Documentos de Académico
Documentos de Profesional
Documentos de Cultura
Persistent phantom pain syndromes American literature, with graphic descrip- (31, 37) suggest that in the year after am-
are difficult to manage, leading to frus- tions of Captain Ahab’s phantom limb in putation, 60% to 70% of amputees expe-
tration of physicians and patients alike. Moby Dick. rience phantom limb pain, but it dimin-
Phantom sensation or pain is the persis- Phantom sensation, phantom pain, ishes with time (14, 31). The incidence of
tent perception that a body part exists or and stump pain are the three most com- phantom limb pain increases with more
is painful after it has been removed by am- monly utilized terms. Phantom sensa- proximal amputations. The reports of
putation or trauma. Ambrose Paré (1, 2), tions may occur in any part of the body phantom limb pain after hemipelvecto-
a french military surgeon, provided the but are most often described in the ex- my ranged from 68% to 88% and follow-
first medical description of postamputa- tremities (4-11). ing hip disarticulation, 40% to 88% (28,
tion phenomena. He noticed, as early as 30). However, wide variations exist with
1551, that amputees may complain of se- EPIDEMIOLOGY reports of phantom limb pain after lower
vere pain in the missing limb a long time Phantom limb sensation in 85% to extremity amputation as high as 72% (21)
after amputation. Civil War surgeon, Silas 98% of amputees is seen in the first 3 and as low as 51% after upper limb ampu-
Weir Mitchell (3) in 1871 popularized the weeks after amputation (12), whereas in a tation (22). Further, 0% prevalence was
concept of phantom limb pain and coined small proportion of the patients (approxi- reported in below knee amputations com-
the term phantom limb with publication mately 8%), phantom limb sensation may pared to 19% in above the knee amputa-
of a long-term study on the fate of Civ- not occur until 1 to 12 months following tions (30). Phantom limb pain has been
il War amputees. However, Herman Mel- amputation (13). Most phantom sensa- reported to occur as early as one week af-
ville immortalized phantom limb pain in tions generally resolve after 2 to 3 years ter amputation and as late as 40 years af-
without treatment, except in the cases ter amputation (4, 33, 34). Phantom pain
where phantom pain develops. Phantom may diminish with time and eventual-
From Pain Management Center of Paducah, limb sensation is strongest in amputations ly fade away. However, some prospec-
Paducah, Kentucky, and Pain Diagnostic Associ- above the elbow and weakest in amputa- tive studies indicate that even 2 years af-
ates, Niagara, Wisconsin. Address Correspondence:
Laxmaiah Manchikanti, MD, 2831 Lone Oak Road, tions below the knee (14), and is more ter amputation, the incidence is almost
Paducah, Kentucky 42003 frequent in the dominant limb of double the same as at onset (31, 37). It is report-
E-mail: drm@apex.net amputees (15). ed that almost 60% of patients continue
Funding: There was no external funding in prepara-
tion of this manuscript.
The incidence of phantom limb to have phantom limb pain (24, 31) after
Conflict of Interest: None pain has been reported to vary from 0% one year, whereas in the first month fol-
to 88% (16-32). Prospective evaluations lowing amputation, 85% to 97% of pa-
tients experience phantom limb pain (24, perimental support has been provided for reduced by the NMDA receptor antago-
29, 30). While phantom limb pain may these clinical observations. Peripherally, nist ketamine (49). Besides functional
begin months to years after an amputa- spontaneous and abnormal evoked activ- changes in the dorsal horn, an anatomi-
tion, pain starting after one year follow- ity following mechanical or neurochemi- cal reorganization also has been described
ing amputation occurs in fewer than 10% cal stimulation are observed in nerve-end (50). It has been shown that peripheral
of patients (4). neuromas (41, 42). This increased activity nerve transection results in a substantial
Stump pain is reported in up to 50% is assumed to be the result of a novel ex- degeneration of afferent C-fiber terminals
of amputees (16, 18, 21-23, 35-37). Re- pression or upregulation of sodium chan- in lamina II, thus reducing the number of
ports showed that 50% to 88% of the pa- nels (43, 44). Thus, the increased sensitiv- synaptic contacts with second-order neu-
tients with phantom pain also reported ity of neuromas to norepinephrine may in rons in lamina II, which normally respond
stump pain (25, 30). part explain the exacerbation of phantom best to noxious stimulation. Consequent-
Phantom limb pain is also associat- pain by stress and other emotional states ly, central terminals of Aβ mechanorecep-
ed with multiple pain problems in other associated with increased catecholamine tive afferents, which normally terminate
areas of the body, with reports indicat- release from sympathetic efferent termi- in deeper laminae, sprout into lamina II
ing headache or pain in joints in 35% of nals which are in close proximity to af- and may form synaptic contacts with va-
the patients, sore throat in 28% of the pa- ferent sensory nerves and sprouts (40). cant nociceptive second-order neurons.
tients, abdominal pain in 18%, and back It was shown that cell bodies in the dor- As a result of this organization, evocation
pain in 13% (38). sal root ganglion show similar abnormal of pain is seen with simple touch, etc., by
spontaneous activity and increased sen- Aβ-fiber input.
