradation. II. Gravimetric evaluation of enzymatic hydrolysis. Journal of Endodontics 31, 73741. Tay FR, Loushine RJ, Monticelli F et al. (2005d) Effectiveness of resin-coated gutta-percha cones and a dual-cured, hydrophilic methacrylate resin-based sealer in obturating root canals. Journal of Endodontics 31, 65964. Taylor JK, Jeansonne BG, Lemon RR (1997) Coronal leakage: effects of smear layer, obturation technique, and sealer. Journal of Endodontics 23, 50812. Teixeira FB, Teixeira EC, Thompson J, Leinfelder KF, Trope M (2004) Dentinal bonding reaches the root canal system. Journal of Esthetic and Restorative Dentistry 16, 34854. Timpawat S, Amornchat C, Trisuwan WR (2001) Bacterial coronal leakage after obturation with three root canal sealers. Journal of Endodontics 27, 369.
Torabinejad M, Ung B, Kettering JD (1990) In vitro bacterial penetration of coronally unsealed endodontically treated teeth. Journal of Endodontics 16, 5669. Torabinejad M, Rastegar AF, Kettering JD, Pitt Ford TR (1995) Bacterial leakage of mineral trioxide aggregate as a root-end lling material. Journal of Endodontics 21, 10912. Trope M, Chow E, Nissan R (1995) In vitro endotoxin penetration of coronally unsealed endodontically treated teeth. Endodontics & Dental Traumatology 11, 904. Vertucci F (1984) Root canal anatomy of human permanent teeth. Oral Surgery, Oral Medicine, and Oral Pathology 58, 58999. Whitworth JM, Baco L (2005) Coronal leakage of sealer-only backll: an in vitro evaluation. Journal of Endodontics 31, 2802. Wu MK, Wesselink PR (1993) Endodontic leakage studies reconsidered. Part 1. Methodology, application and relevance. International Endodontic Journal 26, 3743. Wu MK, De Gee AJ, Wesselink PR, Moorer WR (1993) Fluid transport and bacterial penetration along root canal llings. International Endodontic Journal 26, 2038. Wu MK, Wesselink PR, Boersma J (1995) A 1-year follow-up study on leakage of four root canal sealers at different thicknesses. International Endodontic Journal 28, 1859. Wu MK, Tigos E, Wesselink PR (2002) An 18-month longitudinal study on a new silicon-based sealer, RSA RoekoSeal: a leakage study in vitro. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontolog 94, 499502. Xu Q, Fan MW, Fan B, Cheung GS, Hu HL (2005) A new quantitative method using glucose for analysis of endodontic leakage. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology 99, 10711. Yiu CK, Garcia-Godoy F, Tay FR et al. (2002) A nanoleakage perspective on bonding to oxidized dentin. Journal of Dental Research 81, 62832. Yu , Gu E (2006) Bacterial cel AC ler E, Gu ler AU, Ertas penetration after obturation with four different root canal sealers. Journal of Endodontics 32, 8903. F (2007) Zehnder M, Baumgartner G, Marquardt K, Paque Prevention of bacterial leakage through instrumented root canals by bioactive glass S53P4 and calcium hydroxide suspensions in vitro. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology 103,4238.
312
doi:10.1111/j.1365-2591.2008.01511.x
Effect of educational intervention on adoption of new endodontic technology by general dental practitioners: a questionnaire survey
.Tegelberg4,5 M. Koch1, H. G. Eriksson2, S. Axelsson3 & A

1 Department of Endodontics, Public Dental Service, Stockholm County Council, Stockholm, Sweden; 2R&D Centre/Centre for Clinical Research, So rmland County Council, Eskilstuna, Sweden; 3The Swedish Council on Technology Assessment in Health s, Sweden; and 5Faculty of Odontology, Care, Stockholm, Sweden; 4Centre for Clinical Research, Uppsala University, Va stera Malmo University, Malmo , Sweden
Abstract
. Effect of Koch M, Eriksson HG, Axelsson S, Tegelberg A
educational intervention on adoption of new endodontic technology by general dental practitioners: a questionnaire survey. International Endodontic Journal, 42, 313321, 2009.
Objectives To survey the clinical endodontic protocols of general dental practitioners (GDPs) in public dental clinics and to assess the effect of an educational intervention on the adoption of a nickeltitanium (Ni Ti) rotary system. Methods General dental practitioners in a Swedish Intervention County (IC), underwent an educational programme in endodontics. A follow-up questionnaire was posted to 98 GDPs in the IC and to 97 GDPs in a Control County (CC), where no specic training had been provided. The questionnaire concerned demographics, clinical endodontic protocols and instrumentation techniques. Results The response rate to the questionnaire was 87%. More than 90% of all GDPs reported they always or generally used rubber dam, determined working length, used the canal irrigant 0.5% buffered NaOCl and calcium hydroxide as an interappointment
dressing. Two of three GDPs reported, they generally or always informed the patient of the prognosis. Every second GDP reported routines for postoperative recall and follow-up. The NiTi rotary technique was reported to be completely adopted by 77% of the GDPs in the IC, signicantly higher than in the CC (6%), P < 0.001. In the IC 79% of the GDPs reported they completed instrumentation in one treatment session, compared with only 32% in the CC, P < 0.001. The singlecone mode of canal lling was reported to be signicantly more frequent amongst GDPs in the IC, P < 0.001. Conclusions General dental practitioners in both counties reported using contemporary clinical endodontic protocols. GDPs who had undergone an educational programme in NiTi rotary instrumentation reported they had successfully integrated the technique into daily clinical practice. Keywords: dentistry, endodontics, education, implementation, NiTi rotary technique.
Received 16 August 2008; accepted 4 November 2008
Introduction
A high healing rate following root canal treatment is reported when root llings meet optimal technical desjo standards (O et al. 1990, Sjo gren et al. 1990, Kirkevang et al. 2000). However, Scandinavian studies report poor technical standards of root llings performed by general dental practitioners (GDPs)
Correspondence: Dr Margaretha Koch, Special Dental Clinic, Department of Endodontics, Public Dental Service, Stockholm County Council, Polhemsgatan 48, S-112 82 Stockholm, Sweden (Tel.: 46 8 12315600; fax: 46 8 6536011; e-mail: m.koch@telia.com).
313
NiTi rotary instrumentation in general practice Koch et al.
(Kirkevang et al. 2000, Ridell et al. 2006). A high frequency of apical periodontitis (AP) is observed with root lled teeth in general populations and is correlated to the technical quality of the root lling; a high frequency of AP is frequently reported in teeth with desjo inadequately lled root canals (O et al. 1990, Kirkevang et al. 2000). As the cause of AP is root canal infection, prerequisites for successful treatment outcomes are strict aseptic procedures and high technical standards (Ber ngberg 2004). Internationally, there is genholtz & Spa considerable variation in the quality of clinical endodontic procedures by GDPs. Studies have disclosed poor aseptic control, expressed as low use of rubber dam and underestimation of the microbiological factors inuencing the prognosis for root canal treatment (McColl et al. 1999, Jenkins et al. 2001, Slaus & Bottenberg 2002, Hommez et al. 2003, Bjrndal & Reit 2005, Bjrndal et al. 2007). Root canal instruments made of superelastic nickel titanium alloy (NiTi) are reported to be superior to stainless steel hand instruments in following the root canal path and reducing procedural errors, thereby providing better treatment outcomes than instrumentation with stainless steel les (Pettiette et al. 2001, Scha fer & Lohmann 2002, Guelzow et al. 2005). NiTi instruments are available for both hand and engine-driven instrumentation (NiTi rotary instru ngberg 2001). Compared with prepamentation) (Spa ration with 0.02 tapered NiTi hand instruments, the NiTi rotary technique, however, is reported to clean and shape the root canals more effectively, particularly when used in curved canals (Sonntag et al. 2003, Peters et al. 2004). Studies conrm that adoption of the NiTi rotary technique by GDPs is limited (Hommez et al. 2003, Parashos & Messer 2004, Bjrndal & Reit 2005). However, adoption rates as high as 80% are reported for GDPs who had attended courses in the NiTi rotary technique, including hands-on training (Barbakow & Lutz 1997, Reit et al. 2007). For several years, continuing education courses in endodontics in Sweden have included the concept of NiTi rotary instrumentation. It is not known whether such initiatives promote sustainable change in clinical practice, or what type of intervention is the most effective. The aim of this study was therefore to survey clinical endodontic protocols amongst GDPs employed by the Public Dental Service in two similarly organized Swedish counties. A further aim was to assess the rate of
adoption of a NiTi rotary system after an educational intervention in one of the counties.
Materials and methods Study design

In the Swedish county of So rmland, denoted the Intervention County (IC), an educational programme in endodontics was conducted over 2 years, targeting all GDPs employed in public dental clinics in the county. The education included seminars and handson training in NiTi rotary technique. It was followed by a post-intervention questionnaire to all the participants. For comparison, the same questionnaire was sent to all GDPs in the county of Va stmanland, denoted the Control County (CC), where no educational programme had been provided. The specialist dental clinics, (departments of endodontics) in the two counties were similarly organized, with one Senior Consultant in Endodontics. The counties were equally exposed to the advertising of the technique, continuing education courses run by the Swedish Dental Association and by the manufacturers, and lectures held at The Annual Scientic Congress of the Swedish Dental Society.
The educational programme

The educational programme was conducted in the IC by the district Senior Consultant in Endodontics, during 2003 and 2004. At the time there were 16 public dental clinics in the county. Because of staff turnover, the exact number of GDPs varied, but at the time of the post-intervention evaluation, 87 GDPs had participated in the education. To promote incorporation of the new technique, the intervention comprised a number of implementation strategies, as shown in Table 1. The introductory visit to each public dental clinic lasted 2 h. To ensure that all GDPs had the opportunity to attend, the one-day seminar was conducted four times during a one-year period. The practical training was carried out at the participants home-clinic, where the concept of the NiTi rotary technique was also presented to assisting personnel, who were encouraged to make practical contributions with respect to adoption of the new technique, such as organization, timeplanning and hygiene. A list including descriptions and prices of the equipment was made available to the clinics purchasing ofcer.
314
Koch et al. NiTi rotary instrumentation in general practice
Table 1 Intervention elements in the educational programme

Step Initial meeting to inform heads of Public Dental Service One introductory outreach visit to each Public Dental Clinic A one day seminar for all GDPs, covering the pathogenesis of apical periodontitis, root lling technique and the theory underlying the NiTi rotary technique Introductory seminar for assisting personnel in theory of NiTi rotary technique Educational outreach visits including hands-on practical training of NiTi rotary technique on extracted teeth with the ProFile-system One follow-up visit to each clinic Initiation of personally chosen individual interventions by all participants Reminder emails Monthly presentation of a practical endodontic recommendation Publication of guidelines on management of endodontic emergencies Aim To ensure organizational co-operation To ensure compliance with the educational programme To update GDPs knowledge
To promote understanding among the staff To establish basic practical skills in the technique
To ensure sustained incorporation by repeated feed-back and reminders To emphasize GDPs individual control of the change To promote compliance To improve clinical practice To prevent procedural errors
Each dentist was instructed in the practical application of the ProFile NiTi rotary system (Dentsply Maillefer, Baillagues, Switzerland) on extracted teeth. After access cavities were prepared, working length of the canals was determined with radiographs and stainless steel K-les, size 10. The canals were manually instrumented with stainless steel K-les to apical size 20. The ProFile instruments were used in a crown down sequence (30.06, 25.06, 25.04) in a 128 : 1 hand piece (WD74M, W&H, Bu rmoos, Austria) until working length was reached. A nal apical size of 35, taper 0.06 was recommended and achieved in a stepback mode with the instruments (30.04, 30.06, 35.04, 35.06). Depending on root anatomy, a greater apical size was sometimes required and achieved with a size 40.06 instrument. The canals were lled in single-cone mode using greater taper points cut to the appropriate apical size in a Maillefer Gauge tester (Dentsply Maillefer, Ballaigues, Switzerland), and sealer. Radiographs were taken postoperatively to check the quality of the root lling. At the end of the session there was a general discussion about the participants experience of using the technique. This training session lasted for 4 h. The individual interventions covered a range of issues, e.g. suggestions for rationalizing the instrumentation or adjustments in time-planning. Throughout the intervention, practical recommendations and guidelines were published monthly on the Public Dental Service Intranet. As participation in the education was compulsory for all public dental clinics in the IC, loss of production and
the cost of educational materials were nanced by the Public Dental Service, So rmland County Council.
Post-intervention evaluation
In 2005, 1 year after completion of the educational programme, a postal questionnaire and a return address envelope were sent to 195 GDPs in the IC and CC, preceded by a personal letter to each dentist, explaining the purpose of the study. Each questionnaire was marked with identication numbers. After 4 weeks, nonresponders received a reminder questionnaire, sent by a person unrelated to the study, in order to guarantee anonymity. No further reminders were sent.
Questionnaire
The questionnaire had been modied following testing of a pilot questionnaire in which four GDPs participating in the nal study had been included. The nal version could be completed in approximately 20 min. There were 25 questions about demographics, quality protocols applied in clinical endodontics and experience of and attitudes to different endodontic techniques. The demographic variables included: gender, age, years in profession, experience of continuing education and if endodontic treatments were either regularly, or not/very seldom, part of everyday clinical practice. The following variables relevant to quality protocols in endodontic practice were evaluated: the assessment
315
of prognosis, the use of rubber dam, determination of working length, the use of irrigation and interappointment dressing and routines for postoperative recall and follow-up. Adoption of the new endodontic technique was assessed in terms of the following variables: instrumentation technique, number of treatment sessions needed to complete instrumentation, root lling technique and attitudes to conventional and NiTi rotary techniques. The closed-format questions offered six responses: always, generally, frequently, sometimes, rarely or never. In the open-format questions there were choices as well as space for the participants to make additional comments in case their usual practice had not been adequately covered. The last page of the survey offered space for the participant to make additional remarks.
Two of three GDPs in both counties reported, they always used rubber dam. Another 20% used rubber dam routinely, but made exceptions occasionally. GDPs in both counties who did not inform their patients of the prognosis reported that they were constrained by time pressure or put the patients needs rst in an emergency situation, or that they didnt know what outcome they could expect or that they forgot to inform the patient. Postoperative radiographic control was not routine, but restricted mainly to two indications: before crown therapy or in cases of clinical symptoms.
Patterns of rotary NiTi use

Rate of adoption The reported use of NiTi rotary instrumentation amongst GDPs differed signicantly between the two counties (P < 0.001) (Fig. 2). The technique was reported to be used at all public dental clinics in the IC: 77% of the GDPs reported full integration of the technique, and another 12% used a combination of techniques. Treatment sessions and root lling technique With respect to the number of treatment sessions required to complete instrumentation, signicantly fewer sessions were reported by GDPs in the IC (P < 0.001) (Fig. 3). The use of sealer was established in both groups. However, due to different instrumentation methods, the root lling technique differed between the groups. Canal lling with a single gutta-percha cone was signicantly more frequent in the IC than in the CC (P < 0.001), where the use of sealer was combined with the cold lateral compaction technique (P < 0.001) (Fig. 4).
Statistical analysis
Descriptive data were presented as percentages. Differences between the Intervention and Control Counties were tested with Pearsons chi-squared test and Fischer Exact Test. Statistical tests were two-tailed at the 5% signicance level. All data analysis was performed in the spss version 13 (SPSS, Chicago, IL, USA).
Results
Of the original 195 GDPs in the two counties, 11 had relocated, retired or become specialists. Of the remaining 184, 12 declined to participate in the study, for unknown reasons. The response rate to the questionnaire was 92% of GDPs in the IC (n = 91) and 83% (n = 81) in the CC. The questions were answered by 99100% of the participants. The frequency of answers to questions about reasons for using different techniques, 085%, depended on the participants attitudes and clinical experience. There were no signicant differences between the counties regarding distribution of gender, mean age or mean years in profession (Fig. 1). In the IC 100% and in the CC 94% of the participants regularly undertook endodontic treatment procedures.
Reasons for using different techniques

Respondents using conventional and mixed techniques gave the following main reasons: the conventional technique offered a sense of control over instrumentation, fear of NiTi-instrument fracture and the lack of practical education in rotary techniques (Table 3). Respondents using the NiTi rotary technique and mixed techniques gave the following reasons: greater root lling quality, less physically tiring technique, fast and easy procedures. They agreed that adoption of the rotary technique was dependent on practical education (Table 4).
Quality procedures in clinical practice

The frequency of quality procedures in endodontics applied by general dental practitioners in the IC and CC is shown in Table 2.
316
Figure 1 Flow-chart of the study. Gender, mean age and mean years in profession among public general dental practitioners (GDPs) in the Intervention and Control Counties.
Table 2 Frequency of quality procedures in endodontics

Quality procedures Inform patient of the prognosis The use of rubber dam Determination of working length 0.5% NaOCl (Dakins solution) as canal irrigant Calcium hydroxide as intra canal dressing Routine postoperative control n 91 90 90 91 90 91 IC (%) 54 81 90 90 90 46 (59) (90) (100) (99) (100) (51) n 80 80 80 80 80 79 CC (%) 58 73 78 79 80 38 (73) (91) (98) (99) (100) (48)
Comparison, %, between public general dental practitioners (GDPs) in the Intervention County (IC) who underwent an educational programme in endodontics and public GDPs in the Control County (CC) where no organized training had been provided. The frequency (%) of responses is shown in parentheses.
Discussion
The results are based on self-reported endodontic routines by 172 GDPs in two Swedish counties. A possible explanation for the very high response rate to the questionnaire in general and to all the
questions in particular, might be that the survey offered an opportunity to express everyday practical experiences, besides a growing interest in NiTi rotary instrumentation. The answers can thus be regarded as representative for the GDPs in both counties.
317
Figure 2 The use of different techniques. Comparison (%) between public general dental practitioners in the Intervention County (IC) and in the Control County (CC), where no programmed education had been provided.
Figure 4 The frequency (%) of choice of medicament and obturation technique in root lling procedures among public general dental practitioners in the Intervention County (IC) and in the Control County (CC), where no programmed education had been provided.
Table 3 Reasons for using conventional hand instruments

IC n = 21 (%) 10 (48) 2 (10) 5 (24) 5 (24) 3 (14) 1 (5) 1 (5) 2 (10) 7 (33) 0 (0) CC n = 75 (%) 41 (55) 24 (32) 18 (24) 16 (21) 15 (20) 11 (15) 2 (3) 3 (4) 11 (15) 6 (8) Difference between groups P-value 0.05
Reasons for use of hand instruments It offers a feeling of control No practical education in rotary technique Concern about le fractures with rotary technique Satised with the present quality of my root llings No theoretical education Still dubious about the NiTi rotary technique Been advised not to adopt the NiTi rotary technique I perform endodontics too seldom to change technique Other reasons Equipment unavailable at my clinic
Figure 3 The frequency (%) of treatment sessions required to
complete instrumentation, depending on number of canals, by public general dental practitioners in the Intervention County (IC) and in the Control County (CC), where no programmed education had been provided.
A signicantly higher adoption of the NiTi rotary technique was reported by the GDPs who had undergone an educational programme in its application. Participation in the above programme does not seem to have inuenced the already high standard of clinical endodontic routines reported by GDPs in both counties. It was satisfying to note that the routines reported comply with the consensus report of quality standards of root canal treatment (European Society of Endodontology 2006): more than 90% of the GDPs in both counties reported they used rubber dam routinely, irrigated with Dakins solution (NaOCl), used calcium hydroxide as an
Comparison between public general dental practitioners (GDPs) in the Intervention County (IC) who had undergone a supervised educational programme in endodontics and GDPs in the Control County (CC) where no programmed education had been provided. The frequency (%) of responses is shown in parentheses. The respondents were free to choose multiple answers.
interappointment medicament, and determined the working length of the root canals. However, although a majority of the GDPs in this study informed their
318
Table 4 Reasons for using NiTi rotary technique

IC n = 81 (%) 69 (85) 67 (83) 65 (80) 40 (49) 53 (65) 40 (49) 56 (69) 35 (43) 26 (32) CC n = 26 (%) 15 (58) 21 (81) 13 (50) 10 (38) 14 (54) 3 (12) 10 (38) 7 (27) 10 (38) Difference between groups P-value 0.05 0.05 0.05 0.05
Reasons for use of NiTi rotary technique Improved quality of the root lling Less physically demanding Easier canal lling Converted after theoretical education Canal preparation is faster Fewer treatment sessions required Converted after practical instruction Equipment is available at my clinic Not satised with the quality that the conventional technique offers Colleagues recommended the technique Other reasons
11 (14) 3 (4)
8 (31) 0 (0)
Comparison between public general dental practitioners (GDPs) in the Intervention County (IC) who underwent a supervised educational programme in endodontics and public GDPs in the Control County (CC) where no programmed education had been provided. The frequency (%) of responses is given in parentheses. The respondents were free to choose multiple answers.
patients of the prognosis for the planned treatment, only one in two reported established protocols for postoperative recall and follow-up. Primarily, postoperative recall was limited to cases where crown therapy was planned, or because of symptoms. Without routine postoperative follow-up of all root lled teeth, asymptomatic, unhealed cases may remain undetected. In the absence of signs of symptoms, GDPs probably assume a successful outcome. This approach implies overestimation of the importance of clinical symptoms as preoperative prognostic factors (Bjrndal et al. 2007). The high standard of quality control reported in the clinical routines presented above can be related to the impact of undergraduate education: the GDPs adhere to treatment principles they were taught as undergraduates. However, to translate scientic progress into clinical practice, as with the NiTi rotary concept, GDPs need to be able to review their current professional knowledge and adapt their clinical practice according to new evidence (Parashos & Messer 2006). Besides the clinical reorganization that is often necessary, all personnel involved need to understand the purpose and the changes required (Shojania & Grimshaw 2004).
Evidence-based methods of implementing change are interactive and combine strategies to improve professional practice, and to overcome barriers to change (Lomas 1993, Oxman et al. 1995, Wensing et al. 1998). In this study, emphasis was therefore placed on pre-intervention acceptance of the educational programme, repeated active support and reminders to all the participants (McGlone et al. 2001), which may have inuenced the adoption of the technique by 89% of the GDPs in the IC. Adoption of the technique may, however, be incomplete, as disclosed by a Belgian study (Hommez et al. 2003) where only 1.6% used rotary instruments exclusively, i.e. without the additional use of conventional les. In the study by Barbakow & Lutz 1997, only around 50% of the adopters treated all types of teeth, and Reit et al. 2007 described an uneven distribution of the adopters amongst the clinics. In comparison, in the current study, the NiTi-rotary technique was integrated at all clinics in the IC and a majority of the GDPs, 77% reported they completely implemented the technique. A major thrust of the educational programme was to guide the GDPs towards full adoption of the technique: the expectation was that the focus would shift from the technical difculties of hand instrumentation to greater awareness of the biological issues underlying clinical endodontic procedures. Moreover, it was expected that complete integration of the technique would promote more sustainable change. The benets of the NiTi rotary technique were well recognized by the adopting GDPs in this study. In both counties a majority cited ergonomic reasons for using the technique. However, in the IC the GDPs also noted such benets as improved quality of the root lling and simplied canal lling, conrmed by the signicantly higher reported use of the single-cone technique in the IC. Quicker canal preparation with the NiTi rotary technique is reported in previous studies (Barbakow & Lutz 1997, Parashos & Messer 2004, Reit et al. 2007) but a correlation with signicantly fewer treatment sessions, as disclosed in the present study, is conrmed only in the Danish study by Bjrndal & Reit (2005) reporting that with the NiTi rotary technique root treatments on molar teeth could be completed in signicantly fewer sessions. Whether the correlation between technique and treatment sessions per case was due to increased understanding in treatment planning amongst the GDPs in the IC, or improved technical skill, could not
319
be analysed in this study. It is, however, apparent that the GDPs in the IC were able to take full advantage of all aspects of the new technique: in contrast, the use of the NiTi rotary system by GDPs in the CC was not as well-integrated. General dental practitioner who used conventional techniques perceived no advantages with the new technique, emphasized the sense of control in using hand instruments and expressed concern about the risk of le fractures with NiTi rotary instrumentation. It is apparent that the major reason why GDPs in the CC preferred conventional hand instruments is lack of supervised hands-on instruction in NiTi rotary technique. This may explain the signicant differences in the level of integration between the counties and the attitudes to the technique. Hence, to encourage competence in NiTi rotary instrumentation amongst graduate dentists, professional education should be provided by specialists. To achieve sustainable change, it is likely that interventional steps such as those described in this study are necessary.
Conclusions
General dental practitioners in the two Swedish counties in this study reported a high quality of clinical endodontic protocols. Following a supervised educational programme, GDPs successfully integrated the NiTi rotary technique into everyday clinical practice. There was a correlation between adoption of the new technology and rational instrumentation and lling methods.
Acknowledgements
Eva Nohlert, DDS, MPH, Centre for Clinical Research, s and Kristina Gelin and Uppsala University, Va stera n, County Council of So Inger af Sille rmland, Nyko ping, are kindly acknowledged for their contributions to this study. The nancial support of the So rmland County Council Public Dental Service, Sweden and the Uppsala s, SweUniversity Centre for Clinical Research, Va stera den, is also gratefully acknowledged.
References
Barbakow F, Lutz F (1997) The Lightspeed preparation technique evaluated by Swiss clinicians after attending
continuing education courses. International Endodontic Journal 30, 4650. ngberg L (2004) Controversies in endodonBergenholtz G, Spa tics. Critical Reviews in Oral Biology and Medicine 15, 99 114. Bjrndal L, Reit C (2005) The adoption of new endodontic technology amongst Danish general dental practitioners. International Endodontic Journal 38, 528. Bjrndal L, Laustsen MH, Reit C (2007) Danish practitioners assessment of factors inuencing the outcome of endodontic treatment. Oral surgery, oral medicine, oral pathology, oral radiology, and endodontics 103, 5705. European Society of Endodontology (2006) Quality guidelines for endodontic treatment: consensus report of the European Society of Endodontology. International Endodontic Journal 39, 92130. Guelzow A, Stamm O, Martus P, Kielbassa AM (2005) Comparative study of six rotary nickel-titanium systems and hand instrumentation for root canal preparation. International Endodontic Journal 38, 74352. Hommez GMG, Braem M, De Moor RJG (2003) Root canal treatment performed by Flemish dentists. Part 1. Cleaning and shaping. International Endodontic Journal 36, 16673. Jenkins SM, Hayes SJ, Dummer PMH (2001) A study of endodontic treatment carried out in dental practice within the UK. International Endodontic Journal 34, 1622. Kirkevang LL, rstavik D, Ho rstedt-Bindslev P, Wenzel A (2000) Periapical status and quality of root llings and coronal restorations in a Danish population. International Endodontic Journal 33, 50915. Lomas J (1993) Retailing research: increasing the role of evidence in clinical services for childbirth. The Milbank quarterly 71, 43975. McColl E, Smith M, Whitworth J, Seccombe G, Steele J (1999) Barriers to improving endodontic care: the views of NHS practitioners. British dental journal 186, 5648. McGlone P, Watt R, Sheiham A (2001) Evidence-based dentistry: an overview of the challenges in changing professional practice. British dental journal 190, 6369. desjo n L, Salonen L, Langeland K (1990) PrevaO B, Hellde lence of previous endodontic treatment, technical standard and occurrence of periapical lesions in a randomly selected adult, general population. Endodontics & dental traumatology 6, 26572. Oxman AD, Thomson M, Davis DA, Haynes RB (1995) No magic bullets: a systematic review of 102 trials of interventions to improve professional practice. CMAJ : Canadian Medical Association Journal 153, 142331. Parashos P, Messer HH (2004) Questionnaire survey on the use of rotary nickel-titanium endodontic instruments by Australian dentists. International Endodontic Journal 37, 24959. Parashos P, Messer HH (2006) The diffusion of innovation in dentistry: a review using rotary nickel-titanium technology
320
as an example. Oral surgery, oral medicine, oral pathology, oral radiology, and endodontics 101, 395401. Peters OA, Barbakow F, Peters CI (2004) An analysis of endodontic treatment with three nickel-titanium rotary root canal preparation techniques. International Endodontic Journal 37, 84959. Pettiette MT, Delano EO, Trope M (2001) Evaluation of success rate of endodontic treatment performed by students with stainless-steel K-les and nickel-titanium hand les. Journal of endodontics 27, 1247. Reit C, Bergenholtz G, Caplan D, Molander A (2007) The effect of educational intervention on the adoption of nickeltitanium rotary instrumentation in a Public Dental Service. International Endodontic Journal 40, 26874. ` re I (2006) Periapical Ridell K, Petersson A, Matsson L, Meja status and technical quality of root-lled teeth in Swedish adolescents and young adults. A retrospective study. Acta odontologica Scandinavica 64, 10410. Scha fer E, Lohmann D (2002) Efciency of rotary nickeltitanium FlexMaster instruments compared with stainless steel hand K-Flexole Part 2. Cleaning effectiveness and
instrumentation results in severely curved root canals of extracted teeth. International Endodontic Journal 35, 51421. Shojania KG, Grimshaw JM (2004) Still no magic bullets: pursuing more rigorous research in quality improvement. The American journal of medicine 116, 77880. Sjo gren U, Ha gglund B, Sundqvist G, Wing K (1990) Factors affecting the long-term results of endodontic treatment. Journal of endodontics 16, 498504. Slaus G, Bottenberg P (2002) A survey of endodontic practice amongst Flemish dentists. International Endodontic Journal 35, 75967. Sonntag D, Delschen S, Stachniss V (2003) Root-canal shaping with manual and rotary Ni-Ti les performed by students. International Endodontic Journal 36, 71523. ngberg L (2001) The wonderful world of rotary root canal Spa preparation. Oral surgery, oral medicine, oral pathology, oral radiology, and endodontics 92, 479. Wensing M, van der Weijden T, Grol R (1998) Implementing guidelines and innovations in general practice: which interventions are effective? The British journal of general practice 48, 9917.
321
doi:10.1111/j.1365-2591.2008.01512.x
Evaluation of the strength and radiopacity of Portland cement with varying additions of bismuth oxide
E. Saliba1, S. Abbassi-Ghadi1, R. Vowles1, J. Camilleri2, S. Hooper1 & J. Camilleri2,3

