Cardiac arrest in pregnancy and somatic support after brain death Antara Mallampalli, MD; Elizabeth Guy, MD

Objective: To review the important causes of cardiopulmonary arrest during pregnancy and the recommended modifications to resuscitation protocols when applied to pregnant patients, including the indications for perimortem cesarean section and the expected fetal outcomes, and to review the literature regarding extended somatic support after brain death during pregnancy. Data Sources: MEDLINE review of publications relating to cardiac arrest and resuscitation in pregnancy, physiologic changes after brain death, and attempted somatic support of brain-dead pregnant women. Conclusions: Cardiac arrest during pregnancy is rare, but it is important to recognize the causes, which may be either unrelated to pregnancy or unique to the pregnant woman. For the most part, the resuscitation protocol is the same as for nonpregnant victims of cardiac arrest, with a few important mod-

C

ardiac arrest during pregnancy occurs at approximately 1 per 30,000 pregnancies (1). Although infrequent, it is a complication for which the outcome is critically dependent both on the underlying cause of the arrest and on the speed of resuscitation efforts. Table 1 lists the major reported causes of cardiac arrest during pregnancy that have been reported in the literature. In order of decreasing frequency, these include venous thromboembolism, severe pregnancyinduced hypertension (preeclampsia/ eclampsia), sepsis, amniotic fluid embolism, hemorrhage, trauma, iatrogenic causes including complications of anesthesia and drug errors or allergies, and congenital or acquired heart disease (2– 4). Several of these specific entities are discussed in detail elsewhere in this issue. The cardiovascular and respiratory changes that normally occur during pregnancy are important considerations in the resuscitation efforts. These are

From the Section of Pulmonary and Critical Care Medicine, Department of Medicine, Baylor College of Medicine, Houston, TX. Copyright © 2005 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/01.CCM.0000182788.31961.88

Crit Care Med 2005 Vol. 33, No. 10 (Suppl.)

ifications, including especially the need for relieving aortocaval compression by the gravid uterus, the need for rapid intubation, and the importance of rapid perimortem cesarean delivery when indicated. In those rare cases of brain death occurring in a pregnant patient, continued somatic support of the mother may be possible, even for prolonged periods, to extend the pregnancy and further fetal maturation. The expected physiologic changes after brain death, challenges to successful somatic support, and specific recommendations regarding organ support of the brain-dead pregnant woman are reviewed. (Crit Care Med 2005; 33[Suppl.]:S325–S331) KEY WORDS: cardiac arrest; pregnancy; resuscitation; cardiopulmonary resuscitation, brain death; somatic support; pregnancy maintenance

outlined in Table 2. Circulating blood volume and cardiac output increase gradually, by up to 30 –50% as compared with the pregravid state. The increase in cardiac output results from increases in heart rate and stroke volume together with a decrease in systemic vascular resistance. The gravid uterus receives up to 30% of cardiac output, as compared with ⬍2% in the nonpregnant state. After spontaneous delivery, cardiac output increases by 60 – 80% from prelabor values. This increase is smaller after cesarean delivery, about 30% of prelabor values, possibly due to the effects of anesthesia and blood loss (5– 8). Aortocaval compression caused by the gravid uterus occurs by approximately 20 wks of gestation (8). About 10% of pregnant women will exhibit “supine hypotensive syndrome,” which is characterized by syncope, hypotension, and bradycardia when supine and is due to decreased venous return caused by compression of the vena cava (4, 5, 9, 10). Simply moving the late-term pregnant patient from a supine to lateral decubitus position can augment stroke volume and cardiac output by 25– 30%. This has important implications in the conduct of effective cardiopulmonary resuscitation (CPR). Compensated mild respiratory alkalosis with an average PaCO2 of 28 –32 mm

Hg is the norm during late pregnancy. This increased minute ventilation is due to the effects of progesterone on respiratory drive. Resting oxygen consumption is increased during pregnancy. Functional residual capacity is reduced secondary to diaphragmatic elevation by the enlarging uterus and the hypertrophied breasts, causing reduced chest wall compliance. The combination of reduced functional residual capacity and increased oxygen consumption can result in rapid decline in oxygen saturation when alveolar hypoventilation occurs (9, 11, 12). CPR and Advanced Cardiac Life Support in Pregnant Patients.The resuscitation algorithms for basic and advanced cardiac life support (ACLS) are similarly applicable to the pregnant patient, with few important exceptions. The major modifications recommended in late pregnancy are as follows: 1) more aggressive or prompt airway management, 2) attention to left lateral displacement of the uterus, 3) cautious use of sodium bicarbonate, and 4) early consideration of cesarean section delivery of the fetus (9). Cardiac output during optimal CPR has been estimated to be only about 30% of normal, meaning that uteroplacental blood flow will be markedly depressed during chest compressions after cardiac S325

