Perimortem Cesarean Delivery: Its Role in Maternal Mortality

Vern L. Katz, MD*,,

Since Roman times, physicians have been instructed to perform postmortem cesarean deliveries to aid in funeral rites, baptism, and in the very slim chance that a live fetus might still be within the deceased mothers womb. This procedure was disliked by physicians being called to a dying mothers bedside. As births moved to hospitals, and modern obstetrics evolved, the causes of maternal death changed from sepsis, hemorrhage, and dehydration to a greater incidence of sudden cardiac arrest from medication errors or embolism. Thus, the likelihood of delivering a viable neonate at the time of a mothers death increased. Additionally, as cardiopulmonary resuscitation (CPR) became widespread, physicians realized that during pregnancy, with the term gravid woman lying on her back, chest compressions cannot deliver sufcient cardiac output to accomplish resuscitation. Paradoxically, after a postmortem cesarean delivery is performed, effective CPR was seen to occur. Mothers were revived. Thus, the procedure was renamed the perimortem cesarean. Because brain damage begins at 5 minutes of anoxia, the procedure should be initiated at 4 minutes (the 4-minute rule) to deliver the healthiest fetus. If a mother has a resuscitatable cause of death, then her life may be saved as well by a prompt and timely cesarean delivery during CPR. Sadly, too often, we are paralyzed by the horror of the maternal cardiac arrest, and instinctively, we try CPR for too long before turning to the perimortem delivery. The quick procedure though may actually improve the situation for the mother, and certainly will save the child. Semin Perinatol 36:68-72 2012 Elsevier Inc. All rights reserved. KEYWORDS perimortem cesarean, maternal mortality

our patient, a 29-year-old G3, P0 high school English teacher at 39-4/7 weeks gestation, without other comorbidities, went into spontaneous labor after rupture of membranes early this morning. She requested and received an epidural approximately 1 hour ago. Her status is 6 cm dilated, 100% effaced, and 1 station. The fetus had a reassuring tracing when you left a few minutes ago to grab lunch. At 12:30 PM, you are in the hospital cafeteria. You are about the pay the cafeteria cashier, when you hear a page overhead Code Blue, Labor and Delivery. Code Blue, Labor and Delivery. You run to Labor and Delivery and see a crowd gathered around your patients room. The family and friends are in the
*Clinical Professor, Department of Obstetrics and Gynecology, Oregon Health and Science University, Portland, OR. Clinical Professor, Department of Human Physiology, University Oregon, Eugene, OR. Peace Health Medical Center, Oregon Health Science University, Springeld, OR. Address reprint requests to Vern L. Katz, MD, Peace Health Medical Center, Oregon Health Science University, 181 W 22nd Ave, Eugene, Oregon 97405. E-mail: vkatz@peacehealth.org

hall. Some are ashen-faced; some are covering their eyes and crying. You push through the crowd of onlookers and staff. A terried labor and delivery aide sees you and says, She just sat up, grabbed her chest, and gasped that she couldnt breath, and then fell over. There is a lot of noise, and people in the room, most of whom you do not recognize, are performing code tasks. The team leader of the code loudly announces, Still no pulses. Prepare to shock her again. There are no fetal heart tones on the monitor because it has been disconnected and moved to the corner to allow space for chest compressions. The Labor and Delivery charge nurse turns to you and says, Its been about 6 minutes since we called the code. What do you do? Do you take a scalpel, squeeze between the code team, and perform a laparotomy right there and then in the bed? The correct action, as terrifying as it is, is to do exactly that. This discussion will present the history, development, and scientic rationale of the perimortem cesarean delivery, emphasizing its role as a vital part of the management of maternal cardiac arrest and maternal mortality.