1 Department of Oral and Dental Science, Bristol Dental School, University of Bristol, Bristol, UK; 2Department of Dental Surgery, University of Malta, Malta; and 3Department of Building and Civil Engineering, Faculty for the Built Environment, University of Malta, Malta
Abstract
Saliba E, Abbassi-Ghadi S, Vowles R, Camilleri J, Hooper S, Camilleri J. Evaluation of the strength and radiopacity of
Portland cement with varying additions of bismuth oxide. International Endodontic Journal, 42, 322328, 2009.
Aim To study the effect of addition of various proportions of bismuth oxide on compressive strength and radiopacity of Portland cement. Methodology The compressive strength of white Portland cement and cement replaced with 10, 15, 20, 25 and 30% bismuth oxide was evaluated by testing cylinders 6 mm in diameter and 12 mm high. Twelve cylinders were tested for each material under study. The radiopacity of the cements tested was evaluated using an aluminium step-wedge and densitometer. The optical density was compared with the relevant thickness of aluminium (Al). Statistical analysis was
performed using Analysis of Variance (anova) with P = 0.05 and Tukey test to perform multiple comparison tests. Results Various additions of bismuth oxide had no signicant effect on the strength of the material when compared with the unmodied Portland cement (P > 0.05). The radiopacity of the cements tested ranged from 2.02 mm Al for Portland cement to 9.79 mm Al for the highest bismuth replacement. Conclusions Addition of bismuth oxide did not affect the compressive strength of Portland cement. All the bismuth oxide cement mixtures had radioopacities higher than 3 mm thickness of aluminium. Keywords: bismuth oxide, compressive strength, mineral trioxide aggregate, Portland cement, radiopacity.
Received 15 August 2008; accepted 31 October 2008
Introduction
Mineral trioxide aggregate (MTA) has been used as a root-end lling material for the past decade. The ideal properties of a root-end lling material have been well documented (Grossman et al. 1988) and include sufcient radiopacity to be able to distinguish the material from surrounding structures. Clearly a lack of radiopacity when small quantities of material are employed
Correspondence: Dr Josette Camilleri, PhD, Faculty of Architecture and Civil Engineering, Department of Building and Civil Engineering, University of Malta, Malta (Tel.: 00356 2340 2870; fax: 00356 21330190; e-mail: josette.camilleri @um.edu.mt).
can result in the material not being distinguished from the surrounding anatomical structures (Beyer-Olsen & Orstavik 1981). Grey MTA (ProRootTM) has been shown to have a radiopacity equivalent to 5.34 mm (Danesh et al. 2006) and 6.4 mm (Laghios et al. 2000) aluminium (Al), respectively. The prototype version of MTA used by Torabinejad et al. (1995) was shown to have a radiopacity of 7.17 mm Al. MTA is less radiopaque than Super-EBA (9.9 mm Al), IRM (9.3 mm Al), gutta-percha (11.0 mm Al) or amalgam (15.6 mm Al) (Laghios et al. 2000), but is more radiopaque than Viscosity Enhanced Root Repair Material (4.5 mm Al) (Chng et al. 2005). It but is in the same range as zinc oxideeugenol based canal sealers (Torabinejad et al. 1995).
322
Saliba et al. Strength and radiopacity of Portland cement
Most of the materials used in endodontics have radiopacifying materials added to them. These materials usually include barium or bismuth compounds as they are radiopaque. There have been few investigations on the effect that radiopaciers have on the properties of materials. However, it has been shown that addition of barium sulphate at low concentrations reduced working and initial setting times, but further addition delayed the setting reaction of glass ionomer cements (Prentice et al. 2006). Both compressive strength and surface hardness decreased with increasing concentrations of the radiopacifying agent (Prentice et al. 2006). The British Standard for Dental Root Canal Sealing Materials [BS EN ISO 6876 (British Standard Institution 2002)] does not specically recommend minimum limits for compressive strength. However, there is a minimum recommendation of 50 MPa for compressive strength of low strength dental water-based cements [BS EN ISO 9917-1 (British Standard Institution 2003)]. Compressive strength may not be of paramount importance when MTA is used as a root-end lling, but any subsequent replacement of an associated orthograde root canal lling would require the root-end lling to have adequate strength. The applications for the use of MTA have broadened and sufcient strength must be considered important. Mineral trioxide aggregate is composed of Portland cement and 4 : 1 addition of bismuth oxide (Torabinejad & White 1995). The main constituent phases in MTA are tricalcium and dicalcium silicates, which are the same phases constituting Portland cement (Camilleri et al. 2005). The bismuth oxide added to the MTA has been shown to affect the hydration mechanism of the MTA; it forms part of the structure of calcium silicate hydrate, which is the main by-product of cement hydration and also affects the precipitation of calcium hydroxide in the hydrated paste (Camilleri 2007). Mineral trioxide aggregate is now widely used to repair lateral root perforations (Lee et al. 1993, Pitt Ford et al. 1995), for the repair of cervical resumption (BarattoFilho et al. 2005), as a pulp capping agent (Pitt Ford et al. 1996, Bakland 2000) and as a dressing over pulpotomies (Holland et al. 2001). These extended uses place additional demands on the development of MTA and the compressive strength of the material must be viewed as a more critical property. In addition, it has been demonstrated that additions of bismuth oxide to Portland cement may reduce the compressive strength of the material (Coomaraswamy et al. 2007, Camilleri 2008). The aim of this preliminary study was to investigate the effect of bismuth oxide on the physical properties of
Portland cement thus evaluating the effect on the properties of MTA.

The materials used in this study included white Portland cement [WPC; Italcementi SPA, Bergamo, Italy manufactured to BS EN 197-1; (British Standard Institution 2000), type CEM I, Portland cement strength class 52,5N] and bismuth oxide (Fischer Scientic, Leicester, UK). Bismuth oxide was added to the Portland cement by replacing 10, 15, 20, 25 and 30% of the cement powder by weight. White Portland cement without bismuth oxide was used as a control. The original cement and the cements with bismuth oxide replacements were mixed with water at a powder to water ratio of 1 : 0.37 by weight. This ratio was chosen after performing a standard consistence test [BS EN 196-3; (British Standard Institution 2005)] to determine the amount of water required by this particular cement.
Compressive strength
The compressive strength testing was conducted in accordance to EN 9917-1 (2003) with a modication in specimen dimensions. Cylinders 6 mm in diameter and 12 mm high were cast rather than using the suggested 4 mm diameter and 6 mm height. The cements were cast in three increments and vibrated for a minute per increment to avoid entrapped air. The specimen and mould assemblies were stored in an incubator at 37 C and 100% humidity for 1 day after which the cylinders were removed from the moulds and cured in water at 37 C for 28 days. The specimens were tested after 28 days to allow adequate cement curing. Twelve cylinders were prepared for each material tested. The cylinders were compressed using a compression machine (Lloyd Instruments Ltd, Fareham, UK) with a cross-head speed of 1 mm min)1 until failure in accordance with BS EN 9917-1 (British Standard Institution 2003). The maximum load required to fracture the samples was noted. Compressive strength was calculated using the formula: Compressive strength Applied loadN Areamm2
Radiopacity
The experimental protocol was based on ISO 6876, Section 7.8 (2002). The Portland cement was mixed with distilled water, at a powder to water ratio of 1 : 0.37
323
Strength and radiopacity of Portland cement Saliba et al.
by weight. Cements with different bismuth oxide additions as in the previous experiment were also prepared. The mixes were compacted into stainless steel ring moulds 10 mm in diameter and 1 mm high, and pressed against two glass cover slips to make the specimen 1 mm thick. Two specimens of each material were prepared. The cements were allowed to cure for 24 h at 37 C covered by a plastic sheet to avoid cement desiccation after which they were removed from the moulds. The cements were placed directly on a cassette loaded with a cephalostat type lm with an intensifying screen (Kodak, Rochester, NY, USA) adjacent to an aluminium step wedge with 11 steps each step being 3 mm high (Everything X-ray, High Wycombe, UK) and irradiated with X-rays using a Cephalostat (Orthoralix S, Gendex, Dental Systems, Medivance Instruments Ltd, London, UK) at tube voltage of 64 kV, and current 3 mA and exposure time of 0.5 s. This procedure was used in order to maintain a set target to lm distance of 505 mm. Two specimens per material under test were arranged on the cephalostat lm and two x-rays were taken of the specimens (Film 1 and Film 2). Eight layers of lead foil were used with each lm to ensure that a small section of the lm received no exposure. The radiographs were processed in an automatic processing machine (Clarimat 300, Gendex Dental Systems, Medivance Instruments Ltd, London, UK). A photographic densitometer (PTW densix, Freiburg, Germany) was used to take readings of the density of the radiographic image of the specimens, each step of the step wedge and the un-exposed part of the lm. Three readings of each material were taken for each radiograph of each specimen and the mean density calculated. The net radiographic density was calculated by subtracting the base and fog value from the gross radiographic density. The base and fog value is the inherent optical transmission density (lowest density) of a lm base plus the nonimage density contributed by the developed emulsion. Graphs were plotted for net radiographic density (NRDAL) of the aluminium steps (NRDAL) versus the logarithm of the thickness of aluminium (log d) for each radiograph taken. From the resultant plots, the gradient and the intercept were calculated for each lm. Linear regression of the data was obtained using the following formula: NRDAL m:logd I where NRDAL was the net radiographic density of the aluminium step wedge, m was the gradient, log d was the logarithm of the step height, I was the intercept. By rearranging the above equation into:
log d
(I - NRD) m
the logarithm of the relevant thickness of aluminium for each material could be calculated from its net radiographic density for each lm taking into consideration that specimen thickness was 1 mm. Logarithms of step height were then converted to thicknesses of aluminium (Watts & McCabe 1999).
The compressive strength data was evaluated using spss (Statistical Package for the Social Sciences) software (SPSS Inc., Chicago, IL, USA). The distribution was rst evaluated to select the appropriate statistical analysis. The data was plotted and the distribution curve was analysed together with the Kolmogorov Zmirnov test with P = 0.05. A normal distribution was signied by P > 0.05, thus parametric tests were performed. Analysis of Variance (anova) with P = 0.05 was rst performed to evaluate variation between the means. In addition, once a signicant variance was detected between the data analysed, the Tukey test was used to perform multiple comparison tests to determine which means differed.
Results Compressive strength

The results for compressive strength testing of the cements are shown in Fig. 1. All the cements tested including the control showed compressive strength values higher than 50 Nmm)2, which is the minimum recommended by BS EN 9917-1 (British Standard Institution 2003) for dental water-based cements. There is no minimum requirement for root-end lling materials. There was no statistically signicant difference between the cements with various proportions of bismuth oxide and the white Portland cement (P > 0.05) cured for 28 days. There was a difference between the 10% bismuth oxide replacement and the 30% bismuth oxide replacement with the 30% bismuth oxide replacement showing signicantly higher (P = 0.0006) compressive strength values.
Radiopacity
The base and fog value for Film 1 was 0.23 and that for Film 2 was 0.24. Subtraction of these values from the gross radiographic density resulted in the net
324
130 120
Compressive strength N/mm2
110 100 90 80 70 60 50 40 30 20 10 0 WPC WPC & 10% Bi2O3 WPC & 15% Bi2O3 WPC & 20% Bi2O3 WPC & 25% Bi2O3 WPC & 30% Bi2O3
Cement composition
Figure 1 Compressive strength testing of cement cylinders with varying additions of bismuth oxide SD (n = 12).
radiographic density of the aluminium step wedge. The plots of resultant net radiographic density against the logarithm of the step height for both lms had a gradient: )1.91 for Film 1 and )1.7 for Film 2 with an intercept of 2.74 and 2.4, respectively. The results for mean thickness of aluminium compared with the relevant bismuth oxide replacement cements are shown in Fig. 2. All the cements tested with the exception of un-modied Portland cement had a radiopacity of at least 3 mm Aluminium, which is the
12 11
minimum recommended by BS EN ISO 6876 (British Standard Institution 2002).
Discussion
In the present study the effect of additions of varying bismuth oxide concentrations to Portland cement was investigated. The main constituent phases of MTA are the same as those of Portland cement (Camilleri et al. 2005). MTA is composed of ASTM type 1 Portland
Mean thickness of aluminium (mm)
10 9 8 7 6 5 4 3 2 1 0
0 2.02 10 3.64 15 4.68 20 6.62 25 7.61 30 9.79
Addition of bismuth oxide to portland cement (%)

Figure 2 Radiopacity of cement with varying additions of bismuth oxide expressed as mean thicknesses of Aluminium SD (n = 2). The dotted line shows the minimum value for radiopaque restorative material.
325
cement and bismuth oxide (Torabinejad & White 1995). ASTM Type 1 cements and CEM 1 cements are the same; the only difference being the mode of classication (Neville 1981). Thus, the results of the present study have implications for MTA and the likely effects on its physical properties. The bismuth oxide was added by replacing the amount by weight in percentages varying from 1030%. Mineral trioxide aggregate (ProRootTM) has been reported as having a 20% replacement of bismuth oxide (Torabinejad & White 1995). The compressive strength of cements under study was conducted using the method described in BS EN ISO 9917-1 (British Standard Institution 2003). This standard species the testing of physical properties of dental water-based cements and advises the use of moulds producing cylinders 4 mm in diameter and 6 mm high. In this experiment, cylinders 6 mm in diameter and 12 mm in height were used giving a height to diameter ratio of 2. This method was adopted in previous studies testing MTA and related materials (Camilleri et al. 2006). The strength of cylinders with a height to diameter ratio 2 is not inuenced by the restraining effects of the loading plates. Higher values than 2 may lead to buckling of the specimens and lower values require the use of a correction factor when calculating the compressive strength (Neville 1981). No packing material was used to minimize friction between the loading plates and the specimen. If the surfaces tested are not even, the effective contact area between the specimen and the bearing plate can be reduced. The optical radiographic density of the materials under study was evaluated using a densitometer (Higginbotham 1967) and an aluminium step-wedge as reference (Eliasson & Haasken 1979). The use of digitized X-rays and comparison of the grey pixel value has been reported (Tagger & Katz 2003, CarvalhoJunior et al. 2007). The problem with the use of aluminium step wedges lies in the composition of the aluminium as pure aluminium is very difcult to machine due to its softness. The use of aluminium alloys which are easier to machine makes comparisons difcult (Watts & McCabe 1999). Other variables include the target to lm distance, although it has been reported that varying the target to lm distance does not affect the radiopacity as long as the samples were properly exposed (Gu et al. 2006). Bismuth oxide is added to MTA as a radiopacifying agent in 1 : 4 proportions (Torabinejad & White 1995). In the present study 20% bismuth oxide was added to CEM 1 Portland cement to mimic ProRootTM MTA. The
compressive strength values of the 20% bismuth oxide replacement were similar to that reported by other researchers for white MTA (Islam et al. 2006, Holt et al. 2007, Nekoofar et al. 2007). It has been shown that bismuth affects the hydration mechanism of MTA (Camilleri 2007). SEM analysis of the MTA set structure revealed that bismuth formed part of the structure of silicate hydrate gel, which is the main by-product of cement hydration and also affected the precipitation of calcium hydroxide in the hydrated paste (Camilleri 2007). The results of this study are at variance with a previous study (Camilleri 2008) where it was demonstrated that 20% addition of bismuth oxide to cement clinker reduced the strength considerably at all curing times. Other researchers (Coomaraswamy et al. 2007) also reported a reduction in compressive strength of the cement with increasing levels of bismuth oxide. In this study (Coomaraswamy et al. 2007) the specimens were tested after 10 days of curing in comparison to the present study where the specimens were tested after 28 days. In the present study there seemed to be no changes in the results of compressive strength testing on addition of various percentages of bismuth oxide. The variation in the results of compressive strength testing lies in the different testing regimes used. These variables include the use of different specimen sizes, specimen shape, the loading rate and end preparation of the specimen as discussed by Camilleri et al. (2006). Mineral trioxide aggregate has been shown to have a radiopacity equivalent to 7.17 mm (Torabinejad et al. 1995), 6.4 mm (Laghios et al. 2000) and 5.34 mm (Danesh et al. 2006) aluminium. These values are similar to those reported in this experiment for 20% bismuth oxide replaced Portland cement. The values reported for Portland cement in this experiment were also in agreement with other publications (Danesh et al. 2006) who reported the radiopacity of CEM 1 cement to be 3.32 mm of Al. All the bismuth oxide replaced cements in this study had radiopacities greater than 3 mm of aluminium, which is the minimum requested by the BS EN ISO 6876 (British Standard Institution 2002). The present study suggests that 20% bismuth oxide is higher than is required for radiopacity whilst maintaining other physical properties of the material at appropriate levels.
Conclusions
Addition of bismuth oxide did not seem to affect the compressive strength of Portland cement. All the
326
bismuth oxide replaced cements had radiopacities higher than 3 mm thickness of aluminium.
Acknowledgements
The study was partially funded by a grant from the Faculty of Dental Surgery, University of Malta and by the Bristol Dental School Alumni. Mr. Edward Grupetta of the Radiology Department, Mater Dei Hospital Malta for access to equipment.
References
Bakland LK (2000) Management of traumatically injured pulps in immature teeth using MTA. Journal of the Californian Dental Association 28, 8558. Baratto-Filho F, Limongi O, Arau jo Cde J, Neto MD, Maia SM, Santana D (2005) Treatment of invasive cervical resorption with MTA: case report. Australian Endodontic Journal 31, 76 80. Beyer-Olsen EM, Orstavik D (1981) Radiopacity of root canal sealers. Oral Surgery, Oral Medicine, Oral Pathology 51, 320 8. British Standard Institution. (2000) Cement. Part 1: Composition, Specications and Conformity Criteria for Common Cements. BS EN 197-1. British Standard Institution (2002) Dental Root Canal Sealing Materials. BS EN ISO 6876-7.8. British Standard Institution. (2003) Dentistry: Water-Based Cements. Part 1: Powder/liquid AcidBase Cements. BS EN ISO 9917-1. British Standard Institution. (2005) Methods of Testing Cements- Part 3: Determination of Setting Time and Soundness. BS EN 196-3. Camilleri J (2007) Hydration mechanisms of mineral trioxide aggregate. International Endodontic Journal 40, 46270. Camilleri J (2008) The physical properties of accelerated Portland cement for endodontic use. International Endodontic Journal 41, 1517. Camilleri J, Montesin FE, Brady K, Sweeney R, Curtis RV, Pitt Ford TR (2005) The constitution of mineral trioxide aggregate. Dental Materials 21, 297303. Camilleri J, Montesin FE, Curtis RV, Pitt Ford TR (2006) Characterization of Portland cement for use as a dental restorative material. Dental Materials 22, 56975. Carvalho-Junior JR, Correr-Sobrinho L, Correr AB, Sinhoreti MAC, Consani S, Sousa-Neto MD (2007) Radiopacity of root lling materials using digital radiography. International Endodontic Journal 40, 51420. Chng HK, Islam I, Yap AUJ, Tong YW, Koh ET (2005) Properties of a new root-end lling material. Journal of Endodontics 31, 6658. Coomaraswamy KS, Lumley PJ, Hofmann MP (2007) Effect of bismuth oxide radioopacier content on the material
properties of an endodontic Portland cement-based (MTAlike) system. Journal of Endodontics 33, 2958. Danesh G, Dammaschke T, Gerth HU, Zandbiglari T, Schafer E (2006) A comparative study of selected properties of ProRoot mineral trioxide aggregate and two Portland cements. International Endodontic Journal 39, 2139. Eliasson ST, Haasken B (1979) Radiopacity of impression materials. Oral Surgery, Oral Medicine and Oral Pathology 47, 48591. Grossman LI, Oliet S, Del Rio CE (1988) Endodontic Practice, 11th edn. Philadelphia: Lea & Febiger, 242. Gu S, Rasimick BJ, Deutsch AS, Musikant BL (2006) Radiopacity of dental materials using a digital X-ray system. Dental Materials 22, 76570. Higginbotham TL (1967) A comparative study of the physical properties of ve commonly used root canal sealers. Oral Surgery, Oral Medicine and Oral Pathology 24, 89101. Holland R, de Souza V, Murata SS et al. (2001) Healing process of dog dental pulp after pulpotomy and pulp covering with mineral trioxide aggregate and Portland cement. Brazilian Dental Journal 12, 10913. Holt DM, Watts JD, Beeson TJ, Kirkpatrick TC, Rutledge RE (2007) The anti-microbial effect against enterococcus faecalis and the compressive strength of two types of mineral trioxide aggregate mixed with sterile water or 2% chlorhexidine liquid. Journal of Endodontics 33, 8447. Islam I, Chng HK, Yap AU (2006) Comparison of the physical and mechanical properties of MTA and Portland cement. Journal of Endodontics 32, 1937. Laghios CD, Benson BW, Gutmann JL, Cutler CW (2000) Comparative radiopacity of tetracalcium phosphate and other root-end lling materials. International Endodontic Journal 33, 3115. Lee SJ, Monsef M, Torabinejad M (1993) Sealing ability of a mineral trioxide aggregate for repair of lateral root perforations. Journal of Endodontics 19, 5414. Nekoofar MH, Adusei G, Sheykhrezae MS, Hayes SJ, Bryant ST, Dummer PMH (2007) The effect of condensation pressure on selected physical properties of mineral trioxide aggregate. International Endodontic Journal 40, 45361. Neville AM (1981) Properties of Concrete, 3rd edn. Harlow, UK: Longman Scientic and Technical Essex. Pitt Ford TR, Torabinejad M, McKendry DJ, Hong CU, Kariyawasam SP (1995) Use of mineral trioxide aggregate for repair of furcal perforations. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 79, 75663. Pitt Ford TR, Torabinejad M, Abedi HR, Bakland LK, Kariyawasam SP (1996) Using mineral trioxide aggregate as a pulp-capping material. Journal of the American Dental Association 127, 14914. Prentice LH, Tyas MJ, Burrow MF (2006) The effect of ytterbium uoride and barium sulphate nanoparticles on
327
the reactivity and strength of a glass-ionomer cement. Dental Materials 22, 74651. Tagger M, Katz A (2003) Radiopacity of endodontic sealers: development of a new method for direct measurement. Journal of Endodontics 29, 7515. Torabinejad M, White DJ (1995) Tooth Filling Material and Use. Alexandria, VA, USA: United States Patent and Trademark Ofce. US Patent Number 5,769,638.
Torabinejad M, Hong CU, McDonald F, Pitt Ford TR (1995) Physical and chemical properties of a new root-end lling material. Journal of Endodontics 21, 34953. Watts DC, McCabe JF (1999) Aluminium radiopacity standards for dentistry: an international survey. Journal of Dentistry 27, 738.
328
doi:10.1111/j.1365-2591.2008.01513.x
Topographical evaluation of the major apical foramen in permanent human teeth
lez-Rodr guez2 & L. A. S. Castro3 J. Martos1, C. M. Ferrer-Luque2, M. P. Gonza

1 Department of Semiology and Clinics, School of Dentistry, Federal University of Pelotas, RS, Brazil; 2Department of Dental Pathology and Therapeutics, School of Dentistry, University of Granada, Spain; and 3Laboratory of Immunology and Microscopy, EMBRAPA-CPACT, Pelotas, RS, Brazil
Abstract
lez-Rodr guez MP, Martos J, Ferrer-Luque CM, Gonza Castro LAS. Topographical evaluation of the major apical
foramen in permanent human teeth. International Endodontic Journal, 42, 329334, 2009.
Aim To determine the distance from the anatomical root apex to the major apical foramen and the position of the major foramen on the root apex. Methodology Crowns of 926 human teeth were sectioned at the cementum-enamel junction. Specimens were mounted on microscope slides for measurement parallel to the long axis of the teeth. The major foramen was identied as the largest-diameter opening at the root apex. A total of 1331 root specimens were evaluated using an optical stereomicroscope to an accuracy of 0.01 mm at 40 (10) magnication. The distance from the anatomical apex to the most apical point of the major foramen was measured, and its location (central, buccal, lingual, mesial and distal) was recorded.
Results The mean distance between the major foramen and the anatomical root apex was 0.69 mm; the mean distance was larger in posterior teeth (0.82 mm) and smaller in anterior teeth (0.39 mm). A wide range of anatomical apex to major foramen distances were observed in all tooth groups: the greatest distance was in maxillary molars (0.95 mm) followed by mandibular pre-molars (0.87 mm) and mandibular molars (0.80 mm). The major foramen was at the tip of the root in 40% of teeth. The most frequent deviations of the foramen were to the buccal (20%) and distal (14%). Conclusion In this sample of teeth without apical resorption the distance between the major foramen and the anatomical root apex was always <1 mm. Deviation of the major foramen from the anatomic apex varied widely amongst tooth groups. Keywords: anatomical root apex, apical foramen, dental anatomy, major foramen.
Received 8 February 2008; accepted 9 November 2008
Introduction
Knowledge of apical anatomy and accurate radiographical interpretation during root canal treatment are essential to avoid damage to the periodontal ligament (Dummer et al. 1984, Olson et al. 1991, -Subat et al. 1992). Two measurements are Blaskovic considered important for working length determination: distance from the apical foramen to the apical constriction; and distance from the apex to the apical
Martos, Department of Semiology and Correspondence: Josue Clinics, Pelotas Dental School, UFPel, R. Gonc alves Chaves, 457, 96.015-560, Pelotas, RS, Brazil (Tel.: 55 53 3222 4162; e-mail: josue.sul@terra.com.br).
foramen (Olson et al. 1991). An accurate clinical determination of the root canal terminus and its distance from the anatomical root apex is almost impossible (Gutierrez & Aguayo 1995). Apical foramina can be asymmetrical under physiological and pathological conditions, for example as a consequence of tooth adaptation to functional activity (Kuttler 1955, Mizutani et al. 1992, Mors et al. 1994). Constant remodelling of the root apex by external root resorption and cementum apposition appear to be the most common causes of deviation of -Subat et al. 1992). the major foramen (Blaskovic The distance between the major foramen and anatomical root apex and the frequency of deviation of the major foramen have been studied in extracted teeth by
329
Knowledge of the root apex is essential in endodontic therapy Martos et al.
stereomicroscopy (Kuttler 1955, Green 1960, Chapman 1969, Burch & Hulen 1972, Kerekes & Tronstad -Subat et al. 1977, Dummer et al. 1984, Blaskovic et al. 1997, 1992, Mizutani et al. 1992, Brau Aguade Marroqu n et al. 2004), dye injection into root canal et al. 1998), scanning (Kasahara et al. 1990, Martic electron microscopy (Mors et al. 1994, Gutierrez & Aguayo 1995) and radiography (Palmer et al. 1971, Tamse et al. 1988, Basrani et al. 1997). Mean distances reported ranged from 0.2 to 3.8 mm and the frequency of major foramen deviation from the anatomic apex has been reported to be 34% to 92%. Von der Lehr & Marsh (1973) reported that the point where the endodontic le exits the foramen can be visualized on X-ray when it opens mesially or distally. However, when the foramen opens buccally or lingually, the root structure is often superimposed, masking its radiographic visualization. Some authors reported that the most frequent location of the major foramen was on the buccal surface in anterior teeth (Chapman 1969, Burch & Hulen 1972, Kasahara et al. -Subat et al. 1992) and on the distal 1990, Blaskovic surface in posterior teeth (Burch & Hulen 1972, Tamse -Subat et al. 1992). et al. 1988, Blaskovic The apical constriction is considered on the apical terminus of the root canal preparation. Nevertheless, further research is required to identify situations in which this is not the case (Dummer et al. 1984), e.g. in the presence of apical pathosis and root resorption (Simon 1994, Leonardo et al. 2007), when the apical foramen may be a more useful landmark. There have been relatively few reports to date based on large samples of teeth. The aim of this study was to determine the distance from the apical foramen to the anatomical apex and the location of the foramen in relation to the apex.