arrest, even in the best case (9, 13). Chest compressions should be performed in the same way on pregnant patients as on nonpregnant patients, except that in the second half of pregnancy, relief of aortocaval compression in the supine position is essential to restoring effective circulation. In a study examining the effectiveness of CPR on patients in left lateral decubitus position, Rees and Willis (14) measured the force achieved with chest compressions performed on a manikin in the decubitus position at various angles of inclination. They found that the resuscitative force on the manikin in the full lateral decubitus position was half of that in the supine position and 80% when positioned at 27 degrees of inclination. These findings led to the development of a special wooden frame inclined at a 27degree angle specifically designed for CPR in pregnant patients, called the Cardiff resuscitation wedge (14). In the Table 1. Major causes of cardiac arrest during pregnancy Venous thromboembolism Pregnancy-induced hypertension Sepsis Amniotic fluid embolism Hemorrhage Placental abruption Placental previa Uterine atony Disseminated intravascular coagulation Trauma Iatrogenic Medication errors or allergy Anesthetic complications Hypermagnesemia Preexisting heart disease Congenital Acquired Reproduced from Mallampalli A, Powner DJ, Gardner MO: Cardiopulmonary resuscitation and somatic support of the pregnant patient. Crit Care Clin 2004; 20:747–761; Copyright © 2004, with permission from Elsevier.

situation in which the Cardiff wedge is not available, however, the following maneuvers are recommended during CPR in pregnant patients: manual displacement of the uterus to the left and upward while the patient is supine, use of a foam wedge or pillow placed under the right hip, use of a “human wedge” (i.e., one rescuer kneeling on the floor or other surface with the victim’s back positioned against his or her thighs), or emergent cesarean delivery of the fetus when otherwise appropriate (10, 15, 16). Only one study has directly addressed the question of whether defibrillation energy requirements change significantly during pregnancy. Nanson et al. (17) measured transthoracic impedance registered by a defibrillator in 45 women at term pregnancy, and they repeated the measurements at 6 to 8 wks postpartum in 42 of the women. They found no significant difference in the mean transthoracic impedance between the two groups. Currently, the same defibrillation protocols recommended in the general ACLS algorithm for specific cardiac arrhythmias, including ventricular fibrillation or pulseless ventricular tachycardia, are also recommended for the pregnant patient (1). Immediate mask ventilation and administration of supplemental oxygen at a concentration of 100%, followed by rapid control of the airway through endotracheal intubation early in the resuscitation effort, are especially important in the pregnant cardiac arrest victim. In addition to increased susceptibility of the mother and fetus to rapid development of hypoxia due to the reduced maternal functional residual capacity and increased oxygen consumption, pregnant patients are at increased risk of aspiration of gastric contents due to normally delayed gastric emptying and reduced lower

Table 2. Normal cardiorespiratory changes of pregnancy and effect on resuscitation

Physiologic Parameter

Change During Pregnancy

Cardiac output

Increase (by 50%)

Blood volume

Increase (by 30–50%)

Systemic vascular resistance Minute ventilation

Decrease (variable) Increase (by 50%)

Functional residual capacity Oxygen consumption

Decrease (by 20%) Increase (by 20%)

CPR, cardiopulmonary resuscitation. Adapted from Murphy and Reed (1).

S326

Effect on Resuscitation Increased circulation demands during CPR Dilutional anemia, decreased oxygen carrying capacity Sequestration of blood during CPR Compensated respiratory alkalosis, decreased buffering capacity Rapid development of hypoxia Rapid development of hypoxia

esophageal sphincter tone. This risk of aspiration may be further exacerbated by gastric distention from air insufflation during prolonged bag-mask ventilation. Fetal asphyxia results in progressive hypoxemia and hypercapnia, tissue oxygen debt, and metabolic acidosis. Protective mechanisms in the fetus against the adverse effects of temporary asphyxia include the elevated fetal hemoglobin concentration and higher fetal hemoglobin oxygen saturation (80 –90% at an umbilical vein PO2 of 35 mm Hg due to the leftward shift of the fetal oxyhemoglobin dissociation curve) (8). Shunting of blood flow to vital organs, including the brain, and reduced oxygen consumption have been demonstrated in animal studies of fetal hypoxia (18). It has been estimated that a fetus can survive for ⱖ10 mins in the face of asphyxia because of these compensatory mechanisms (8, 18). The prevalence and severity of newborn complications, especially encephalopathy and respiratory complications, have been correlated with increasing degrees of fetal metabolic acidosis. Umbilical arterial blood base deficit of ⬎12 mmol/L was identified in one study to correlate with increasing prevalence of newborn complications (18). Unfortunately, this cannot be readily correlated with a particular duration of maternal hypoxia in the setting of cardiopulmonary arrest. There are little data specific to pregnant patients regarding pharmacologic interventions during ACLS. Although alpha-adrenergic agents may theoretically reduce uteroplacental blood flow, their actual clinical effect on blood flow to the fetus in this setting is unknown. In general, it is recommended that the same medication protocols for ACLS be used in pregnant victims of cardiac arrest as in nonpregnant patients (1, 16). The use of sodium bicarbonate to treat metabolic acidosis during cardiac arrest has been controversial, and this applies to its role in maternal cardiac arrest as well (19). Animal studies suggest that bicarbonate crosses the placenta poorly, although it is unclear whether this is the case in humans. Correction of maternal (but not fetal) acidosis with sodium bicarbonate might be expected to reduce compensatory hyperventilation, causing normalization of maternal PaCO2, which could result in a concomitant increase in fetal PaCO2 and potential worsening of fetal acidosis (9, 20). Restoration of effective maternal circulation and correction of hypoxia are ultimately the most effecCrit Care Med 2005 Vol. 33, No. 10 (Suppl.)