ETIOLOGY sitivity to mechanical and neurochemical The third step in the process is the
stimulation (45). Thus, abnormal activi- supraspinal or central mechanism. Based
Pathophysiology ty from neuromas and dorsal root gangli- on peripheral and spinal cord mecha-
Among phantom sensations, phan- on cell bodies may contribute to the phan- nisms, it is reasonable to assume that am-
tom pain, and stump pain, phantom sen- tom limb percept, including pain. putation not only produces a cascade of
sations are the easiest to explain. It is be- The second mechanism is consid- events in the periphery and in the spinal
lieved that, throughout life an individual’s ered to be at the spinal cord level. The in- cord, but these changes eventually sweep
body image develops from proprioceptive, creased barrage from neuromas and from more centrally and alter neuronal activi-
tactile, and visual inputs (39). Thus, once dorsal root ganglia cells is thought to in- ty in cortical and subcortical structures. It
a cortical representation of the body im- duce long-term changes in central pro- has been shown that thalamic stimulation
age is established, it is unchanged follow- jecting neurons in the dorsal horn, in- results in phantom sensation and pain in
ing limb amputation (4, 7). cluding spontaneous neuronal activity, amputees (51). This suggests that plastic
The etiology and pathophysiologi- induction of immediate early genes, in- changes in the thalamus are involved in
cal mechanisms of phantom pain are not creases in spinal cord metabolic activi- the generation of chronic pain, as normal-
clearly defined. However, both peripher- ty, and expansion of receptive fields (46, ly such stimulation does not evoke pain.
al and central neuronal mechanisms are 47). Nikolajsen and Jensen (40) described Other studies in humans have document-
likely to occur. In addition, psychological that the pharmacology of spinal sensiti- ed a cortical reorganization after amputa-
mechanisms have been proposed. How- zation involves increased activity in N- tion using multiple cerebral imaging tech-
ever, none of the theories independently, methyl-D-aspartate (NMDA) receptor- niques (52-67).
fully explain the clinical characteristics of operated systems (48), and many aspects Finally, psychological theories have
this condition. of the central sensitization can be reduced been forwarded as the explain phantom
Nikolajsen and Jensen (40) described by NMDA receptor antagonists. This pain. While a biopsychosocial mechanism
several clinical observations (Table 1) that was further confirmed in human ampu- may be involved in the development and
suggest that mechanisms in the periphery, tees with one aspect of such central sen- persistence of phantom pain, no consis-
either in the stump or in the central parts sitization, the evoked stump or phantom tent pattren of personality disorders or
of sectioned primary afferents may play pain produced by repetitive stimulation clinical syndromes have been shown to be
a role in the phantom limb percept. Ex- of the stump by non-noxious pin prick, increased in patients with phantom limb
pain. However, psychological disturbanc-
Table 1. Clinical observations of phantom limb pain es related to the loss of a limb or feelings
of dependence, as well chronic pain and
Phantom limb sensations can be modulated by various stump manipulations. disability, may lead to a host of psycho-
Phantom limb sensations are temporarily abolished after local stump anesthesia. logical problems in these patients (68-
73). Patients reporting phantom limb
Stump revisions and removal of tender neuromas often reduce pain, at least transiently. pain have been shown to be more rigid,
Phantom pain is significantly more frequent in those amputees with long-term stump pain compulsive, and self-reliant than their co-
than in those without persistent pain. horts (14).
Although obvious stump pathology is rare, altered cutaneous sensibility in the stump is a Etiology of stump pain is often as-
common if not universal feature. sociated with definite pathological find-
ings that may account for the pain in the
Changes in stump blood flow alter the phantom limb perception.
stump and/or the phantom limb, such as
Adapted from Nikolajsen and Jensen (40) skin pathology, circulatory disturbances,
• Angina may be presented as only 12 trials. Of the 12 trials, only 3 ran- on the basis of the analyzed data. They
exacerbated phantom limb pain (96, domized, controlled studies with parallel concluded that perioperative epidural an-
97) groups and 3 randomized cross-over tri- esthesia has been shown to be an effec-
• Finally, in patients undergoing als were identified. They also mentioned tive prophylaxis of phantom limb pain.
amputation secondary to malignant particular challenges associated with ex- However, perioperative epidural anes-
disease, if phantom limb pain amining phantom limb pain, with an ex- thesia does not completely abolish phan-
increases significantly, metastatic tremely low rates of amputations, high tom limb pain, but increases the number
disease should be evaluated. mortality rates among the amputees, and of patients with a mild form of phantom
finally, interventions designed to examine pain.
MANAGEMENT operative and perioperative treatments Investigators in 4 trials (108-111)
Treatment of phantom limb pain may be ethically unacceptable. assessed preoperative epidural pain re-
or stump pain is difficult and has gen- lief and were unable to provide definitive
erally not been very successful. Halbert Prevention evidence to support its routine use. The
et al (98) conducted a systematic review An increasing knowledge about the results of two studies involving a small
to evaluate the evidence for the opti- mechanisms involved in the develop- number suggested that epidural analge-
mal management of acute and chron- ment and perpetuation of neuropathic sia may help but were inconsistent: one
ic phantom pain. They concluded that pain theoretically should allow us a ratio- showed relief at 7 days, 6 months, and
there is currently a gap between research nal approach to its prevention. However, 1 year postoperatively (109), the second
and practice in the area of phantom limb the initially hopeful attempts like the use study (108) showed less phantom limb
pain. Nevertheless, in the past decade, of pre and postsurgical epidural block- pain in the intervention group at 1 week,
clinical trials have examined treatments ade have been questioned and its real util- 6 months, and 1 year, and the difference
for phantom limb pain. Surveys suggest ity now appears to be controversial (106). reached significance only at 6 months.
that although physicians believe treat- Advances in neuroimaging techniques are The largest of the studies (110) showed
ments are effective (99), fewer than 10% just now unveiling some keys to the prob- no difference in phantom pain at 7 days,
of patients with phantom limb pain re- lem. The current emphasis is put on the 3 months, 6 months, and 12 months. In
ceive lasting relief from prescribed med- adaptive processes taking place in the cen- a randomized prospective study by Lam-
ical treatments (27). Even then, clini- tral nervous system following a deaffer- bert et al (111), 30 patients scheduled for
cians have been restricted by the lack entation. In this sense, it seems that our lower limb amputation were randomly as-
of clinical trials that would aid in treat- ability to prevent post-amputation pain signed epidural bupivacaine or an intra-
ment decisions and by the absence of ev- will depend on our capability to modulate operatively placed perineural catheter for
idence-based treatment guidelines. In a the plasticity of the central nervous sys- intra and postoperative administration of
literature review in 1980, 43 methods for tem. Feria (106) suggested that the prob- bupivacaine. All patients had general an-
treating phantom limb pain were identi- lem needs a broad-based approach in- esthesia. The results showed there was no
fied. However, it was concluded that few cluding control of perioperative pain and significant difference between periopera-
produced relief and that placebo respons- inflammation, adequate follow-up of the tive epidural block and perineural infu-
es were common (100). Multiple authors patients, correct surgical technique, long- sion of local anesthetic. Phantom pain af-
also have recommended treatment for term rehabilitation, and the use of phar- ter 3 days in the epidural group was 29%,
phantom limb pain in line with the man- macological and behavioral approaches at 6 months it was 63%, and at 12 months
agement of neuropathic pain states (101- reflecting current knowledge. it was 38%. Thus, it is not known wheth-
103). However, literature review suggests Multiple authors have attempted er epidural anesthesia reduced the preva-
that trials of treatments for neuropathic psychological preparation, drug therapy, lence of phantom limb pain.
