Amniotic Fluid Embolism

Quisqueya T. Palacios

Introduction

In 1987, maternal mortality rates were reported to be 6.6 deaths per 100,000 live births by the Health Resources and Services Administration and held for a period of more than 10 years (1). In 2010, the World Health Organization (WHO) estimated maternal mortality rate to be approximately 17/100,000 pregnancies in the United States. This was significantly higher than the goal of 3.3/100,000 live births set by the US Department of Health and Human Services in healthy people for 2010. In the triennium 2003 to 2005, the Centre for Maternal and Child Enquiries (CMACE) reported that the maternal mortality rate was 13.95/100,000 maternities. In 2011, CMACE reported that the maternal mortality rate for the most recent triennium 2006 to 2008 was 11.39/100,000 maternities.

According to a study by the Centers for Disease Control and Prevention (CDC) of pregnancy-related mortality in the United States between 1991 and 1998, the leading causes of maternal deaths are hemorrhage, pregnancy-related hypertensive disorders, pulmonary embolism, amniotic fluid embolism (AFE), infection, and pre-existing chronic conditions, such as, cardiovascular disease (2). Clark confirmed that AFE is one of the leading causes of maternal death in addition to preeclampsia, pulmonary thromboembolism, obstetric hemorrhage, and cardiac disease (3). Combined AFE and pulmonary thromboembolism account for approximately 25% of maternal deaths (Table 22-1). An estimated 5% to 15% of all maternal deaths in Western countries are due to AFE. In the triennium 2006 to 2008, CMACE reported AFE, the fourth leading cause of direct maternal deaths in the United Kingdom. Although the majority of deaths from AFE are not preventable, it is up to the anesthesia care provider to have a thorough understanding of the anesthetic implications of AFE in order to immediately diagnose and treat AFE during pregnancy and delivery.

AFE is thought to be a rare but often fatal complication of pregnancy whose onset can neither be predicted nor prevented (4). However, early diagnosis, expeditious resuscitation and delivery, and management of sequelae by a team approach including anesthesiologist, obstetrician, and intensivist may improve maternal and fetal outcomes. Despite aggressive and early management, maternal and fetal morbidity and mortality remained unacceptably high between 60% and 80% through the mid-1990s (5). However, new strategies and innovative approaches to management and treatment of AFE, including the use of an intra-aortic balloon pump with extracorporeal membrane oxygenation (ECMO) (6,7,8), cardiopulmonary bypass (9), inhaled nitric oxide (10), right ventricular assist devices (8), recombinant factor VIIa (rFVIIa) (7,11), have been reported with success and should be considered when all established, “standard of care” management approaches have failed. Although manifestations of AFE may be along a continuum from mild and transient to severe and fulminant cardiopulmonary collapse, prompt diagnosis and aggressive management improve maternal outcomes (12).

Although Meyer first reported a case of stillbirth associated with maternal death and AFE to the lungs in 1926 (13), Steiner and Lushbaugh first described the AFE syndrome in a case report of eight women who died unexpectedly from obstetric shock associated with pathologic evidence of AFE of fetal material in maternal lung blood vessels in 1941 (14). They theorized that amniotic fluid was forced into the maternal circulation during contractions. Steiner and Lushbaugh also described experimental evidence of the syndrome in dogs and rabbits following intravenous injection of human amniotic fluid rich in vernix or meconium which leads to a similar clinical presentation as described in the autopsies of eight parturients associated with plugging of the pulmonary vessels by squamous cells, presumably of fetal origin. On the basis of experimental animal studies supporting detailed pathologic findings in eight cases of unexpected maternal death resulting from a physical obstruction of the pulmonary vasculature by fetal material, Steiner and Lushbaugh proposed a new obstetrical disease, AFE.