tive ways to correct fetal acidosis during cardiac arrest. Perimortem Cesarean Section and Outcomes. A critical question confronting the physicians involved in resuscitation of pregnant victims of cardiac arrest when initial resuscitative efforts are not immediately successful is whether to attempt emergent cesarean delivery of the fetus. Prompt recognition of the appropriateness of delivery when resuscitation efforts are failing and rapid performance of the procedure are essential to optimize the outcome and limit adverse neurologic sequelae in the survivors. Katz et al. (13) reviewed the published cases of perimortem cesarean sections from 1900 through 1985. They were able to identify 61 of 188 surviving infants for whom the data allowed determination of the time interval from maternal death to delivery of the infant. They found that 70% of surviving neonates had been delivered within 5 mins of maternal death, with 93% delivered within 15 mins. None of the infants delivered within 5 mins of maternal arrest had any reported adverse neurologic sequelae. There were cases of infant survival when delivery occurred ⬎21 mins after maternal death, but neurologic deficits in these infants were more frequent and more severe (13). Based on these findings, Katz et al. (13) recommended initiation of cesarean section within 4 mins of maternal cardiac arrest if circulation has not been restored, aiming for fetal delivery within 5 mins. This recommendation has been supported by other authors and forms the basis of the so-called 4-min rule (1). However, because there are also a number of published reports of neonatal survival without adverse neurologic sequelae when delivery occurred well after 5 mins postmaternal arrest, this rule should not be taken as absolute. Lopez-Zeno et al. (21) reported a case of perimortem cesarean delivery after 22 mins of CPR after maternal cardiac arrest secondary to a fatal gunshot wound. The infant survived and was described as clinically normal at 18 months of age. Nevertheless, most experts recommend that if perimortem delivery is being considered for a potentially viable fetus, it should be performed as soon as possible, and preferably within 5 mins of the arrest, to maximize the chances of a favorable outcome for the newborn. Gestational age is an important factor in predicting prognosis for the newborn after a perimortem cesarean section. The Crit Care Med 2005 Vol. 33, No. 10 (Suppl.)

threshold for expected fetal viability is generally considered to be around 24 wks of gestation (16). In situations in which the gestational age is unknown from the available medical or prenatal history, practical and rapid methods of assessing gestational age in the emergency setting include calculation based on the mother’s last menstrual period or from measurement of fundal height (22). Between 20 and 36 wks of gestation, the fetal age in weeks is approximated by the distance in centimeters from the pubic symphysis to the top of the uterine fundus when the mother is supine. Another rule of thumb is that a fundal height at the level of the umbilicus corresponds to 20 wks of gestation (4). However, these methods of estimating gestational age may be inaccurate in the setting of multiparity, extreme obesity, or abdominal distention from other causes or intrauterine growth retardation (22). Although ultrasound examination can provide an accurate estimation of gestational age and confirmation of fetal viability, it is not always rapidly available during emergencies and may be difficult to perform in the setting of maternal cardiac arrest. When available, ultrasound may be helpful in guiding the decision to perform an emergent delivery, particularly in the case of a fetus whose estimated gestational age is borderline in terms of potential viability. However, it is not recommended that perimortem cesarean delivery be delayed for an attempt to perform an ultrasound examination in the setting of maternal cardiac arrest when there has been no return of spontaneous circulation within 4 mins (1). Although the primary goal of cesarean section in the perimortem period has been survival of the fetus, the procedure may also have a role in saving both mother and infant in some cases. Cesarean delivery may significantly improve maternal cardiac output after relieving aortocaval compression. There are now numerous case reports of emergent cesarean section in the setting of apparently refractory maternal cardiac arrest leading to survival of both the infant and mother because of improved maternal circulation after delivery (5, 23–25). Finegold et al. (25) reported a case of a previously healthy 35-yr-old woman at 39 wks of gestation who had a cardiac arrest soon after rupture of membranes. Emergent cesarean section was performed 15 mins after arrest, with immediate recovery of maternal pulse and blood pressure, and