pain rarely included patients with phan- epidural anesthesia, and regional nerve
tom limb pain. blocks, among others, to reduce the oc- Regional Anesthesia
Early trials concentrated on reduc- currence of phantom limb pain and to de- Multiple trials have assessed perineu-
tion of established postoperative phan- lay or stop the process of progressing from ral (111-113), and intraneural (114) bupi-
tom limb pain, but newer approaches acute to chronic pain. At least some of the vacaine blocks, either at the time of sur-
have used analgesic agents administered postamputation pain may be prevented by gery or immediately postoperatively. De-
before amputation (104). Treatment ap- appropriate psychological preparation of spite some early benefits, there was no
proaches continue to be based on the as- the patients. difference in pain between the interven-
sumption that long-term phantom limb tion and control groups in the postoper-
pain is the result of functional or struc- Epidural Anesthesia ative period (112, 113). Perineural block
tural changes in the central nervous sys- Gehling and Tryba (107) showed was similar to infusion of local anesthetic
tem in response to noxious somatosen- that pre-, intra-, and postoperative epi- through an epidural catheter (111). Eval-
sory input (105). Thus, therapies are di- dural anesthesia was associated with a sig- uation of continuous brachial plexus an-
rected at early reduction of pain. nificant reduction of phantom limb pain algesia showed prevention of phantom
Halbert et al (98) noted that their 12 months after amputation. However, limb pain, which did not reappear during
review was limited by the poor quality they concluded that a reduction of phan- follow-up of 1 year (115). Nerve sheath
of the included trials. While they identi- tom limb pain by postoperative epidural catheter analgesia also showed reduced
fied 186 articles, they were able to utilize anesthesia alone could not be confirmed prevalence (116).
Other Interventions triptyline versus nefazodone for the man- arms of the study. Both placebo and gaba-
Other treatments assessed for pre- agement of neuropathic pain. Of the 120 pentin treatments resulted in reduced VAS
vention of phantom limb pain includ- patients included in this study, less than scores compared with baseline. However,
ed administration of calcitonin, ket- 10 patients suffered with phantom limb the pain intensity difference was signifi-
amine, intervenous lidocaine, and trans- pain. The quality of pain was burning and cantly greater than placebo for gabapen-
cutaneous electrical nerve stimulation cutting in 62.3% of the cases, lancinating tin therapy at the end of the treatment.
(117-121). Intravenous calcitonin in one in 40%, and sharp in 25%. The results They concluded that after 6 weeks, gaba-
study (121) evaluating 8 patients showed showed that after 3 months of therapy, pentin monotherapy was better than pla-
that only 2 of 8 patients developed phan- the amitriptyline group showed a pain se- cebo in relieving postamputation phan-
tom limb pain after 10 days of intrave- verity of 2 + 0.9 and in nefazodone group, tom limb pain. There were no significant
nous treatment with salmon calcitonin, 3 + 1.1. Pain relief was greater than 75% differences in mood, sleep interference, or
with prevalence of phantom limb pain re- (excellent) in 42 patients treated with am- activities of daily living. Serpell et al (130)
maining at 25% in systematic follow-up itriptyline and in 36 patients treated with evaluated the use of gabapentin in neu-
at 3, 6, and 12 months. However, in an- nefazodone, between 50% to 75% (good) ropathic pain in a randomized, double-
other study (117), intravenous calcitonin in 18 patients treated with amitriptyline blind, placebo-controlled trial of 305 pa-
reduced phantom limb pain in the early and in 12 patients treated with nefazo- tients in a wide range of neuropathic pain
postoperative period, but phantom limb done, and below 50% (poor) in 3 patients syndromes, including phantom limb pain
pain on longer-term follow-up was not treated with amitriptyline and 3 patients in 2% of these patients. They concluded
adequately controlled. The effectiveness treated with nefazodone. They conclud- that at an average dose of 900 mg to 2400
of ketamine was studied in a prospective, ed that both drugs were effective for the mg per day, gabapentin was well tolerated
observational study with historical con- management of neuropathic pain. The and was associated with significant pain
trols with 14 patients in each group (120). group treated with nefazodone showed control with few secondary effects – dizzi-
However, the results showed that phan- least incidence of side effects, except for ness and somnolence, most of which were
tom limb pain remained high at 72%, nausea and vomiting. The amitriptyline transient and occurred during the titra-
even though only 9% of the patients af- group showed a significant incidence of tion phase.
ter ketamine compared to 71% of the pa- orthostatic hypotension, dry mouth, nau- Analgesic effects of intravenous li-
tients in the control group, complained of sea, and vomiting. docaine and morphine on postamputa-
severe phantom limb pain. Transcutane- Historically, carbamazepine is the tion pain were evaluated in a randomized
ous electrical nerve stimulation was as- most commonly used anti-convulsant double-blind, active placebo-controlled,
sessed in the 2-week postoperative peri- (126, 127). Elliott et al (126) and Pat- cross-over trial by Wu et al (119) An intra-
od, with the treated group reporting less terson (127) reported cases of lancinat- venous bolus followed by an intravenous
pain at 4 weeks (118). However, by 12 ing phantom limb pains that improved infusion of morphine, lidocaine, and the
months, there was no difference between with oral carbamazepine. Logan (128) re- active placebo (diphenhydramine), were
the groups. ported incomplete relief with carbamaze- performed on three consecutive days.