Although the diagnosis of AFE is clinically based, the presence of sequelae including but not limited to respiratory arrest, cardiac shock, coagulopathy and DIC, and nonreassuring fetal heart tones in association with the presence of anucleated squamous cells in the pulmonary artery blood are not pathognomonic of AFE. In 1995, Clark confirmed that fetal squamous cells were found in the pulmonary circulation in 73% of fatal cases of AFE (5). In addition, fetal squamous cells were detected in only 50% of patients with a diagnosis of AFE during aspiration of pulmonary arterial blood. Also, the presence of squamous cells in the circulation during the peripartum period is not always associated with AFE (15). Contamination of the maternal blood by fetal squames may occur during pulmonary artery insertion and may be minimized by following the method suggested by Masson (16). However, the demonstration of fetal debris is highly significant and consistent with the diagnosis of AFE. However, the high variability in symptoms, the lack of characteristic findings on radiologic examination, the absence of a dose–response effect
on symptoms, and the occurrence of coagulopathies are not entirely consistent with a physical block to the circulation as the main mechanism of disease. Alternatively, it might be the result of complement activation initiated by fetal antigen leaking into the maternal circulation. The rare immune response may be initiated by a rare pathologic antigen, or by common antigens presented uncommonly (17).

Table 22-1 Leading Causes of Maternal Death

Cause of Death	Number	(%)
Complications of preeclampsia	15	(16)
Amniotic fluid embolism	13	(14)
Obstetric hemorrhage	11	(12)
Cardiac disease	10	(11)
Pulmonary thromboembolism	9	(9)
Obstetric infection	7	(7)
Adapted from: Clark SL, Belfort MA, Dildy GA, et al. Maternal death in the 21^st century: causes, prevention, and relationship to cesarean delivery. Am J Obstet Gynecol 2008;199(1):36.e1–36.e5; discussion 91–92.e7–e11.

In 1995, Clark established a National Registry of AFE cases with 46 entries and in 2005, Tuffnell established a UK registry of AFE cases with 44 entries, both depended on self-reporting and had similar established entry criteria (5,18) (Table 22-2). Although the majority of cases of AFE occurred during labor, Clark confirmed that 19% of cases of AFE in their registry women became symptomatic during cesarean delivery when not in labor. In an analysis of the National Registry of cases of AFE, Clark reports a history of allergies in 41% of patients. Clark also reported a similarity between the clinical course, biphasic response, and hemodynamic changes of AFE to patients with anaphylactic shock and proposed that AFE was immunologic versus nonimmunologic and supported changing the name from AFE to anaphylactoid syndrome of pregnancy. Although Clark used the term anaphylactoid versus anaphylactic syndrome of pregnancy suggesting that the process associated with mast cell degranulation was not associated with an antigen and antibody, Benson suggested that placenta, fetus, and meconium-stained amniotic fluid could potentially be either sources of foreign antigens or lead to exposure to large quantities of nonantigenic materials and a fatal nonimmune anaphylaxis (19,20,21).

Table 22-2 Amniotic Fluid Embolism National and UK Registries’ Entry Criteria

Acute hypoxia
Acute hypotension/cardiac arrest
Coagulopathy
Onset of symptoms
   During labor
   Cesarean delivery
   Dilation and evacuation
   Within 30 min postpartum
Other possible diagnosis have been excluded
Occurrence within 5 yr of registry opening

Adapted from: Clark SL, Hankins GVD, Dudley DA, et al. Amniotic fluid embolism: analysis of the national registry. Am J Obstet Gynecol 1995;172(4 Pt 1):1158–1167; discussion 1167–1169.

There are many unanswered questions regarding the etiology and mechanism of AFE if an “all or nothing” mechanism is considered in the presence of conflicting results of animal models and the pathologic absence of mechanical obstruction from fetal debris. Alternatively, one may consider “a response continuum” to AFE in pregnant women with increased immunologic reactivity, many of whom may have an associated underlying subclinical sepsis, trauma, or other risk factors during labor and the immediate postpartum period as the possible etiology and mechanisms of AFE. Romero describes two such cases of maternal deaths associated with subclinical intra-amniotic infection. He proposes that the mechanism of peripartum cardiovascular collapse may be infection and systemic inflammation instead of AFE. On the other hand, I propose that infection and systemic inflammation may lead to lowering of the threshold for cardiovascular collapse seen with AFE. Therefore, in these cases laboratory tests should include specific immunologic testing, multiple blood cultures, and specifically directed antibiotic therapy in addition to following the management protocols of AFE. Further improvements in maternal outcomes may be seen if recombinant human-activated protein C is included in the management of the subset of patients with clinical evidence of sepsis, prolonged rupture of membranes, prolonged labor, meconium staining of amniotic fluid, and fever associated with cardiovascular collapse at delivery (22).