both mother and infant survived with no adverse neurologic sequelae. Given the nature of case reports and the probable publication bias in favor of cases with favorable outcomes, however, it should be noted that such reports cannot permit any firm conclusion about whether cesarean delivery truly improves outcomes in the setting of late-term maternal cardiac arrest (9). There is little information regarding whether cesarean section to deliver a previable fetus would be beneficial to maternal outcome. The hemodynamic benefits to the mother of delivering a smaller fetal-placental mass in this setting would not be expected to be as significant as after delivery occurring later in pregnancy. In general, in cases of ⬍24 wks of estimated gestational age, perimortem cesarean delivery is not recommended, and efforts should be focused on optimizing ACLS performance to achieve restoration of spontaneous circulation in the mother to provide the best hope of recovery for both mother and fetus (1, 4, 26). There are case reports in the literature of successful resuscitation from cardiac arrest in early pregnancy even after a prolonged period of CPR and ACLS before return of spontaneous circulation, with subsequent continuation of the pregnancy to term and favorable neurologic outcomes for both mother and baby (26). The technique of perimortem cesarean section is beyond the scope of this review, but it has been well described elsewhere (22, 27). As noted above, speed is essential once the decision is made to undertake delivery. The procedure should be performed by the available provider most skilled in cesarean delivery. CPR must be continued during delivery, and the procedure should not be delayed to attempt to obtain consent from next of kin. Most experts agree that in this setting, the doctrine of emergency or implied consent applies and the best interests of the potentially viable fetus take precedence (13, 22). According to Katz et al. (13), there have been no legal findings of liability against physicians in the United States for performing a postmortem cesarean section. Finally, when perimortem cesarean section is being considered, the availability of neonatal intensive care facilities and staff must also be taken into account because surviving infants will frequently be premature and may need extensive support immediately after delivery. S327

Table 3. Expected physiologic changes after brain death Hemodynamic instability Panhypopituitarism Respiratory arrest Temperature lability (poikilothermia) Reduced resting energy expenditure

Somatic Support After Brain Death in a Pregnant Patient In rare cases, a pregnant patient may suffer brain death as a result of complications of severe traumatic brain injury, a catastrophic cerebrovascular accident, or other critical illness. In such a case, the physicians must consider whether to 1) attempt immediate delivery of the fetus (when past the age of viability), 2) continue full support of the mother’s body in an attempt to prolong the pregnancy and allow the fetus to mature further, or 3) immediately discontinue mechanical ventilation and other supportive measures with the understanding that the fetus will then also die. The prevalence of brain death in pregnancy is not known, but it fortunately seems to be low. In one study, during a 6-yr period, only 11 of 252 brain-dead women referred to one organ procurement organization as potential donors were pregnant (28). The term “somatic support” is sometimes used to refer to the nonneurologic care provided after brain death (29). In a pregnant patient, the goal of such support is to extend the pregnancy to improve fetal outcome. Although there is now fairly extensive experience with short-term somatic support after brain death in the case of heart-beating cadaveric organ donors, support of the braindead pregnant woman is more complicated, in part, because of the longer duration of support usually being attempted. A recent comprehensive review by Powner and Bernstein (29) summarized the ten cases that have been published to date in which somatic support was provided after maternal brain death until successful delivery occurred. The causes of maternal brain death in these reported cases of successful somatic support were as follows: intracranial hemorrhage (six cases, one related to cocaine use), closed head injury (one case), meningitis (two cases), and intracranial mass lesion (one case) (29 –37). Bernstein et al. (30) reported the case of the longest duration of successful somatic support after maternal brain death (107 days); this was S328

also the case of the earliest gestational age (15 wks) at which such support was begun and led to successful delivery (at 32 wks). It should be noted that the frequency of unsuccessful support of pregnancies after maternal brain death is really unknown, as there are few such reports in the published literature. There are predictable physiologic changes (Table 3) that occur after brain death that must be managed to successfully extend a pregnancy in such cases. These changes and some recommendations for somatic support of the pregnant woman after brain death are discussed below. Cardiovascular System.After brain death, there is massive sympathetic discharge from cerebral herniation such that this early period is characterized by severe hypertension. Normotension and, more predictably, hypotension requiring vasopressor support ensues when there is loss of sympathetic output. Severe vasoconstriction and end-organ hypoperfusion can lead to decreased cardiac output and worsen hypotension (38). Vasomotor instability is encountered in practically all cases after brain death. Hemodynamic support is critical in maintaining adequate perfusion of the uteroplacental unit and fetal viability. The fetus is at particularly high risk for adverse effects from hypotension because of lack of autoregulation in uterine blood flow. The management of hypotension is directed toward maintaining adequate end-organ perfusion. This generally will require a combination of intravenous fluids, inotropic medications and one or more vasopressor agents. Dobutamine, dopamine, and norepinephrine were the medications used most frequently in cases reported to date. Low doses of vasopressin used in combination with dopamine or epinephrine have been associated with improved hemodynamic stability and lower dose requirements of the other pressor agents (39, 40). However, it must be remembered that in nonpregnant patients, vasopressin has been shown to cause uterine vasoconstriction (41). Intravascular volume should be maintained, and determination of adequate preload using central venous pressure or pulmonary artery occlusion pressure measurement has been advocated, albeit the “optimum” values in this setting are not known (29). The mother should be maintained in the left lateral recumbent position. Volume expansion with combi-