pine but complete relief with chlorproma- The results showed that 31 of 32 subjects
PAIN MANAGEMENT zine in long-standing phantom limb pain. enrolled completed the study. Eleven sub-
There is no evidence that carbamazepine jects had both stump and phantom pains,
Drug Therapy is effective for pains that are not of the in- 11 and 9 subjects had stump and phantom
Medical therapy is the most com- tense, brief, lancinating type. pain alone, respectively. They conclud-
monly utilized modality of treatment Currently, gabapentin is the most ed that stump pain was diminished both
for phantom pain syndromes. The most commonly anti-convulsant used for phan- by morphine and lidocaine, while phan-
commonly used classes of medications are tom limb pain. Other than sedation, side tom pain was diminished only by mor-
anti-depressants and anti-convulsants. A effects are rare and patients become toler- phine, suggesting that the mechanisms
large number of randomized, controlled ant to sedation with time. Since there is and pharmacological sensitivity of stump
clinical trials have shown a beneficial ef- no known long-term toxicity, monitoring and phantom pains are different.
fect of tricyclic anti-depressants and sodi- of blood levels, as with other anti-convul- The effect of an NMDA receptor an-
um channel blockers under different neu- sants is not necessary. The effectiveness tagonist have been examined in different
ropathic pain conditions. Even though no of gabapentin in postamputation phan- studies (51, 131-134). In a double-blind,
controlled trials in phantom pain have tom limb pain was studied in a random- placebo-controlled study, intravenous
been performed, the drugs are general- ized, double-blind, placebo-controlled, ketamine reduced pain, hyperalgesia, and
ly considered to be effective – at least in cross-over study by Bone et al (129). They “wind-up” like pain in 11 amputees with
some patients (122-127). Tricyclic anti- evaluated analgesic efficacy of gabapentin stump and phantom pain (52). In an-
depressants have been thoroughly studied in phantom limb pain in patients attend- other controlled trial by the same authors
in other denervation syndromes, such as ing a multidisciplinary pain clinic. Each (132), 19 patients received memantine, an
post herpetic neuralgia and diabetic neu- treatment was for 6 weeks separated by a NMDA receptor antagonist available for
ropathy (125). However, there have been 1-week washout. The daily dose of gaba- oral use, in a blinded, placebo-controlled,
no studies of their use in treatment of pentin was titrated in increments of 300 cross-over fashion. Memantine failed to
phantom limb pain specifically. mg to 2400 mg or the maximum tolerat- have any effect on spontaneous pain, allo-
Canovas et al (122) assessed the an- ed dose. Nineteen eligible patients were dynia, and hyperalgesia. In another ran-
algesic effectiveness and tolerance of ami- randomized, of whom 14 completed both domized, double-blinded, placebo-con-
trolled trial (134), memantine failed to shown to reduce pain significantly. in pain lasting less than 10 hours (154).
demonstrate a significant clinical bene- Yet, other authors reported good to ex-
fit of the NMDA receptor antagonist in Neural Blockade cellent results in only 25% of the patients
chronic phantom limb pain. Nerve blocks are commonly used treated with TENS (155). Stimulation of
Beta-adrenergic blockers have also in the treatment of phantom limb pain, the contralatateral extremity with TENS
been suggested for treatment of phan- and physicians performing these blocks also has been shown to have a favorable
tom limb pain based on three cases (135). report a high success rate, though it has response in some patients (156, 157).
However, in a double-blind cross-over tri- not been substantiated (99). These range Recent evaluations of spinal cord
al of propranolol up to 240 mg daily, the from trigger point injections to neuro- stimulation have shown encouraging re-
authors were unable to show significant lytic sympathetic blocks with stump in- sults in neuropathic pain, including reflex
improvement in post-traumatic neural- jections, sympathetic blocks, peripheral sympathetic dystrophy (158, 159). Thus,
gias (136). nerve blocks, and epidural or subarach- stimulation of posterior columns of the
Salmon calcitonin has been shown noid blocks. However, it has been shown spinal cord is the most common neuro-
to provide analgesic effect in a series of that only 14% of patients with phantom surgical technique used for the treatment
painful conditions, including phantom limb pain report even a significant tem- of phantom limb pain. The selection pro-
limb pain (117, 137-139). However, there porary change, whereas less than 5% re- cess is crucial. Response to transcutane-
are no controlled trials available to show port a large permanent change or cure ous stimulation or percutaneous electri-
the effectiveness of calcitonin in chronic (27). The use of neural blockade in the cal stimulation may predict a response to
phantom limb pain. Dextromethorphan treatment of phantom limb pain is large- dorsal column stimulation (160). How-
was studied for attenuation of phantom ly based on anecdotal reports in the litera- ever, even with appropriate patient selec-
pain in cancer amputees in a double-blind ture (148-150). tion, it has been reported that only 65% of
cross-over trial involving 3 patients (140). Blankenbaker (148) reported that the patients receive a greater than 25% re-
Results showed that oral dextrometho- sympathetic blocks are successful if am- duction in pain immediately after surgical
rphan effectively reduced postamputa- putees are treated soon after the onset of implantation (161). Further, the success
tion phantom limb pain, bestowing im- phantom limb pain. Halbert et al (98) in rate of dorsal column stimulation steadi-
provement in feeling and minimizing se- a systematic review to evaluate evidence ly declines over time, and greater than
dation in comparison with the pre-treat- for the optimal management of acute and 50% long-term pain reduction is pres-
ment or placebo conditions, with no side chronic phantom pain was unable to find ent in only one-third of patients original-
effects. Capsaicin also was tried in phan- any trials that met criteria for inclusion. ly showing improvement (162, 163). Spi-
tom limb pain (141, 142). In this study, Lesions of the dorsal root entry zone nal cord stimulation may not be effective
which was done in a double-blind fashion have been reported to provide long-term with pain or phantom limb sensations. In
with 24 patients, the authors concluded pain relief in patients with phantom limb one case report, it was shown that good
that capsaicin may be used as an alterna- pain following avulsion of nerve roots or to excellent results were observed in five
tive treatment for the phantom limb pain. amputation (83, 151, 152). It has been re- patients, as judged by decreased pain and
Some have reported a beneficial effect of ported that 36% of patients had pain re- increased functional status with decrease
benzodiazepines (143) however, the gen- lief on follow-up at 6 months to 4 years in medication (164). However, in an-
eral impression is that benzodiazepines do following dorsal root entry zone lesions other report, dorsal column stimulation
not produce substantial pain relief. Mexi- (83, 152). However, they reported very provided minimal relief in patients with
letine (the oral congener of lidocaine) also poor relief in patients with stump pain phantom limb pain (165). Another re-
has been reported to be effective (144). alone. port showed that dorsal column stimula-
Finally, opioid analgesics with or tion provided improvement in only 25%
without other drugs are considered as Neurostimulation of the patients (166). Thus, one should
the mainstay of treatment in modern Transcutaneous electrical nerve weigh the risk-benefit ratio with caution
medicine. Generally, it is quoted in text- stimulation (TENS) has been used with and diligence.