68

0146-0005/12/$-see front matter 2012 Elsevier Inc. All rights reserved. doi:10.1053/j.semperi.2011.09.013

Perimortem cesarean delivery

69 cian. In 1860, the famous obstetrician Scanzoni wrote, The fetus in utero is the same as any other internal organ of the woman. If we recognize the instance of the death of the mother when all organic expression is gone and organic laws yield to chemical laws, that instance we must recognize also the death of the child.3-6 One of the reasons for the high frequency of postmortem sections was the high rate of maternal mortality. Throughout the 19th century, the maternal mortality rate in Europe ranged between 2% and 4% of all births.7 The causes for maternal mortality were often sepsis, dehydration, and maternal shock. As might be expected, fetuses delivered from mothers in sepsis or hemorrhagic shock would usually die before the mother died, thereby contributing to the bad reputation of the postmortem section. However, as the causes of maternal mortality changed in the 20th century, and the incidence of infection and hemorrhagic shock declined, the potential salvage of infants from a postmortem section began to increase. Of recorded cases of postmortem sections between 1879 and 1956, infection represented 40% of the causes of maternal mortality. Between 1970 and 1985, infection accounted for only 3% of the causes of maternal mortality, and the proportion of cases of cardiac, anesthetic, and embolic deaths increased signicantly. Sudden deaths, with the fetus being in better health before the maternal cardiac arrest, led to an increase in anecdotal reports of live infants and a gradual improvement in the attitude toward postmortem section. By the mid 1900s, the attitudes toward postmortem section had evolved remarkably.3-6 The second issue that led to the evolution of physician attitudes was the realization of the difculties of cardiopulmonary resuscitation (CPR) in late pregnancy. As witnessed cardiac arrests in the hospital increased, and modern CPR became widespread, more and more pregnant women had documented attempts at resuscitations. These resuscitations were rarely, if ever, successful. Corke and Spielman pointed out the difculty when cardiac arrest occurred secondary to anesthetic complications.8 DePace et al, in a landmark report published in 1982, described a woman who developed a cardiac arrest during bronchoscopy while being evaluated for hemoptysis. After 20 minutes of unsuccessful CPR, a postmortem section was performed. As soon as the surgery was performed, the patient was able to be successfully resuscitated. Pulsations returned with the chest compressions. Both mother and baby did well, with no long-term sequelae.9 In 1985, we were called to see a morbidly obese mother with severe preeclampsia who developed a cardiac arrest in the Emergency Department parking lot. After 22 minutes of unsuccessful CPR, a postmortem section was performed. Immediately after the viable baby was delivered, we were able to achieve pulsations with chest compressions. However, we could not resuscitate the mother. The observation of the temporal relationship between emptying of the uterus, relieving the aortocaval compression, and successful chest compressions in achieving arterial pulses led us to the concept of the perimortem section.