Nine hundred twenty-six nonrestored permanent maxillary and mandibular teeth (central and lateral
incisors, canines, pre-molars and molars) with completely formed apices were used. The teeth were extracted during treatment of adult patients at the School of Dentistry, Federal University of Pelotas, Brazil. A total of 1331 roots were studied. Their distribution amongst tooth groups is shown in Table 1. The study was approved by the Ethics Committee of the University of Pelotas. Before storage, calculus was removed from teeth by an ultrasonic scaler (Sonic Borden 2000N, Kavo Equipaments, Joinville, Brazil), using a 2% sodium hypochlorite solution to remove periodontal tissue remnants. Teeth were then kept in buffered formalin solution until analysis. Teeth were cut transversely at the cementumenamel junction using an Accutom-50 diamond cutter (Accutom Hard Tissue Microtome, Struers, Denmark) under copious water cooling. The apical area was stained with blue ink applied with a cotton swab to facilitate identication of the major foramen, i.e. the largest-diameter opening at and around the root apex (Fig. 1). Individually labelled bottles were used to keep roots separated, and each tooth was given a code number. Root sections were mounted on a microscope slide for measurements parallel to the long axis of the teeth. The apices were examined with a Leitz optical stereomicroscope (Ernst Leitz GmbH, Wetzlar, Germany) at 40 (10) magnication. Numerical values were obtained from the micrometric scale on the ocular stage of the stereomicroscope. All measurements were made to an accuracy of 0.01 mm and evaluated by two examiners together under a direct light source. The criterium to characterize the major foramen as opposed to other minor foramina in this investigation was the opening of the largest diameter found at the apical level, previously observed under the stereomicroscope by the examiners. In cases of disagreement a size 6 Kle (DentsplyMaillefer, Ballaigues, Switzerland) especially trimmed for this purpose, was inserted into the canal until it emerged from the major apical foramen.
Table 1 Mean distance (mm) and standard deviation of the major apical foramen and the anatomical root apex
Maxillary Tooth Incisors Canines pre-molars Molars n 78 33 104 107 x SD 0.37 0.48 0.67 0.95 0.27 0.39 0.40 0.79 Range 01.20 01.70 01.20 0.101.70 Mandibular n 183 87 164 170 x SD 0.32 0.42 0.87 0.80 0.29 0.32 0.78 0.54 Range 01.50 01.90 02.20 03.10
Mean apical foramen distance for all tooth types = 0.69 mm.
330
Martos et al. Knowledge of the root apex is essential in endodontic therapy
tively evaluated by SEM using a digital scanning microscope (Zeiss DSM 940A, Oberkochen, Germany). Using spss 8.0 software (SPSS Incorporated, Chicago, IL, USA) the mean values and standard deviations of the distances and frequencies of deviation of the major foramen with respect to the anatomical apex were calculated for each group of specimens.
Results
The mean distances from the major apical foramen to the anatomical apex are summarized in Table 1. The mean distance from the major foramen to the anatomical root apex was 0.69 mm; it was greater in posterior (0.82 mm) versus anterior (0.39 mm) teeth. The greatest mean distance was observed in the maxillar molar group (0.95 mm), followed by the mandibular pre-molar (0.87 mm) and mandibular molar (0.80 mm) groups. The major foramen was in a central location on the root apex in 40% of specimens and deviated from the anatomical apex in 61%. The frequency of deviation was higher in posterior (43%) versus anterior (17%) teeth, and higher in mandibular (35%) versus maxillary (25%) teeth. The most frequent locations were buccal (20%) and distal (14%) surfaces, followed by lingual (13%) and mesial (13%) (Table 2).
Figure 1 Root apex with a stained major foramen.
The standard le was used merely as a probe. Therefore, the major foramen was identied in two ways. It was the opening with the largest diameter found in the root apex conrmed by the visualization of the endodontic le tip. After identication of the major foramen, the instrument was removed and the root fragment was carefully mounted on a microscope slide for evaluation. By this means, the distance from the apex to the most apical point of the major foramen and the position of the major foramen were determined (Fig. 2). The position of the major foramen was classied as central, buccal, lingual, mesial, or distal. Additionally, the surface morphology of each two randomly selected specimens for group was qualita-
Discussion
Anatomical knowledge of the root apex is essential for accurate determination of root canal working length because it contains the apical foramen, which is often the reference point for root canal treatment. In this study, 1331 roots from 926 adult teeth were examined. The measurements were performed by two different examiners; with respect to the tooth groups, one was not blinded whilst the other was. The results obtained by each examiner were compared and when differences were noted between them, the average of values found was calculated. Data were collected on the distance from a perpendicular point of the anatomic apex to the most apical point of the major foramen (Gutierrez & Aguayo 1995). The most apical point of the major foramen was used for the measurement because of the difculty in determining its midpoint. Globally, the mean distance from the apical point of major apical foramen to the anatomic apex in these roots was 0.69 mm. In agreement with previous reports (Kuttler 1955, Green 1956, 1960, Burch & Hulen 1972, Dummer et al.
Figure 2 Schematic drawing of the measurements performed in the present investigation. The black lines under the rule represent the distance from the anatomic apex to the most apical point of the major foramen.
331
Table 2 Position of major foramen in root apex: n(%)

Maxillary(%) Tooth Incisors Canines pre-molars Molars central 23 9 86 103 (29.5) (27.3) (49.1) (38) buccal 18 12 17 52 (23.1) (36.4) (9.7) (19.2) lingual 1 6 27 57 (1.3) (18.2) (15.4) (21) mesial 17 5 24 30 (21.8) (15.2) (13.7) (11.1) distal 19 1 21 29 (24.4) (3) (12) (10.7) Mandibular(%) central 84 33 74 114 (45.9) (37.9) (45.1) (33.4) buccal 54 21 20 69 (29.5) (24.1) (12.2) (20.2) lingual 11 13 15 40 (6) (14.9) (9.1) (11.7) mesial 20 15 33 41 (10.9) (17.2 (20.1) (12) distal 14 5 22 76 (7.7) (5.7 (13.4) 22.3)
1984, Tamse et al. 1988, Kasahara et al. 1990, Blask -Subat et al. 1992, Mizutani et al. 1992, Mors ovic et al. 1994), the distance never exceeded 1 mm in any root. When resorption or eroded apical areas were observed, the specimen was excluded from the study because this would impair the accuracy during measurement. The areas of apical root cementum resorption in teeth with periapical lesion are generally extensive and deep and involve the entire surface of the root apex (Malueg et al. 1996, Leonardo et al. 2007). This resorption could occur both in a periforaminal and foraminal locations to different degrees (Vier & Figueiredo 2002). Radiographical and morphological studies of different teeth groups have described mean distances between the apical foramen and the most apical end of the root in the range of 0.203.80 mm (Kuttler 1955, Green 1956, 1960, Burch & Hulen 1972, Dummer et al. 1984, Tamse et al. 1988, Kasahara -Subat et al. 1992, Mizutani et al. et al. 1990, Blaskovic 1992, Mors et al. 1994, Gutierrez & Aguayo 1995, et al. 1998). Variations amongst ndings may Martic be explained by differences in examination methods, number of teeth evaluated, racial differences and origin of samples. In the maxillary and mandibular anterior teeth, the mean distance (0.33 mm) was similar to reports by Green (1960) and Dummer et al. (1984). It differed from the ndings of 0.54 mm by Burch & Hulen (1972) -Subat et al. (1992), and 0.65 mm by Blaskovic although these authors used the most cervical point of the anatomical foramen as measurement reference and studied small numbers of roots. In maxillary and mandibular pre-molar groups, the mean distance was -Subat et al. (1992) described val0.76 mm. Blaskovic ues of 1 mm in lingual and 1.3 mm in buccal roots of pre-molars, but in a sample of only 40 roots (from 20 teeth) compared with the 175 maxillary and 164 mandibular pre-molars in the present study. The mean distance was 0.95 mm in maxillary and 0.80 mm in mandibular molars. The more pronounced variations
in posterior teeth may be explained by the higher chewing force applied, which would produce apical remodelling by resorption and cementum apposition (Palmer et al. 1971, Burch & Hulen 1972). Racial et al. 1998) and pathological conditions aspects (Martic such as hypercementosis (Malueg et al. 1996, Vier & Figueiredo 2002, Leonardo et al. 2007) may also be responsible. In the present study, 60.5% of specimens showed deviation of the main foramen. Green (1960) found that the major foramen did not open directly to the apex in 69% of anterior teeth. Higher frequencies of -Subat 76% and even 92% were reported by Blaskovic et al. (1992) and Burch & Hulen (1972), respectively. Although there is a close relationship between the apical foramen and the root apex, they frequently do not coincide. Deviations ranging from 34 to 92% have been reported (Kuttler 1955, Chapman 1969, Burch & -Subat Hulen 1972, Kasahara et al. 1990, Blaskovic et al. 1992, Mizutani et al. 1992, Basrani et al. 1997, et al. 1997, Martic et al. 1998). Brau Aguade Gutierrez & Aguayo (1995) reported that foramen openings never coincided with the long axis of the root, dening apical deviation as the extension of a small portion of the foramen from the centre of the long axis. In the present study, because of the different shapes and peripheral contours of the apical foramen, deviation was only dened when a large (>50%) portion of the major foramen extended to the centre of the long axis. Buccal deviation of the major foramen was observed in 25% of maxillary and 29% of mandibular incisors, similar to ndings by other authors (Chapman 1969, Kasahara et al. 1990) but different from reports by -Subat et al. (1992) and Burch & Hulen Blaskovic (1972) of a predominantly lingual deviation in maxillary incisors. In canine groups, the most frequent deviation was buccal (36%), as observed by others (Chapman 1969, -Subat et al. 1992). Burch & Hulen (1972) Blaskovic reported a predominance of buccal deviation in mandibular (41%) and distal deviation in maxillary (31%) canines.
332
Martos et al. Knowledge of the root apex is essential in endodontic therapy
A buccal location of the foramen has the potential to cause an incorrect clinical measurement of the canal. Radiographically, an apical foramen located buccally or lingually is superimposed over the root structure, making it difcult to view the exit point of the instrument (Von der Lehr & Marsh 1973, Olson et al. 1991). The most frequent deviation of the major foramen was lingually in maxillary pre-molars (15.4%) and mesially in mandibular pre-molars (20.1%). In contrast, Burch & Hulen (1972) reported a predominance of distal deviation in both maxillary and mandibular -Subat et al. (1992) found a pre-molars. Blaskovic predominance of distal deviation in single-rooted premolars (60%) but of mesial in bifurcated pre-molars (45%), whilst a lingual location was most frequent in mandibular pre-molars (33%). In the maxillary molars the most frequent deviations were to the buccal and lingual positions, as found by -Subat et al. 1992, whilst Burch & Hulen Blaskovic (1972) described a predominance of distal deviation in buccomesial and buccodistal maxillary molar roots. In the mandibular root, the most frequent deviation was distal, as also reported by Burch & Hulen (1972) and -Subat et al. (1992) Tamse et al. (1988). Blaskovic found a higher frequency of distal deviation in distal roots but lingual deviation in mesial roots. A typical pattern of the apical foramen shape was not found. However, in the present study the greatest prevalence was of round and ovoid shapes (Figs 3 and 4). In most cases, the surface of the apical area was irregular or rough and in some others the apex appeared to have been eroded. All SEM observations enabled the identication of accessory foraminas
Figure 4 Typical ovoid shape of a foramen from a rst maxillary molar (original magnication 74).
Figure 5 Appearance of accessory foramina around the apex of a maxillary pre-molar. (original magnication 31).
(Fig. 5). Green (1956) observed in 400 anterior teeth, three types of apex congurations: infundibular, tapered and deected. The differences in this case could have been due to nomenclature as the apex shapes were similar to those of the present study.
Conclusion
The mean distance between the major apical foramen and the anatomical root apex was 0.69 mm. The frequency of deviation of the major foramen was 60% and varied amongst tooth groups.
Acknowledgements
Figure 3 SEM micrograph of the deviation of the major apical foramen from the root apex (original magnication 21).
This investigation was supported by the Coimbra Group Scholarship Program for Researchers from Latin
333
America Universities between the University of Granada (Spain) and Federal University of Pelotas (Brazil).
References
n del foramen Basrani B, Revah S, Robinson C (1997) Ubicacio n Odontolo gica Argentina 85, apical. Revista de la Asociacio 2302. -Subat V, Maricic B, Sutalo J (1992) Asymmetry of Blaskovic the root canal foramen. International Endodontic Journal 25, 15864. E, Roig Cayo n M, Canalda Sahli C (1997) Brau Aguade pico de la morfolog Estudio estereomicrosco a apical. Revista Operatoria Dental y Endodoncia 1, 12. Burch JG, Hulen S (1972) The relationship of the apical foramen to the anatomic apex of the tooth root. Oral Surgery, Oral Medicine and Oral Pathology 34, 2628. Chapman CE (1969) A microscopic study of the apical region of human anterior teeth. Journal of the British Endodontic Society 3, 528. Dummer PMH, McGinn JH, Rees DG (1984) The position and topography of the apical canal constriction and apical foramen. International Endodontic Journal 17, 1928. Green D (1956) A stereomicroscopic study of the root apices of 400 maxillary and mandibular anterior teeth. Oral Surgery, Oral Medicine and Oral Pathology 9, 1224 32. Green D (1960) Stereomicroscopic study of 700 root apices of maxillary and mandibular posterior teeth. Oral Surgery, Oral Medicine and Oral Pathology 13, 72833. Gutierrez JH, Aguayo P (1995) Apical foraminal openings in human teeth. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 79, 76977. Kasahara E, Yasuda E, Yamamoto A, Anzai M (1990) Root canal system of the maxillary central incisor. Journal of Endodontics 16, 15861. Kerekes K, Tronstad L (1977) Morphometric observations on root canals of human anterior teeth. Journal of Endodontics 3, 249. Kuttler Y (1955) Microscopic investigation of root apexes. Journal of the American Dental Association 50, 54452.
Leonardo MR, Rossi MA, Bonifacio KC, da Silva LA, Assed S (2007) Scanning electron microscopy of the apical structure of human teeth. Ultrastructural Pathology 31, 3215. Malueg LA, Wilcox LR, Johnson W (1996) Examination of external apical root resorption with scanning electron microscopy. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics 82, 8993. Marroqu n BB, El-Sayed MAA, Zo nnchen BW (2004) Morphology of the physiological foramen: I. Maxillary and mandibular molars. Journal of Endodontics 30, 3218. D, Prpic -Mehic G, Simeon P, Pevalek J (1998) Martic ic Morphometrical analysis of main and accessory canals in apical root portion of frontal teeth. Collegium Antropologicum 22, 1539. Mizutani T, Ohno N, Nakamura H (1992) Anatomic study of the root apex in the maxillary anterior teeth. Journal of Endodontics 18, 3447. Mors A, Sylaras SN, Georgopoulou M, Kernani M, Proutzos F (1994) Study of the apices of human permanent teeth with the use of scanning electron microscope. Oral Surgery, Oral Medicine and Oral Pathology 77, 1726. Olson AK, Goerig AC, Cavataio RE, Luciano J (1991) The ability of the radiograph to determine the location of the apical foramen. International Endodontic Journal 24, 2835. Palmer MJ, Weine FS, Healey HJ (1971) Position of the apical foramen in relation to endodontic therapy. Journal Canadian Dental Association 8, 3058. Simon JHS (1994) The apex: how critical is it? General Dentistry 42, 3304. Tamse A, Kaffe I, Littner MM, Moskona D, Gavish A (1988) Morphological and radiographic study of the apical foramen in distal roots of mandibular molars. Part II. The distance between the foramen and the root end. International Endodontic Journal 21, 2117. Vier FV, Figueiredo JAP (2002) Prevalence of different periapical lesions associated with human teeth and their correlation with the presence and extension of apical external root resorption. International Endodontic Journal 35, 7109. Von der Lehr WN, Marsh RA (1973) A radiographic study of the point of endodontic egress. Oral Surgery, Oral Medicine and Oral Pathology 35, 1059.
334
doi:10.1111/j.1365-2591.2008.01514.x
Effects of ethylenediaminetetraacetic, etidronic and peracetic acid irrigation on human root dentine and the smear layer
S. Lottanti1, H. Gautschi2, B. Sener1 & M. Zehnder1

1 Department of Preventive Dentistry, Periodontology and Cariology, 2Center for Electron Microscopy and Image Analysis, University of Zu rich, Zu rich, Switzerland
Abstract
Lottanti S, Gautschi H, Sener B, Zehnder M. Effects of
ethylenediaminetetraacetic, etidronic and peracetic acid irrigation on human root dentine and the smear layer. International Endodontic Journal, 42, 335343, 2009.
Aim To evaluate the effects of ethylenediaminetetraacetic (EDTA), etidronic (EA) and peracetic acid (PA) when used in conjunction with sodium hypochlorite (NaOCl) as root canal irrigants on calcium eluted from canals, smear layer, and root dentine demineralization after instrumentation/irrigation. Methodology Single-rooted human premolars were irrigated as follows (n = 12 per group): (1) 1% NaOCl during instrumentation, deionized water after instrumentation, (2) 1% NaOCl during, 17% EDTA after instrumentation, (3) a 1 : 1-mixture of 2% NaOCl and 18% EA during and after instrumentation, and (4) 1% NaOCl during, 2.25% PA after instrumentation. Irrigant volumes and contact times were 10 mL/ 15 min during and 5 mL/3 min after instrumentation. The evaluated outcomes were eluted calcium by atomic absorption spectroscopy, smear-covered
areas by scanning electron microscopy in secondary electron mode and apparent canal wall decalcications on root transsections in backscatter mode. For the smear layer analysis, sclerotic dentine was taken into consideration. Results were compared using appropriate parametric and nonparametric tests, alpha = 0.05. Results The statistical comparison of the protocols regarding calcium elution revealed that protocol (1) yielded less calcium than (3), which yielded less than protocols (2) and (4). Most of the instrumented canal walls treated with one of the decalcifying agents were free of smear layer. Protocols (1) and (3) caused no decalcication of root dentine, whilst (2) and (4) showed substance typical demineralization patterns. Conclusions The decalcifying agents under investigation were all able to remove or prevent a smear layer. However, they eroded the dentine wall differently. Keywords: EDTA, etidronic acid, peracetic acid, smear layer.
Received 17 September 2008; accepted 9 November 2008
Introduction
Root canal irrigation is important for proper debridement of infected root canals. Ideally, all microorganisms, necrotic tissue remnants, and the smear layer
Correspondence: Dr Matthias Zehnder, PD Dr med dent PhD, Department of Preventive Dentistry, Periodontology and Cariology, University of Zu rich Center for Dental Medicine, Plattenstrasse 11, Zu rich CH 8032, Switzerland (Tel.: +41 44 634 3284; fax: +41 44 634 4308; e-mail: matthias.zehnder @zzmk.uzh.ch).
that is created during mechanical canal instrumentation should be removed. The last issue, however, is not based on evidence. Current clinical concepts rely on sodium hypochlorite for necrotic soft tissue management and infection control (Naenni et al. 2004, Vianna et al. 2006). When canals are rinsed with hypochlorite during preparation, instrumented root dentine walls are more or less free of organic remnants, but covered with inorganic shavings forming a smear layer (Lester & Boyde 1977, Koskinen et al. 1980). Depending on the materials used for root lling, it is not clear how much of this inorganic smear layer should be removed
335
Canal wall decalcication Lottanti et al.
(De-Deus et al. 2008a, Saleh et al. 2008). Specically, when strong chelators are employed to completely remove the smear layer, the decalcication of the root canal wall is a side effect that could have a negative impact on canal sealability (De-Deus et al. 2008a). Yet, decalcication of root dentine by endodontic irrigants has received little attention in the dental literature (Garcia-Godoy et al. 2005). Moreover, when an attempt is made to remove the smear layer, the use of chemicals in conjunction with sodium hypochlorite is not straightforward, as hypochlorite reacts with most moieties that can be oxidized (Zehnder 2006). Consequently, it has been recommended that a sodium hypochlorite irrigant be used during instrumentation of the root canal to prolong disinfection and tissue dissolution time, and then a chelator solution be administered to clean the canal system of inorganic debris. Finally, sodium hypochlorite or another antiseptic can be applied to optimize disinfection (Yamada et al. 1983). In theory, there are two ways to simplify this protocol: i) to use a chelator that does not interfere with sodium hypochlorite or ii) to use a chelator with a strong disinfecting capacity as nal irrigant. Etidronic acid (also known as 1-hydroxyethylidene-1, 1-bisphosphonate or HEBP) is a biocompatible chelator that can be used in combination with sodium hypochlorite without short-term loss of the desired properties of either compounds (Girard et al. 2005, Zehnder et al. 2005b). This could have the advantage that a sodium hypochlorite-etidronic acid combination could be used as a single irrigant during and after instrumentation so that a smear layer is never created. However, the chelating capacity of etidronic acid is relatively weak (De-Deus et al. 2008b), and it is not known whether its use results in root canals that are as clean as counterparts irrigated with NaOCl followed by EDTA. As for the second option, peracetic acid is a strong candidate to be used as a nal irrigant. This peroxygen is sporicidal, bactericidal, virucidal and fungicidal at low concentrations of less than 0.5%, even in the presence of protein (Lensing & Oei 1985). It decomposes to safe by-products, acetic acid and oxygen. Peracetic acid actually does not exist in pure form in aqueous solution, but occurs in equilibrium with hydrogen peroxide, acetic acid and acetylhydroperoxide. The fact that acetic acid is liberated from/present in a peracetic acid solution poses the possibility that a peracetic acid solution could be used after instrumentation to dissolve the smear layer and provide a thorough disinfection of the root canal system
pre-treated with NaOCl. Acetic acid is a weak chelator that forms water-soluble complexes with calcium (Martell & Motekaitis 1992). Interestingly, irrigants containing peracetic acid were used throughout Eastern Europe and the former eastern block to disinfect root canal systems (Ku hluck & Klammt 1980). However, with recent political changes, their potential usefulness in endodontics has somehow been forgotten. When comparing different irrigating regimens on the smear layer that is created during instrumentation, the canal walls of fractured root specimens have traditionally been inspected using scanning electron microscopy (Cameron 1983). This method, however, is prone to bias, because it largely compares the amount of open dentinal tubules between groups. Dentine is a heterogeneous structure. In addition, it undergoes changes during ageing. First in the apical and later in middle and coronal root aspects, tubules become sclerotic (Vasiliadis et al. 1983a,b). It is possible, that in the studies on root canal smear layer that have been published so far, smear layer may not have been differentiated from sclerotic dentine. It was the goal of this study to compare the following irrigating protocols on extracted human premolars: 1% NaOCl throughout instrumentation followed by water (negative control); 1% NaOCl throughout instrumentation followed by 17% EDTA (gold standard, positive control); a fresh 1 : 1 mixture of 2% NaOCl and 18% etidronic acid throughout instrumentation and as nal irrigant; or 1% NaOCl throughout instrumentation followed by 2.25% peracetic acid. The outcome variables evaluated were: calcium eluted from the canal system, smear layer on instrumented canal walls and dentine decalcication observed on root transsections. A new way to calculate smear layer taking dentine sclerosis into consideration is presented.
Materials and methods Solutions

The 1% (wt/vol) and 2% NaOCl as well as the 17% EDTA solutions were bought from a commercial source (Kantonsapotheke, Zu rich, Switzerland), as was the peracetic acid solution (Uterofertil, Kesla Pharma Wolfen GmbH, Greppin, Germany). According to the manufacturer, this solution contained 4.5% (wt/vol) peracetic acid, 3.5% acetic acid and 7.3% hydrogen peroxide. It was diluted 1 : 1 with deionized water, resulting in a 2.25% peracetic acid solution. The 18% (saturated) etidronic acid solution was prepared from
336
Lottanti et al. Canal wall decalcication
HEBP salt (Cublen K8514P, Zschimmer & Schwarz, Burgsta dt, Germany) in deionized water. Using a calibrated pH meter (827 pH Lab, Metrohm, Herisau, Switzerland), the pH values of these solutions was determined. All solutions were stored at 5 C in air-tight dark containers between experiments. On experimental days, the solutions were taken from the refrigerator and stored for 60 min at room temperature prior to being used. A fresh 1 : 1 mixture of 2% NaOCl and 18% etidronic acid was prepared immediately before the experiments, resulting in a solution that contained both 1% NaOCl and 9% etidronic acid (Zehnder et al. 2005b).
irrigation time during instrumentation was 15 min, the volume was 10 mL. After instrumentation, 5 mL of the nal irrigant was administered 1 mm from working length over 3 min. Subsequently, the canal was rinsed with 5 mL of deionized water. In the positive control group for the root canal wall demineralization analysis, the canal was irrigated during instrumentation as described before, and subsequently, 10 mL of 17% EDTA was administered over 30 min.
Atomic absorption spectroscopy

The 20 mL of total eluate per specimen were collected in individual glass vials (Duran, Schott, Mainz, Germany). Each time after instrumentation and irrigation of one specimen per group, the eluates were centrifuged (Z320, BHG Hermle, Wehringen, Germany) at 4000 g for 10 min. Subsequently, 10 mL of the supernatant was transferred to a polypropylene tube (TPP, Trasadingen, Switzerland) with a lid and stored at )20 C until further analysis. Once all the eluates had been collected, they were thawed and then analysed for their calcium content using an atomic absorption spectrophotometer (Model 2380, Perkin-Elmer, Norwalk, CT, USA) with an airacetylene ame. Measurements were performed in duplicates. Each eluate was measured against a standard series of Ca2+. Phosphate was masked with strontium chloride. Results are expressed as ppm Ca2+ in the eluate.
Tooth selection and preparation

A total of 51 single-rooted premolars from the departments collection of extracted teeth were used for this study. Teeth had been stored in 0.1% thymol solution at 5 C for no more than 1 year. Teeth were selected based on their appearance suggesting one single root canal, which was veried using digital radiography (Digora, Soredex, Helsinki, Finland). The crowns of all teeth were shortened to a full root length of 12 mm as measured from the apex using a laboratory hand piece and a diamond-coated microsaw. To prevent calcium from being eluted from outer root surfaces, these were covered with nail varnish. Three premolars were used as positive controls for the canal wall decalcication analysis (see below).
Simulated clinical procedures

A step-down procedure was performed with Gates Glidden drills (Dentsply Maillefer, Ballaigues, Switzerland). Canals were instrumented using ProFile instruments (Dentsply Maillefer) in a crown-down manner, so that nally a size 45, 0.04 taper instrument reached working length, which was set at 11 mm for all teeth. Using a random sequence generator (http:// www.random.org), teeth were allocated to four experimental (n = 12) and one positive control group (n = 3) for root canal wall decalcication (see below). Teeth were irrigated as follows: group 1: 1% NaOCl during instrumentation, deionized water after instrumentation; group 2: 1% NaOCl during, 17% EDTA after instrumentation; group 3: the 1 : 1-mixture of 2% NaOCl and 18% etidronic acid during and after instrumentation; group 4: 1% NaOCl during, 2.25% peracetic acid after instrumentation. The total
Electron microscopy
Longitudinal grooves, which did not penetrate into the canal, were placed in the facial and lingual surfaces of the roots to facilitate and guide their fracture. Roots were fractured along these grooves by dipping them in water and then in liquid nitrogen. Root sections were dehydrated using an ascending ethanol series up to 100%. Root transsections One root half was inltrated and nally embedded in ascending concentrations of an isobornyl methacrylate resin (Technovit 7200 VLC, EXAKT, Norderstedt, Germany). Resin blocks polymerized with white and blue light were cut to expose root transsections in the coronal, middle and apical root third in each specimen using a diamond-coated saw (Isomet, Bu hler, Switzerland) under water cooling. Root sections were mounted on individual stubs and raw polished using 1200 and
337
then 2400 grit silicon paper. Thereafter, specimens were polished using diamond pastes of particle sizes down to 0.5 lm. The exposed surfaces were carboncoated in an electron-beam evaporator (BAL-TEC MED 020, Baltec Union, Balzers, Liechtenstein). These specimens were examined in the scanning electron microscope (Supra 50 VP, Zeiss, Oberkochen, Germany) equipped with an annular mono-crystal scintillation type (YAG) backscatter detector in backscatter mode at an accelerating voltage of 11 kV and 9 to 12 mm working distance. Digital images were taken at magnications of 100 to 500 (overviews) followed by 2000 for the detailed views at a resolution of 1024 768 pixels. For control reasons, the specimens that were exposed to EDTA for 30 min were used to compare images obtained in backscatter mode to counterparts from the same specimen obtained by energy-dispersive x-ray (EDX) analysis sampled over 300 min (Fig. 1). Because EDX analysis takes half a working day per specimen and the dark areas in the backscattered electron micrographs exactly matched the decalcied areas, backscatter images were used to compare apparent decalcications between groups. High-resolution images taken from the whole canal wall area (2000 magnication, typically eight per specimen) were analysed by one observer. Furthermore, the length of sclerotic dentine per total canal wall length on each specimen was determined using an image analysis software (imagej, rsbweb.nih.gov/ij/). Smear layer on canal wall The opposing half of each root was prepared for the smear layer analysis by mounting it on a stub and goldsputtering (BAL-TEC SCD 030, Baltec Union). Prepared root canal surfaces (i.e. areas without calcospherites) were observed in a scanning electron microscope
(Supra 50 VP, Zeiss) equipped with an SE (secondary electron) detector, and three photomicrographs were taken at 1000 magnication from typical areas of the coronal, middle, and apical root canal thirds at an accelerating voltage of 10 kV and a working distance of 7 to 9 mm. For the analysis of the smear layer, canal wall images were compared with the pattern of sclerotic dentine from the opposing root half (i.e. the corresponding root transsection specimen). This made it possible to discern between smear-covered and sclerotic areas (Fig. 2). The imagej software was used to calculate the area that was covered with a smear layer in percent of the whole image. Mean values of the three images per specimen were used for further calculations.
Data presentation and analysis