books that narcotic analgesics are not ef- some success in the treatment of phantom Intracranial neurostimulation dem-
fective in producing long-term pain relief pain. However, the results are inconclu- onstrated initial pain relief in 80% of pa-
in patients with phantom limb pain (27). sive and not encouraging and inconclu- tients with sensory thalamic stimulation
However, modern evidence suggests that sive. Spinal cord stimulation (SCS), deep (167) and 86% had significant relief with
opioids can be used safely for years with a brain stimulation (DBS) of the thalam- deep brain stimulation (168) Thalam-
limited risk of drug dependence (4, 27, 43, ic nucleus ventralis caudalis, and motor ic stimulation, in contrast to spinal cord
66, 102, 145-147). Further, patients un- cortex stimulation (MCS) are all used in stimulation, may block spontaneous neu-
dergoing amputation related to systemic managing phantom limb pain with vari- ronal activity, which has been proposed to
medical diseases have only a 42% 5-year able success. mediate phantom sensation in some mod-
survival rate, thus the risk of opioid ad- Some authors have reported excel- els (54). Thus, some believe that it may be
diction may be weighed against quality- lent relief with transcutaneous electrical more effective than spinal cord stimula-
of-life issues (36). In a review of five pa- nerve stimulation. One author reported tion, however, it has not been proven thus
tients, a 50% to 90% reduction in pain at success in 5 of 6 patients with phantom far. Percutaneous stimulation of the peri-
12 to 26 months was reported with meth- pain following treatment with transcuta- osteum has been used, even though it has
adone 10 to 20 mg per day (146). In a pla- neous electrical nerve stimulation (153). not been well studied (169).
cebo-controlled trial (66), morphine was Another author reported a 66% reduction
14. Weiss SA, Fishman S. Extended and tele- Phantom pain and risk factors: A multi- Spontaneous discharge originates in the
scoped phantom limb in unilateral ampu- variate analysis. J Pain Symptom Manage dorsal root ganglion at the onset of a pain-
tees. J Abnorm Soc Psychol 1963; 66:489- 2002; 24:578-585. ful peripheral neuropathy in the rat. Neu-
497. 30. Bailey AA, Moersch FP. Phantom limb. rosci Lett 1992; 138:225-228.
15. Almagor M, Jaffe Y, Lomranz J. The relation Can Med Assoc J 1941; 45:37-42. 46. Price DD, Mao J, Mayer DJ. Central conse-
between limb dominance, acceptance of 31. Jensen TS, Krebs B, Nielsen J et al. Imme- quences of persistent pain states. Jensen
disability, and the phantom limb phenom- diate and long-term phantom limb pain in TS, Turner JM, Wiesenfeld-Hallin Z (eds.)
enon. J Abnorm Psychol 1978; 87:377- amputees: Incidence, clinical character- In Proceedings of the 8th World Congress
379. istics and relationship to pre-amputation on Pain, Progress in Pain Research and
16. Alamo Tomillero F, Rodriguez de la Torre R, limb pain. Pain 1985; 21:267-278. Management, Vol. 8. Seattle, IASP Press,
Caba Barrientos F et al. Prospective study 1997, pp 155-184.
32. Sternbach T, Nadvorna H, Arazi D. A five-
of prevalence and risk factors for painful year follow-up study of phantom limb pain 47. Cook AJ, Woolf CJ, Wall PD et al. Dynamic
phantom limb in the immediate postop- in post-traumatic amputees. Scand J Re- receptive field plasticity in rat spinal cord
erative period of patients undergoing am- hab Med 1982; 14:203-207. dorsal horn following c-primary afferent
putation for chronic arterial ischemia. Rev input. Nature 1987; 325:151-153.
33. Ribera H, Cano P, Dora A et al. Phantom
Esp Anestesiol Reanim 2002; 49:295-301. 48. Doubell TP, Mannion RJ, Woolf CJ. The dor-
limb pain secondary to post-trauma stump
17. Lacoux PA, Crombie IK, Macrae WA. Pain hematoma 40 years after amputation: De- sal horn: State-dependent sensory pro-
in traumatic upper limb amputees in Sier- scription of one case. Revista de la Socie- cessing, plasticity and the generation of
ra Leone. Pain 2002; 99:309-312. dad Espanola del Dolor 2001; 8:217-220. pain. In Wall PD, Melzack R (eds.) Text-
18. Gallagher P, Allen D, Maclachlan M. Phan- book of Pain, 4th ed. Edinburgh, Churchill
34. Rajbhandari SM, Jarett JA, Griffiths PD et
tom limb pain and residual limb pain fol- Livingstone, 1999, pp 165-181.
al. Diabetic neuropathic pain in a leg am-
lowing lower limb amputation: A descrip- putated 44 years previously. Pain 1999; 49. Nikolajsen L, Hansen CL, Nielsen J et al.