nation of crystalloids and colloids is used, keeping in mind that low oncotic pressure and hypoalbuminemia predispose to pulmonary edema. Respiratory System. Mechanical ventilation is integral to continuing support of the brain-dead pregnant patient. The goals of mechanical ventilation are similar to that of a nonpregnant patient, with a few additional considerations that are important to fetal respiration. Maintaining the diffusion gradient for transfer of CO2 from the fetus to maternal circulation dictates that the PaCO2 should be maintained at 30 –35 mm Hg (29, 31). However, more profound respiratory alkalosis has also been shown in animal studies to result in uterine hypoperfusion (31, 42), which is potentially more deleterious to fetal well-being. The normal PaO2 during pregnancy is 105 mm Hg, and although the fetus has protective mechanisms against hypoxia, adequate maternal oxygenation is defined differently compared with the nonpregnant state. Endocrine System. Disruption of the hypothalamic-pituitary axis after brain death frequently results in panhypopituitarism. Central diabetes insipidus occurred in the majority (⬎70%) of cases of somatic support of the pregnant woman after brain death (29). The resulting intravascular volume depletion, if uncorrected, can contribute to hemodynamic instability. In addition to hypotonic fluids, administration of vasopressin (intramuscularly, intravenously, or intranasally) is indicated and should be titrated to control the polyuria and hypernatremia. Adrenal insufficiency and hypothyroidism requiring treatment also occur frequently after brain death, but the need for hormone replacement should be guided by measurement of adrenal and thyroid function in the individual case (29). Studies in nonpregnant organ donors after brain death have shown reduced T3 levels, variably reduced T4 levels, frequently normal thyroid-stimulating hormone levels, and mostly normal reverse T3 levels (43). This may represent a variant of the sick euthyroid syndrome and may not warrant intervention. In one study, no clear correlation was found between T3 levels and the need for inotropic support (43). In normal pregnancy, total serum cortisol is elevated as a result of increased cortisol-binding globulin and placental production of adrenocorticotropic hormone and corticotropin-releasing horCrit Care Med 2005 Vol. 33, No. 10 (Suppl.)

mone (29, 44). After brain death, variable levels of serum cortisol have been reported (29). Before replacement hormone is administered, it is recommended that maternal cortisol levels be measured. The choice of corticosteroid preparation may be dictated by its ability to cross the placental barrier and the need to avoid prolonged fetal exposure to corticosteroid (29). Hyperglycemia in the mother is also commonly encountered, and just as in treatment of gestational diabetes, close attention to blood sugar and lowering to the euglycemic range with use of insulin therapy is an integral part of the therapeutic armamentarium. Temperature Regulation. Poikilothermia (a state in which body temperature is dependent on that of the environment) is a frequently described complication after brain death due to the loss of hypothalamic function after brain death. Labile temperatures and hyperthermia have been described, but hypothermia is far more common after brain death. Passive rewarming with blankets and other standard methods is recommended when hypothermia occurs. Similarly, hyperthermia, if not caused by infectious complication, can be treated with acetaminophen and passive cooling measures (29, 31). In animal models, fetal shivering, altered breathing patterns, and fetal bradycardia reversible with rewarming have been described (45, 46). The true effect of prolonged maternal hypothermia on the fetus is not well understood. Nutritional Support. Normal pregnancy is accompanied by a weight gain of 25–35 pounds. Only 40% of this is attributable to the fetus, placenta, and amniotic fluid (47). This is an important consideration in the nutritional support of these patients, as maternal malnutrition will have lasting adverse effects on the fetus. Aggressive nutritional intervention by either enteral, parenteral, or combination is highly recommended. Nuutinen et al. (33) provided a detailed description of nutritional variables and supplementation using a combination of enteral and parenteral alimentation in their description of the successful somatic support of a brain-dead pregnant patient for 10 wks. After brain death, resting energy expenditure is generally less than is usually estimated from the Harris-Benedict equation, averaging 15% lower than predicted in one study (48). Indirect calorimetry may help in determining the appropriate caloric goals during extended support afCrit Care Med 2005 Vol. 33, No. 10 (Suppl.)