Data pertaining to calcium eluted from root canals (ppm) were compared between groups using one-way analysis of variance (anova). Because of their normal distribution, these data are presented as means and standard deviations. Sclerotic dentine per canal wall and smear score comparisons were done using KruskalWallis analysis of variance followed by the Mann Whitney U test. These values are presented as medians and ranges. Bonferronis correction for multiple testing was applied for all individual comparisons. The alphatype error was set at 0.05.
Results
The pH values of the solutions used in this study were: 1% NaOCl: 12.0; 17% EDTA: 8.0; 18% etidronic acid: 10.5; the 1 : 1 mixture of 18% etidronic acid with 2% NaOCl; 10.5, 2.25% peracetic acid: 2.5.
Figure 1 Comparison between images from a sample treated for 30 min with ethylenediaminetetraacetic acid (EDTA) (positive demineralization control) obtained by energy-dispersive x-ray (EDX) analysis (left, calcium in yellow) and in backscatter mode (right). The decalcication reached 50100 lm into the dentine and followed the tubules.
338
Figure 2 Root transsections (left) and corresponding canal walls from the opposing root segment (right) from the coronal, middle and apical root third (top to bottom) of a typical specimen rinsed with ethylenediaminetetraacetic acid (EDTA). Note that despite the fact that there is no smear layer, the tubules appear occluded in middle and apical root sections because they are sclerotic and hence not visible. Sclerotic areas have a typical surface structure and show few tubular openings. This is in contrast to smearcovered instrumented areas that are covered with a homogenous layer (not shown).
Irrigation with the sodium hypochlorite irrigant during instrumentation followed by water as a nal rinse hardly eluted any calcium from the canal system: 1 1 ppm (mean SD). By contrast, the protocols with EDTA and peracetic acid dissolved 62 13 and
57 10 ppm respectively (P between these two groups >0.05). The irrigating protocol with the 2% NaOCl/ 18% etidronic acid mixture resulted in 32 14 ppm calcium in the eluate, which was signicantly (P < 0.05) less than with the other two decalcifying
339
agents, but signicantly more than with the NaOCl/ water treatment. The median portion of sclerotic dentine per instrumented canal wall length was 37% (range: 0%100%), 49% (range: 0%100%) and 69% (range: 0%100%) in coronal, middle and apical root thirds respectively. Coronal and middle root thirds did not differ signicantly regarding the percentage of sclerotic dentine, whilst the apical root third had signicantly more sclerotic areas compared to coronal and middle aspects (P < 0.05). All the protocols with a decalcifying agent resulted in statistically similar reduction of smear-covered dentine, whilst the specimens irrigated with NaOCl and then with water were almost completely covered with a smear layer (Table 1). In coronal root areas, EDTA removed slightly more smear layer than peracetic acid (P < 0.05), whilst in middle root and apical root areas, no differences between EDTA, etidronic and peracetic acid protocols were detected. There were no differences
Table 1 Smear-covered instrumented canal wall area in % of
in the amount of smear layer between coronal, middle and apical root aspects (KruskalWallis, P = 0.95). Apparent decalcications were not observed in specimens irrigated with sodium hypochlorite and then water. Few if any demineralizations were seen in the NaOCl/etidronic acid-treated specimens. Specimens irrigated with EDTA for 3 min as a nal rinse showed typical decalcication patterns: the tubules were enlarged at their canal wall opening, and the decalcication followed the tubular walls up to 20 lm into the dentine. EDTA decalcications showed no gradual increase in mineral, but rather 15 lm of complete decalcication into the intertubular dentine with a sharp demarcation between demineralized and unaffected dentine (Fig. 3). By contrast, peracetic acid decalcications typically showed a gradual mineral increase from the exposed canal wall into the root dentine, with fewer decalcied areas into the tubules.
Discussion
The current study showed that the smear layer could be reduced in instrumented root canals by a protocol employing either etidronic or peracetic acid to a similar extent as with the conventional EDTA treatment. Both protocols could be clinically advantageous over the use of EDTA. An etidronic acid/sodium hypochlorite mixture could be administered as the sole irrigant. Alternatively, peracetic acid could be left in the canal after instrumentation for combined disinfection and smear layer dissolution. Whilst many articles on the appearance of root dentine walls after different irrigating protocols in the scanning electron microscope have been published, surprisingly few studies have addressed what is actually happening to the underlying dentine (Garcia-Godoy
total area (median values, ranges) in coronal and middle root thirds according to the irrigating protocol (during instrumentationnal rinse)
Treatment Coronal third
a
Middle third 100 (60100) 0 (055)b 0 (035)b 0 (04)b

a
Apical third 100 (100100)a 0 (041)b 8 (020)b 0 (00)b
NaOCl Water 97 (55100) NaOCl EDTA 0 (01)b NaOCl & 0 (054)b,c EANaOCl & EA NaOCl PA 4 (029)c
NaOCl: sodium hypochlorite, EDTA: ethylenediaminetetraacetic acid, EA: etidronic acid, PA: peracetic acid. Data sets that share an identical superscript letter in a column did not differ signicantly at the 5% level from each other (KruskalWallis, MannWhitney U test, Bonferroni correction).
Figure 3 Backscatter images of typical root transsections rinsed with 1% NaOCl throughout instrumentation followed by 2.25% peracetic acid (left) and 1% NaOCl followed by 17% ethylenediaminetetraacetic acid (EDTA) (right). Note how the pattern of the demineralization in the EDTA-treated specimen differs from that in the peracetic acid group.
340
et al. 2005, Marending et al. 2007). To the best of our knowledge, this would be the rst investigation to specically assess the effects of root canal irrigation with decalcifying agents on the appearance of root dentine transsections. Furthermore, the amount and distribution of sclerotic dentine in the studied tooth material was taken into consideration. The results reported here regarding sclerotic dentine in the different root thirds are in line with published observations; tubular sclerosis is most pronounced in the apical area (Vasiliadis et al. 1983a). Using a split-root analysis as was done here, the amount of smear layer on root canal walls could be assessed with more certainty compared to conventional investigations that relied solely on the amount of open tubules in a given canal wall area. This was possible because in single-rooted teeth, tubular sclerosis around the main canal is symmetrical (Vasiliadis et al. 1983a). Based on the current data, it may be surmised that previous reports on smear layer in the apical root area were prone to a systematic error. In other words, it may not be the case that there is more smear layer in apical areas than in coronal or mid-root counterparts, although this has been reported in most studies. The backscatter analysis that was performed to screen for apparent decalcications in root dentine walls can be regarded as sound. Many studies have used a similar methodology to investigate dentinal caries (Angker et al. 2004). A backscatter detector identies high-energy electrons whilst the EDX detector can spot specic elements. As indicated in the Material and methods section, however, EDX analysis takes too long to screen such a large study material. Furthermore, the backscatter images show more contrast than the corresponding EDX micrographs (see Fig. 1) and apparent decalcications are easier to detect. The current study was performed in a laboratory environment, and direct clinical conclusions should therefore not be drawn from this report. The main concern regarding the use of any new compound clinically is the question of its potential side effects. Etidronic acid is biocompatible and is used as an additive in various personal care products such as soaps (Licata 1993). It is also used in swimming pools because of its compatibility with hypochlorite to prevent stains from metal ions. However, etidronic acid is a bisphosphonate, and the systemic administration of bisphosphonates has been linked to osteonecroses of the jaws (Migliorati et al. 2005). On the other hand, no such reports have appeared related to etidronate, only to other bisphosphonates with a much
higher capacity to inhibit osteoclast function (Krueger et al. 2007). Peracetic acid is relatively cytotoxic (Sagripanti & Bonifacino 2000). Nevertheless, it is considered to be an alternative to sodium hypochlorite for drinking water disinfection (Marabini et al. 2006). Moreover, despite its ubiquitous use in Eastern Europe in the 1980s, no adverse effects have been reported other than the acidic smell that permeated the dental ofce. In the current study, a 2.25% peracetic acid solution was used, which is probably as caustic as a hypochlorite solution of the same concentration. In the only published clinical study on root canal disinfection with a peracetic acid irrigant, a solution containing 0.4% peracetic acid (1% Wofasteril) was used, based on tests that were previously made on oral mucosa (Ku hluck & Klammt 1980). Consequently, it may be necessary to investigate the effect of lower concentrations of peracetic acid on the outcome variables studied here. In this study, a protocol with 3 min of peracetic acid treatment was used, based on the observation that this should be enough for the EDTA to dissolve the smear layer (Saito et al. 2008). However, low concentrations of peracetic acid could be left in the root canal system for extended times. It has been suggested to leave an iodine solution in the canal system for 10 min as an intra-appointment dressing (Kvist et al. 2004). Conceivably, this could be done with peracetic acid. Whilst the current data look promising, more research is necessary to further investigate the possible usefulness of peracetic acid preparations in endodontics. Studies regarding the antimicrobial effectiveness under the specic conditions of the root canal system should be performed to further evaluate irrigation protocols with either etidronic or peracetic acid. Calcium ppm contained in EDTA and etidronic acid eluates as determined by AAS were comparable to published material (Zehnder et al. 2005a,b). Because of the acidity of peracetic acid, the calcium stays in solution and does not reprecipitate (Gubler et al. 2008). This could explain that despite the weak chelating power of this agent, similar amounts of calcium were eluted from the root canal as compared to EDTA, which is a much stronger chelator but can only be in solution at a slightly alkaline pH. Apparently, the decalcication kinetics differ between EDTA and acetic acid. The results regarding the mineral proles after EDTA and peracetic acid treatment presented here are in line with a published report on chemically induced articial caries in dentine (Kawasaki et al. 2000). EDTA left no mineral on the outer half of the lesion, whilst acetic acid produced a more gradual demineralization
341
pattern. The consequences of these differences on the sealability of root canals with the different types of root lling materials that are on the market remain to be investigated. As has already been shown, the minimal demineralization promoted by the sodium hypochlorite/etidronic acid treatment may result in a better bond of resin-based materials to the canal wall (De-Deus et al. 2008a).
Conclusions
Irrigation protocols employing 1% NaOCl and then 17% EDTA, 1% NaOCl and then 2.25% peracetic acid, or a combined solution containing 1% NaOCl and 9% etidronic acid left similar amounts of smear layer on instrumented root canal walls. Three minutes of 17% EDTA caused complete, 3 min of 2.25% peracetic acid a gradual demineralization of the rst few micrometers of root dentine that were exposed to the irrigant, whilst a combined NaOCl/ etidronic acid irrigant during canal instrumentation and as a nal rinse did not decalcify the canal walls. Irrigating protocols employing either etidronic or peracetic acid showed potential to replace the conventional treatment with EDTA.
References
Angker L, Nockolds C, Swain MV, Kilpatrick N (2004) Quantitative analysis of the mineral content of sound and carious primary dentine using BSE imaging. Archives of Oral Biology 49, 99107. Cameron JA (1983) The use of ultrasonics in the removal of the smear layer: a scanning electron microscope study. Journal of Endodontics 9, 289292. De-Deus G, Namen F, Galan JJ, Zehnder M (2008a) Soft chelating irrigation protocol optimizes bonding quality of Resilon/Epiphany root llings. Journal of Endodontics 34, 703705. De-Deus G, Zehnder M, Reis C et al. (2008b) Longitudinal cosite optical microscopy study on the chelating ability of etidronate and EDTA using a comparative single-tooth model. Journal of Endodontics 34, 7175. Garcia-Godoy F, Loushine RJ, Itthagarun A et al. (2005) Application of biologically-oriented dentin bonding principles to the use of endodontic irrigants. American Journal of Dentistry 18, 281290. F, Badertscher M, Sener B, Zehnder M (2005) Girard S, Paque Assessment of a gel-type chelating preparation containing 1-hydroxyethylidene-1, 1-bisphosphonate. International Endodontic Journal 38, 810816. Gubler M, Brunner TJ, Zehnder M, Waltimo T, Sener B, Stark WJ (2008) Do bioactive glasses convey a disinfecting
mechanism beyond a mere increase in pH? International Endodontic Journal 41, 670678. Kawasaki K, Ruben J, Tsuda H, Huysmans MC, Takagi O (2000) Relationship between mineral distributions in dentine lesions and subsequent remineralization in vitro. Caries Research 34, 395403. Koskinen KP, Meurman JH, Stenvall H (1980) Appearance of chemically treated root canal walls in the scanning electron microscope. Scandinavian Journal of Dental Research 88, 505 512. Krueger CD, West PM, Sargent M, Lodolce AE, Pickard AS (2007) Bisphosphonate-induced osteonecrosis of the jaw. Annals of Pharmacotherapy 41, 276284. Ku hluck I, Klammt J (1980) Untersuchung zur Eignung von Peressigsa ure zur Wurzelkanaldesinfektion. Stomatologie der DDR 30, 558563. n G, Reit C (2004) Microbiological Kvist T, Molander A, Dahle evaluation of one- and two-visit endodontic treatment of teeth with apical periodontitis: a randomized, clinical trial. Journal of Endodontics 30, 572576. Lensing HH, Oei HL (1985) Investigations on the sporicidal and fungicidal activity of disinfectants. Zentralblatt fu r Bakteriologie, Mikrobiologie und Hygiene. 1. Abt. Originale B, Hygiene 181, 487495. Lester KS, Boyde A (1977) Scanning electron microscopy of instrumented, irrigated and lled root canals. British Dental Journal 143, 359367. Licata AA (1993) From bathtub ring to osteoporosis: a clinical review of the bisphosphonates. Cleveland Clinic Journal of Medicine 60, 284290. Marabini L, Frigerio S, Chiesara E, Radice S (2006) Toxicity evaluation of surface water treated with different disinfectants in HepG2 cells. Water Research 40, 267272. Marending M, Luder HU, Brunner TJ, Knecht S, Stark WJ, Zehnder M (2007) Effect of sodium hypochlorite on human root dentine- mechanical, chemical and structural evaluation. International Endodontic Journal 40, 786793. Martell AE, Motekaitis RJ (1992) Determination and Use of Stability Constants. New York: John Wiley & Sons. Migliorati CA, Schubert MM, Peterson DE, Seneda LM (2005) Bisphosphonate-associated osteonecrosis of mandibular and maxillary bone: an emerging oral complication of supportive cancer therapy. Cancer 104, 8393. Naenni N, Thoma K, Zehnder M (2004) Soft tissue dissolution capacity of currently used and potential endodontic irrigants. Journal of Endodontics 30, 785787. Sagripanti JL, Bonifacino A (2000) Cytotoxicity of liquid disinfectants. Surgical Infections 1, 314. Saito K, Webb TD, Imamura GM, Goodell GG (2008) Effect of shortened irrigation times with 17% ethylene diamine tetra-acetic acid on smear layer removal after rotary canal instrumentation. Journal of Endodontics 34, 1011 1014. Saleh IM, Ruyter IE, Haapasalo M, rstavik D (2008) Bacterial penetration along different root canal lling materials in the
342
presence or absence of smear layer. International Endodontic Journal 41, 3240. Vasiliadis L, Darling AI, Levers BG (1983a) The amount and distribution of sclerotic human root dentine. Archives of Oral Biology 28, 645649. Vasiliadis L, Darling AI, Levers BG (1983b) The histology of sclerotic human root dentine. Archives of Oral Biology 28, 693700. Vianna ME, Horz HP, Gomes BP, Conrads G (2006) In vivo evaluation of microbial reduction after chemomechanical preparation of human root canals containing necrotic pulp tissue. International Endodontic Journal 39, 484492.
Yamada RS, Armas A, Goldman M, Lin PS (1983) A scanning electron microscopic comparison of a high volume nal ush with several irrigating solutions: Part 3. Journal of Endodontics 9, 137142. Zehnder M (2006) Root canal irrigants. Journal of Endodontics 32, 389398. Zehnder M, Schicht O, Sener B, Schmidlin P (2005a) Reducing surface tension in endodontic chelator solutions has no effect on their ability to remove calcium from instrumented root canals. Journal of Endodontics 31, 590592. Zehnder M, Schmidlin P, Sener B, Waltimo T (2005b) Chelation in root canal therapy reconsidered. Journal of Endodontics 31, 817820.
343
doi:10.1111/j.1365-2591.2008.01518.x
A comparison of the efcacy of conventional and new retreatment instruments to remove gutta-percha in curved root canals: an ex vivo study
nal, B. U reyen Kaya, A. G. Tac G. C elik U & A. D. Kec eci

Department of Endodontics, Su leyman Demirel University, Isparta, Turkey
Abstract
nal G, U reyen Kaya B, Tac C elik U AG, Kec eci AD.
A comparison of the efcacy of conventional and new retreatment instruments to remove gutta-percha in curved root canals: an ex vivo study. International Endodontic Journal, 42, 344350, 2009.
Aim To compare the efcacy of conventional and new retreatment instruments when removing gutta-percha root llings in curved root canals. Methodology A total of 56 curved molar roots were instrumented with ProFile instruments and lled using system B and Obtura II. The root llings were removed with manual K-les and Hedstro m les (Dentsply Maillefer), ProFile (Dentsply Maillefer), R-Endo (MicroMega) or ProTaper Universal retreatment les (Dentsply Maillefer). Eucalyptol was used as a solvent with all techniques. Bucco-lingual and proximal radiographs of the roots were exposed and the percentage area of the remaining material was calculated by dividing the area of remaining lling material by the area of canal wall. Data were statistically analysed with KruskalWallis and MannWhitney U tests (P = 0.05).
Results None of the techniques completely removed the root lling materials. No signicant differences were found amongst the coronal, middle and apical thirds in both radiographic projections (P > 0.05). In the buccolingual direction, the remaining lling material was signicantly less following manual instrumentation than R-Endo and ProTaper instrumentation (P < 0.05). In the proximal view, it was signicantly less following manual and ProFile instrumentation than R-Endo (P < 0.05). Complete removal of lling material occurred only in three specimens (with manual instruments). Manual instruments were signicantly faster than R-Endo and ProFile (P < 0.05). More procedural errors (ve fractured instruments and two perforation) were noted when using ProTaper (P < 0.05). Conclusions In this laboratory study in curved molar roots, ProTaper Retreatment and R-Endo instruments were less effective in removing lling material from canal walls than manual and ProFile instruments. Keywords: canal wall cleanliness, gutta-percha removal, ProFile, ProTaper, R-Endo, retreatment.
Received 6 March 2008; accepted 17 November 2008
Introduction
The techniques used to remove gutta-percha from root canals include manual endodontic hand instruments (Imura et al. 1996, Schirrmeister et al. 2006a), ultrasonics (Ladley et al. 1991), lasers (Farge et al. 1998),
nal, Su Correspondence: Yrd. Doc . Dr Gu elik U lC leyman Demirel University, Department of Endodontics, 32260 Cunur-Isparta/ Turkey (Tel.: +90 246 2113286; fax: +90 246 2370607; e-mail: celikunalgul@yahoo.com, bureyen@hotmail.com).
heat carrying instruments (Wolcott et al. 1999), as well as NiTi rotary instruments (Imura et al. 2000, Sae-Lim et al. 2000, Hu lsmann & Bluhm 2004, Schirrmeister et al. 2006b,c, Gergi & Sabbagh 2007, Saad et al. 2007). In curved root canals, the removal of lling materials, and further cleaning and shaping are more difcult than in straight canals and more likely to cause instrument distortion or breakage. Whilst there are many studies on straight root canals (Imura et al. 1996, Imura et al. 2000, Betti & Bramante 2001,
344
nal et al. Retreatment techniques in curved root canals C elik U
Barrieshi-Nusair 2002, Masiero & Barletta 2005, Kosti et al. 2006), studies on the efciency of removing root lling in curved root canals are rare (Ferreira et al. 2001, Schirrmeister et al. 2006c, Barletta et al. 2007, Gergi & Sabbagh 2007). Recently, new rotary instruments have been designed for the removal of lling materials in curved root canals: the ProTaper Universal retreatment system (Dentsply Maillefer, Ballaigues, Switzerland) and the R-Endo retreatment (MicroMega, Besanc on, France). Gergi & Sabbagh (2007) and Tas demir et al. (2008) reported that ProTaper and R-Endo rotary instruments were inadequate for the complete removal of lling material from curved root canals as were Hedstro m and M-Two instruments. Gu et al. (2008) and Somma et al. (2008) observed that ProTaper instruments left remnants of lling material from straight root canals as did Hedstro m, M-Two and K-Flex instruments. The purpose of this ex vivo study was to compare remaining root lling material after the retreatment of curved root canals using manual, ProFile rotary and two new retreatment instruments (R-Endo and ProTaper Universal retreatment les).
that the roots were always positioned in the same orientation (Fig. 1). A size 10-K le was introduced to working length and standard digital radiographs (Visualix Gendex Dental Systems, Monza, Italy) were taken in buccolingual and proximal directions at 65 kVp, using an exposure time of 0.16 s. The images were transferred to a computer and the canal curvatures were calculated according to Schneider (1971). The distance from the orice of the canal to the apical foramen was calculated as the root length (autocad 2000, San Rafael, CA, USA).
Preparation of the teeth

All canals were instrumented by the same operator using a crown-down technique with a series of ProFile Ni-Ti rotary instruments (Dentsply Maillefer, Ballaigues, Switzerland) in an electric motor (Technika, Dentsply Maillefer) rotating at 300 rpm. The preparation of root canals was completed when a ProFile 0.06 taper with a tip equivalent to size 30 at the working length. Glyde File Prep (Dentsply Maillefer) was used as a lubricant, and 3 mL of 2.6% NaOCl was used as an irrigant at each change of instrument. After the biomechanical preparation, a nal rinse with 3 mL NaOCl and 17% EDTA was used for 1 min, followed by distilled water for 1 min. Each canal was dried using size 30 paper points.
Materials and methods Specimen preparation

A total of 56 extracted maxillary human molar teeth with fully formed apices and no calcication, internal resorption, or previous root canal treatment were used. Soft tissue and calculus were removed mechanically from the surface of the roots. Root canals with apical diameters no greater than a size 15-K le and with a curvature of 2042 were selected. Only teeth in which a size 10-K le could just be seen through the apex and a size 15-K le tightly tted at the apical foramen were included. The crowns were attened to stabilize the reference point and following access cavity preparation the root canal contents were removed with a barbed broach, apical patency was controlled with size 10-K les. The canal length was established visually when the tip of the 10-K le was seen at the apex and the working length was recorded as being 1 mm shorter. The roots were then embedded into acrylic blocks (Orthoacril, Dentarium, Ispringen, Germany) leaving the root apices exposed. A lm holder (Hawe-Neos Dental, Bioggio, Switzerland) was modied to allow to exposure of standard digital radiographs. A stainless steel cube was attached to the device in order to ensure
Root lling
A ne-medium non standardized gutta-percha cone (SybronEndo, Orange, CA, USA) was trimmed to t at the working length with tug back. The cone was lightly coated with sealer (AH Plus, De-trey-Dentsply,
Figure 1 A modied lm holder to obtain standard radiographic images.
345
Konstanz, Germany) prior to the placement into the canal to the working length. Fine-medium sized system B plugger (Analytic Technology, Redmond, VA, USA) set at 200 C was introduced into the canal and the cone was seared off at the canal orice. The tip of the plugger was reactivated and condensation was terminated within 3 mm of working length. The plugger was held in position for 10 s before the system B was activated for 1 s and withdrawn from the tooth. Filling of the coronal portion of the canal was achieved with Obtura II (Obtura Spartan, Fenton, MO, USA) using 23 gauge needle tips at a temperature of 185 C and condensed with Buchanan Hand Plugger Size 1 (SybronEndo, Orange, CA, USA). The lling was assessed on digital radiographs in bucco-lingual and proximal directions. When the lling appeared to be dense and contained no voids, it was deemed adequate. Inadequately lled canals were recondensed. The coronal aspect of the root llings was covered with a composite resin (Valux Plus, 3M, ESPE, St Paul, MN, USA). The roots were stored in gauze dampened with aqueous solution containing 0.1% sodium azide (NaN3) for 2 weeks at 37 C to allow the sealer to set. The roots were randomly assigned to four experimental groups for the different procedures.
was used at working length. Eucalyptol was replenished up to twice or thrice. In total, six instruments were used in this group. R-Endo retreatment les (MicroMega, Besanc on, France) R-Endo retreatment les were used with a rotary electric motor and handpiece (Inget 06 contraangle; MicroMega, Besanc on, France) at 300 rpm. A size 25, 0.04 taper Rm hand le (K-File) was used with 1/4 turn pressure directed towards the apex to create a pathway thus allowing the centering and the alignment of the next instrument. A size 25, 0.12 taper Re NiTi rotary le was used 1 to 3 mm beyond the pulp chamber oor with circumferential ling. Again, 0.1 mL of eucalyptol was deposited into the reservoir created for 3 min. A size 25, 0.08 taper R1 NiTi rotary le was used to penetrate from the coronal third to the beginning of the middle third through repeated apically directed pushing actions. A size 25, 0.06 taper R2 NiTi rotary le was used from the middle third to the beginning of the apical third. A size 25, 0.04 taper R3 NiTi rotary le was used at the working length with circumferential ling action. Finally, the retreatment procedure was concluded with the use of a size 30, 0.04 taper Rs NiTi rotary le at the working length. In total, six instruments were used in this group. ProTaper Universal retreatment les (Dentsply Maillefer, Ballaigues, Switzerland) The D1 ProTaper le was used to remove the lling material from the cervical third of the root canal and 0.1 mL of eucalyptol was deposited for 3 min into the reservoir created. A D2 ProTaper le was used in the coronal two thirds of the root canal. The D3 ProTaper le was used with light apical pulses of pressure until the working length was reached and no further lling material could be removed. In total, three instruments were used in this group. Copious irrigation with 2 mL of 5.25% NaOCl was performed throughout the procedures at each change of instrument. Final irrigation was performed with 10 mL of 5.25% NaOCl. Criteria for the assessment of removal of the lling material were the absence of gutta-percha or sealer on the instrument used last. To provide similar conditions for all groups, 300 rpm was selected for all rotary les that were used in an electric motor (Technika, Dentsply Maillefer). Each instrument was used for a maximum of ve canals. If any deformation or fracture occurred, it was recorded, and the instrument or tooth was replaced.
Retreatment procedures
Manual instruments GatesGlidden (Dentsply Maillefer) size 3 and subsequently size 2 were used to remove coronal lling material and create a reservoir for eucalyptol. After 0.1 mL of eucalyptol was deposited for 3 min, the lling material was removed with K les (Dentsply Maillefer) and Hedstro m les (Dentsply Maillefer) sizes 3015 (in descending order) to the working length using a circumferential lling motion. Eucalyptol was replenished up to twice or thrice. Once the working length had been reached with a size 15 le, sizes 20, 25 and 30 were used at the working length. In total, 10 instruments were used in this group. ProFile ISO rotary instruments (Dentsply Maillefer) ProFile size 4 and 3 orice shapers were used to remove the coronal lling material and create a reservoir for 0.1 mL of eucalyptol. After 3 min, size 30, 0.06 taper and size 25, 0.06 taper ProFile NiTi rotary instruments were used in the middle third of the canal. Then, 0.04 tapered size 30 and 25 instruments were used in the apical third; nally a size 30, 0.06 taper instrument
346
Evaluation criteria
The roots were digitally radiographed in bucco-lingual and proximal directions. The images were transferred to a PC and autocad 2000 was used to evaluate the area of lling material remaining on the root canal walls. Remaining lling material was identied and outlined by a trained operator through a difference in radio-opacity (Fig. 2). The area of remaining lling material as well as the canal wall in each portion was measured in both directions using autocad 2000. Area fractions of root canal wall covered by remaining lling material was calculated in percentage terms by dividing the area of remaining lling material with the total area of canal wall. Each image was outlined thrice by the same operator to increase the reproducibility and to
decrease the intra-operator variability, and then the average of the obtained measurements was taken. The total removal time was noted for each root. It began with the initial gutta-percha removal and ended when canals were deemed to be clean. Extrusion of debris or lling material through the apical foramen was evaluated visually and scored as; no debris 0, only sealer 1, sealer and gutta-percha 2. Fractures and deformations of instruments and ledging and perforations were also recorded.
Root canal lengths and the root canal curvatures were analysed using one-way anova and KruskalWallis tests at a signicance level of P < 0.05 respectively. For the analysis of cleanliness of root canal walls, working time, and extruded materials, KruskalWallis and MannWhitney U tests were performed at a signicance level of P < 0.05.
Results
The mean root canal lengths of the teeth were 12.58 mm 1.47 mm. The root curvatures of the teeth ranged between 20 and 42 (mean = 32 4). There were no statistically signicant differences amongst the experimental groups in terms of the root canal length and the degree of curvature (F = 0.405, P = 0.415, v2 = 2.850, P = 0.750 respectively). The working length was obtained in all groups in all teeth except seven teeth that were excluded due to fractured instruments.
Evaluation of remaining lling material

Complete removal of lling material occurred in only three specimens retreated with manual instruments when evaluated radiographically; the differences were not signicant (P > 0.05). Considering the whole canal, there were statistically signicant differences amongst the groups in terms of the remaining lling material (P < 0.05). The remaining lling material in the bucco-lingual direction was less in the group using manual instruments than that in R-Endo and ProTaper groups (P < 0.05). In the proximal direction, the remaining lling material was less in groups using manual and ProFile instruments than that in R-Endo (P < 0.05) (Table 1). There were no signicant differences in the coronal, middle, and apical thirds in both bucco-lingual and
Figure 2 Images of gutta-percha and sealer remaining on the root canal walls and the calculation of the ratios of areas as the percentage of remaining debris using a software program (autocad 2000).
347
Table 1 The area (%) of remaining lling material imaged in bucco-lingual and proximal directions (mean SD)*
Groups Manuel instruments ProFile R-Endo ProTaper Bucco-lingual Proximal mean SD (n = 14) mean SD (n = 14) 14 15 27 24 25a 20a,b 30b 28b 14 11 23 20 25a,c 15c 27b 23a,b,c
Table 3 The area (%) of remaining lling material in each third of root canals (mean SD)*
MinimumMaximum Coronal third Middle third Apical third 067 078 0100 Mean SD 7 13a 14 21b 34 29c
*Groups identied by different letters are signicantly different (P > 0.05).
Table 4 Time(s) required for the removal of the lling material* and the number of procedural errors
Groups Manuel instruments ProFile R-Endo ProTaper Time Deformed Fractured instrument instrument Perforation
proximal directions amongst the groups (P > 0.05) (Table 2) when evaluating the remaining lling material. On the other hand, there were statistically signicant differences amongst the coronal, middle and apical thirds of the canal walls irrespective of the technique used. A greater amount of lling material remained in the apical third than in the middle and cervical (P < 0.05) (Table 3).
5.08 2.99a 8.81 2.65b 8.03 3.40b 1 6.02 3.02a
2 R3 3 D3, 2 D2 2
Amount of apically extruded material

No signicant differences were found amongst the groups in terms of the apically extruded material (v2 = 1.538, P > 0.05).
observed in two roots of ProTaper groups. Seven teeth in ProTaper and two in R-Endo groups were replaced. More procedural errors were noted in ProTaper group (P < 0.05) (Table 4).
Discussion Time required for removal