tive analysis. Disabil Rehabil 2001; 23: 83:627-629. The effect of ketamine on phantom pain:
522-530. A central neuropathic disorder maintained
35. Sherman RA, Sherman CJ. A comparison
19. Fraser CM, Halligan PW, Robertson IH et al. by peripheral input. Pain 1996; 67:69-77.
of phantom sensations among amputees
Characterizing phantom limb phenomena whose amputations were of civilian and 50. Woolf CJ, Shortland P, Coggeshaal RE.
in upper limb amputees. Prosthet Orthot military origins. Pain 1985; 21:91-97. Peripheral nerve injury triggers central
Int 2001; 25:235-242. sprouting of myelinated afferents. Nature
36. Helm P, Engel T, Holm A et al. Function af-
20. Araya E, Pernía J, Teran P et al. Phantom 1992; 355:75-78.
ter lower limb amputation. Acta Orthop
limb pain following traumatic upper limb Scand 1986; 57:154-157. 51. Davis KD, Kiss ZH, Luo L. Phantom sensa-
amputation. Cirugia Plastica Ibero-Latino- tions generated by thalamic microstimula-
37. Nikolajsen L, Ilkjaer S, Kroner K et al. The
americana 2001; 27:159-163. tion. Nature 1998; 391:385-387.
influence of pre-amputation pain on post-
21. Ehde DM, Czerniecki JM, Smith DG et al. amputation stump and phantom pain. 52. Willoch F, Rosen G, Tolle TR et al. Phantom
Chronic phantom sensations, phantom Pain 1997; 72:393-405. limb pain in the human brain: Unraveling
pain, residual limb pain, and other region- neural circuitries of phantom limb sensa-
38. Houser SA. Phantom limb pain. In Warf-
al pain after lower limb amputation. Arch tions using positron emission tomogra-
ield CA, Fausett HJ (eds.) Manual of Pain
Phys Med Rehabil 2000; 81:1039-1044. phy. Ann Neurol 2000; 48:842-849.
Management, 2nd ed. Philadelphia, Lip-
22. Kooijman CM, Dijkstra PU, Geertzen JH et pincott Williams & Wilkins, 2002, pp 181- 53. Mackert BM, Sappok T, Grüsser S et al.
al. Phantom pain and phantom sensations 183. The eloquence of silent cortex: Analysis
in upper limb amputees: An epidemiologi- of afferent input to deafferented cortex
39. Frazier SH. Psychiatric aspects of causal-
cal study. Pain 2000; 87:33-41. in arm amputees. Neuroreport 2003; 14:
gia, the phantom limb, and phantom pain.
23. Wilkins KL, McGrath PJ, Finley GA et al. 409-412.
Dis Nerv Syst 1966; 27:441-451.
Phantom limb sensations and phantom 54. Dettmers C, Adler T, Rzanny R et al. In-
40. Nikolajsen L, Jensen TS. Postamputation
limb pain in child and adolescent ampu- creased excitability in the primary motor
pain. In Melzack R, Wall PD (eds.) Hand-
tees. Pain 1998; 78:7-12. cortex and supplementary motor area in
book of Pain Management. Edinburgh,
24. Hagberg K, Brånemark R. Consequences patients with phantom limb pain after up-
Churchill Livingstone, 2003, pp 247-257.
of non-vascular trans-femoral amputation: per limb amputation. Neurosci Lett 2001;
41. Wall PD, Gutnick M. Ongoing activity in 307:109-112.
A survey of quality of life, prosthetic use
peripheral nerves: The physiology and
and problems. Prosthet Orthot Int 2001; 55. Karl A, Birbaumer N, Lutzenberger W et al.
pharmacology of impulses originating
25:186-194. Reorganization of motor and somatosen-
from a neuroma. Exp Neurol 1974; 43:
25. Jensen TS, Krebs B, Nielsen J et al. Phan- sory cortex in upper extremity amputees
580-593.
tom limb, phantom pain and stump pain in with phantom limb pain. J Neurosci 2001;
42. Devor M, Seltzer Z. Pathophysiology of 21:3609-3618.
amputees during the first 6 months follow-
damaged nerves in relation to chronic
ing limb amputation. Pain 1983; 17:243- 56. Wei F, Zhuo M. Potentiation of sensory re-
pain. In Wall PD, Melzack R (eds.) Text-
256. sponses in the anterior cingulate cortex
book of Pain, 4th ed. Edinburgh, Churchill
26. Sherman RA, Sherman CJ. Prevalence and following digit amputation in the anaes-
Livingstone, 1999, pp 129-164.
characteristics of chronic phantom limb thetised rat. J Physiol 2001; 532:823-833.
43. Devor M, Govrin-Lippman R, Angelides K.
pain among American veterans: Results 57. Flor H. The modification of cortical reorga-
Na+ channels immunolocalization in pe-
of a trail survey. Am J Phys Med 1983; 62: nization and chronic pain by sensory feed-
ripheral mammalian axons and changes
227-238. back. Appl Psychophysiol Biofeedback
following nerve injury and neuroma for-
27. Sherman RA, Sherman CJ, Parker L. Chron- 2002; 27:215-227.
mation. J Neurosci 1993; 135:1976-1992.
ic phantom and stump pain among Amer- 58. Huse E, Larbig W, Birbaumer N et al. Cor-
44. Novakovic SD, Tzoumaka E, McGivern JG
ican veterans: Results of a survey. Pain tical reorganization and pain. Empirical
et al. Distribution of the tetrodotoxin-re-
1984; 18:83-95. findings and therapeutic implication using
sistant sodium channel PN3 in rat sen-
28. Wall R, Novotny-Joseph P, MacNamara TE. the example of phantom pain. Schmerz
sory neurons in normal and neuropathic
Does preamputation pain influence phan- 2001; 15:131-137.
pain conditions. J Neurosci 1998; 18:2174-
tom limb pain in cancer patients? South 2187. 59. Roux FE, Ibarrola D, Lazorthes Y et al.