The Development of the Perimortem Cesarean Section

Almost all ancient mythologies have references to heroes and gods whose births occurred miraculously with their mother dying in childbirth.1 English, Egyptian, Persian, Pacic Islanders, Caribbean, Greek, Roman, Native American, Central American, and Scandinavian cultures all refer to such a birth, with the infant, a hero or god, being delivered from the dying or dead mother. Notable Greek mythological deities were delivered in this way. Dionysus (also known as Bacchus in the Roman pantheon) was delivered from his dying mothers abdomen, after Zeus set her on re. The father of medicine, Asclepius, was said to have been delivered by his father, Apollo, from the abdomen of the dead nymph Koronis. Shakespeares Macbeth is killed by Macduff, who was untimely ripped from his mothers womb, as she was dying, and thus was not of women born.2 In less ancient times, notables such as Scipio Africanus (the Roman general who defeated Hannibal), Robert II of Scotland, Andrea Doria (the Genoese admiral), and Pope Gregory XIV were all said to have been delivered by postmortem cesarean section after their mother had died. Many think that the origin of the term cesarean section is derived from the postmortem section. In 715 BC, the Roman king Numus Pompilius decreed that no child should be buried within its mother. This was known as the Lex Regis, and was later translated into the Lex Cesare (the law of Caesar), and hence the term cesarean section. The law was carried into the Christian era by church decrees. These edicts to perform a postmortem procedure were issued repeatedly throughout the middle ages to aid with baptism. Royal and religious degrees in Cologne (1280), Langres (1404), and Sens (1514) continued to reinforce the postmortem cesarean decree. The description and instructions for postmortem sections are found in some of the earliest printed obstetrics textbooks. In 1363, Guy de Chauliac described a postmortem cesarean as the de extractione foetus in La Chirurgia, the midwifery textbook.3 Physicians who would be called to see women after they had died in childbirth disliked the procedure. Thus, many of the European city states, Venice (1608), Frankfurt (1786), and Bavaria (1816), enacted laws forcing physicians to perform the operation. Indeed, 1 physician in Sicily was condemned to death in 1747 for not performing a postmortem cesarean. Physician resistance rapidly evolved. Analysis of the medical literature in the 1800s demonstrates a vigorous debate for the pros and cons of the procedure. Because some live infants were delivered, some physicians believed any infant life was worth saving. Medical reports noted the frequency and outcomes of the procedure: in 1837 in Paris, 49 operations were performed, of which 7 infants survived. In 1864, 147 cases were reported from the Berlin Obstetric Society, with 3 infant survivors; in the early 19th century, a hospital in London noted 330 operations with 19 survivors.3-6 Detractors of the procedure believed that survival was minimal and the procedure was inappropriate for a physi-

70

V.L. Katz
other 60%. Ueland et al in a series of radiologic experiments showed that stroke volume in a healthy woman at term and lying in a supine position was approximately 30% of that of a nonpregnant woman.11 Thus, chest compressions, at best, produce 10% of normal cardiac output. Kerr documented that immediately on emptying the uterus, aortocaval compression stops.12

From Postmortem to Perimortem Section and the 4-Minute Rule

The evolution of the 4-minute rule occurred from the inability to perform successful CPR in the pregnant woman at term, and from the importance of timely emptying of the uterus during CPR for both the baby and the mother.3 Suppose we have 2 patients, and if one is dying, we cannot let the other die while trying for a prolonged time to save the rst patient. Too often, there is a futile continued resuscitation of the mother even if she is unresuscitable, such as in the case of a lethal trauma. The horror of a maternal cardiac arrest paralyzes the code team, the obstetrician, and the nursing staff. In these cases, the baby dies while a futile resuscitation effort is continued in the mother. This is incorrect medical ethics. We know from multiple medical studies (reference) that brain damage begins at about 4-5 minutes after the cessation of blood ow. Fetuses probably have a few more minutes if the mother has appropriate oxygenation before her cardiac arrest. In that sense, we should start the cesarean delivery within 4 minutes, so the baby can be delivered within 5 minutes. Thus, the procedure is really a perimortem delivery, not a postmortem one. Importantly, healthy infants have been delivered 30 minutes after an arrest, so the procedure should be attempted even if signicant time has expired. Perimortem cesarean section should be carried out any time a baby could be viable, that is, at any time beyond 24 weeks. While teaching nonobstetric providers about resuscitation, we remind them about the ngerbreadth rule. The fundus is a ngerbreadth above the umbilicus for every 2 weeks past 20 weeks; 2 ngerbreadths from the fundus to the umbilicus is consistent with a 24-week pregnancy. This is the point of viability, and an attempt should be made to deliver that fetus. The Emergency Department physician can use this bedside test to assess viability. Some might argue that we might deliver a baby who will have neurologic decits. That has never been part of the ethical approach to deciding who should be born. Just as critical as the delivery of a healthy baby is the issue of aortocaval compression in the third trimester, which renders CPR ineffective. After the pregnancy reaches 24-25 weeks, the fundus begins to compress the vena cava and aorta. Because CPR must be done with the patient lying at on her back, the effect of the gravid third trimester uterus causes grave problems. If the mother is not lying at during chest compression, as the instructions for CPR explain, the torso has a tendency to roll.10 From the American Heart Associations manual on basic life support: The patient must always be in the supine position when external chest compression is performed. . . If the thrust is other than straight down, the torso has a tendency to roll, part of the force is lost and chest compression may be less effective. Indeed, effective chest compression only produces approximately 30% of normal cardiac output in the best of circumstances.10 However, with the mother lying at on her back, with aortocaval compression, cardiac output is reduced by an-