The time required for removal with manual instruments was signicantly lower than when R-Endo and ProFile instrument were used (P < 0.05). ProTaper and manual instruments were more rapid than the ProFile (P < 0.05) (Table 4). Different techniques have been used to evaluate remaining lling material (Friedman et al. 1993, Imura et al. 2000, Schirrmeister et al. 2006b,c) and radiographs have been used extensively (Ferreira et al. 2001, Masiero & Barletta 2005, Gergi & Sabbagh 2007). Observer performance can vary because root canal wall cleanliness evaluation is subjective and semi-quantitative. However, the remaining lling material is not disturbed, which might otherwise be lost by splitting the roots (Ferreira et al. 2001, Masiero & Barletta 2005, Schirrmeister et al. 2006a). As radiographs are two dimensional, they cannot distinguish sealer from gutta-percha and may be subjected to magnication and distortion. It
Procedural errors
No fracture occurred in the manual and ProFile instruments. Five instruments in the ProTaper (3 D3 and 2 D2) and two in R-Endo (R3) groups fractured. Lateral perforation at the outer side of the root was
Table 2 The area (%) of remaining lling material imaged in bucco-lingual and proximal directions in each third of root canals (mean SD)
Bucco-lingual direction Groups Manuel instruments ProFile R-Endo ProTaper Coronal 2 5 14 10 5 11 20 17 Middle 8 10 30 16 13 13 29 24 Apical 32 29 36 45 35 24 35 29 Proximal direction Coronal 2 4 10 7 7 11 15 10 Middle 6 6 24 15 11 11 26 22 Apical 34 23 37 38 34 16 31 22
348
is also known that radiographs do not reveal all remaining material (Schirrmeister et al. 2006a). In this study, roots were radiographed digitally in bucco-lingual and proximal directions. In addition, each image was outlined thrice by a trained operator who was unaware of the group assignment and had previously identied and outlined 300 images according to their differences in radio-opacity. It was not possible to remove all the traces of guttapercha and sealer from the root canal walls with any of the techniques, which has also been reported in other studies (Wilcox et al. 1987, Hu lsmann & Bluhm 2004, Tas demir et al. 2008). Complete removal of all lling material as detected radiographically was obtained only in three of the root canals, instrumented with manual instruments. Ferreira et al. (2001) and Gergi & Sabbagh (2007) reported that removal of root lling materials in curved canals using Hedstro m, 0.04 ProFile and R-Endo instruments produced similar levels of remaining material. As demonstrated in previous studies (Masiero & Barletta 2005, Bueno et al. 2006), it was also observed in this study that a greater amount of lling material remained in the apical third than in the middle and cervical thirds irrespective of the technique used. Anatomical variations are greater in the apical third. Although Masiero & Barletta (2005) reported that a rotary system was more effective in removing the lling materials in the apical third, no difference was found in this study amongst the various techniques in curved root canals. This nding was also reported in a previous study (Schirrmeister et al. 2006c). The manufacturer of R-Endo instruments claims that the instrument is designed specially for retreatment as they have three equally spaced cutting edges, no radial land and active tip. Tas demir et al. (2008) reported that R-Endo retreatment les and Hedstro m les have similar effectiveness in removing lling material in straight root canals. Similarly, Gergi & Sabbagh (2007), in curved root canals, reported no signicant difference between R-Endo retreatment and Hedstro m les. In this study, K-les were used in combination with Hedstro m les to remove the gutta-percha mass and this combination may have advantages. The good performance of ProTaper Universal retreatment instruments in straight root canals in the study of Gu et al. (2008) was attributed to the three progressive tapers and length design of D1, D2 and D3 les. They mentioned that these features may enable the retreatment instruments to cut not only gutta-percha but also the supercial layer of dentine during root lling removal. This aggressive design and
active cutting tip of D1 may have caused the lateral perforation at the outer side of two roots in the ProTaper group in this study. As the aim of this study was to assess the efciency of the instruments designed for removing gutta-percha only, conventional ProTaper rotary instruments were not used. If the nishing instruments had been used, the performance of the ProTaper group would have improved. In previous studies, rotary instruments were reported to be more rapid in removing gutta-percha than manual instruments (Hu lsmann & Stotz 1997, Betti & Bramante 2001, Ferreira et al. 2001, Hu lsmann & Bluhm 2004). In contrast, Imura et al. (2000) found that Hedstro m les required less time for retreatment than the Quantec group and they attributed it to the removal of guttapercha in larger pieces. Similarly, it was concluded in this study that the combination of the Hedstro m and K les required less time than the other techniques. The number and effectiveness of the instruments in the removal process of lling material inuenced the working time. Despite the similar effectiveness of manual and ProFile instruments, the removal of gutta-percha in larger pieces by Hedstro m les may shorten the working time. The number of instruments in ProTaper and REndo groups also affected the working time even though they have similar effectiveness. ProTaper groups where three instruments were used, were signicantly faster than R-Endo groups. Tas demir et al. (2008), By contrast, reported that R-Endo was faster than Hestro m les in straight root canals. This conclusion can be explained with the number of les (three les) used in R-Endo groups. Five instruments fractured in the ProTaper and two in the R-Endo groups. Ha kel et al. (1999) noted the taper was a signicant factor in determining fracture probability for rotary instruments. The higher rate of fractured ProTaper may be related to the greater tapers of (0.07) the D3 les. In an earlier study investigating the retreatment of curved canals, only F2 and F3 ProTaper nishing instruments were used because of the high fracture rate of the F1 ProTaper nishing les. As reported in earlier studies (Imura et al. 2000, Betti & Bramante 2001, Schirrmeister et al. 2006c), apical extrusions were observed in all groups but there was no signicant difference amongst the groups.
Conclusions
Under the experimental conditions, it was impossible to remove all traces of gutta-percha and sealer from the root canal walls with any of the techniques used.
349
However, the full working length was achieved in all root canals. ProTaper Retreatment and R-Endo instruments were less effective in the removal of lling material from curved root canal walls than the Manual and ProFile instruments.
References
Barletta FB, Rahde Nde M, Limongi O, Moura AA, Zanesco C, Mazocatto G (2007) In vitro comparative analysis of 2 mechanical techniques for removing gutta-percha during retreatment. Journal of the Canadian Dental Association 73, 6565e. Barrieshi-Nusair KM (2002) Gutta-percha retreatment: effectiveness of nickeltitanium rotary instruments versus stainless steel hand les. Journal of Endodontics 28, 4546. Betti LV, Bramante CM (2001) Quantec SC rotary instruments versus hand les for gutta-percha removal in root canal retreatment. International Endodontic Journal 34, 5149. Bueno CE, Delboni MG, de Arau jo RA, Carrara HJ, Cunha RS (2006) Effectiveness of rotary and hand les in gutta-percha and sealer removal using chloroform or chlorhexidine gel. Brazilian Dental Journal 17, 13943. Farge P, Nahas P, Bonin P (1998) In vitro study of a Nd:YAP laser in endodontic retreatment. Journal of Endodontics 24, 35963. Ferreira JJ, Rhodes JS, Ford TR (2001) The efcacy of guttapercha removal using ProFiles. International Endodontic Journal 34, 26774. Friedman S, Moshonov J, Trope M (1993) Residue of guttapercha and a glass ionomer cement sealer following root canal retreatment. International Endodontic Journal 26, 16972. Gergi R, Sabbagh C (2007) Effectiveness of two nickel-titanium rotary instruments and a hand le for removing guttapercha in severely curved root canals during retreatment: an ex vivo study. International Endodontic Journal 40, 5327. Gu LS, Ling JQ, Wei X, Huang XY (2008) Efcacy of ProTaper Universal rotary retreatment system for gutta-percha removal from root canals. International Endodontic Journal 41, 28895. Ha kel Y, Serfaty R, Bateman G, Senger B, Allemann C (1999) Dynamic and cyclic fatigue of engine-driven rotary nickel titanium endodontic instruments. Journal of Endodontics 25, 43440. Hu lsmann M, Bluhm V (2004) Efcacy, cleaning ability and safety of different rotary NiTi instruments in root canal retreatment. International Endodontic Journal 37, 46876. Hu lsmann M, Stotz S (1997) Efcacy, cleaning ability and safety of different devices for gutta-percha removal in root canal retreatment. International Endodontic Journal 30, 22733. Imura N, Zuolo ML, Ferreira MO, Novo NF (1996) Effectiveness of the canal nder and hand instrumentation in removal of gutta-percha root llings during root canal retreatment. International Endodontic Journal 29, 3826.
Imura N, Kato AS, Hata GI, Uemura M, Toda T, Weine F (2000) A comparison of the relative efcacies of four hand and rotary instrumentation techniques during endodontic retreatment. International Endodontic Journal 33, 3616. Kosti E, Lambrianidis T, Economides N, Neotou C (2006) Ex vivo study of the efcacy of H-les and rotary NiTi instruments to remove gutta-percha and four types of sealer. International Endodontic Journal 39, 4854. Ladley RW, Campbell AD, Hicks ML, Li SH (1991) Effectiveness of halothane used with ultrasonic or hand instrumentation to remove gutta-percha from the root canal. Journal of Endodontics 17, 2214. Masiero AV, Barletta FB (2005) Effectiveness of different techniques for removing gutta-percha during retreatment. International Endodontic Journal 38, 27. Saad AY, Al-Hadlaq SM, Al-Katheeri NH (2007) Efcacy of two rotary NiTi instruments in the removal of Gutta-Percha during root canal retreatment. Journal of Endodontics 33, 3841. Sae-Lim V, Rajamanickam I, Lim BK, Lee HL (2000) Effectiveness of ProFile .04 taper rotary instruments in endodontic retreatment. Journal of Endodontics 26, 1004. Schirrmeister JF, Hermanns P, Meyer KM, Goetz F, Hellwig E (2006a) Detectability of residual Epiphany and gutta-percha after root canal retreatment using a dental operating microscope and radiographsan ex vivo study. International Endodontic Journal 39, 55865. Schirrmeister JF, Wrbas KT, Meyer KM, Altenburger MJ, Hellwig E (2006b) Efcacy of different rotary instruments for gutta-percha removal in root canal retreatment. Journal of Endodontics 32, 46972. Schirrmeister JF, Wrbas KT, Schneider FH, Altenburger MJ, Hellwig E (2006c) Effectiveness of a hand le and three nickeltitanium rotary instruments for removing guttapercha in curved root canals during retreatment. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 101, 5427. Schneider SW (1971) A comparison of canal preparations in straight and curved root canals. Oral Surgery, Oral Medicine, and Oral Patholology 32, 2715. Somma F, Cammarota G, Plotino G, Grande NM, Pameijer CH (2008) The effectiveness of manual and mechanical instrumentation for the retreatment of three different root canal lling materials. Journal of Endodontics 34, 4669. Tas demir T, Er K, Yildirim T, Celik D (2008) Efcacy of three rotary NiTi instruments in removing gutta-percha from root canals. International Endodontic Journal 41, 1916. Wilcox LR, Krell KV, Madison S, Rittman B (1987) Endodontic retreatment: evaluation of gutta-percha and sealer removal and canal reinstrumentation. Journal of Endodontics 13, 4537. Wolcott JF, Himel VT, Hicks ML (1999) Thermal retreatment using a new System B technique or a solvent. Journal of Endodontics 25, 7614.
350
doi:10.1111/j.1365-2591.2008.01532.x
Effectiveness of different laser systems to kill Enterococcus faecalis in aqueous suspension and in an infected tooth model
M. A. Meire1, K. De Prijck2, T. Coenye2, H. J. Nelis2 & R. J. G. De Moor1

1 Department of Operative Dentistry and Endodontology, Dental School, Ghent University, Gent, Belgium; and 2Laboratory for Pharmaceutical Microbiology, Ghent University, Gent, Belgium
Abstract
Meire MA, De Prijck K, Coenye T, Nelis HJ, De Moor RJG. Effectiveness of different laser systems to kill Enterococcus faecalis in aqueous suspension and in an infected tooth model. International Endodontic Journal, 42, 351359, 2009.
Aim To assess the antibacterial action of laser irradiation (Nd:YAG, KTP), photo activated disinfection (PAD) and 2.5% sodium hypochlorite (NaOCl) on Enterococcus faecalis, in an aqueous suspension and in an infected tooth model. Methodology Root canals of 60 human teeth with single straight canals were prepared to apical size 50, autoclaved, inoculated with an E. faecalis suspension and incubated for 48 h. They were randomly allocated to four treatment and one control groups. After treatment, the root canals were sampled by ushing with physiological saline, and the number of surviving bacteria in each canal was determined by plate count
and solid phase cytometry. The same experimental or control treatments were completed on aqueous suspensions of E. faecalis, and the number of surviving bacteria was determined in the same way. Results In aqueous suspension, PAD and NaOCl resulted in a signicant reduction in the number of E. faecalis cells (P < 0.001), whilst Nd:YAG or KTP had no effect. In the infected tooth model, only the PAD and NaOCl treated teeth yielded signicantly different results relative to the untreated controls (P < 0.001). Conclusions The laser systems as well as PAD were less effective than NaOCl in reducing E. faecalis, both in aqueous suspension and in the infected tooth model. Keywords: disinfection, laser, photodynamic therapy, root canal.
Received 27 March 2008; accepted 2 December 2008
Introduction
The role of micro-organisms and their by-products in the pathogenesis of apical periodontitis has clearly been established (Kakehashi et al. 1965, Sundqvist et al. 1998). Likewise, disinfection of the root canal system has been recognized as an essential aspect of root canal treatment. Traditionally, this is accomplished by chemo-mechanical cleaning, a combination of (i)
Correspondence: Prof. Dr Roeland De Moor, Department of Operative Dentistry and Endodontology, Dental School, Ghent University, Ghent University Hospital, De Pintelaan 185/P8, B-9000 Gent, Belgium (Tel.: +32 9 2404000 or 4001; fax: +32 9 403851; e-mail: roeland.demoor@ugent.be).
mechanical instrumentation; (ii) use of disinfecting solutions for irrigation of the root canal space; and (iii) placement of intracanal medication between appointments (Bystro m & Sundqvist 1981, 1983, 1985, Sjo gren et al. 1991). However, despite meticulous chemo-mechanical cleaning, eradication of all microorganisms from the root canal system is difcult (Sjo gren et al. 1997, Nair et al. 2005). Micro-organisms have been shown to persist in the anatomical complexities of the root canal system and to be the cause of treatment failure (Lin et al. 1991, Sundqvist et al. 1998). Therefore, various laser systems have been examined as adjuncts to currently used disinfection methods in root canal treatment. The laser light is thought to be able to reach areas that are inaccessible
351
Laser-assisted killing of E. faecalis in vitro and ex vivo Meire et al.
with the traditional techniques (e.g. bacteria located deeply in ns, isthmuses, lateral canals and in the dentinal tubules). Indeed, it has been demonstrated that dentinal tubules effectively act as bre optic channels, redirecting the light in multiple directions (Odor et al. 1996). Lasers such as CO2 (wavelength of 10 600 nm), Nd:YAG (neodymium-doped yttrium aluminium garnet) (1 064 nm), Er:YAG (erbium-doped yttrium aluminium garnet) (2 940 nm) and diode (810 or 980 nm) have been tested for disinfection of root canals. However, the reported success rate varies between studies. The Nd:YAG laser is probably the best documented laser for this application. Moritz et al. (1999) found a 99.16% reduction of bacterial numbers in inoculated root canals after Nd:YAG irradiation (Moritz et al. 1999). Folwaczny et al. (2002) and Piccolomini et al. (2002), in comparable experiments, also found the Nd:YAG laser to reduce the bacterial content, but they reported NaOCl to be more effective. The KTP (potassium titanyl phosphate) laser, emitting at 532 nm, a new wavelength for dental applications, has been used primarily for tooth bleaching procedures (De Moor & Vanderstricht 2009). Recently, Schoop et al. (2006) reported on the antibacterial effectiveness of this laser. Using a dentine slice model, they observed a considerable reduction in the number of bacteria in most samples and concluded that the KTP laser was a suitable tool for root canal disinfection. The bactericidal effects of these high-power lasers are the result of dose-dependent heat generation. The amount of heat delivered may have undesirable effects such as charring and cratering of dentine (Depraet et al. 2005) or possible thermal injury to the periodontal ligament, resulting in root resorption, ankylosis or periradicular necrosis (Bahcall et al. 1992). However, these disadvantages may be overcome by the use of a low-power laser to activate a dye (photosensitizer), which in turn exerts a lethal effect on bacteria (Wilson et al. 1992). This photo-activated disinfection (PAD) technique can be undertaken with a range of visible red and near infrared lasers. Systems using low power (100 mW) visible red semiconductor lasers in conjunction with tolonium chloride dye are now commercially available. It has been shown that the PAD technique kills several bacterial species commonly found in the oral cavity, even if located within biolms (Dobson & Wilson 1992). Experimental studies reporting on the antimicrobial effectiveness of various chemicals and techniques are numerous in endodontics. Most of them are based on
comparison of bacterial counts before and after disinfection. For enumeration of viable bacteria, culture-based techniques are mostly used, such as the plate count method or examination of turbidity in a broth. At present, nonculture based approaches are becoming increasingly important (Malacrino et al. 2001). Amongst these, solid phase cytometry (SPC) offers the advantages of a low detection limit, high speed and the ability to detect viable nonculturable bacteria (DHaese & Nelis 2002). In the infected root canal, micro-organisms can be found within the canal lumen, attached to the root canal wall, and invading the dentinal tubuli (Nair 1987, Siqueira et al. 2002a). The bacteria sticking to the root canal wall (i.e. a bacterial biolm, Costerton et al. 1999) are of particular interest, as it is known that a signicant part of the root canal wall often remains untouched by instruments during conventional treatment (Peters et al. 2003). In these areas, the bacterial biolm cannot be removed mechanically. The aim of this study was to assess the antibacterial action of Nd:YAG and KTP laser irradiation, PAD and sodium hypochlorite (NaOCl) on Enterococcus faecalis. This study was carried out both in vitro and on the uninstrumented infected root canal walls of extracted teeth (ex vivo), by means of a culture-dependent and a culture-independent microbiological technique.
Materials and methods Bacteria and culture conditions

A pure culture of E. faecalis ATCC 10541 was grown in brain heart infusion (BHI) broth and incubated overnight at 37 C. Ten-fold serial dilutions were then prepared in physiological saline (PS, 0.9% (w/v) NaCl), to obtain working concentrations of 105 to 108 CFU mL)1.
Preparation of teeth
A total of 60 extracted, single-rooted human teeth, stored in physiologic saline, were decoronated following an informed consent protocol approved by the Ethics Committee of the Ghent University Hospital (ref. B67020084251). The external root surface was cleaned with curettes to remove calculus and periodontal soft tissues. Root canal preparation commenced with Gates Glidden Burs 4-3-2 (Dentsply Maillefer, Ballaigues, Switzerland) in a crown-down mode. Working length was established by passing a size
352
Meire et al. Laser-assisted killing of E. faecalis in vitro and ex vivo
10 K-le (Dentsply Maillefer) in the canal until visible at the apex and subtracting 1 mm. Instrumentation was continued until MAF 50 (taper 0.02) under irrigation with physiologic saline. The apical foramen was sealed by composite resin (Z250, 3M ESPE, Seefeld, Germany) and the root surface covered with bonding agent, to prevent bacterial leakage. Teeth were then placed in test tubes and sterilized by autoclaving (134 C, 10 min).
Ex vivo experiments
Ten microlitre aliquots of a freshly prepared E. faecalis suspension were transferred into the root canals, resulting in approximately 103104 cells per root canal. Sterile PS was added until the entire canal space was lled with uid. The tooth specimens were then enclosed in sterile Eppendorf tubes and incubated in a humid atmosphere at 37 C for 48 h. Any residual medium in the root canals after incubation was carefully removed with sterile paper points and the teeth were randomly allocated to the experimental or control groups. In the rst group, teeth were treated with pulsed Nd:YAG laser (Fidelis, Hightechlaser, Herzele, Belgium) irradiation at 1064 nm (n = 10), using the following lasing parameters: power: 1.5 W; frequency: 10 Hz. The light was delivered through a 200 lm exible bre. The bre tip was applied with a spiral movement from apical to coronal, for ve times 5 s with 20 s intervals (total energy: 37.5 J). In the second experimental group, teeth were treated with KTP laser (Smartlite KTP laser, Deka, Calenzano, Italy) irradiation at 532 nm (n = 10) in pulsed mode. The power was set at 1 W and the pulse frequency at 10 Hz, again applying a spiral movement from apical coronal with the 200 lm bre, for ve times 5 s with 20 s intervals (total energy: 25 J). The actual output of both lasers was controlled with a power meter (Ophir Nova, Ophir Optronics Ltd, Jerusalem, Israel) before each experiment, and regularly checked during the experiments. In the third group, teeth (n = 10) were subjected to PAD treatment. To this end, a photosensitizing agent (PAD solution: i.e. toluidine blue O (TBO) at a concentration of 12.7 mg mL)1, pH 5) was added to ll the canal. After a 2 min pre-irradiation time; laser light (635 nm) generated by a diode laser (Denfotex, Inverkeithing, UK) was applied through an endodontic handpiece consisting of a exible cylinder (15 mm length, 400 lm diameter), emitting 70% of the light radially as a cylinder uniformly along the length and
30% of the total light intensity at the tip. Power was set at 100 mW and irradiation time was 150 s (total energy: 15 J), according to the manufacturers recommendations. The endotip was gently moved up and down the canal during irradiation. In the fourth group (n = 10), canals were lled to the orice with NaOCl solution (2.5% w/v). After 5 min, the NaOCl was removed from the canal with sterile paper points and fresh NaOCl solution was added. Five minutes later, this procedure was repeated, resulting in a total contact time of 15 min. The fth group (n = 20) served as a positive control; i.e. teeth received no treatment. Uninoculated negative control teeth (n = 3) were also included in the experiment.
Root canal sampling

In all groups, root canals were ushed with 990 lL sterile PS by means of a 3 mL syringe with a 27-gauge endodontic needle (Monoject, Sherwood Medical, St Louis, MO, USA) of which the apical 4 mm had been removed. This needle was gently moved up and down in the canal space whilst applying rm pressure on the syringe plunger. To this end, the teeth were held upside down, to collect the sampling liquid in a test tube, resulting in a volume of 1 mL. Following aspiration of the sampling liquid, ushing was repeated. A 1/10 dilution series of the collected bacterial suspension was prepared in sterile PS.
In vitro experiments
Sterile Eppendorf tubes (n = 24) were lled with 10 lL of a freshly prepared E. faecalis suspension (105106 bacterial cells). The following treatment protocols were used. A rst treatment consisted of pulsed Nd:YAG laser irradiation at 1064 nm (n = 6) with a power of 1.5 W and a frequency of 15 Hz. Using the 200 lm bre, the suspension was uniformly irradiated by applying a scanning movement just above the liquid surface. This was done ve times 5 s, with 20 s intervals. In the second group, KTP laser irradiation at 532 nm (n = 6) was used (1 W; 10 Hz). Using the 200 lm bre, the aliquot was uniformly irradiated by applying a scanning movement just above the uid surface. This was done ve times 5 s, with 20 s intervals. For the PAD treatment (n = 6), 20 lL of PAD solution was added to each tube. This solution was ltered over a 0.22 lm membrane lter (Millipore, Bedford, MA, USA) prior to use. After a 2 min pre-irradiation time; the
353
bacterial suspension was uniformly irradiated for 150 s at 100 mW (according to the manufacturers recommendations), holding the spherical (caries) tip in the centre of the liquid. In the NaOCl group (n = 6), 10 lL of NaOCl solution (2.5% v/v) was added to each tube. A contact time of 15 min was allowed. Sterile PS was added to each tube to obtain a working volume of 1 mL, except for the NaOCl group, where 980 lL of a 5% sodium thiosulphate solution was added to inactivate the NaOCl. Prior experiments had proven this solution to effectively inactivate sodium hypochlorite whilst not inuencing the enterococcal viability (data not shown). A 1/10 dilution series of the nal bacterial suspension was prepared in sterile PS.
the labelled cells. The signals produced were processed by a PC using a series of software discriminants, to differentiate between valid signals (labelled bacteria) and background. Scan results were displayed as coloured spots on a membrane lter image. These spots were visually conrmed using epiuorescence microscopy.
Plate count
Aliquots of each (diluted) bacterial suspension were transferred to petridishes and mixed with molten (45 C) brain heart infusion agar (BHIA). After solidication of the agar, the plates were incubated aerobically at 37 C for at least 48 h. Finally, the number of colony forming units (CFUs) in the undiluted suspension was calculated. Cell counts were logarithmically transformed to normalize the data prior to statistical comparison (one-way anova, post hoc Scheffe test, level of significance set at 5%). The mean and standard deviation of each group were calculated.
Solid phase cytometry (SPC)

In order to count the surviving micro-organisms using SPC, 100 lL of each bacterial suspension was ltered through a black polyester membrane lter (Cycloblack CB 04, 25 mm diameter, 0.4 lm pore size, Chemunex, Ivry-sur-Seine, France). Subsequently, these lters were incubated with a labelling solution (1% dilution of ChemChrome V6 in ChemSol B16, Chemunex) for 30 min at 30 C. The uorescent labelling of viable bacteria in SPC is based on the cleavage of a uorescein type ester by intracellular esterases and retention of the free uorescein in cells with an intact cytoplasmic membrane. Consequently, only viable bacteria with an intact membrane will be detected by this technique. After labelling, the membrane lter was scanned in the ChemScan (Chemunex). This apparatus was equipped with an argon laser, emitting light of 488 nm with a (variable) power of 25 mW that scanned the 25-mm diameter membrane lter in 3 min. Two photomultiplier tubes with wavelength windows set for the green (500530 nm) and amber (540585 nm) regions of the emission spectrum of uorescein detected the uorescence light emitted by
Results In vitro experiments