Med J 1985; 78:34-36. Chronic motor cortex stimulation for phan-
45. Kajander KC, Wakisaka S, Bennett GJ.
29. Dijkstra PU, Geertzen JH, Stewart R et al. tom limb pain: A functional magnetic reso-
nance imaging study: Technical case. Neu- ship to etiology. J Nerv Ment Dis 1956; 1984; 54:373-375.
rosurgery 2001; 48:681-688. 123:365-368. 96. Cohen H. Anginal pain in a phantom limb.
60. Flor H, Elbert T, Knecht S. Phantom limb 78. Carlen PL, Wall PD, Nadvorna H et al. Br Med J 1976; 2:475.
pain as a perceptual correlate of cortical Phantom limbs and related phenomena in 97. Mester SW, Cintron GB, Long C. Phantom
reorganization following arm amputation. recent traumatic amputations. Neurology angina. Am Heart J 1988; 116:1627-1628.
Nature 1995; 375:482-484. 1978; 28:211-217.
98. Halbert J, Crotty M, Cameron ID. Evidence
61. Flor H, Elbert T, Mühlnickel W. Cortical re- 79. Katz J, Melzack R. Pain ‘memories’ in for optimal management of acute and
organization and phantom phenomena in phantom limbs: Review and clinical ob- chronic phantom pain: A systematic re-
congenital and traumatic upper-extremity servations. Pain 1990; 43:319-336. view. Clin J Pain 2002; 18:84-92.
amputees. Exp Brain Res 1998; 119:205- 80. Krane EJ, Heller LB. The prevalence of 99. Sherman RA, Sherman CJ, Gall NG. A sur-
212. phantom sensation and pain in pediatric vey of current phantom limb pain treat-
62. Birbaumer N, Lutzenberger W, Montoya amputees. J Pain Symptom Manage 1995; ment in the United States. Pain 1980; 8:
P et al. Effects of regional anesthesia on 10:21-29. 85-99.
phantom limb are mirrored in changes in 81. Wartan SW, Hamann W, Wedley JR et al. 100. Sherman R. Published treatments of
cortical reorganization in upper limb am- Phantom pain and sensation among Brit- phantom limb pain. Am J Phys Med 1980;
putees. J Neurosci 1997; 17:5503-5508. ish veteran amputees. Br J Anaesth 1997; 59:232-244.
63. Huse E, Larbig W, Flor H et al. The effect of 78:652-659.
101. McQuay H, Carroll D, Jadad AR et al. Anti-
opioids on phantom limb pain and cortical 82. Montoya P, Larbig W, Grulke N. Relation- convulsant drugs for management of pain:
reorganization. Pain 2001; 90:47-55. ship of phantom limb pain to other phan- A systematic review. BMJ 1995; 311:1047-
64. Matzer O, Devor M. Contrasting thermal tom limb phenomena in upper extremity 1052.
sensitivity of spontaneously active A- and amputees. Pain 1997; 72:87-93.
102. Baron R, Wasner G, Lindner V. Optimal
C-fibers in experimental nerve-end neuro- 83. Saris SC, Iacono RP, Nashold BS Jr. Dorsal treatment of phantom limb pain in the el-
mas. Pain 1987; 30:373-384. root entry zone lesions for post-amputa- derly. Drugs Aging 1998; 12:361-376.
65. Kolb LC. The Painful Phantom. Spring- tion pain. J Neurosurg 1985; 62:72-76.
103. Esquenazi A, Meier RH. Rehabilitation
field, Charles C. Thomas, 1954. 84. Appenzeller O, Bicknell JM. Effects of ner- in limb deficiency; IV. Limb amputation.
66. University of California. Progress Report vous system lesions on phantom experi- Arch Phys Med Rehabil 1996; 77:S18-S28.
to the Advisory Committee on Artificial ence in amputees. Neurology 1969; 19:
104. Katz J. Prevention of phantom limb pain
Limbs, 2nd ed. Berkeley, University of Cali- 141-146.
by regional anesthesia. Lancet 1997; 349:
fornia Press, 1952. 85. Houghton AD, Nicolls G, Houghton AL et 519-520.
67. Ewalt JR, Randall GC, Morris H. The phan- al. Phantom pain: Natural history and as-
105. McQuay HJ, Dickenson AH. Implications
tom limb. Psychosom Med 1947; 9:118- sociation with rehabilitation. Ann R Coll
of nervous system plasticity for pain man-
123. Surg Engl 1994; 76:22-25.
agement. Anaesthesia 1990; 45:101-102.
68. Lindesay J. Validity of the general health 86. Henderson WR, Smyth GE. Phanton limbs.
106. Feria M. Can we prevent post-amputation
questionnaire (GHQ) in detecting psychi- J Neurol Neurosurg Psych 1948; 11:88-112.
pain. DOLOR 2001; 16:73-77.
atric disturbance in amputees with phan- 87. Sherman RA, Barja RH, Bruno GM. Ther-
tom pain. J Psychosom Res 1986; 30:277- 107. Gehling M, Tryba M. Prophylaxis of phan-
mographic correlates of chronic pain:
281. tom pain: Is regional analgesia ineffec-
Analysis of 125 patients incorporating
tive? Schmerz 2003; 17:11-19.
69. Whyte AS, Niven CA. Psychological dis- evaluations by a blind panel. Arch Phys
tress in amputees with phantom limb Med Rehabil 1987; 68:273-279. 108. Bach S, Noreng MF, Tjellden NU. Phantom
pain. J Pain Symptom Manage 2001; 22: limb pain in amputees during the first 12
88. Mackenzie N. Phantom limb pain during
938-946. months following limb amputation, after
spinal anaesthesia. Recurrence in ampu-
preoperative lumbar epidural blockade.