The Paradox: To Achieve Effective Chest Compressions With CPR, the Uterus Must Be Emptied and the Baby Delivered
Most maternal cardiac arrests are from etiologies in which the mother cannot be resuscitated. However, if she can be saved, then there is even greater reason to perform a timely delivery. Delivery within 4-5 minutes would protect the mothers central nervous system as well as the babies. Gerty Marx, the mother of obstetric anesthesia, published an illustrative report of 5 cases of cardiac arrest that occurred at the time of induction of anesthesia for elective cesarean deliveries. These women were at term and on the operating table for cesarean delivery, with their obstetricians scrubbed and the pediatricians in attendance. Three of the mothers had immediate cesarean sections performed at the same time that CPR was initiated. All 3 did well. In 2 other mothers, CPR was continued from 6 to 9 minutes before the cesarean sections were begun. Both mothers had irreversible brain damage from waiting the extra time with no cerebral perfusion.13 In the modern era, more than 200 cases of maternal cardiac arrest with CPR have been published.3-6,14-18 There is obviously a selection bias, with a tendency to publish cases with survival. However, in a review of the literature of the cases of maternal cardiac arrest, we could only nd 3 cases where effective CPR was actually achieved. If pulses can be obtained with CPR, then there is no reason to perform a perimortem section. The exception is when the mother has a nonresuscitable cause. In that case, one would perform the perimortem section to immediately deliver and salvage the baby. If the mother has pulses that can be palpated, then effective CPR negates the need for the perimortem delivery. Again, that in itself has been reported, to our knowledge, rarely. In contrast, multiple case reports and series have noted successful CPR only after the cesarean section was performed. Those reports continue to this day. After the introduction of the 4-minute rule and the perimortem cesarean delivery in 1986, we reexamined the issue in 2005.14 Of the more than 100 cases of perimortem cesarean sections that had subsequently been documented in the medical literature, reports of 38 cases with 34 surviving infants (3 sets of twins and 1 set of triplets) could be extracted to provide insights about the course of the arrests. The question in that review was whether there was a positive impact on CPR from the cesarean delivery; 20 of the 34 women had potentially resuscitable causes, and 13 of the 20 mothers were successfully resuscitated after perimortem cesarean and discharged from the hospital in good condition. One mother who was also able to be resuscitated died later of complications from her

Perimortem cesarean delivery

amniotic uid embolism. Interestingly, in the case reports that documented cardiovascular parameters, 12 of 18 women had dramatic return of pulses and blood pressure immediately after the perimortem cesarean was performed and the uterus emptied. This review reinforced and supported the theory of perimortem delivery.14 Since that review was published in 2005, there have been multiple additional case reports of successful perimortem deliveries. Two exemplary cases were published by McDonnell in 200916. One patient was a 36-year-old woman who collapsed in labor and delivery, became pulseless, and for whom CPR was initiated. When no cardiac output, no pulses, could be produced with chest compressions, she underwent a Pfannenstiel incision, with delivery of the baby within 5 minutes. Immediately after the cesarean delivery was concluded, the mother had normal cardiac output and pulses. The mother and infant were discharged from the hospital without sequelae. The second patient was a 32-year-old woman in labor who had a presumed eclamptic seizure and then a cardiopulmonary arrest (this was later found to be due to a magnesium overdose). Four minutes after her arrest, the physicians could not document cardiac output, and a perimortem cesarean section was performed, again through a Pfannenstiel incision. The neonate and the mother were discharged without sequelae, several days later. Importantly, that mother developed a palpable pulse 1 minute after her perimortem cesarean delivery, thus again showing the importance of emptying the uterus in achieving successful CPR.16