In Table 1 the reductions in the number of viable (SPC) and culturable E. faecalis cells after the various treatments are listed. The results represent the mean values calculated with reference to the nontreated control samples. No signicant differences between SPC and plate count values were observed. Nd:YAG or KTP laser irradiation did not result in a reduction in the number of surviving Enterococci. PAD treatment using the ltersterilized photosensitizer resulted in a signicant reduction (2.7 log units), whilst contact with NaOCl killed all the cells present. The use of the PAD photosensitizer, as provided by the manufacturer, in combination with low power laser
Table 1 Effect of Nd:YAG and KTP laser irradiation, photo activated disinfection and sodium hypochlorite on the survival of Enterococcus faecalis in vitro
Before SD Treatment Nd:YAG KTP PAD NaOCl SPC 5.20 5.20 6.13 5.13 0.18 0.18 0.07 0.11 Culture 5.27 5.27 6.14 5.14 0.12 0.12 0.07 0.09 After SD SPC 5.45 0.08 5.37 0.25 3.47 0.68 0 Culture 5.42 0.14 5.28 0.28 3.44 0.49 0 Log reduction SPC )0.25 )0.17 2.66 5.13 Culture )0.15 )0.01 2.69 5.14
Results are expressed as mean (logarithmic) before and after treatment gures ( SD) and as reductions in viable (SPC) and culturable (plate count) cells. n = 6 in each group.
354
(Fig. 2). None of the two high power lasers reduced the number of E. faecalis (log reduction <1; P > 0.05). PAD application in prepared root canals diminished the SPC counts and the total plate count by approximately 1 and 1.5 log units respectively. This result was signicantly different from that of the control group (P < 0.001). Largest reductions were obtained in the sodium hypochlorite treated group, i.e. 1.72 and 2.10 log reductions respectively, depending on the quantication method (P < 0.001). The negative controls yielded no culturable cells.
Figure 1 Effect of lethal photosensitization on the viability of Enterococcus faecalis cells in vitro. PAD (+) indicates the use of lter-sterilized photosensitizer, whilst in PAD ()), the photosensitizer was not sterilized. n = 6 in each group.
Discussion
Enterococcus faecalis was chosen as the test organism in this study because it is the species most often associated with persistent endodontic infections (Molander et al. 1998, Sundqvist et al. 1998, Peciuliene et al. 2000, Pinheiro et al. 2003) and it can also be found in primary root canal infections (Siqueira et al. 2002b). This Gram-positive, facultative anaerobe coccus has been used by many other investigators, probably due to its ease of growth and laboratory manipulation (Haapasalo et al. 2000, Hems et al. 2005, Tandjung et al. 2007). The effectiveness of the laser applications and the NaOCl control against E. faecalis was tested on bacterial suspensions (in vitro) and in an infected tooth model (ex vivo). The results of the in vitro experiments show that both high-power lasers (Nd:YAG and KTP) did not affect the viability of planktonic E. faecalis cells. It has been demonstrated that laser-induced bacterial killing is due to different mechanisms: thermal heating of the bacterial environment above lethal values, local heating inside bacteria (by laser light-sensitive chromophores inside the bacteria) or light-induced modulation of enzymatic activity (Hellingwerf et al. 1996). As virtually all bacteria survived, none of the above-mentioned mechanisms took place, or the amount of energy delivered was insufcient. This might be due to poor absorbance of the Nd:YAG and KTP wavelengths in water. Probably, transmission of the laser beams rather than absorption by the bacterial suspension occurred, explaining the survival of bacteria. Rooney et al. (1994) investigated the bactericidal effect of the Nd:YAG laser on E. faecalis suspensions in glass capillary tubes. No or a minimal bactericidal effect was observed at energy doses below 30 J. Above this value, they noted increasing bacterial killing with increasing doses of irradiation, with reductions of
Figure 2 Effect of Nd:YAG and KTP laser irradiation, photo activated disinfection and sodium hypochlorite treatment on the survival of Enterococcus faecalis in the infected tooth model. Vertical bars represent the mean number of recovered cells SD.
light resulted only in one log reduction as indicated by SPC results. By contrast, the plate count revealed a higher reduction for suspensions treated identically (Fig. 1). Microscopic validation of the SPC results by means of an epiuorescence microscope showed the presence of rod shaped micro-organisms indicating nonsterility of the photosensitizer. Filtration of pure (nonlasered) dye solution over a polyester membrane lter, labelling and scanning conrmed this observation. The exact extent of contamination of the dye solution was batch-dependent (data not shown).
Ex vivo experiments
In the ex vivo experiments, no signicant differences between SPC and plate count values were observed
355
>4 log units above 54 J. When a black dye was added, similar reductions were obtained at much lower energies (Rooney et al. 1994). In the present Nd:YAG experiment, energy doses of 37.5 J were delivered (1.5 W during 25 s) The latter parameters were chosen to be identical with those used in the ex vivo experiments, which were adopted from Moritz et al. (1999) and Schoop et al. (2006). The data obtained in this study are more or less in accordance with those from Rooney et al. (1994). The fact that in the latter study, Enterococci were suspended in broth compared with saline in this study, may have led to an increased absorbance of laser energy. Application of the PAD system in vitro revealed some methodological problems. Many batches of the photosensitizer for the PAD application, as received from the manufacturer, proved to be nonsterile. Therefore, part of the PAD solution was sterilized by ltration over a 0.22 lm membrane lter. When using the lter-sterilized photosensitizer, PAD application in vitro led to a 2.7 log reduction of E. faecalis cells in both SPC and plate counts. This reduction was highly signicant compared with the nontreated controls. The best results in the in vitro experiments were obtained in the NaOCl treated group, as no Enterococci survived this treatment. This is in accordance with results obtained by Radcliffe et al. (2004), who found that a 5 min contact time between an E. faecalis suspension and NaOCl 2.5% (w/v) eradicated all bacteria. The results of the infected tooth model experiments show that it is hard to eradicate E. faecalis from the root canal. Comparison of the inoculum size (103 104) with the number of recovered cells (105106) indicates that the E. faecalis cells did grow well in the teeth. Sealing of the apical foramen with composite resin did not seem to interfere with bacterial growth. This observation is substantiated by several studies that show that composite resins have minimal antibacterial effects (Karanika-Kouma et al. 2001, Beyth et al. 2007). Both high-power lasers gave a slight reduction of the intracanal ora, but the differences with the nontreated controls were not signicant. Moritz et al. (1999), in a similar study, found a 99.16% (2 log units) reduction of bacteria in inoculated root canals after Nd:YAG irradiation. In this study, it was not possible to achieve such reductions. The Moritz group, however, used a suspension of E. faecalis and E. coli to inoculate their root canals. It has been shown that the susceptibility of the gramnegative rod E. coli to Nd:YAG laser irradiation is
substantially higher than that of E. faecalis (Moritz et al. 2000). This might explain the higher log reductions achieved in the latter study. Schoop et al. (2006) irradiated E. faecalis inoculated dentine disks at the bacteria free side with the KTP laser after 4 h of incubation. They observed minor changes in bacterial count at a power setting of 1 W. Using the higher setting of 1.5 W, no growth was observed in 10 out of 20 samples. However, the lower level of detection was rather high (5 102 CFU mL)1), so that the effect may have been overestimated. In this experiment, the power setting of 1 W appeared to be too low to affect the viability of the E. faecalis cells. In the PAD treated teeth, a 1.42 log reduction was observed with the culture-based method, but the difference from the control group was not signicant. Williams et al. (2006) used the same PAD device in a comparable set-up and obtained a mean log reduction of 2.01 in teeth. However, other test bacteria were used in their study. Although the results were statistically different, the observed log reductions are not clinically relevant for the treatment of endodontic disease. The counts in the NaOCl treated group were significantly different from those in the nontreated control teeth. Nevertheless, only an average 2 log reduction was obtained. The poorer results in the infected tooth model compared with those of the tests on suspensions (100% bacterial kill) could be attributed to the inactivation of NaOCl by the organic component in root dentine and the small volume of irrigant used. The root canal was completely lled with NaOCl, so the volume of irrigant that was able to act on the canal content was not more than a few 100 lL, which is rather low compared with the volumes that are used during root canal treatment. The exposure times in the various treatments (KTP and Nd:YAG laser for 25 s; PAD for 150 s and NaOCl for 15 min), but also the total energy delivered in case of laser treatment differ. These conditions were chosen according to the best available evidence and because they are clinically relevant. The differences demonstrated in the killing of E. faecalis may therefore not only be due to the different methods but also the different exposure times used. Consequently, the use of other laser settings in each group can inuence the outcome of the experiment. It has been demonstrated that E. faecalis is able to form a bacterial biolm in the root canal (George et al. 2005). Biolm-grown (sessile) cells differ from their planktonic counterparts in a number of aspects,
356
including the presence of an extracellular polymeric substance, cell wall composition, growth rate, metabolic activity and gene expression (Costerton et al. 1999). Since at the end of the inoculation time, all liquid content was removed from the canals using paper points, planktonic cells were also removed. The presence of an E. faecalis biolm in the teeth probably explains the poor efcacy of PAD and NaOCl in the ex vivo model. Solid phase cytometry proved to be a fast and valid technique for enumeration of Enterococci. In this study no difference between SPC and plate counts was observed, except when using PAD. Application of PAD on the in vitro model led to a 2 log reduction of bacteria as indicated by plate count results, but only a 1 log reduction as obtained with SPC. However, visual inspection of the lter by epiuorescence microscopy revealed that other micro-organisms than the inoculating species were present, as evidenced from the different morphology of the cells. It was found that rod-shaped microorganisms were present next to cocci. This observation raised the assumption of a contaminated photosensitizer, which was subsequently conrmed. It also shows that not all micro-organisms detected by SPC could be cultured using standard culture conditions. Neither of the three laser systems yielded a clinically relevant reduction of the number of intracanal bacteria. Highest reductions were obtained after NaOCl irrigation but still this 2 log reduction of a 6 log population in the infected root canal is insignicant in practice. Overall, the effectiveness of the treatment modalities in this study to eliminate root canal infection was disappointing. It can be concluded that the root canal areas that escape mechanical cleaning cannot be effectively disinfected with the present treatment protocols. These ndings do not necessarily mean that lasers cannot be benecial in the eld of root canal disinfection. The antibacterial effect of Nd:YAG has been proven in nonendodontic settings (Ward et al. 1996). Further research is necessary to establish the appropriate laser parameters permitting adequate antimicrobial action without harmful thermal side effects. Also, there is a need for randomized, blind, clinical trials. Besides, other applications such as laser activation of root canal irrigants have been reported (Blanken & Verdaasdonk 2007) that could benet root canal treatment.
and in the infected tooth model. The search for laser applications in endodontics should however be continued and different laser wavelengths and settings should be examined.
Acknowledgement
This work was supported by the European Society of Endodontology annual research grant in 2005.
References
Bahcall J, Howard P, Miserendino L, Walia H (1992) Preliminary investigation of the histological effects of laser endodontic treatment on the periradicular tissues in dogs. Journal of Endodontics 18, 4751. Beyth N, Domb AJ, Weiss EI (2007) An in vitro quantitative antibacterial analysis of amalgam and composite resins. Journal of Dentistry 35, 2016. Blanken J, Verdaasdonk R (2007) Cavitation as a working mechanism of the Er,Cr:YSGG laser in endodontics: a visualization study. Journal of Oral Laser Applications 7, 97106. Bystro m A, Sundqvist G (1981) Bacteriologic evaluation of the efcacy of mechanical root canal instrumentation in endodontic therapy. Scandinavian Journal of Dental Research 89, 3218. Bystro m A, Sundqvist G (1983) Bacteriologic evaluation of the effect of 0.5 percent sodium hypochlorite in endodontic therapy. Oral Surgery, Oral Medicine and Oral Pathology 55, 30712. Bystro m A, Sundqvist G (1985) The antibacterial action of sodium hypochlorite and EDTA in 60 cases of endodontic therapy. International Endodontic Journal 18, 3540. Bystro m A, Claesson R, Sundqvist G (1985) The antibacterial effect of camphorated paramonochlorophenol, camphorated phenol and calcium hydroxide in the treatment of infected root canals. Endodontics and Dental Traumatology 1, 1705. Costerton JW, Stewart PS, Greenberg EP (1999) Bacterial biolms: a common cause of persistent infections. Science 284, 131822. DHaese E, Nelis HJ (2002) Rapid detection of single cell bacteria as a novel approach in food microbiology. Journal of AOAC International 85, 97983. De Moor RJ, Vanderstricht K (2009) The use of the KTP laser, an added value for tooth bleaching. Aesthetic Dentistry Today 3, 1720. Depraet FJ, De Bruyne MA, De Moor RJ (2005) The sealing ability of an epoxy resin root canal sealer after Nd:YAG laser irradiation of the root canal. International Endodontic Journal 38, 3029. Dobson J, Wilson M (1992) Sensitization of oral bacteria in biolms to killing by light from a low-power laser. Archives of Oral Biology 37, 8837.
Conclusion
The laser systems, at the current settings, were less effective than NaOCl in reducing E. faecalis both in vitro
357
Folwaczny M, Mehl A, Jordan C, Hickel R (2002) Antibacterial effects of pulsed Nd:YAG laser radiation at different energy settings in root canals. Journal of Endodontics 28, 249. George S, Kishen A, Song KP (2005) The role of environmental changes on monospecies biolm formation on root canal wall by Enterococcus faecalis. Journal of Endodontics 31, 86772. Haapasalo HK, Siren EK, Waltimo TM, Orstavik D, Haapasalo MP (2000) Inactivation of local root canal medicaments by dentine: an in vitro study. International Endodontic Journal 33, 12631. Hellingwerf KJ, Hoff WD, Crielaard W (1996) Photobiology of microorganisms: how photosensors catch a photon to initialize signalling. Molecular Microbiology 21, 68393. Hems RS, Gulabivala K, Ng YL, Ready D, Spratt DA (2005) An in vitro evaluation of the ability of ozone to kill a strain of Enterococcus faecalis. International Endodontic Journal 38, 229. Kakehashi S, Stanley HR, Fitzgerald RJ (1965) The Effects of Surgical Exposures of Dental Pulps in Germ-Free and Conventional Laboratory Rats. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 20, 3409. Karanika-Kouma A, Dionysopoulos P, Koliniotou-Koubia E, Kolokotronis A (2001) Antibacterial properties of dentin bonding systems, polyacid-modied composite resins and composite resins. Journal of Oral Rehabilitation 28, 15760. Lin LM, Pascon EA, Skribner J, Gangler P, Langeland K (1991) Clinical, radiographic, and histologic study of endodontic treatment failures. Oral Surgery, Oral Medicine and Oral Pathology 71, 60311. Malacrino P, Zapparoli G, Torriani S, Dellaglio F (2001) Rapid detection of viable yeasts and bacteria in wine by ow cytometry. Journal of Microbiological Methods 45, 12734. Molander A, Reit C, Dahlen G, Kvist T (1998) Microbiological status of root-lled teeth with apical periodontitis. International Endodontic Journal 31, 17. Moritz A, Schoop U, Goharkhay K et al. (1999) The bactericidal effect of Nd:YAG, Ho:YAG, and Er:YAG laser irradiation in the root canal: an in vitro comparison. Journal of Clinical Laser Medicine & Surgery 17, 1614. Moritz A, Jakolitsch S, Goharkhay K et al. (2000) Morphologic changes correlating to different sensitivities of Escherichia coli and Enterococcus faecalis to Nd:YAG laser irradiation through dentin. Lasers in Surgery and Medicine 26, 25061. Nair PN (1987) Light and electron microscopic studies of root canal ora and periapical lesions. Journal of Endodontics 13, 2939. Nair PN, Henry S, Cano V, Vera J (2005) Microbial status of apical root canal system of human mandibular rst molars with primary apical periodontitis after one-visit endodontic treatment. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 99, 23152.
Odor TM, Watson TF, Pitt Ford TR, McDonald F (1996) Pattern of transmission of laser light in teeth. International Endodontic Journal 29, 22834. Peciuliene V, Balciuniene I, Eriksen HM, Haapasalo M (2000) Isolation of Enterococcus faecalis in previously root-lled canals in a Lithuanian population. Journal of Endodontics 26, 5935. Peters OA, Peters CI, Schonenberger K, Barbakow F (2003) ProTaper rotary root canal preparation: effects of canal anatomy on nal shape analysed by micro CT. International Endodontic Journal 36, 8692. Piccolomini R, DArcangelo C, DErcole S, Catamo G, Schiaffino G, De Fazio P (2002) Bacteriologic evaluation of the effect of Nd:YAG laser irradiation in experimental infected root canals. Journal of Endodontics 28, 2768. Pinheiro ET, Gomes BP, Ferraz CC, Sousa EL, Teixeira FB, Souza-Filho FJ (2003) Microorganisms from canals of rootlled teeth with periapical lesions. International Endodontic Journal 36, 111. Radcliffe CE, Potouridou L, Qureshi R et al. (2004) Antimicrobial activity of varying concentrations of sodium hypochlorite on the endodontic microorganisms Actinomyces israelii, A. naeslundii, Candida albicans and Enterococcus faecalis. International Endodontic Journal 37, 43846. Rooney J, Midda M, Leeming J (1994) A laboratory investigation of the bactericidal effect of a NdYAG laser. British Dental Journal 176, 614. Schoop U, Kluger W, Dervisbegovic S et al. (2006) Innovative wavelengths in endodontic treatment. Lasers in Surgery and Medicine 38, 62430. Siqueira JF Jr, Rocas IN, Lopes HP (2002a) Patterns of microbial colonization in primary root canal infections. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 93, 1748. Siqueira JF Jr, Rocas IN, Souto R, de Uzeda M, Colombo AP (2002b) Actinomyces species, streptococci, and Enterococcus faecalis in primary root canal infections. Journal of Endodontics 28, 16872. Sjo gren U, Figdor D, Spangberg L, Sundqvist G (1991) The antimicrobial effect of calcium hydroxide as a short-term intracanal dressing. International Endodontic Journal 24, 11925. Sjo gren U, Figdor D, Persson S, Sundqvist G (1997) Inuence of infection at the time of root lling on the outcome of endodontic treatment of teeth with apical periodontitis. International Endodontic Journal 30, 297306. Sundqvist G, Figdor D, Persson S, Sjo gren U (1998) Microbiologic analysis of teeth with failed endodontic treatment and the outcome of conservative re-treatment. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 85, 8693. Tandjung L, Waltimo T, Hauser I, Heide P, Decker EM, Weiger R (2007) Octenidine in root canal and dentine disinfection ex vivo. International Endodontic Journal 40, 84551.
358
Ward GD, Watson IA, Stewart-Tull DE, Wardlaw AC, Chatwin CR (1996) Inactivation of bacteria and yeasts on agar surfaces with high power Nd:YAG laser light. Letters in Applied Microbiology 23, 13640. Williams JA, Pearson GJ, Colles MJ (2006) Antibacterial action of photoactivated disinfection {PAD} used on endodontic
bacteria in planktonic suspension and in articial and human root canals. Journal of Dentistry 34, 36371. Wilson M, Dobson J, Harvey W (1992) Sensitization of oral bacteria to killing by low-power laser radiation. Current Microbiology 25, 7781.
359
doi:10.1111/j.1365-2591.2008.01534.x
Quality of life after microscopic periradicular surgery using two different incision techniques: a randomized clinical study
M. Del Fabbro, S. Taschieri & R. Weinstein

Department of Health Technologies, IRCCS Istituto Ortopedico Galeazzi, University of Milan, Milan, Italy
Abstract
Del Fabbro M, Taschieri S, Weinstein R. Quality of life after
microscopic periradicular surgery using two different incision techniques: a randomized clinical study. International Endodontic Journal, 42, 360367, 2009.
Aim To monitor the quality of life of patients after periradicular surgery when two different ap designs were used. Methodology Forty patients with teeth having a periradicular lesion of endodontic origin were included according to specic selection criteria. Patients were randomly assigned to two groups. In one group a sulcular incision (SI) with complete papilla mobilization was made, and in the other group a papilla-base incision (PBI) was used. Periradicular surgery was performed using a surgical microscope. Parameters related to life quality were recorded daily in the rst week post-surgery using a questionnaire. Pain was evaluated with a 0100 visual analog scale (VAS). Other symptoms (swelling, bleeding and nausea), plus functions (chewing, speaking, sleeping, daily routine
and work) were assessed using a ve-point scale. Analgesic intake was recorded. Fishers test and unpaired t-test were used to assess the difference between groups. Results The VAS score for pain, and the scores for swelling, chewing and phonetic impairment, peaked on days 1 and 2 postoperatively. A signicant difference in favour of the PBI group was found for chewing and swelling in the rst 4 days. Starting from day 3 postsurgery, the PBI group reported a signicantly more rapid decrease in pain levels and analgesics use than the SI group (P < 0.05). The other parameters were similar in the two groups. Conclusions The papilla-base incision technique may be preferred as reduction of pain levels, swelling and drug intake were more rapid in the rst week postoperatively compared with cases in which a sulcular incision was used. Keywords: ap design, periradicular surgery, quality of life.
Received 15 September 2008; accepted 1 December 2008
Introduction
After the introduction of microsurgical techniques in endodontics, there has been an increased interest in developing protocols for improving the root-end management of teeth (Kim 2002). Conversely, less attention has been paid to the surgical management
Correspondence: Massimo Del Fabbro, Department of Health Technology, IRCCS Istituto Ortopedico Galeazzi, University of Milan, Via R. Galeazzi 4, Milano 20161, Italy (Tel.: +39 02 50319950; fax: +39 02 50319960; e-mail: massimo. delfabbro@unimi.it).
of soft tissues and to patient-related outcomes in the early post-operative phase (Velvart 2002, Velvart et al. 2003, 2004a,b, Velvart & Peters 2005, von Arx et al. 2007). Post-operative quality of life of patients is dependent on the degree of pain, tissue swelling, chewing ability, phonetics, and can be of importance for the overall assessment of the treatment success as well as to its acceptance. Pain and swelling are secondary effects that may occur in the immediate post-surgical period (Gutmann & Harrison 1991). About two thirds of the patients treated by the traditional technique without using magnication devices require analgesics during the early
360
Del Fabbro et al. Life quality using different aps in apical surgery
post-operative phase (Seymour et al. 1986, Meechan & Blair 1993). Recently the goal of endodontic surgery has shifted from the mere reduction or elimination of existing pathosis to the achievement of successful outcomes regarding function and aesthetics as well as to periodontal tissue preservation (Carr & Bentkover 1998, Kim 2004). The introduction of microscopic techniques in periradicular surgery allows the surgeon to control both components with a high level of precision. Peri-operative pain management is fundamental in any surgical procedure for preserving the patients psychological welfare. Reducing pain-related discomfort in the immediate postoperative period may enhance the quality of life of the patient (Iqbal et al. 2007). Few studies have examined postoperative discomfort after endodontic surgery, reporting that the use of microsurgical technique is associated with less postoperative pain compared with the traditional technique (Pecora & Andreana 1993, Tsesis et al. 2003, 2005). It may be hypothesized that proper soft tissue management also could be of importance in the control of postsurgical discomfort. The aim of the present study was to assess and compare patient quality of life after microscopic periradicular surgery when two different ap designs were used.
Material and methods

This randomized study was conducted according to the principles embodied in the World Medical Association Helsinki Declaration of 1975 for biomedical research involving human subjects, as revised in 2000 (World Medical Association Declaration of Helsinki 2000). Ethical approval was obtained from the Institutional Review Board of Milan University. All patients were informed of the nature of the study and gave their written consent. Patients requiring endodontic surgical treatment were recruited during a period of 24 months (from December 2004 to December 2006) in a University clinic and in a private practice setting. A single experienced surgeon performed all surgeries.
periradicular lesion of strictly endodontic origin (chronic apical periodontitis) of size not exceeding 10 mm; the nonsurgical re-treatment was judged unfeasible or had previously failed; the tooth had an adequate nal restoration with no clinical evidence of coronal leakage; the apical root canal was devoid of the presence of a post for at least 6 mm; no acute symptoms were present. Both single-rooted and multirooted teeth, located in the aesthetic regions (maxillary anterior and pre-molar teeth), were included. The following exclusion criteria were applied: presence of any kind of pathosis associated with vertical root fracture; perforation of the furcation area or lateral canal walls; presence of traumatic injuries; periodontal bone loss, detected with a periodontal probe (>4 mm probing depth); bone defects involving both the buccal and lingual cortical bone; presence of a thin gingival biotype. Based on sample size calculation it was planned to enrol at least 16 patients for each treatment group, in order to detect a between-group 10% difference in postoperative pain (that was considered the most important symptom affecting quality of life), with a power of 0.8 and a signicance level equal to 0.05. According to the above criteria 40 teeth in 40 consecutive patients (23 women and 17 men), were included in the study. Each patient was given written information about the surgical procedure and the necessary follow-up care. They were also given the opportunity to withdraw from the study at any time. A consent form was signed if they agreed. Each patient received one session of professional oral hygiene on the day before surgery.
Allocation to groups
The choice of using one or the other kind of surgical ap for each patient was made by a computer-generated randomized table. A closed opaque envelope containing the indication of which surgical ap had to be used was opened before the start of each surgical operation. Twenty patients (20 teeth) were allocated to the group using a sulcular incision (SI group) with a complete mobilization of the entire papilla and 20 patients (20 teeth) to the group using a papilla-base incision (PBI group).
Patient selection
The following criteria were adopted for case selection: the patients had no general medical contra-indications for oral surgical procedures (they were ASA-1 or ASA-2); the patients had only one tooth that required periradicular surgery; the tooth treated surgically had a
Surgical procedure
Preoperatively, the patient rinsed with an antiseptic mouthwash containing 0.2% chlorhexidine
361
Life quality using different aps in apical surgery Del Fabbro et al.
digluconate (Curasept, Curaden Healthcare s.r.l., Saronno, Milan, Italy) to reduce the risk of contamination of the surgical eld. Treatment was provided under local anaesthesia with lidocaine 2% and epinephrine 1 : 1 00 000.
Soft and hard tissue management

The aps were rectangular and consisted of two releasing vertical incisions and a horizontal incision. The vertical incisions were placed in at least one tooth both distal and mesial to the tooth being treated. The initial portion of the vertical incision was placed perpendicular to the marginal course of the gingiva toward the mid section of the papilla and gradually turning the incision parallel to the tooth axis. Subsequently it ran vertical, parallel to the tooth axis and to supraperiosteal blood vessels in the mucosa and gingiva with paramedian releasing incision (Fig. 1). In the SI group, the two releasing incisions were connected by a sulcular horizontal incision involving interproximal spaces to free the buccal from the palatal papilla with a complete mobilization of the buccal papilla (Fig. 2).
Figure 2 An example of the incision performed in the sulcular incision (SI) group, with complete mobilization of the papilla.
In the PBI group, two different incisions were performed at the base of papilla resulting in a splitthickness ap, as described by Velvart (2002). Buccally, over the tooth, the interproximal spaces were joined by an intrasulcular incision in a curvilinear fashion (Fig. 3). The sulcular incision reached from the releasing incisions to the start of the nearest PBI. A periosteal release incision was made for releasing residual muscle tension and facilitates the passive coronal displacement of the ap. In the SI group a 15c surgical blade (Kai Europe, GmbH, Solingen, Germany) was used for the incision, whilst a CK-2 microsurgical scalpel (Analytic, Glendora, CA, USA) was used in the PBI group. In both groups surgical loupes were used as a magnication device for ap elevation procedure. The full mucoperiosteal ap was mobilized, reected and carefully retracted during the
Figure 1 An example of the distal releasing vertical incision, common to both groups.
Figure 3 An example of the papilla-base incision (PBI group).
362
root-end management. After ap reection, a retractor was positioned on the exposed cortical bone with light but rm pressure acting as a passive mechanical barrier to the reected tissues. The ap was frequently irrigated with sterile saline to prevent dehydration of the periosteal surface. Surgical access to the root was then made through the cortical bone using a round bur. Shaving of the bone was performed with a brush stroke approach, low rotary speed and constant sterile water irrigation. The periradicular lesion was removed with sharp bone curettes and angled periodontal curettes. The curetted tissue was placed in 10% formalin solution for pathological diagnosis.
tooth brushing during the day of surgery (Gutmann & Harrison 1985). Ice packs were provided after surgery. The patients were instructed to rinse their mouth twice daily with chlorhexidine digluconate 0.2% (Curasept , Curaden Healthcare s.r.l.) for plaque control, up to 10 days after surgery. All patients were prescribed nonsteroidal analgesics after the surgical procedure for pain relief and/or swelling control if needed. No antibiotic therapy was prescribed. Sutures were removed 5 days after surgery.
Evaluation parameters
A questionnaire similar to that used in previous studies (Shugars et al. 1996, Tsesis et al. 2005) was used to evaluate postoperative limitations in function (chewing, talking, sleeping, daily routine and work), as well as pain and the presence of other symptoms (swelling, bleeding, nausea, bad taste/breath). For pain assessment a visual analog scale (VAS) was adopted, where 0 = no pain, and 100 = unbearable pain. For other symptoms and functional activities the answers were based on a 5-point Likert-type scale, ranging from 1 (none) to 5 (very much). Finally, patients were asked whether they had taken any analgesics on each postoperative day. Patients received the questionnaire to ll out on each day starting on the day of surgery, for 7 days. Questionnaires were returned postage-paid.
Management of the resected root end

After exposure of the root end, a straight ssure bur in a hand-piece was positioned perpendicular to the long axis of the root and then beginning from the apex, cutting coronally, 2.53 mm of the root-end was removed. Prior to root-end preparation, local haemostasis was achieved using bone wax. The root-end cavities were prepared using zirconium nitride retrotips (Dentsply Maillefer Instruments, Ballaigues, Switzerland) driven by an ultrasonic device unit (Piezon Master 600, EMS, Nyon, Switzerland). All root-end cavities were created with the setting of the ultrasonic device unit at no more then half power, under constant copious sterile water irrigation to avoid over-heating. The retro-tips allowed a well-dened parallel preparation of 2.5 to 3 mm deep. Root-end cavities were then dried using paper points. Finally a zinc oxide EBAreinforced cement (Super Seal, Ogna Pharmaceuticals, Milan, Italy) was used as the root-end lling material. For performing root-end management procedures an operating microscope was used as the magnication device in both groups. The reected tissues were re-approximated to their original position, compressed and stabilized, and sutured with polyamide 6-0 (Ethicon Inc., Johnson & Johnson, Piscataway, NJ, USA). The ap closure was initiated from the releasing incisions and in the papillabase incision two interrupted sutures were used. The time needed to complete each surgical procedure was recorded, starting from the rst incision to nishing the last suture.
Fishers exact test was used to assess statistically the difference between groups for analgesics and for any variable related to function and symptoms on each postoperative day. For simplicity, the responses corresponding to 1 (none) and 2 (little) were combined in a single category. The patients experience of pain was evaluated using analysis of variance (anova) for repeated measures. The difference between the two groups for pain on each postoperative day was assessed using an unpaired t-test. The patient was considered as the unit of analysis. A probability P = 0.05 was considered as the level of signicance. The software statistica (StatSoft, Inc., Tulsa, OK, USA) version 5.0 was used for statistical analysis.
Results Postoperative instructions

The patients were advised to avoid mouth rinsing, hard and hot food, hot drinks, heavy physical work and Forty patients were initially treated with periradicular surgery. One patient (belonging to the SI group) failed to ll out the questionnaire completely and another one
363
Table 1 Distribution of the surgically treated teeth in the two study groups
Tooth location Anterior Pre-molar Total RF group 15 (1)a 5 20 (1) PBI group 16 (1)a 4 20 (1) Total 31 (2)a 9 40 (2)
a The number of teeth excluded from the analysis is indicated between parentheses
(belonging to the PBI group) failed to return the questionnaire; both were excluded from the analysis. Thus, a nal of 38 patients (22 women and 16 men) were evaluated. Fifteen of them were smokers (six in the SI group and nine in the PBI group) with an average daily consumption of 7.2 cigarettes. Nineteen patients were evaluated in the SI group (10 women and nine men, ranging in age from 22 to 59 years, average 36.4 years), and 19 were evaluated in the PBI group (12 women and seven men, ranging in age from 29 to 56 years, average 33.7 years). Teeth consisted of maxillary anterior and pre-molar teeth (Table 1). The average size of the periradicular lesion was 6 mm (range 49 mm). No statistically signicant difference was found in the distribution of patients according to age, gender, smoking habits and lesion size between the two groups.
The average time needed to complete the surgical procedure was 42 min (range 3558 min) for the PBI group, and 35 min (range 2841 min) in the SI group; the difference was signicant (P = 0.01). Tables 2 and 3 report the results of the evaluation for symptoms and functional activities, respectively. Figure 4 reports the levels of pain reported for 7 days postsurgery. Figure 5 reveals the percentage of patients taking analgesics during the postoperative period. All patients reported some degree of discomfort because of pain, swelling, chewing impairment and difculties in phonetics on days 1 and 2 postoperatively. In the PBI group, a more rapid decrease in pain levels (Fig. 4) and analgesics taken (Fig. 5) was observed compared with the SI group starting from day 3 (P < 0.05). Such difference became negligible after 56 days. In the SI group, swelling was signicantly higher than in the PBI group from days 1 to 4 (Table 2). Chewing impairment also was signicantly greater in the SI group as compared with the PBI group from days 2 to 4 (Table 3). No signicant difference was found at any time between smokers and non smokers for pain levels and tissue swelling. Bleeding, nausea and bad taste/breath were occasionally reported in the rst 2 days and were negligible
Table 2 Occurrence of symptoms in the rst week postoperatively

day 1 (%) Symptom SI PBI day 2 (%) SI PBI day 3 (%) SI PBI day 4 (%) SI PBI day 5 (%) SI PBI day 6 (%) SI PBI day 7 (%) SI PBI
Swelling Very much 15.8 Quite a bit 47.4 Some 36.8 Little/none Bleeding Very much Quite a bit 5.3 Some 47.4 Little/none 47.4 Nausea Very much Quite a bit Some 15.8 Little/none 84.2 Bad taste/breath Very much Quite a bit 21.1 Some 42.1 Little/none 36.8
47.4 52.6 5.3 47.4 47.4 100 10.5 57.9 31.6
42.1 31.6 21.1 5.3 5.3 94.7 100 47.4 52.6
5.3 47.4 42.1 5.3 10.5 89.5 100 42.1 57.9
10.5 42.1 42.1 5.3 100 100 21.1 78.9
47.4 52.6 100 100 10.5 89.5
21.1 52.6 26.3 100 100 5.3 94.7
31.6 68.4 100 100 5.3 94.7
15.8 84.2 100 100 100
100 100 100 100
100 100 100 100
100 100 100 100
100 100 100 100
100 100 100 100
The gray area indicates signicant differences between groups
364
Table 3 Impairment of common activities in the rst week postoperatively