ter brain death. Accurate weight measurement should be part of daily care. The recommended daily allowance for protein during pregnancy is 0.8 g·kg⫺1·day⫺1 (the normal intake for an average healthy adult) plus an additional 1.3, 6.1, or 10.7 g/day for the first, second, or third trimesters, respectively (47). However, in the setting of brain death, the balance of protein synthesis and breakdown is not well understood. A catabolic state is frequently seen in severe critical illness but unknown after brain death. Serial measurement of 24-hr urine urea nitrogen excretion and serum prealbumin levels can be used to guide the adequacy of protein replacement (29, 33), as is frequently done in critical illness. Deficiency of essential fatty acids should be prevented; 20 –25% of nonprotein calories in the form of fat is generally adequate. The increased requirement for vitamins and trace elements to support the growing fetus may not be adequately achieved by commercially available formulas and necessitates supplementation (29, 47). Infectious Complications. The prolonged duration of support likely contributes to the universal occurrence of infectious complications in all published cases of somatic support of brain-dead pregnant patients. As in other intensive care unit patients, urinary tract infections and ventilatorassociated pneumonia are most common. Gram-negative and Gram-positive organisms that are frequent pathogens in the intensive care unit are encountered, but occasionally, fungal pathogens have been reported (36). Strict adherence to infection control measures, surveillance, and aggressive treatment of infectious complications are important to ensure successful outcome. The choice of antibiotics should be guided by the potential toxicity to the fetus. Ensuring Fetal Outcome. Throughout the period of somatic support, monitoring of the fetus for proper intrauterine growth and maturation is performed by various methods. These include fetal heart rate monitoring and nonstress testing, serial ultrasounds, and amniocentesis. In the review by Powner and Bernstein (29), events that precipitated delivery of the fetus were: persistent hemodynamic instability, preterm labor not responsive to tocolytic therapy, intrauterine growth retardation, and achievement of adequate fetal lung maturity. The use of drugs to control uterine contractions was successful in at least two of the cases reported in the literature. In one, a

magnesium sulfate infusion combined with indomethacin was used to control uterine contractions and allowed prolongation of the pregnancy for 25 more days (36). Other agents available for tocolysis include beta-2agonists, calcium channel blockers, and oxytocin antagonists; however, the hemodynamic effects of beta-2-agonists and calcium channel blocking agents may limit their usefulness in this setting in which maternal hemodynamic instability is common (36). In all ten published cases of somatic support after maternal brain death, the fetuses survived with good outcomes. There was one case of fetal hydantoin syndrome thought to be due to previous chronic phenytoin use by the mother (32); no other congenital abnormalities were reported. In four cases, the infants required a period of mechanical ventilation due to neonatal respiratory distress syndrome or pneumonia. In all six cases for which there were follow-up data, the infants were noted to be developmentally normal. Despite these reported favorable outcomes, it must be said that no firm conclusions can be drawn about the likelihood of success of prolonged somatic support of a pregnancy in a given case of maternal brain death or about the expected prevalence of favorable fetal outcomes with such support because there are little or no data on the frequency of cases of unsuccessful support or unfavorable fetal outcome. This may be due in part to the small number of cases overall, but it undoubtedly also reflects a publication bias against the cases in which extended support could not be achieved or a reluctance in many cases on the part of providers and families to even attempt such support after brain death of the mother (30, 34). Ethical and Legal Considerations. The idea of maintaining a brain-dead pregnant woman on artificial life support for an extended period of time, effectively serving as an “incubator” for the maturing fetus, raises a number of ethical questions that previously have been discussed in the literature. Some have argued that the situation is no different from the somatic support of brain-dead organ donors, in that the efforts to maintain the body after brain death are directed toward the benefit of another (in this case, the fetus). Veatch (49) argued that the relevant legal and ethical literature support this position so that if the mother had previously indicated her wishes in favor of being an organ donor, the physicians would be justified in using her body as an S329

incubator, but that without such knowledge of her wishes, consent from the next of kin would be needed. Another view has been that because the interventions required to extend a pregnancy successfully after brain death are much more extensive and prolonged than the brief period of support required for brain-dead organ donors, this is more akin to an experimental therapy in the sense of requiring careful consideration of the individual case and informed consent from the next of kin— especially when brain death occurs far from fetal viability (50). Unfortunately, the wishes of the mother in such cases are rarely if ever known because these cases invariably involve sudden catastrophic neurologic injuries, often in previously healthy young women. Moreover, in the rare cases in which a maternal advance directive is available, it should be recognized that some state laws nullify such directives when the individual is pregnant. The gestational age at the time of maternal brain death is no longer the primary consideration in deciding whether to attempt somatic support of the woman, although certainly the farther the fetus is from viability, the more uncertain is the probability of successful delivery of a healthy infant. In early reports, it was suggested that consideration of intensive somatic support be limited to cases in which the fetus was ⱖ24 wks of gestation because of the perceived difficulty at that time of maintaining somatic support beyond 2– 4 wks after brain death (32). Subsequent reports have described successfully extending the pregnancy for much longer periods after brain death; thus, there is presently no clearly defined lower limit of gestational age before which support should not be attempted. As discussed by Bernstein et al. (30), however, this also should not be taken to mean that physicians must always undertake such support. Most authors agree on the importance of involving the next of kin—particularly the father of the fetus—in making this decision (30, 49, 50). This requires, especially, informing the family of the risk of delivery of a severely premature fetus, with the associated emotional, physical, and economic costs. Another ethical perspective in this situation argues that the fetus also has interests or rights that could take precedence over the next of kin’s decision to terminate support. In cases in which no authorized surrogate decision maker is available or in which there is disagreeS330