Perhaps once a certain threshold is reached, especially in the presence of other immunologic factors, such as an intrauterine infection, an early first response is respiratory with clinical dyspnea and hypoxia, associated with pulmonary arterial hypertension and severe transient vasospasm. This usually leads to right ventricular failure and cardiac arrest. If the patient survives, left ventricular failure develops. However, in patients with an associated patent foramen ovale, ASD, VSD, or PDA, perhaps this initial phase may be more transient or absent and instead replaced by an immediate, sustained, and severe left heart failure and/or DIC. Alternatively, left ventricular dysfunction may be the direct effect of endogenous mediators causing cardiac depression and coagulopathy. This might explain the presence or lack of commonly presenting clinical symptoms, such as dyspnea, hypotension, seizures, DIC, and nonreassuring fetal status. Immune tolerance may explain why all mothers do not develop an immune response to their fetus as seen in Rh-negative mothers who do not develop isoimmunization on subsequent pregnancies or why all mothers do not reject the fetus (17). In addition, a better understanding of immune tolerance may help elucidate the pathophysiology of AFE and other diseases such as, preeclampsia and recurrent miscarriage.

Incidence and Mortality

Historically, the incidence of AFE in the United States is estimated to occur between 1 in 8,000 and 1 in 80,000 deliveries. However, more recently in 2008, Abenhaim in a retrospective population-based study on 3 million birth records in the United States from 1999 to 2003 estimated an incidence of approximately 7.7/100,000 deliveries or 1:13,000 deliveries (23).

In 2006, Kramer in a retrospective population-based hospital database in Canada estimated the incidence of AFE in 6.1 cases per 100,000 births (24). In a prospective national cohort study which included data from 3 million hospital deliveries from 1991 to 2002 and utilized information from the UK Obstetric Surveillance System (UKOSS) in the United Kingdom, Knight estimated a significantly lower incidence of 2/100,000 deliveries (25). Recently in the UKOSS, Dawson reported an incidence of AFE of 2 cases per 100,000 maternities for the 4-year period, 2005 to 2009 (26). In 2010, Roberts reported the AFE incidence rate of 3.3/100,000, maternal mortality rate of 35%, and perinatal mortality rate of 32%
in an Australian population-based cohort study (27). Newly identified risk factors included induction with vaginal prostaglandin and manual removal of the placenta.

In 1979, Morgan reviewed 272 documented cases of AFE in UK medical literature and reported a mortality rate of 86%. Of those, 25% of the deaths occurred within the first hour of the onset of symptoms. In 1995, Clark published the national registry of AFE and reviewed 46 cases of AFE and reported the maternal mortality rate of 61% (5). In Clark’s study, more than 50% of patients died within the first hour and two-thirds of these deaths occurred within 5 hours of the AFE and only 15% of survivors remain neurologically intact. However, more recently, Gilbert reported a lower mortality rate of 27% in a population-based study in 1999 (28).

In 2005, Tuffnell also suggested a lower mortality rate of 37% in UK Registry (18,25). In 2011, CMACE confirmed AFE is the fourth leading cause of maternal mortality in the triennium 2006 to 2008 in the United Kingdom (29). CMACE reported on the deaths of 13 out of 261 women in the United Kingdom who died directly or indirectly related to pregnancy giving a maternal death rate of 0.57/100,000 directly associated with AFE. Although this represents a decline in maternal mortality, consistent with the report of the latest morbidity study from the UKOSS, it is not statistically significant. In addition, substandard care was implicated in 62% (major 15%, minor 46%) of these cases as the cause of deaths (30). Implicated in substandard care were poor organization of transfers, communication breakdowns, poor documentation, ineffective resuscitation, and avoidable delays in performing perimortem cesarean delivery within 5 minutes of collapse, which may have contributed to poor outcomes and deaths. In addition, mortality was associated increasingly with minority ethnicities and the Black African group. In 2006, Kramer reported a maternal mortality rate of 13% in a Canadian population-based cohort study (24).