day 1 (%) Activity Sleeping Very much Quite a bit Some Little/none Chewing Very much Quite a bit Some Little/none Phonetics Very much Quite a bit Some Little/none Daily routine Very much Quite a bit Some Little/none Missed work Very much Quite a bit Some Little/none SI PBI day 2 (%) SI PBI day 3 (%) SI PBI day 4 (%) SI PBI day 5 (%) SI PBI day 6 (%) SI PBI day 7 (%) SI PBI
5.3 42.1 52.6 42.1 26.3 26.3 5.3 21.1 42.1 36.8 36.8 47.4 15.8 5.3 36.8 42.1 15.8
15.8 36.8 47.4 26.3 36.8 36.8 15.8 57.9 26.3 10.5 63.2 26.3 21.1 63.2 15.8
15.8 84.2 15.8 36.8 47.4 47.4 52.6 10.5 36.8 52.6 10.5 42.1 47.4
5.3 21.1 73.7 42.1 52.6 5.3 21.1 78.9 42.1 57.9 52.6 47.4
10.5 89.5 10.5 31.6 47.4 10.5 10.5 89.5 36.8 63.2 36.8 63.2
10.5 89.5 47.4 52.6 100 26.3 73.7 26.3 73.7
5.3 94.7 5.3 42.1 52.6 100 21.1 78.9 100
10.5 89.5 5.3 94.7 100 15.8 84.2 100
5.3 94.7 15.8 84.2 100 100 100
10.5 89.5 100 100 100 100
100 100 100 100 100
100 100 100 100 100
100 100 100 100 100
100 100 100 100 100
The gray areas indicate signicant differences between groups
Figure 4 Diagram showing the trend of pain levels in the two
groups, assessed by means of a visual analog scale (VAS). Asterisks indicate signicant difference between groups.
Figure 5 Diagram showing the trend of the percentage of patients using analgesics in the two groups. Asterisks indicate signicant difference between groups.
thereafter. No signicant difference between the two groups was found for these symptoms. The recovery of normal speech and sleeping was similar in the two groups. No patient reported the maximum score for these two activities. Moderate impairment for routine daily activities, function and loss of work was reported in both groups, especially in the rst 3 days (signicance was detected
only for day 1). A gradual recovery was observed during the rst postoperative week for these two parameters.
Discussion
The preservation of soft tissues represents a challenge in any surgical and reconstructive procedure. When the success of a surgical treatment has to be evaluated,
365
not only healing and function should be assessed. The aesthetic outcome and the patients subjective symptoms such as post-treatment discomfort should also be taken into account as they may affect their quality of life and acceptance of treatment. Treatment of the soft tissues with adequate surgical and reconstructive techniques as well as ongoing maintenance is mandatory in modern dentistry, and preservation of the dentition is no longer acceptable without considering the aesthetic consequences (Allen 1988). The prognosis of endodontic periradicular surgery is dependent on a myriad of factors (Rud et al. 1972). According to Friedman (1988), these factors can be divided into preoperative, intra-operative and postoperative. Amongst the latter there is the degree of tissue shrinkage. Some studies showed that a full-thickness marginal ap is related to considerable retraction of the papilla especially during the initial healing phase that may often lead to scar formation (Velvart & Peters 2005). By contrast, the PBI technique allows predictable recession-free healing of the interdental papilla, without scar formation (Velvart & Peters 2005). For these reasons such a technique should be preferred, to avoid opening of the proximal space, when periradicular surgical treatment is necessary. In the present study, the two types of incisions were compared by examining the patients quality of life postoperatively, without considering the aesthetic issue. The latter in fact should be evaluated later than 1 week post-surgery, when complete soft tissue healing has occurred. Both SI and PBI groups revealed that age, gender, smoking, site of operation and size of the lesion had no inuence on postoperative sequelae. This result is in agreement with other studies (Tsesis et al. 2005, Penarrocha et al. 2006, Christiansen et al. 2008). Another recent study reported conicting results regarding smoking effects on postoperative pain and swelling (Garc a et al. 2007), suggesting that the actual inuence of smoking habits on postoperative symptoms, if any exists, is yet to be determined. Amongst the various possible confounding factors, preoperative oral hygiene status was also recently claimed to negatively affect pain and swelling after periapical surgery (Garc a et al. 2007). Conversely, another study reported no inuence of such factor on the postoperative period (Penarrocha et al. 2006). This factor was not considered in the present study because in addition to the presurgical rinse with chlorhexidine, all the patients underwent a session of professional oral hygiene the day before surgery.
All patients of both groups reported in the rst 2 days the greatest discomfort with speaking and chewing, whilst sleeping was only moderately affected. The postoperative symptoms were similar to those reported in the study by Tsesis et al. (2005). Some studies have revealed a lower incidence of postoperative pain and swelling following periapical surgery using operating microscopes versus periapical surgery performed using traditional techniques. Pecora & Andreana (1993) hypothesized that the microscope allowed for a better control of soft tissues, minimizing surgical trauma. In the present study, the same anaesthesia and the same microsurgical procedure during the root-end management was used in both groups. Pain experience peaked in both groups in the rst 2 days, but the decrease in pain levels was more rapid and the amount of analgesics taken, as well as tissue swelling, was lower in patients belonging to the PBI group from days 3 to 7. Using the full thickness ap the papilla is mobilized and becomes part of the ap. The buccal papilla should be dissected from the lingual papilla, but in narrow interproximal space the separation process is technically difcult and damage to the papilla occurs easily due to the elevation process. Residual tissue fragments after the ap elevation process are often too small to survive and may necrotize leading to recession (Gutmann & Harrison 1991, Velvart & Peters 2005). Often there is insufcient adaptation of the papilla to the underlying tissue surface and this might pre-dispose to recession. The dimension of the papilla might also have an effect on the healing pattern after surgery. The PBI technique, which leaves the body of the papilla in place, would eliminate any visible opening of the interproximal space during the healing process. With this technique split ap thickness can be preserved thereby maintaining tissue vitality. Hence, the risk of recession defects and loss of papilla height can be reduced. The data of the present study showed that a more rapid decrease in pain levels is achieved with the PBI incision. This may affect the quality of life of patients and treatment acceptance. Further prospective studies with larger sample size are needed to conrm the present results.
Conclusions
The type of incision performed inuenced pain levels, swelling and drug intake in the rst postoperative week. A papilla-base incision technique, when applicable,
366
may be preferred to other ap designs for the management of soft tissues in periradicular surgery.
References
Allen EP (1988) Use of mucogingival surgical procedure to enhance esthetics. Dental Clinics of North America 32, 307 30. von Arx T, Vinzens-Majaniemi T, Bu rgin W, Jensen SS (2007) Changes of periodontal parameters following apical surgery:prospective clinical study of three incision techniques. International Endodontic Journal 40, 95969. Carr G, Bentkover S (1998) Surgical endodontics. In: Cohen S, Burns R, eds. Pathways of the Pulp, 7th edn. St. Louis, MO: Mosby Inc., pp. 60856. Christiansen R, Kirkevang L-L, Hrsted-Bindslev P, Wenzel A (2008) Patient discomfort following periapical surgery. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 105, 24550. Friedman S (1988) Treatment outcome and prognosis of endodontic therapy. In: rstavik D, Pitt Ford T, eds. Essential Endodontology: Prevention and Treatment of Apical Periodontitis, 1st edn. Oxford: Blackwell Science, pp. 38891. Garc a B, Penarrocha M, Mart E, Gay-Escodad C, von Arx T (2007) Pain and swelling after periapical surgery related to oral hygiene and smoking. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 104, 2716. Gutmann JL, Harrison JW (1985) Posterior endodontic surgery: anatomical considerations and clinical techniques. International Endodontic Journal 18, 834. Gutmann J, Harrison JW (1991) Surgical Endodontics, 1st edn. Boston, MA, USA: Blackwell Scientic Publications. Iqbal MK, Kratchman SI, Guess GM, Karabucak B, Kim S (2007) Microscopic periradicular surgery: perioperative predictors for postoperative clinical outcomes and quality of life assessment. Journal of Endodontics 33, 23944. Kim S (2002) Endodontic microsurgery. In: Cohen S, Burns RC, eds. Pathways of the Pulp, 8th edn. St. Louis, MO: Mosby Inc., pp. 683721. Kim S (2004) Modern endodontic practice. Instruments and techniques. Dental Clinics of North America 48, 19. Meechan JG, Blair GS (1993) The effect of two different local anaesthetic solutions on pain experience following apicectomy. British Dental Journal 175, 4103.
Pecora G, Andreana S (1993) Use of dental operating microscope in endodontic surgery. Oral Surgery, Oral Medicine, Oral Pathology 75, 7518. Penarrocha M, Garc a B, Marti E, Balaguer J (2006) Pain and inammation after periapical surgery in 60 patients. Journal of Oral & Maxillofacial Surgery 64, 42933. Rud J, Andreasen JO, Mo ller Jensen JE (1972) A multivariate analysis of various factors upon healing after endodontic surgery. International Journal of Oral Surgery 1, 25871. Seymour RA, Meechan JG, Blair GS (1986) Postoperative pain after apicoectomy. A clinical investigation. International Endodontic Journal 19, 2427. Shugars DA, Benson K, White RP Jr, Simpson KN, Bader JD (1996) Developing a measure of patient perceptions of short-term outcomes of third molar surgery. Journal of Oral & Maxillofacial Surgery 54, 14028. Tsesis I, Fuss Z, Lin S, Tilinger G, Peled M (2003) Analysis of postoperative symptoms following surgical endodontic treatment. Quintessence International 34, 75660. Tsesis I, Shoshani Y, Givol N, Yahalom R, Fuss Z, Taicher S (2005) Comparison of quality of life after surgical endodontic treatment using two techniques: a prospective study. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 99, 36771. Velvart P (2002) Papilla base incision: a new approach to recession-free healing of the interdental papilla after endodontic surgery. International Endodontic Journal 35, 45360. Velvart P, Peters CI (2005) Soft tissue management in endodontic surgery. Journal of Endodontics 31, 416. Velvart P, Ebner-Zimmermann U, Ebner JP (2003) Comparison of papilla healing following sulcular full-thickness ap and papilla base ap in endodontic surgery. International Endodontic Journal 36, 6539. Velvart P, Ebner-Zimmermann U, Ebner JP (2004a) Comparison of long term papilla healing following sulcular full thickness ap and papilla base ap in endodontic surgery. International Endodontic Journal 37, 68793. Velvart P, Ebner-Zimmermann U, Ebner JP (2004b) Papilla healing following sulcular full thickness ap in endodontic surgery. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontics 98, 3659. World Medical Association Declaration of Helsinki (2000) Ethical principles for medical research involving human subjects. Journal of American Medical Association 284, 30435.
367
doi:10.1111/j.1365-2591.2008.01535.x
Combination of apex locator and endodontic motor for continuous length control during root canal treatment
M. J. Altenburger, Y. C enik, J. F. Schirrmeister, K.-T. Wrbas & E. Hellwig

Department of Operative Dentistry and Periodontology, University Medical Hospital and Dental School, Albert-Ludwigs-University, Freiburg, Germany
Abstract
Altenburger MJ, C enik Y, Schirrmeister JF, Wrbas K-T, Hellwig E. Combination of apex locator and endodontic motor
for continuous length control during root canal treatment. International Endodontic Journal, 42, 368374, 2009.
Aim To compare ex vivo an experimental setup consisting of an electronic apex locator (EAL) and endodontic motor with an established product (Tri Auto ZX) for accuracy of length control during root canal treatment with three different types of les. Methodology An experimental setup consisting of porous spongy material and an electrolyte was used. Sixty anterior teeth were randomly assigned to six groups. Access cavities were prepared. During root canal treatment, constant length monitoring was performed either with the Tri Auto ZX or the Raypex5 apex locator attached to an endodontic motor (Endo IT professional) using ProTaper, Mtwo or FlexMaster les. After root canal preparation the distances between le tip and major apical foramen and le tip and minor
apical foramen were measured using a microscope and analysed using two-way anova to evaluate the accuracy of the two systems. Results Distances between the le tip and the major apical foramen were not signicantly different between the le systems and the two EALs. In cases treated with FlexMaster signicantly larger distances between le tip and minor apical foramen were found compared to Mtwo and ProTaper. No signicant differences were observed between the two EALs. After preparation of the root canals with the Tri Auto ZX, multiple minor apical foramina were mechanically widened. Conclusion With the limitation of this laboratory study the combination of EAL and endodontic motor was as accurate as the Tri Auto ZX system in terms of length control during root canal preparation. Keywords: accuracy, apical foramen, electronic apex locator, endodontic motor, working length.
Received 15 November 2007; accepted 1 December 2008
Introduction
A recent review (Lin et al. 2005) concluded that the absence of bacterial contamination and sufcient removal of infected necrotic tissue were the main factors for a positive outcome following root canal treatment. The review also concluded that the accurate
Correspondence: Dr Markus Jo rg Altenburger, Universita tsklinik fu r Zahn-, Mund- und Kieferheilkunde, Abteilung fu r Zahnerhaltungskunde und Parodontologie, Hugstetter Strae 55, 79095 Freiburg, Germany (Tel.: +49 761 2704957; fax: +49 761 2704762; e-mail: markus.altenburger@uniklinikfreiburg.de).
determination of working length played a major role in reducing contamination and the bacterial load in the root canal system. Under-instrumentation of root canals, particularly in cases of infected necrotic pulps and asymptomatic apical periodontitis, leads to significantly lower success rates compared cases where an accurate working length was achieved (Sjo gren et al. 1990, Chugal et al. 2003). On the other hand, overinstrumentation with enlargement of the apical constriction, trauma to the apical tissues, extrusion of infected material apically and destruction of the apical binding point for the root lling can affect the outcome of root canal treatment negatively (Sjo gren et al. 1990, Chugal et al. 2003, Souza 2006).
368
Altenburger et al. Setup of EAL and endodontic motor
In practice, the determination of working length and its control remains a challenge. The assessment of working length by tactile sensation alone is not advisable (Seidberg et al. 1975). Radiographic examination is appropriate for diagnostic purposes and evaluation of the root morphology, but is not able to determine the working length consistently due to the anatomical variations between teeth (Dummer et al. 1984, Olson et al. 1991, Gutierrez & Aguayo 1995). Electronic apex locators (EAL) are an alternative for the determination of correct working length (Gordon & Chandler 2004, Nekoofar et al. 2006). The continuous monitoring of working length is important during canal preparation as the working length may vary during the procedure, especially in curved canals. The endodontic instrument causes increased dentine removal from the inner wall of curved canals that straightens the root canal (Caldwell 1976, Farber & Bernstein 1983). Failures to adjust the working length can lead to negative side effects (Bhaskar & Rappaport 1971, Seltzer et al. 1973). Therefore, combinations of EAL and low-speed endodontic handpieces have been introduced to achieve the accuracy of conventional EALs during canal shaping (Grimberg et al. 2002, Alves et al. 2005). Besides the length measurement function these devices also have torque control and speed settings. As there are a large number of EALs and endodontic motors on the market, the question arises as to whether these two stand-alone devices might be used in combination as an apex locating endodontic motor. As the impedance of a root is complex (Tipler & Mosca 2007), electronic devices or the earthing of the endodontic motor may interfere with the electrical circuit of the EAL. Consequently, the hypothesis of the study was that this experimental setup is able to determine and to monitor the correct working length in combination with three different le systems at least as accurately as a commercially available product (Tri Auto ZX, J. MORITA, Kyoto, Japan).
Material and methods Specimen preparation and experimental setup

Sixty extracted human anterior teeth were collected from a pool of extracted teeth. The teeth were stored under moist conditions in a thymol solution. Only teeth with an overall length of 2025 mm, fully formed apices and with no caries, coronal restorations, signs of resorptions or cracks were chosen. The teeth were
radiographed in two dimensions to assure a single root canal was present with a canal curve of <7 using the goniometry function of an imaging software according to (Schneider 1971) (Scanner: Digora Gendex; software: VixWin 2000 v1.8; Dentsply Gendex Dental Systeme, Hamburg, Germany). The pulp chambers were accessed using a water-cooled, cylindrical diamond bur (Brasseler, Lemgo, Germany) in a handpiece. The canal system was ooded with 3.0% sodium hypochlorite using a syringe (Injekt LL 10 mL, B. Braun, Melsungen, Germany) and a 30-gauge needle (NaviTip, Ultradent Products, South Jordan, UT, USA). The coronal and middle section of the root canals were shaped using size 15, 20, 25 stainless steel Hedstro m and K-les (VDW, Munich, Germany) to provide access to the apical third of the root canal. To standardize the size of the apical preparation, root canals were instrumented until a size 08 K-le was visible at the major apical foramen but a size 15 K-le bound approximately 2 mm short of the major apical foramen. Teeth not matching these criteria were discarded. To simulate the tissues surrounding the teeth a laboratory setup was used. A porous plastic block, normally used for ower arrangements (OASIS Ideal 1, Smithers-Oasis, Cuyahoga Falls, OH, USA) was placed in a VDW-le container (FlexMaster Systembox; VDW). Ringers solution (DeltaSelect, Dreieich, Germany), an isotonic liquid for intravenous injection, was placed in the box to assure ion ow between the electrodes of the EAL. The liquid was introduced into the box until the spongy material was completely soaked and 5 mm uid covered the bottom of the box. ` ne/ Rubber dam (Roeko Flexi Dam non-latex, Colte Whaledent, Langenau, Germany) was stretched over the box, both to simulate a clinical situation and to insulate the spongy material from the user, thereby assuring the completion of the measurement circuit of the EAL. Two holes were made in the rubber dam; in one hole, the teeth were placed through the dental dam into the porous block leaving the coronal 23 mm of the root out of the block. In the second hole, the lip clip of the EAL was clamped (Fig. 1).
Endodontic treatment units

For mechanical root canal preparation the established Tri Auto ZX was used as a control. This contra-angle handpiece consists of a torque-controlled endodontic motor and an integrated EAL. For root canal shaping the Auto Apical Reverse function (AAR) was set to 0.5 as
369
Setup of EAL and endodontic motor Altenburger et al.
Figure 1 Experimental setup: Plastic box containing porous material and electrolyte covered by rubber dam. The lip clip is tucked into the conductive material. The le clip is attached to the le.
described in the instruction manual to indicate the minor apical foramen (apical constriction). Torque settings for each le were set as recommended. For the experimental setup a combination of the Endo IT professional (contraangle handpiece and torque-controlled endodontic motor) and the Raypex5 EAL (both: VDW) were connected. To close the measuring circuit of the EAL, the le clip was clamped to the le shaft. The Auto Stop Reverse function was enabled. The Endo IT professional handpiece was calibrated for each le of each system according to the manufacturers instruction.
Root canal preparation

Sixty of the teeth prepared as described above were randomly assigned to six different groups. Two main groups, representing the two root canal treatment units (Tri Auto ZX and Raypex5) and three subgroups (ProTaper, Mtwo, FlexMaster) in each group according to the three different le types were used in this study.
Before canal preparation and after irrigation (sodium hypochlorite, 3%) working length was determined using and size 08 K-le and either the Raypex5 or the Tri Auto ZX switched to the apex locator mode. In both groups, les were rst taken to over-instrumentation and then pulled back to the apical constriction as indicated in the apex zoom (Raypex5) or 0.5 (Tri Auto ZX) respectively. Subsequently, root canals were then shaped with K-les size 8, 10, 15 with FileCare EDTA (VDW) at full working length to establish a glide path as recommended. Prior to and during mechanical instrumentation, each root canal system was irrigated with 3.0% sodium hypochlorite solution. Latex examination gloves (DermaClean, Ansell Healthcare, Brussels, Belgium) were worn to isolate the user against the Tri Auto ZX and the experimental setup. The lip clip was placed in the spongy material and the clamp was attached to the le (Fig. 1). During subsequent canal preparation les were not adjusted to working length, rather monitoring of the working length took place
370
either by the display of the Raypex5 EAL or via the AAR-function of the Tri Auto ZX. The following rotary le systems were used:
ProTaper (Dentsply Maillefer, Ballaigues, Switzerland)

The ProTaper shaping les S1 and SX were introduced into the canal performing brushing motions to prepare the coronal and middle sections of the root canal. Then les S1 and S2 were used at working length. ProTaper nishing les F1, F2 and F3 were brought to full working length, avoiding excessive pecking motions. Except for shaping le SX, all les reached working length. This procedure was interrupted intermittently by irrigating and recapitulating the canal using a size 15 K-le.
Mtwo (VDW)
A size 10, .04 taper le was brought to full working length either by performing lateral cutting (brushing motion) or a pecking motion (up and down motion). After reaching full working length les size 15, .05 taper, size 20, .06 taper, size 25, .06 taper and a size 30, .05 taper were used as described above. After each le the canal was irrigated and recapitulated.
microscope whilst paying attention to the anatomical canal characteristics (Wrbas et al. 2007). For the analysis of the roots a video camera (AxioCam MRc5; Carl Zeiss, Oberkochen, Germany) connected to a stereomicroscope (Leica WILD M3Z; Leica Mikrosysteme Vertrieb, Bensheim, Germany) and a computer in combination with a calibrated measurement software (software: AxioVs40 vers. 4.5.0.0., Carl Zeiss; calibration grid: Carl Zeiss) were used. The distance between le tip and major apical foramen was measured as well as the distance between le tip and minor apical foramen (accuracy 0.005 mm). Major and minor apical foramina were dened according to (Nekoofar et al. 2006). There the apical portion of the root canal is considered as an inverted cone and its base indicates the major apical foramen. The apex of this inverted cone indicates the location of the minor apical foramen or apical constriction (Fig. 2). Two-way anova (SPSS 16.0.1; SPSS inc., Chicago, IL, USA) was used to determine the effects of the two independent variables used EAL and used le system on the dependent variables distance le tip minor apical foramen and distance le tip major apical foramen. The level of signicance was set to 0.05.
Results Distance le tip major apical foramen
FlexMaster (VDW)
In the coronal part of the root canals, les size 30, .06 taper, size 25, .06 taper, size 20, .06 taper and size 30, .04 taper were used. As recommended, this was carried out in a crown-down manner to a length 23 mm short of the indicated working length by performing pecking motions and interrupted intermittently by irrigation and recapitulation. File size 20, .02 taper reached working length followed by .02 taper les with tip sizes 25 and 30.
In none of the specimens was the le tip found beyond the major foramen, neither in the groups treated with the Raypex5/Endo IT control or with the Tri Auto ZX.
Analysis of the root canal

At the end of the root canal preparation the last instrument was brought to working length as indicated by the respective EAL. Rubber stoppers and the le itself were then xed in the tooth with a light-curing composite (Tetric EvoCeram, Ivoclar Vivadent AG, Schaan, Liechtenstein). The apical 35 mm of the roots were carefully removed using a diamond blade and a scalpel until the instruments and the canal walls were visible. This was performed under a light
Figure 2 Apical region of a tooth, including distances and canal walls.
371
Table 1 Mean values and standard deviations (SD) of the distances between le tip and major apical foramen and le tip and minor apical foramen for the main groups represented by the EALs, their three subgroups represented by the types of les and for each le system regardless of the used EAL
Distance le tip major apical foramen (mm) Raypex5 All le systems ProTaper Mtwo FlexMaster Mean Mean Mean Mean (SD) (SD) (SD) (SD) 0.66 0.63 0.61 0.74 (0.44) (0.21) (0.20) (0.30) TriAutoZX 0.78 0.80 0.82 0.72 (0.47) (0.16) (0.31) (0.30) Both EAL Distance le tip minor apical foramen (mm) Raypex5 0.24 0.12 0.06 0.53 (0.55) (0.16) (0.18) (0.30) TriAutoZX 0.32 0.26 0.20 0.53 (0.58) (0.23) (0.34) (0.29) Both EAL
0.72 (0.20) 0.72 (0.28) 0.73 (0.29)
0.19 (0.20) 0.13 (0.28) 0.53 (0.29)
EAL, electronic apex locator.
Mean distances between le tip and major foramen varied between 0.61 mm (experimental setup and Mtwo) and 0.82 mm (Tri Auto ZX and Mtwo) (Table 1). The analysis of the distances between the respective le systems showed no signicantly different results for groups treated with ProTaper, Mtwo and FlexMaster (P = 0.978). The results of the two length measurement devices, regardless of the used le types, were also not signicantly different (P = 0.068). There was no signicant interaction effect between the used EALs and the used le systems (P = 0.309).
to Mtwo and ProTaper (P < 0.001). Although the distances between le tip and minor apical foramen were generally larger in groups treated with the Tri Auto ZX no signicant difference was found (P = 0.191). Also no signicant interaction effect between the used EALs and le systems was observed (P = 0.627).
Discussion
The use of an endodontic motor in combination with a separate EAL has not yet been described in literature. Therefore, the objective of the present study was to test this combination under laboratory conditions with respect to its function and accuracy. Compounds used for these laboratory setups include agar-agar in different concentrations (Aurelio et al. 1983, Nahmias et al. 1987, Fouad & Krell 1989), gelatine (Donnelly 1993) or alginate (Kaufman et al. 1989, Kaufman & Katz 1993). These materials have numerous disadvantages. They are expensive, designed for single use and must be stored under special conditions. Furthermore, their preparation is time consuming and the results obtained cannot be reproduced (Fouad & Krell 1989). In contrast, the present experimental setup is a less complicated, inexpensive and is reusable. Sodium hypochlorite (3.0%) was used for irrigation as it does not affect the accuracy of the EALs utilized (Meares & Steiman 2002, Nekoofar et al. 2006). To allow an even more comprehensive evaluation, three different le systems were chosen according to their mode of utilization. Hence, le diameters used for root canal treatment varied at all times, but this did not negatively inuence the EAL function (Nguyen et al. 1996, Lee et al. 2002). According to the recommendation of the European Society of Endodontology (2006) the apical constriction is recommended as the end-point of root canal treatment. Instruction manuals and displays of the utilized
Distance le tip minor apical foramen