70. Sherman RA, Sherman CJ, Bruno GM. Psy- tees. Anaesthesia 1983; 38:886-887.
Pain 1988; 33:297-301.
chological factors influencing chronic 89. Martin G, Grant SA, MacLeod DB et al. Se-
phantom limb pain: An analysis of the lit- 109. Jahangiri M, Jayatunga AP, Bradley JWP et
vere phantom leg pain in an amputee af-
erature. Pain 1987; 28:285-295. al. Prevention of phantom pain after ma-
ter lumbar plexus block. Reg Anesth Pain
jor lower limb amputation by epidural in-
71. Frazier SH, Kolb LC. Psychiatric aspects of Med 2003; 28:475-478.
fusion of diamorphine, clonidine, and bu-
pain and the phantom limb. Orthop Clin 90. Murphy JP, Anandaciva SP. Phantom limb pivacaine. Ann R Coll Surg Engl 1994; 76:
North Am 1970; 1:481-495. pain and spinal anesthesia. Anaesthesia 324-326.
72. Ewalt JR. The phantom limb. Ann Intern 1984; 39:188.
110. Nikolajsen L, Ilkjær S, Christensen JH et al.
Med 1956; 44:668-677. 91. Sellick BC. Phantom limb pain and spinal Randomised trial of epidural bupivacaine
73. Larbig W, Montoya P, Flor H et al. Evidence anesthesia. Anesthesiology 1985; 62:801- and morphine in prevention of stump and
for a change in neural processing in phan- 802. phantom pain in lower-limb amputation.
tom limb patients. Pain 1996; 67:275- 92. Lee ED, Donovan K. Reactivation of phan- Lancet 1997; 350:1353-1357.
283. tom limb pain after combined interscalene 111. Lambert AW, Dashfield AK, Cosgrove C et
74. Loeser JD. Pain after amputation: Phan- brachial plexus block and general anes- al. Randomized prospective study com-
tom limb and stump pain. In Bonica JJ thesia: Successful treatment with intrave- paring preoperative epidural and intraop-
(ed.) The Management of Pain, 2nd ed. nous lidocaine. Anesthesiology 1995; 82: erative perineural analgesia for the pre-
Philadelphia, Lea & Febiger, 1990, pp 244- 295-298. vention of postoperative stump and phan-
256. 93. Finneson BE, Haft H, Krueger EG. Phantom tom limb pain following major amputation.
75. Van der Schans CP, Geertzen JH, Schop- limb syndrome associated with herniated Reg Anesth Pain Med 2001; 26:316-321.
pen T et al. Phantom pain and health-re- nucleus pulposus. J Neurosurg 1957; 14: 112. Fisher A, Meller Y. Continuous postoper-
lated quality of life in lower limb ampu- 344-346. ative regional analgesia by nerve sheath
tees. J Pain Symptom Manage 2002; 24: 94. Wilson PR, Person JR, Su DW et al. Herpes block for amputation surgery—A pilot
429-436. zoster reactivation of phantom limb pain. study. Anesth Analg 1991; 72:300-303.
76. Sunderland S. Nerves and Nerve Injuries. Mayo Clin Proc 1978; 53:336-338. 113. Pinzur M, Garla PG, Pluth T et al. Contin-
New York, Churchill Livingstone, 1978. 95. Sugarbaker PH, Weiss CM, Davidson DD uous postoperative infusion of a region-
77. Miles JE. Phantom limb syndrome occur- et al. Increasing phantom limb pain as al anaesthetic after an amputation of the
ring during spinal anesthesia: Relation- a symptom of cancer recurrence. Cancer lower extremity. J Bone Joint Surg Am
165. Wester K. Dorsal column stimulation in in posttraumatic amputees. Scand J Reha- 176. Dougherty J. Relief of phantom limb
pain treatment. Acta Neurol Scand 1987; bil Med 1982; 14:203-207. pain after EMG biofeedback-assisted re-
75:151-155. 171. Monga TN, Jaksic T. Acupuncture in phan- laxation: A case report. Behav Res Ther
166. Hunt WE, Goodman JH. Dorsal column tom limb pain. Arch Phys Med Rehabil 1980; 18:355-357.
stimulation for phantom limb pain. J Neu- 1981; 62:229-231. 177. Seigel EF. Control of phantom limb pain
rosurg 1975; 43:250-251. 172. Levine JD, Gormley J, Fields HL. Observa- by hypnosis. Am J Clin Hypnosis 1979; 21:
167. Levy RM, Lamb S, Adams JE. Treatment tions on the analgesic effects of needle 285-286.
of chronic pain by deep brain stimulation: puncture (acupuncture). Pain 1976; 2: 178. Cedercreutz C. Hypnotic treatment of
Long-term follow-up and review of the lit- 149-159. phantom sensations in 100 amputees.
erature. Neurosurgery 1987; 21:885-893. 173. Fukui S, Shigemori S, Komoda Y, et al. Acta Chir Scand 1954; 107:158-162.
168. Mundinger F, Nermuller H. Programmed Phantom pain with beneficial response 179. Oakley DA, Whitman LG, Halligan PW.
transcutaneous (TNS) and central (DBS) to electroconvulsive therapy (ECT) and re- Hypnotic imagery as a treatment for phan-
stimulation for control of phantom limb gional cerebral blood flow (rCBF) studied tom limb pain: Two case reports and a re-
pain and causalgia: A new method for with xenon-CT. Pain Clinic 2002; 13:355- view. Clin Rehabil 2002; 16:368-377.
treatment. In Siegfriend J, Zimmerman M 359. 180. Belleggia G, Birbaumer N. Treatment of
(eds.) Phantom and Stump Pain. New 174. Rasmussen KG, Rummans TA. Electrocon- phantom limb pain with combined EMG
York, Springer, 1981. vulsive therapy for phantom limb pain. and thermal biofeedback: A case report.
169. Lawrence RM. Persistent limb pain in be- Pain 2000; 85:297-299. Appl Psychophysiol Biofeedback 2001;
low-knee amputee. JAMA 1976; 236:822- 175. Sherman RA, Gall N, Gormley J. Treatment 26:141-146.
823. of phantom limb pain with muscular relax- 181. Frischenschlager O, Pucher I. Psychologi-
170. Steinbach TV, Nadvorna H, Arazi D: A five ation training to disrupt the pain-anxiety- cal management of pain. Disabil Rehabil
year follow-up study of phantom limb pain tension cycle. Pain 1979; 6:47-55. 2002; 24:416-422.