71 dure, if maternal resuscitation is successful, we recommend recovery in the intensive care unit, with the patient remaining intubated until an appropriate time for extubation. Antibiotics may be given afterward as well. The most important point, though, is to keep CPR going until well after the procedure. Sometimes, the question arises about the legality of performing the procedure without obtaining consent. Under basic hospital guidelines, an emergency procedure may be performed if it is in the patients best interest. Given that the perimortem cesarean section is the standard of care, consent would be unnecessary in an emergency. We know of no case, since perimortem section was introduced in 1986, of a physician being charged with either criminal or civil malfeasance for performing a perimortem section. However, we are aware of 2 cases in which a lawsuit was brought against physicians and hospital staff for not performing a perimortem section. Although there is widespread teaching of and acceptance for the perimortem section, in most training programs, the timely performance of the procedure is still problematic. Dijkman et al reviewed recent cases reported in the Netherlands. Since 2003, a program, Managing Obstetric Emergency Trauma, has been introduced to all pregnancy providers.17 The authors noted that the incidence of perimortem cesarean sections has increased over the past several years. However, there still is a lack of timeliness, with many procedures being performed 15 minutes after maternal cardiac arrest. In their series of 12 perimortem cesarean sections, 8 patients who had no cardiac output with CPR had return of pulses immediately after the cesarean section was performed, thereby, in their own words, conrming the hypothesis of the perimortem section.17 They found that there was an association between increased maternal survival and a short interval between arrest and delivery. In their series, many physicians performed Pfannenstiel incisions because of familiarity. The authors also reviewed the maternal morbidity and mortality reports from the United Kingdom from 2003 to 2005, in which 49 instances of perimortem sections were performed. Twenty infants survived. This series was only on mothers who did not survive; therefore, we do not know what the successful ratio of perimortem cesarean section was for surviving cases. An important advancement in maternal cardiac arrest is the development of the Cardiff wedge. This special table allows clinicians to perform CPR by placing the mother in a stabilized lateral tilt. Obviously, the mother has to arrest in the proximity of a Cardiff wedge. These tables are large, and although they may be present in labor and delivery suites, they are not always immediately available. Even with a Cardiff wedge, if pulsations cannot be obtained after 4 minutes, a perimortem section should be performed. In review of the past several decades of perimortem cesarean births, it is clear that the physicians wait too long to perform surgery. Anecdotally, and in case series, physicians who have been trained in simulation of obstetric emergencies seem to move more quickly to the 4-minute rule. We believe that simulation exercises in training programs will help reinforce the optimal initiation of the procedure.

Practical Aspects of the Perimortem Section

After the decision is made to perform a perimortem section, CPR should be continued, given the chance that the emptying of the uterus will allow for adequate chest compressions. The patient should not be moved, nor should there be time spent looking for fetal viability. This only wastes valuable time. The procedure should be done irrespective of whether viability can be assumed. Ultrasonography may be problematic during the time of the CPR. Obtaining fetal heart tones with the Doppler is very difcult, and often the fetus will be in a period of bradycardia. Thus, we would recommend proceeding immediately to laparotomy. In no case does the patient need to be moved to an operating theater. We would recommend performing a vertical incision, as it is faster; however, many clinicians have successfully done perimortem cesarean deliveries with maternal resuscitation using a Pfannenstiel incision.16 Whichever surgical incision works the fastest for the clinician should be used. During and after the procedure, CPR should be continued and the mother should be sewn up quickly. Attention should be taken to not damage the bowel or the bladder in case maternal recovery is obtained. The placenta should be delivered from the uterus as well. The procedure is bloodless, as the patient has no effective cardiac output, but bleeding will occur after cardiac output begins to return. After the proce-