Due to the fact that the canals had been treated, no information was available on the morphology of the constriction types (Dummer et al. 1984). In 11 cases, the minor apical foramen was mechanically widened although the le tips were short of this region when analysed. This appeared only in cases where the root canal treatment was performed with the Tri Auto ZX, whereas the allocation of the cases between the different le systems was equal. In four cases, equally distributed between the two EALs, the le tips were found where the canal walls begin to diverge in the area between minor and major apical foramen. Of the 30 measurements performed by every EAL 25 (83.3%) determined working lengths in the Raypex5 group and 20 (66.7%) measurements in the Tri Auto ZX group were within 0.5 mm of the minor apical foramen. In all groups distances were within 1 mm of the minor apical foramen. The mean distances between le tip and minor apical foramen varied between 0.06 mm (Mtwo and Raypex5/ Endo IT control) and 0.53 mm (FlexMaster and Tri Auto ZX; FlexMaster and Raypex5/ Endo IT control) (Table 1). Regardless of the used EAL the distance between the le tip and the minor apical foramen were signicantly higher in the group treated with FlexMaster compared
372
EALs imply that they are capable of locating the minor and major apical foramen. Kobayashi & Suda (1994) described that two frequencies, impedance ratio measurement based EALs (Raypex5 and TriAutoZX) are able to locate the minor apical foramen. Measured mean discrepancies of le tip and minor apical foramen of around 0.24 mm (experimental setup) and 0.32 mm (Tri Auto ZX) were in similar ranges and in accordance with ndings already reported (Welk et al. 2003). In the literature 75.082.3% of the determined working lengths are accurate to within 0.5 mm of the minor apical foramen (Dunlap et al. 1998, Meares & Steiman 2002, Tselnik et al. 2005) and are believed to be clinically acceptable (Ounsi & Naaman 1999, Grimberg et al. 2002). Working lengths determined by the experimental setup were within these limits in 83.3% of the cases, although fewer measurements conducted with Tri Auto ZX were within this limit (66.7%). Whilst there is doubt about the possibility of locating the minor apical foramen (Hoer & Attin 2004), it has been summarized that modern EALs are able to detect the point where the le tip reaches the tissues of the periodontal ligament (Nekoofar et al. 2006). In none of the groups were the le tips found beyond the major apical foramen. Reviewing the possible negative side effects of over-instrumentation (Sjo gren et al. 1990, Chugal et al. 2003, Souza 2006), this fact is important to improve the quality of root canal treatment. Measured mean distances between le tip and major apical foramen, varying from 0.61 to 0.74 mm (experimental setup) and from 0.72 to 0.82 mm (Tri Auto ZX), are comparable to a mean distance of 0.75 mm found by (Ounsi & Naaman 1999). In a comparative study a previous model of the Raypex5 also determined the major apical foramen signicantly more accurately than an EAL comparable to the Tri Auto ZX, but was discussed as clinically negligible (Kaufman et al. 2002). Although no negative interference between the Endo IT control and the Raypex5 was observed the user had to wear latex gloves to prevent him from being in direct contact with the handpiece or the tooth in order to obtain correct measurements. After shaping, only in root canals treated with Tri Auto ZX were the minor apical foramina damaged by root canal instruments, although the le tips were found short of the minor apical foramen. Although the AAR-function is described as an add-on to avoid accidental over-instrumentation (Grimberg et al. 2002), this phenomenon has already been described in a study evaluating the AAR-function of the Tri Auto ZX (Campbell et al. 1998). A level of one is
recommended to preserve the apical structures (Campbell et al. 1998, Carneiro et al. 2006). The preset 0.5 in this study was chosen according to the instruction manual to determine the apical constriction. The screw-in effect of the les in combination with the lower AAR-function setting in this study probably overstrained the reaction time of the AAR-function and led to this phenomenon. On the other hand, it was very challenging to remain within the indicated area of the minor apical foramen during length monitoring using the Raypex5, especially when pecking motions were performed. This was mainly observed in combination with FlexMaster. The screw-in effect of the le-system in combination with pecking motions and the abovedescribed problems to maintain the correct working length with the respective EAL might explain the higher discrepancies between le tip and major and minor apical foramen when FlexMaster was used.
Conclusion
This laboratory study proved the accuracy of the experimental setup in comparison to the Tri Auto ZX.
References
Alves AM, Felippe MC, Felippe WT, Rocha MJ (2005) Ex vivo evaluation of the capacity of the Tri Auto ZX to locate the apical foramen during root canal retreatment. International Endodontic Journal 38, 71824. Aurelio JA, Nahmias Y, Gerstein H (1983) A model for demonstrating an electronic canal length measuring device. Journal of Endodontics 9, 5689. Bhaskar SN, Rappaport HM (1971) Histologic evaluation of endodontic procedures in dogs. Oral Surgery Oral Medicine Oral Pathology 31, 52635. Caldwell JL (1976) Change in working length following instrumentation of molar canals. Oral Surgery Oral Medicine Oral Pathology Oral Radiology Endodontology 41, 1148. Campbell D, Friedman S, Nguyen HQ, Kaufman A, Keila S (1998) Apical extent of rotary canal instrumentation with an apex-locating handpiece in vitro. Oral Surgery Oral Medicine Oral Pathology Oral Radiology Endodontology 85, 31924. Carneiro E, Bramante CM, Picoli F, Letra A, da Silva Neto UX, Menezes R (2006) Accuracy of root length determination using Tri Auto ZX and ProTaper instruments: an in vitro study. Journal of Endodontics 32, 1424. Chugal NM, Clive JM, Spangberg LS (2003) Endodontic infection: some biologic and treatment factors associated with outcome. Oral Surgery Oral Medicine Oral Pathology Oral Radiology Endodontology 96, 8190.
373
Donnelly JC (1993) A simplied model to demonstrate the operation of electronic root canal measuring devices. Journal of Endodontics 19, 57980. Dummer PM, McGinn JH, Rees DG (1984) The position and topography of the apical canal constriction and apical foramen. International Endodontic Journal 17, 1928. Dunlap CA, Remeikis NA, BeGole EA, Rauschenberger CR (1998) An in vivo evaluation of an electronic apex locator that uses the ratio method in vital and necrotic canals. Journal of Endodontics 24, 4850. ESE (2006) Quality guidelines for endodontic treatment: consensus report of the European Society of Endodontology. International Endodontic Journal 39, 92130. Farber JP, Bernstein M (1983) The effect of instrumentation on root canal length as measured with an electronic device. Journal of Endodontics 9, 1145. Fouad AF, Krell KV (1989) An in vitro comparison of ve root canal length measuring instruments. Journal of Endodontics 15, 5737. Gordon MP, Chandler NP (2004) Electronic apex locators. International Endodontic Journal 37, 42537. Grimberg F, Banegas G, Chiacchio L, Zmener O (2002) In vivo determination of root canal length: a preliminary report using the Tri Auto ZX apex-locating handpiece. International Endodontic Journal 35, 5903. Gutierrez JH, Aguayo P (1995) Apical foraminal openings in human teeth. Number and location. Oral Surgery Oral Medicine Oral Pathology Oral Radiology Endodontology 79, 76977. Hoer D, Attin T (2004) The accuracy of electronic working length determination. International Endodontic Journal 37, 12531. Kaufman AV, Katz A. (1993) Reliability of Root ZX apex locator texted by an in vitro model. Journal of Endodontics 19, 201. Kaufman AY, Szajkis S, Niv N (1989) The efciency and reliability of the Dentometer for detecting root canal length. Oral Surgery Oral Medicine Oral Pathology 67, 5737. Kaufman AY, Keila S, Yoshpe M (2002) Accuracy of a new apex locator: an in vitro study. International Endodontic Journal 35, 18692. Kobayashi C, Suda H (1994) New electronic canal measuring device based on the ratio method. Journal of Endodontics 20, 1114. Lee SJ, Nam KC, Kim YJ, Kim DW (2002) Clinical accuracy of a new apex locator with an automatic compensation circuit. Journal of Endodontics 28, 7069. Lin LM, Rosenberg PA, Lin J (2005) Do procedural errors cause endodontic treatment failure? The Journal of the American Dental Association 136, 18793. Meares WA, Steiman HR (2002) The inuence of sodium hypochlorite irrigation on the accuracy of the Root ZX electronic apex locator. Journal of Endodontics 28, 5958.
Nahmias Y, Aurelio JA, Gerstein H (1987) An in vitro model for evaluation of electronic root canal length measuring devices. Journal of Endodontics 13, 20914. Nekoofar MH, Ghandi MM, Hayes SJ, Dummer PMH (2006) The fundamental operating principles of electronic root canal length measurement devices. International Endodontic Journal 39, 595609. Nguyen HQ, Kaufman AY, Komorowski RC, Friedman S (1996) Electronic length measurement using small and large les in enlarged canals. International Endodontic Journal 29, 35964. Olson AK, Goerig AC, Cavataio RE, Luciano J (1991) The ability of the radiograph to determine the location of the apical foramen. International Endodontic Journal 24, 2835. Ounsi HF, Naaman A (1999) In vitro evaluation of the reliability of the Root ZX electronic apex locator. International Endodontic Journal 32, 1203. Schneider SW (1971) A comparison of canal preparations in straight and curved root canals. Oral Surgery Oral Medicine Oral Pathology 32, 2715. Seidberg BH, Alibrandi BV, Fine H, Logue B (1975) Clinical investigation of measuring working lengths of root canals with an electronic device and with digital-tactile sense. The Journal of the American Dental Association 90, 37987. Seltzer S, Soltanoff W, Smith J (1973) Biologic aspects of endodontics. V. Periapical tissue reactions to root canal instrumentation beyond the apex and root canal llings short of and beyond the apex. Oral Surgery Oral Medicine Oral Pathology 36, 72537. Sjo gren U, Ha gglund B, Sundqvist G, Wing K (1990) Factors affecting the long-term results of endodontic treatment. Journal of Endodontics 16, 498504. Souza RA (2006) The importance of apical patency and cleaning of the apical foramen on root canal preparation. Brazilian Dental Journal 17, 69. Tipler PA, Mosca G (2007) Physics for Scientists and Engineers, 6th edn. New York, NY, USA: W.H. Freeman, pp. 995. Tselnik M, Baumgartner JC, Marshall JG (2005) An evaluation of root ZX and elements diagnostic apex locators. Journal of Endodontics 31, 5079. Welk AR, Baumgartner JC, Marshall JG (2003) An in vivo comparison of two frequency-based electronic apex locators. Journal of Endodontics 29, 497500. Wrbas KT, Ziegler AA, Altenburger MJ, Schirrmeister JF (2007) In vivo comparison of working length determination with two electronic apex locators. International Endodontic Journal 40, 1338.
374
doi:10.1111/j.1365-2591.2008.01539.x
The upregulation of receptor activator NF-jB ligand expression by interleukin-1a and Porphyromonas endodontalis in human osteoblastic cells
S.-C. Chen1,2, F.-M. Huang3, S.-S. Lee4, M.-Z. Li1 & Y.-C. Chang3,5
1 Institute of Medicine, Chung Shan Medical University; 2Section of Metabolism & Endocrinology, Cheng Ching Hospital; 3Oral Medicine Center, Chung Shan Medical University Hospital; 4School of Public Health, Chung Shan Medical University; and 5School of Dentistry, Chung Shan Medical University, Taichung, Taiwan
Abstract
Chen S-C, Huang F-M, Lee S-S, Li, M-Z, Chang Y-C. The
upregulation of receptor activator NF-jB ligand expression by interleukin-1a and Porphyromonas endodontalis in human osteoblastic cells. International Endodontic Journal, 42, 375 380, 2009.
Aim To investigate the receptor activator of nuclear factor-kappa B (NF-jB) ligand (RANKL) in osteoblastic cells stimulated with inammatory mediators. Methodology The expression of RANKL in human osteoblastic cell line U2OS stimulated by pro-inammatory cytokine interleukin (IL)-1a and black-pigmented bacteria Porphyromonas endodontalis was investigated by Western blot and enzyme-linked immunosorbent assay (ELISA). The signicance of the results obtained from control and treated groups was statistically analysed by the paired Students t-test.
Results IL-1a was found to upregulate RANKL production in U2OS cells (P < 0.05). Investigations of the time dependence of RANKL expression in IL-1a-treated cells revealed a rapid accumulation of RANKL protein after 1 h of exposure; it remained elevated throughout the 24-h incubation period shown by Western blot and ELISA. In addition, P. endodontalis also increased RANKL expression in U2OS cells after 4-h incubation period demonstrated by Western blot and ELISA (P < 0.05). Conclusions IL-1a and P. endodontalis may be involved in developing apical periodontitis through the stimulation of RANKL production. Keywords: apical periodontitis, IL-1a, osteoblastic cell, P. endodontalis, RANKL.
Received 9 August 2008; accepted 17 November 2008
Introduction
Apical periodontitis is an inammatory disorder of periradicular tissues caused by persistent microbial infection within the root canal system of the affected tooth (Nair 2006). Continuous ow of bacteria and their products through the apical foramen induces inux, activation and coordinated interaction of immune-inammatory cells within the periapical area.
Correspondence: Dr Yu-Chao Chang DDS. MMS. PhD, Professor, School of Dentistry, Chung Shan Medical University, 110, Sec. 1, Chien-Kuo N. Rd., Taichung, Taiwan (Tel.: 886 4 22015111: fax: 886 4 24759065; e-mail: cyc@csmu.edu.tw).
The pathologic responses involve a complex array of immunologic mechanisms, some of which may act primarily to protect the pulp and periapical region, whereas others mediate periapical tissue destruction, particularly bone resorption (Stashenko et al. 1998). A recently identied tumour necrosis factor (TNF) super family molecule, receptor activator of NF-jB ligand (RANKL), its receptor (RANK) and natural antagonist, osteoprotegerin, have been shown to be the key regulators of bone remodelling and are directly involved in the differentiation, activation and survival of osteoclasts and osteoclast precursors (Simonet et al. 1997, Lacey et al. 1998, Yasuda et al. 1998). Membrane-bound or soluble RANKL is primarily produced in osteoblastic lineages and activated T cells and
375
RANKL induced by IL-1a and P. endodontalis Chen et al.
stimulates osteoclast differentiation (Kong et al. 1999a,b, Suda et al. 1999). Interleukin-1 (IL-1) plays a central role in the regulation of immunological and inammatory reactions. Biological activity of IL-1 molecules seems to be directly relevant to bone resorption (Stashenko et al. 1987). IL-1a has been found to express in periapical lesions (Bando et al. 1993, Honma et al. 1998). IL-1a may play an important role in the pathogenesis of apical periodontitis. Black-pigmented gram-negative anaerobic bacteria, such as Prevotella spp. and Porphyromonas spp. are relative common in infected root canals and endodontic abscesses (Bogen & Slots 1999, Jacinto et al. 2003). Porphyromonas endodontalis (P. endodontalis, formerly Bacteroides endodontalis) is a black-pigmented, anaerobic, rod-shaped bacterium, which is strongly associated with endodontic infection. This species, as well as other black-pigmented anaerobic rods, has been implicated in the aetiology of infected root canals and apical periodontitis (Siqueira et al. 2001, Fouad et al. 2002, Gomes et al. 2005, Tomazinho & AvilaCampos 2007). Porphyromonas endodontalis is capable of degrading several important proteins, such as immunoglobulins, complement factors and heptoglobulin (van Winkelhoff et al. 1992). The invasion of host tissue by P. endodontalis may induce periapical destruction. Surprisingly, there have been relatively few studies addressing the mechanisms of tissue destruction in apical periodontitis. The subsequent reactions leading to periapical injury after the induction of IL-1a and P. endodontalis remains to be elucidated. Menezes et al. (2006) have shown that the presence of RANKL in periapical cyst and granulomas may involve the development of periapical lesions. Moreover, Sabeti et al. (2005) have reported that RANKL may play a role in apical periodontitis-induced bone resorption. However, little information is available about the presence of RANKL and its role in apical periodontitis and associated bone resorption. Osteoblasts are considered as cells primarily concerned with providing physical barriers and structural components in periapical tissues. These cells may be important in the recruitment of immune cells and contribute to the inammation (Yang et al. 2003). In this study, the expression of RANKL in human osteoblastic cell line U2OS cells stimulated by proinammatory cytokine IL-1a and P. endodontalis was investigated by the Western blot and enzyme-linked immunosorbent assay (ELISA).
Materials and methods Chemicals and materials

IL-1a was purchased from Sigma Chemical Co. (St Louis, MO, USA). All culture materials were obtained from GIBCO (Grand Island, NY, USA). IL-1a was directly dissolved in the culture medium. The nal concentration of IL-1a used in this study was 20 ng mL)1.
Bacterial strain and preparation of supernatant

Porphyromonas endodontalis (ATCC 27067) were grown under anerobic conditions and harvested at the end of the logarithmic phase of growth as described previously (Yang et al. 2003, Huang et al. 2005). Briey, they were maintained in brainheart infusion broth prereduced anaerobically, sterilized and supplemented with 5 mg L)1 haemin and 0.5 mg L)1 menadione. The density of the inoculum, prepared in brainheart infusion broth, was adjusted to a turbidity of 2 McFarland standard (6 108 colony-forming units mL)1). After centrifugation, supernatants were ltersterilized using a 0.2-lm lter and stored at )80 C until used. The supernatants of P. endodontalis were directly diluted in culture medium and the nal dilution was 1 : 50 according to a previous study (Chang et al. 2003).
Cell culture
U2OS cells (American Tissue Type Collection HTB 96) were cultured in Dulbeccos modied Eagles medium (DMEM) supplemented with 10% fetal calf serum, 100 lg mL)1 of streptomycin, 100 mg mL)1 of penicillin at 37 C in humidied incubator under ambient pressure air atmosphere containing 5% CO2. Conuent cells were detached with 0.25% trypsin and 0.05% EDTA for 5 min and aliquots of separated cells were subcultured. The cells were subcultured at 1 : 4 splits every third day.
Treatments
Conuent cells were trypsinized, counted and plated at a concentration of 5 104 cells in 60-mm-culture dish and allowed for 48 h to achieve conuence. Prior to treatment, the cells were washed with serum-free DMEM and immediately exposed for the indicated incubation times to 20 ng mL)1 IL-1a and supernatants of P. endodontalis (1 : 50). After different periods
376
Chen et al. RANKL induced by IL-1a and P. endodontalis
of time (0, 1, 2, 4, 8 and 24 h), cell lysates and the conditioned medium samples were collected for Western blot and ELISA respectively.
Triplicate separate experiments were performed throughout this study. The signicance of the results obtained from control and treated groups was statistically analysed by the paired Students t-test. A P-value of <0.05 was considered to be statistically signicant.
Western blot
Cell extracts were solubilized with SDS-solubilization buffer (5 mmol L)1 EDTA, 1 mmol L)1 MgCl2, 50 mmol L)1 TrisHCl, pH 7.5 and 0.5% Trition X100, 2 mmol L)1 phenylmethysulfonyl uoride and 1 mmol L)1 N-ethylmaleimide) for 30 min on ice. Then, cell lysates were centrifuged at 12 000 g at 4 C and the protein concentrations determined with Bradford reagent using bovine serum albumin (BSA) as standards. Equivalent amounts of total protein per sample of cell extracts were run on a 10% SDS-PAGE and immediately transferred to nitrocellulose membranes. The membranes were blocked with PBS containing 3% BSA for 2 h, rinsed and then incubated with primary antibodies anti-RANKL (Chemicon International Inc., Temecula, CA, USA) and diluted 1 : 1000 in PBS containing 0.05% Tween 20 for 2 h. After three washes with Tween 20 for 10 min, the membranes were incubated for 1 h with biotinylated secondary antibody diluted 1 : 2000 in the same buffer, washed again as described above and treated with 1 : 2000 streptavidin-peroxidase solution for 30 min. After a new series of washing steps, the reactions were developed using diaminobenzidine (Zymed, South San Francisco, CA, USA). All steps were done at room temperature. As a loading control b-actin (Santa Cruz Biotechnology, CA, USA) was included. The intensities of the obtained bands were determined using a densitometer (AlphaImager 2000; AlphaInnotech, San Liandro, CA, USA). Each densitometric value, expressed as the mean SD, was obtained from three independent experiments.
Results
IL-1a was found to upregulate RANKL production in U2OS cells (P < 0.05). Investigations of the time dependence of RANKL expression in IL-1a-treated cells revealed a rapid accumulation of the RANKL protein after 1 h of exposure that remained elevated throughout the 24-h incubation period (Fig. 1a). The quantitative measurement by the AlphaImager 2000 was shown in Fig. 1b. The levels of the RANKL increased about 2.1-, 2.4-, 2.7-, 2.8- and 3.7-fold after exposure to IL-1a for 1, 2, 4, 8 and 24 h respectively. In addition, the results of Western blot were conrmed by ELISA. Similar pattern was seen by Western blot. The production of RANKL was enhanced by the IL-1a as compared with control (P < 0.05). As shown in Fig. 2, the amounts of RANKL were about
Enzyme-linked immunosorbent assay

Levels of RANKL antigen were determined by ELISA (R&D systems, Minneapolis, MN, USA). Briey, 20 lL of conditioned media were directly transferred to the microtest strip wells of the ELISA plate. All further procedures were performed following the manufacturers instructions. The absorbance at 495 nm was measured in a microtest plate spectrophotometer and RANKL levels were determined with a calibration curve using human RANKL as a standard. The amounts of RANKL were expressed as pg L)1 protein. Each value was expressed as the mean SD.
Figure 1 (a) Kinetics of receptor activator of nuclear factorkappa B (NF-jB) ligand (RANKL) expression in U2OS cells exposed to IL-1a for 0, 1, 2, 4, 8 and 24 h respectively. b-actin was used as a loading control. (b) Levels of RANKL treated with IL-1a were measured by densitometric gel imaging. *Signicant difference from control values with P < 0.05.
377
Figure 2 Expression the protein levels receptor activator of nuclear factor-kappa B (NF-jB) ligand of conditioned medium from U2OS cells treated with the IL-1a for 0, 1, 2, 4, 8 and 24 h respectively. Each value was expressed as the mean SD of optical density from triplicate independent experiments. *Signicant difference from control values with P < 0.05.
107, 127, 135, 141, 175 and 281 pg mL)1 after exposure to IL-1a for 0, 1, 2, 4, 8 and 24 h respectively. Porphyromonas endodontalis was also found to increase RANKL production in U2OS cells (P < 0.05). The kinetics of this response showed that RANKL was signicantly increased in cell lysates after 4-h post P. endodontalis challenge and remained elevation throughout the 24-h incubation period (Fig. 3a). The quantitative measurement by the AlphaImager 2000 is shown in Fig. 3b. RANKL levels increased about 1.7-, 2.3- and 3.1-fold after exposure to P. endodontalis for 4, 8 and 24 h respectively. The results from ELISA demonstrated that the production of RANKL was enhanced by P. endodontalis after the 4 h treatment period as compared with control (P < 0.05). As shown in Fig. 4, the amounts of RANKL were about 108, 110, 119, 137, 161 and 208 pg mL)1 after exposure to endodontalis for 0, 1, 2, 4, 8 and 24 h respectively.
Figure 3 (a) Kinetics of receptor activator of nuclear factorkappa B (NF-jB) ligand (RANKL) expression in U2OS cells exposed to Porphyromonas endodontalis for 0, 1, 2, 4, 8 and 24 h respectively. b-actin was included to monitor equal protein loading. (b) Levels of RANKL treated with P. endodontalis were measured by densitometric gel imaging. *Signicant difference from control values with P < 0.05.
Discussion
Diagnosis and treatment of inammatory bone resorption induced by apical periodontitis present a clinical challenge. This process involves a local inammatory reaction. However, the exact mechanism and sequence of events associated with the activation of osteoclastic activity to produce bone resorption-induced apical periodontitis has not been fully elucidated. It was suggested that a novel secreted glycoprotein ligand known as RANKL may play a critical role in the
Figure 4 Expression the protein levels receptor activator of nuclear factor-kappa B (NF-jB) ligand of conditioned medium from U2OS cells treated with the Porphyromonas endodontalis for 0, 1, 2, 4, 8 and 24 h respectively. Each value was expressed as the mean SD of optical density from triplicate independent experiments. *Signicant difference from control values with P < 0.05.
development of osteoclasts that result in bone resorption (Fuller et al. 1998). Recently, RANKL was found to be expressed in apical periodontitis (Sabeti et al. 2005). However, there is little information about the RANKL expression in inammatory cytokine and endodontopathic bacteria-induced apical periodontitis.
378
Chen et al. RANKL induced by IL-1a and P. endodontalis
Studies have reported that RANKL is expressed in periodontal ligament cells, osteoblasts, osteoclasts, stromal cells, T lymphocytes, endothelial cells and epithelial cells (Boyle et al. 2003). In this study, it was noted that RANKL was expressed intracellularly by Western blot and extracellularly by ELISA in an osteoblastic cell line U2OS cell. Similar results were reported by Mancini et al. (2007) who reported that U2OS cells can express RANKL mRNA and protein in vitro. Consistently, RANKL was found to be expressed in specimens from apical periodontitis (Sabeti et al. 2005). Thus, osteoblasts might be one of the cells that can express RANKL within apical periodontitis lesions. It is accepted that IL-1 has multiple biological activities, such as the acceleration of bone resorption. IL-1a has been found to play an important role in the pathogenesis of periapical lesions (Bando et al. 1993, Honma et al. 1998). In this study, it was found that the expression of RANKL was upregulated by the IL-1a in U2OS cells. These results were in agreement with those of the previous studies, which IL-1a is known to modulate RANKL expression (Fukushima et al. 2005, Wei et al. 2005). Taken together, these ndings suggest that IL-1a may be involved in developing apical periodontitis through the upregulation of RANKL production. Bacteria and their products are the primary cause of pulpal necrosis and periapical lesions. How bacteria stimulate bone pathosis is still unclear. In this study, it was found initially that the expression of RANKL was upregulated by P. endodontalis in U2OS cells. Consistently, similar results were found evaluating the effects of several components of black-pigmented bacteria P. gingivalis on mRNA expression of RANKL (Okahashi et al. 2004, Kobayashi-Sakamoto et al. 2004). Taken together, P. endodontalis might promote cellular processes that stimulate the degradation of bone via RANKL expression. In interpreting the results presented here, it is interesting to consider that IL-1a and P. endodontalis may lead to the upregulation of RANKL production in U2OS cells. The presence of excessive amounts of RANKL derived from U2OS cells may inuence the inammatory response by virtue of its ability to activate immune cells. Therefore, the capacity of U2OS cells to produce RANKL in response to IL-1a and P. endodontalis suggests that human osteoblastic cells additionally contributes to the orchestration of immuno-participant cells in the host defence network of apical periodontitis. Further studies seem necessary to identify not just the
in situ localization of RANKL in apical periodontitis but also using a functional assay.
References
Bando Y, Henderson B, Meghji S, Poole S, Harris M (1993) Immunocytochemical localization of inammatory cytokines and vascular adhesion receptors in radicular cyst. Journal of Oral Pathology & Medicine 22, 2217. Bogen G, Slots J (1999) Black-pigmented anaerobic rods in closed periapical lesions. International Endodontic Journal 32, 20410. Boyle WJ, Smonet WS, Lacey DL (2003) Osteoclast differentiation and activation. Nature 423, 33742. Chang YC, Huang FM, Yang SF et al. (2003) Induction of cyclooxygenase-2 mRNA and protein expression in human pulp cells stimulated with black-pigmented Bacteroides. Journal of Endodontics 29, 2403. Fouad AF, Barry J, Caimano M et al. (2002) PCR-based identication of bacteria associated with endodontic infections. Journal of Clinical Microbiology 40, 322331. Fukushima H, Jimi E, Okamoto F, Motokawa W, Okabe K (2005) IL-1-induced receptor activator of NFkappaB ligand in human periodontal ligament cells involves ERK-dependent PGE2 production. Bone 36, 26775. Fuller K, Wong B, Fox S, Choi Y, Chambers TJ (1998) Trance is necessary and sufcient for osteoblast-mediated activation of bone resorption in osteoclast. Journal of Experimental Medicine 188, 9971001. Gomes BPFA, Jacinto RC, Pinheiro ET et al. (2005) Porphyromonas gingivalis, Porphyromonas endodontalis, Prevotella intermedia and Prevotella nigrescens in endodontic lesions detected by culture and by PCR. Oral Microbiology & Immunology 20, 2115. Honma M, Hayakawa Y, Kosugi H, Koizumi F (1998) Localization of mRNA for inammatory cytokines in radicular cyst tissue by in situ hybridization, and induction of inammatory cytokines by human gingival broblasts in response to radicular cyst contents. Journal of Oral Pathology & Medicine 27, 399404. Huang FM, Chen YJ, Chou MY, Chang YC (2005) Examination of the signal transduction pathways leading to upregulation of tissue type plasminogen activator by Porphyrmonas endodontalis in human pulp cells. International Endodontic Journal 38, 8605. Jacinto RC, Gomes BPFA, Ferraz CCR, Zaia AA, Souza-Filho FJ (2003) Microbiological analysis of infected root canals from symptomatic and asymptomatic teeth with periapical periodontitis and the antimicrobial susceptibility of some isolated anaerobic anaerobic bacteria. Oral Microbiology & Immunology 18, 28592. Kobayashi-Sakamoto M, Hirose K, Isogai E, Chiba I (2004) NFkappaB-dependent induction of osteoprotegerin by Porphyromonas gingivalis in endothelial cells. Biochemic & Biophysic Research Communication 315, 10712.
379
Kong YY, Yoshida H, Sarosi I et al. (1999a) OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397, 31523. Kong YY, Feige U, Sarosi I et al. (1999b) Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 402, 3049. Lacey DL, Timms E, Tan HL et al. (1998) Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93, 16576. Mancini L, Paul-Clark MJ, Rosignoli G et al. (2007) Calcitonin and prednisolone display antagonistic actions on bone and have synergistic effects in experimental arthritis. The American Journal of Pathology 170, 101827. Menezes R, Bramante CM, da Silva KB et al. (2006) Receptor activator NFjB-ligand and osteoprotegerin protein expression in human periapical cysts and granulomas. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiolology & Endodontics 102, 4049. Nair PNR (2006) On the causes of persistent apical periodontitis: a review. International Endodontic Journal 39, 24981. Okahashi N, Inaba H, Nakagawa I et al. (2004) Porphyromonas gingivalis induces receptor activator of NFkappaB ligand expression in osteoblasts through the activator protein 1 pathway. Infection & Immunity 72, 170614. Sabeti M, Simon J, Kermani V, Valles Y, Rostein I (2005) Detection of receptor activator of NF-jB ligand in apical periodontitis. Journal of Endodontics 31, 178. Simonet WS, Lacey DL, Dunstan CR et al. (1997) Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89, 30919. Siqueira JF Jr, Rocas IN, Oliveira JCM, Santos KRN (2001) Molecular detection of black-pigmented bacteria in infections of endodontic origin. Journal of Endodontics 27, 5636.
Stashenko P, Dewhirst FE, Peros WJ, Kent RL, Ago JM (1987) Synergistic interactions between interleukin 1, tumor necrosis factor, and lymphotoxin in bone resportion. Journal of Immunology 138, 1648. Stashenko P, Teles R, DSouza R (1998) Periapical inammatory responses and their modulation. Critical Review of Oral Biology and Medicine 9, 498521. Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT, Martin TJ (1999) Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocrine Review 20, 34557. Tomazinho LF, Avila-Campos MJ (2007) Detection of Porphyromonas gingivalis, Porphyromonas endodontalis, Prevotella intermedia, and Prevotella nigrescens in chronic endodontic infection. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology & Endodontology 103, 2858. Wei S, Kitaura H, Zhou P, Ross FP, Teitelbaum SL (2005) IL-1 mediates TNF-induced osteoclastogenesis. Journal of Clinical Investigations 115, 28290. van Winkelhoff AJ, van Steenbergen TJM, De Graaff J (1992) Porphyromonas (Bacteroides) endodontalis: its role in endodontal infections. Journal of Endodontics 18, 4314. Yang LC, Huang FM, Lin CS, Liu CM, Lai CC, Chang YC (2003) Induction of interleukin-8 gene expression by black-pigmented Bacteroides in human pulp broblasts and osteoblasts. International Endodontic Journal 36, 774 9. Yasuda H, Shima N, Nakagawa N et al. (1998) Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/ RANKL. Proceedings of National Academy Sciences United State of American 95, 3597602.
380

Iej 04 2009

Cargado por

Información del documento

Descripción original:

Título original

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

Iej 04 2009

Cargado por

Copyright:

Formatos disponibles

doi:10.1111/j.1365-2591.2008.01537.

The mysterious appearance of enterococci in lled root canals

M. Zehnder1 & B. Guggenheim2

2009 International Endodontic Journal

International Endodontic Journal, 42, 277287, 2009

The case of the Enterococcus Zehnder & Guggenheim

Is it really difcult to eliminate enterococci from the root canal?

International Endodontic Journal, 42, 277287, 2009

2009 International Endodontic Journal

Zehnder & Guggenheim The case of the Enterococcus

Enterococci in dentinal caries

2009 International Endodontic Journal

International Endodontic Journal, 42, 277287, 2009

The case of the Enterococcus Zehnder & Guggenheim

Enterococci in primary root canal infections

International Endodontic Journal, 42, 277287, 2009

2009 International Endodontic Journal

Zehnder & Guggenheim The case of the Enterococcus

Enterococcal invasion of the root canal during or after treatment

2009 International Endodontic Journal

International Endodontic Journal, 42, 277287, 2009

The case of the Enterococcus Zehnder & Guggenheim

Are enterococci colonizers of the oral cavity?

International Endodontic Journal, 42, 277287, 2009

2009 International Endodontic Journal

Zehnder & Guggenheim The case of the Enterococcus

2009 International Endodontic Journal

International Endodontic Journal, 42, 277287, 2009

The case of the Enterococcus Zehnder & Guggenheim

Concluding remarks and call for future studies

International Endodontic Journal, 42, 277287, 2009

2009 International Endodontic Journal

Zehnder & Guggenheim The case of the Enterococcus

2009 International Endodontic Journal

International Endodontic Journal, 42, 277287, 2009

The case of the Enterococcus Zehnder & Guggenheim

International Endodontic Journal, 42, 277287, 2009

2009 International Endodontic Journal

Zehnder & Guggenheim The case of the Enterococcus

2009 International Endodontic Journal

International Endodontic Journal, 42, 277287, 2009

The properties and applications of chlorhexidine in endodontics

Z. Mohammadi1,2 & P. V. Abbott3

Received 1 May 2008; accepted 18 November 2008

International Endodontic Journal, 42, 288302, 2009

2009 International Endodontic Journal

Mohammadi & Abbott Chlorhexidine in endodontics

Structure and mechanism of action

2009 International Endodontic Journal

International Endodontic Journal, 42, 288302, 2009

Chlorhexidine in endodontics Mohammadi & Abbott

International Endodontic Journal, 42, 288302, 2009

2009 International Endodontic Journal

Mohammadi & Abbott Chlorhexidine in endodontics

CHX and biolms

2009 International Endodontic Journal

International Endodontic Journal, 42, 288302, 2009

Chlorhexidine in endodontics Mohammadi & Abbott

International Endodontic Journal, 42, 288302, 2009

2009 International Endodontic Journal

Mohammadi & Abbott Chlorhexidine in endodontics

Modulating effect of dentine on CHX