ment among family members (or between family and physicians) about the best course of action, intervention by the courts may be required, and decisions are likely to vary across jurisdictions (30). Finally, although the economic cost of intensive somatic support is undeniably high (30, 31), this must be balanced (at least when gestational age is beyond that of fetal viability) against the cost of caring for an even more severely premature infant who would require prolonged neonatal intensive care unit care if delivery were not delayed to allow the fetus to mature further in utero (29).

REFERENCES 1. Murphy N, Reed S: Maternal resuscitation and trauma. In: Advanced Life Support in Obstetrics (ALSO) Provider Course Syllabus. Damos JR, Eisinger SH (Eds). Leawood, KS, American Academy of Family Physicians, 2000, pp 1–25 2. Lewis G (Ed): Why Mothers Die, 1997–1999: The Fifth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. London, RCOG Press, 2001 3. Chang J, Elam-Evans LD, Berg CJ, et al: Pregnancy-related mortality surveillance, United States, 1991–1999. MMWR Surveill Summ 2003; 52:1– 8 4. Luppi CJ: Cardiopulmonary resuscitation: Pregnant women are different. AACN Clin Issues 1997; 8:574 –585 5. DePace NL, Betesh JS, Kotler MN: ‘Postmortem’ cesarean section with recovery of both mother and offspring. JAMA 1982; 248: 971–973 6. Ueland K, Hansen JM: Maternal cardiovascular hemodynamics: III. Labor and delivery under local and caudal analgesia. Am J Obstet Gynecol 1969; 103:8 –17 7. Ueland K, Akamatsu TJ, Eng M, et al: Maternal cardiovascular hemodynamics: IV. Cesarean section under epidural anesthesia without epinephrine. Am J Obstet Gynecol 1972; 114:775–780 8. Chesnutt A: Physiology of normal pregnancy. Crit Care Clin 2004; 20:609 – 615 9. Dildy GA, Clark SL: Cardiac arrest during pregnancy. Obstet Gynecol Clin North Am 1995; 22:303–314 10. Lee RV, Rodgers BD, White LM, et al: Cardiopulmonary resuscitation of pregnant women. Am J Med 1986; 81:311–318 11. Fillion DN: Being prepared for a pregnant code blue. MCN Am J Matern Child Nurs 1998; 23:240 –245 12. Campbell LA, Klocke RA: Implications for the pregnant patient. Am J Respir Crit Care Med 2001; 163:1051–1054 13. Katz VL, Dotters DJ, Droegemueller W: Perimortem cesarean delivery. Obstet Gynecol 1986; 68:571–576

14. Rees GA, Willis BA: Resuscitation in late pregnancy. Anaesthesia 1988; 43:347–349 15. Mauer DK, Gervais HW, Dick WF, et al: Cardiopulmonary resuscitation (CPR) during pregnancy: Working Group on CPR of the European Academy of Anaesthesiology. Eur J Anaesthesiol 1993; 10:437– 440 16. Whitty JE: Maternal cardiac arrest in pregnancy. Clin Obstet Gynecol 2002; 45: 377–392 17. Nanson J, Elcock D, Williams M, et al: Do physiological changes in pregnancy change defibrillation energy requirements? Br J Anaesth 2001; 87:237–239 18. Low JA, Lindsay BG, Derrick EJ: Threshold of metabolic acidosis associated with newborn complications. Am J Obstet Gynecol 1997; 177:1391–1394 19. Bar-Joseph G, Ambramson NS, JansenMcWilliams L, et al: Clinical use of sodium bicarbonate during cardiopulmonary resuscitation: Is it used sensibly? Resuscitation 2002; 54:47–55 20. Lapinsky SE, Kruczynski K, Slutsky AS: Critical care in the pregnant patient. Am J Respir Crit Care Med 1995; 152:427– 455 21. Lopez-Zeno JA, Carlo WA, O’Grady JP, et al: Infant survival following delayed postmortem cesarean delivery. Obstet Gynecol 1990; 76:991–992 22. Strong TH Jr, Lowe RA: Perimortem cesarean section. Am J Emerg Med 1989; 7:489 – 494 23. Parker J, Balis N, Chester S, et al: Cardiopulmonary arrest in pregnancy: Successful resuscitation of mother and infant following immediate caesarean section in labour ward. Aust N Z J Obstet Gynaecol 1996; 36: 207–210 24. O’Connor RL, Sevarino FB: Cardiopulmonary arrest in the pregnant patient: A report of a successful resuscitation. J Clin Anesth 1994; 6:66 – 68 25. Finegold H, Darwich A, Romeo R, et al: Successful resuscitation after maternal cardiac arrest by immediate cesarean section in the labor room. Anesthesiology 2002; 96:1278 26. Selden BS, Burke TJ: Complete maternal and fetal recovery after prolonged cardiac arrest. Ann Emerg Med 1988; 17:346 –349 27. Stallard T, Burns B: Emergency delivery and perimortem C-section. Emerg Med Clin North Am 2003; 21:679 – 693 28. Suddaby EC, Schaeffer MJ, Brigham LE, et al: Analysis of organ donors in the peripartum period. J Transpl Coord 1998; 8:35–39 29. Powner DJ, Bernstein IM: Extended somatic support for pregnant women after brain death. Crit Care Med 2003; 31:1241–1249 30. Bernstein IM, Watson M, Simmons GM, et al: Maternal brain death and prolonged fetal survival. Obstet Gynecol 1989; 74(3 Pt 2): 434 – 437 31. Field DR, Gates EA, Creasy RK, et al: Maternal brain death during pregnancy: Medical and ethical issues. JAMA 1988; 260:816 – 822 32. Dillon WP, Lee RV, Tronolone MJ, et al: Life