Figure 22-1 Possible sites of AFE into the maternal circulation.

In 2008, Abenhaim estimated a case mortality rate of 21.6% (23). He also found that AFE was associated with maternal age greater than 35, placenta previa, cesarean delivery in addition to preeclampsia, abruptio placenta, and the use of forceps. Abenhaim recommends the continuation of the National Registry of AFE to collect and review differences in management practices and outcomes in order to develop evidence-based algorithms for the treatment of AFE. In addition to newer management strategies and better outcomes, perhaps some of these patients did not present with all of the classic features of AFE and the wide range of maternal mortality and morbidity may describe a response continuum to AFE.

In 1995, Clark reported a neonatal mortality rate of 20% to 25% and only 50% of survivors remain neurologically intact. As maternal resuscitation and maternal outcomes improve, neonatal outcomes should also improve.

Etiology

Usually throughout pregnancy, intact membranes separate the maternal circulation from the amniotic fluid. For AFE to occur amniotic fluid must find a mode of entry into the maternal circulation. This is usually associated with rupture of membranes in 78% of cases. Symptoms of AFE occur in 14% of patients within 3 minutes of rupture of membranes (5). Potential modes of entry include the intrauterine pressure catheter, uterine trauma, small tears in the lower uterine segment, and the endocervix during placental abruption at the placental implantation site (Fig. 22-1). In addition, it has been
assumed that there is a bulk movement of amniotic fluid due to a pressure gradient which facilitates the easy entry of amniotic fluid into blood vessels in the uterus. Karetzky suggests that chemical mediators of AFE move largely down an electrochemical gradient (31). Although some describe induction of labor with oxytocin and strong contractions as possible risk factors of AFE, tetanic contractions would actually impede the entry of amniotic fluid into the maternal circulation. Perhaps, strong and hypertonic contractions are the result of vasospasm of the uterus, myometrium, and vasculature following the previous entry of amniotic fluid.

Typically AFE occurs during labor and delivery or in the immediate postpartum period. Approximately 70% of cases occur before delivery and the remainder can occur as late as 48 hours postpartum. In 13% of cases, AFE occurs prior to the onset of labor. However, there are reported cases of AFE following induced abortion, transabdominal amniocentesis, abdominal and surgical trauma, and cerclage removal.

Risk Factors

Although predisposing factors, such as advanced maternal age, multiparity, and tumultuous labor, were initially considered predisposing to AFE by Steiner, others could not confirm any direct correlation. However, macrosomia, advanced gestational age, amnioinfusion, induction of labor were considered predisposing factors to AFE by Morgan. However, more recently, Abenheim identified only advanced maternal age and cesarean delivery as risk factors in a large population-based cohort study (23). In 1995, Clark published the analysis of the AFE Registry and only confirmed ruptured membranes as a predisposing risk factor to AFE; he included the demographic characteristics of patients with AFE (Table 22-3). Although Clark suggests only one predisposing risk factor associated with AFE, others continue to refer to additional risk factors. However, Knight described a possible increased risk of dying from AFE in older, ethnic-minority women for reasons related to underlying medical problems and/or access to care. Oi suggests fatal factors of clinical manifestations and laboratory testing in patients with fatal AFE: Multiparity, cardiac arrest, dyspnea, or loss of consciousness; higher sialyl levels were also seen in cases of fatal AFE (32).

Table 22-3 Demographic Characteristics of Patients with Amniotic Fluid Embolism

Factor	Mean (+/-SD)
Maternal age	27 (+/-9)
Gravidity	3 (+/-2)
Parity	2 (+/-2)
Maternal weight (kg)	73 (+/-11)
Gestational age (wk)	39 (+/-2)
Birth weight (g)	3519 (+/-732)
Race White Hispanic Black Asian	Number of patients (%) 29 (63) 8 (17) 7 (15) 2 (4)
Male fetus	35/37 (67)
Twin gestation	1 (2)
Prior elective abortion	9 (20)
Prior spontaneous abortion	8 (17)
History of drug allergy or atopy	19 (41)
Adapted from Clark SL, Hankins GVD, Dudley DA, et al. Amniotic fluid embolism: analysis of the national registry. Am J Obstet Gynecol 1995;172(4 Pt 1):1158–1167; discussion 1167–1169.