72

V.L. Katz
2. Shakespeare W. Macbeth: Act V, scene viii, in Harbage A (ed): William Shakespeare: The Complete Works. Baltimore, MD, Penguin Books, 1969, p 1134 3. Katz VL, Cefalo RC: History and evolution of cesarean delivery, in Phelan JP, Clark SL (eds): Cesarean Delivery. New York, NY, Elsevier, 1988, p 1 4. Weber CE: Postmortem cesarean section: Review of the literature case reports. Am J Obstet Gynecol 110:158, 1971 5. Duer EL: Postmortem delivery. Am J Obstet Gynecol 12:1, 1879 6. Katz VL, Dotters DJ, Droegemueller W. Perimortem cesarean delivery. Obstet Gynecol 68:571-576, 1986 7. Hayden FJ: Maternal mortality in history and today. Med J Aust 1:100, 1970 8. Corke BC, Spielman FJ: Problems associated with epidural anaesthesia in obstetrics. Obstet Gynecol 65:837, 1985 9. DePace NL, Betesh SS, Kotter MN: Postmortem cesarean section with recovery of both mother and offspring. JAMA 248:971, 1982 10. Montgomery WH, Herrin TJ, Lewis AJ: Basic Life Support for Physicians. Dallas TX, American Heart Association, 1983, p 8 11. Ueland K, Novy MJ, Peterson EN, et al: Maternal cardiovascular dynamics IV: The inuence of gestational age on the maternal cardiovascular response to posture and exercise. Am J Obstet Gynecol 104:856, 1969 12. Kerr MG: The mechanical effects of gravid uterus in late pregnancy. J Obstet Gynaecol Br Commonw 513, 1965 13. Marx GF: Cardiopulmonary resuscitation of late-pregnant woman. Anesthesiology 56:156, 1982 14. Katz V, Balderston K, DeFreest M: Perimortem cesarean delivery: Were our assumptions correct? Am J Obstet Gynecol 192:1916-1921, 2005 15. Warraich Q, Esen U: Perimortem caesarean section. J Obstet Gynaecol 29:690-693, 2009 16. McDonnell NJ: Cardiopulmonary arrest in pregnancy: Two case reports of successful outcomes in association with perimortem caesarean delivery. Br J Anaesth 103:406-409, 2009 17. Dijkman A, Huisman C, Smit M, et al: Cardiac arrest in pregnancy: Increasing use of perimortem caesarean section due to emergency skills training? Br J Obstet Gynaecol 117:282-287, 2010 18. Capobianco G, Balata A, Mannazzu MC, et al: Perimortem cesarean delivery 30 minutes after a laboring patient jumped from a fourth-oor window: Baby survives and is normal at age 4 years. Am J Obstet Gynecol 198:e15-e16, 2008

Conclusion
The perimortem section evolved from the postmortem section as we moved into the modern era of obstetrics. As CPR developed, and as the causes of maternal mortality shifted toward potentially resuscitable causes, the nature of the procedure changed. The operation changed from being a postmortem to a perimortem delivery to aid in resuscitation. The literature suggests that the procedures should be performed within 4 minutes to achieve the least amount of neurologic damage for the fetus. If the mother cannot be resuscitated, then obviously the sooner the procedure can be performed, the better. If the mother has a resuscitable cause, then the procedure in most cases will aid in resuscitation. Importantly, the surgery should be performed at the site of the arrest or as soon as the mother reaches the hospital. CPR should be continued during the delivery. If there are maternal pulsations achievable with chest compressions, then there is no need to do the CPR unless it is to save the baby, as the mother cannot be salvaged. Instances of gunshot wounds, head trauma, or a recent case of suicide by jumping out of an 8-story window would be instances when the mother cannot be resuscitated, and the fetus should be delivered immediately. Perimortem cesarean delivery is an important tool for successful resuscitation when a tragedy does occur. Our instincts are to attempt CPR for as long as we can, to carry the code through. We need to ght our instincts and use our scientic knowledge. This will lead to more successful resuscitation and the delivery of live infants.

References
1. Thompson S: Motif Index of Folk Literature, 2nd ed. Bloomington, IN, University Press, 1955