Crit Care Med 2005 Vol. 33, No. 10 (Suppl.)

33.

34.

35.

36.

37.

38.

support and maternal death during pregnancy. JAMA 1982; 248:1089 –1091 Nuutinen LS, Alahuhta SM, Heikkinen JE: Nutrition during ten-week life support with successful fetal outcome in a case with fatal maternal brain damage. JPEN J Parenter Enteral Nutr 1989; 13:432– 435 Vives A, Carmona F, Zabala E, et al: Maternal brain death during pregnancy. Int J Gynaecol Obstet 1996; 52:67– 69 Lewis DD, Vidovich RR: Organ recovery following childbirth by a brain-dead mother: A case report. J Transpl Coord 1997; 7:103–105 Catanzarite VA, Willms DC, Holdy KE, et al: Brain death during pregnancy: Tocolytic therapy and aggressive maternal support on behalf of the fetus. Am J Perinatol 1997; 14:431– 434 Iriye BK, Asrat T, Adashek JA, et al: Intraventricular haemorrhage and maternal brain death associated with antepartum cocaine abuse. Br J Obstet Gynaecol 1995; 102: 68 – 69 Pratschke J, Wilhelm M, Kusaka M, et al: Brain death and its influence on donor organ

Crit Care Med 2005 Vol. 33, No. 10 (Suppl.)

39.

40.

41.

42.

43.

quality and outcome after transplantation. Transplantation 1999; 67:343–348 Pennefather SH, Bullock RE, Mantle D, et al: Use of low dose arginine vasopressin to support brain-dead organ donors. Transplantation 1995; 59:58 – 62 Iwai A, Sakano T, Uenishi M, et al: Effects of vasopressin and catecholamines on the maintenance of circulatory stability in braindead patients. Transplantation 1989; 48: 613– 617 Hauksson A, Akerlund M, Melin P: Uterine blood flow and myometrial activity at menstruation, and action of vasopressin and a synthetic antagonist. Ann N Y Acad Sci 1994; 734:47–56 Levinson G, Shnider SM, DeLorimier AA, et al: Effects of maternal hyperventilation on uterine blood flow and fetal oxygenation and acid-base status. Anesthesiology 1974; 40: 340 –347 Robertson KM, Hramiak IM, Gelb AW: Endocrine changes and haemodynamic stability after brain death. Transplant Proc 1989; 21(1 Pt 2):1197–1198

44. Nader S: Other endocrine disorders of pregnancy. In: Maternal-Fetal Medicine. Creasy RK, Resnick R (Eds). Philadelphia, WB Saunders, 1999, pp 1015–1037 45. Aboud E, Neales K: The effect of maternal hypothermia on the fetal heart rate. Int J Gynaecol Obstet 1999; 66:163–164 46. Gluckman PD, Gunn TR, Johnson BM: The effect of cooling on breathing and shivering in unanaesthetized fetal lambs in utero. J Physiol 1983; 343:495–506 47. Hamaoui E, Hamaoui M: Nutritional assessment and support during pregnancy. Gastroenterol Clin North Am 2003; 32:59 –121 48. Dominguez-Roldan JM, Murillo-Cabezas F, Santamaria-Mifsut JL, et al: Changes in resting energy expenditure after development of brain death. Transplant Proc 1995; 27: 2397–2398 49. Veatch RM: Maternal brain death: An ethicist’s thoughts. JAMA 1982; 248:1102–1103 50. Loewy EH: The pregnant brain dead and the fetus: Must we always try to wrest life from death? Am J Obstet Gynecol 1987; 157: 1097–1101

S331