Advanced Maternal Age and Multiple Pregnancies

However, Knight showed evidence of an association of AFE with multiple pregnancies, and older, ethnic-minority women in addition to induction of labor. In addition, Knight found an association of Cesarean delivery with postnatal AFE. Knight carried out a prospective population-based cohort study with case-control analysis, using the UKOSS from which 60 women having AFE in the United Kingdom were identified between February, 2005 and February, 2009. Knight estimated an incidence of AFE of 2/100,000 deliveries. Maternal mortality was estimated 20% and perinatal mortality was estimated 135/1,000 live births (25). Women who died were significantly more likely to be older and from ethnic-minority groups. Twenty-six of sixty women had AFE after delivery; nineteen of twenty-six women or 73% of these had AFE after cesarean delivery. Ten of the nineteen women who had AFE after cesarean delivery were not in labor at the time of cesarean. Fifty-five of sixty women or 92% developed AFE within 45 minutes of ruptured membranes. Women with AFE presented with premonitory symptoms as the first sign of AFE in 30% of cases, followed by shortness of breath in 20% and fetal bradycardia or other nonreassuring fetal heart tones in 20% of cases. Mulder described a case of a 44-year old Mexican multigravid woman who developed fetal bradycardia associated painful contractions following artificial amniotomy with meconium-stained fluid seen. Patient arrested and resuscitation was not successful and male infant subsequently died following cesarean delivery under general anesthesia. Microscopic examination of blood in the right atrium revealed multiple cells of trophoblastic origin and nucleated fetal squames within terminal branches of pulmonary arteries (33).

Amnioinfusion and Insertion of Intrauterine Pressure Catheter

In 2008 and 2010, Matsuo and Harbison described two cases of anaphylactoid syndrome after placement of an intrauterine pressure catheter. Although placement of an intrauterine pressure catheter is a routine procedure in labor and delivery, it is associated with a few complications, such as, trauma to uterus and placenta associated with placental abruption, uterine perforation, and endometritis. Uterine trauma may lead to disruption of the separation between the maternal circulation and amniotic fluid, a risk factor for AFE. Although there are only two documented cases of anaphylactoid syndrome of pregnancy after intrauterine pressure catheter placement, I also had a case of AFE associated with placement of an intrauterine pressure catheter confirmed by pathologic examination (34,35).

In 1994, Maher describes two cases of fatal AFE in two nulliparous patients in labor with an epidural anesthetic who received saline amnioinfusion for thick meconium staining of the amniotic fluid (36). Possible predisposing factors associated with amniotomy include the use of intrauterine pressure catheters for saline infusion which can lead to trauma to the cervical or uterine blood vessels thus providing a mode of entry. In addition, utilization of a pump under pressure which can increase the resting tone of the uterus potentially forces the amniotic fluid into the maternal circulation. The hypotension and possible toxicity associated with the use of epidural anesthesia may have contributed to the lowering of the pressure gradient further facilitating entry of the amniotic fluid into the maternal circulation. In addition, meconium
staining of the amniotic fluid potentially contains leukotrienes and may lead to an anaphylactic reaction.

Amniotomy and Amniocentesis

Mato describes a case of a 40-year-old patient at 41 weeks of gestation admitted for induction of labor for postdates who arrested immediately following amniotomy and an uneventful combined spinal–epidural 3 hours previously (37). Resuscitation was successful following emergency cesarean delivery, and hysterectomy for severe uterine atony and DIC with massive transfusion of 12 units of packed red blood cells, 24 units of pooled platelets, 8 units of cryoprecipitate, and 8 units of fresh frozen plasma. Although surgical hemostasis was achieved following ligation of uterine vessels and hysterectomy, moderate to severe oozing consistent with DIC continued. Resuscitation also included the use of recombinant coagulation factor VIIa 200 μg/kg over 30 minutes which was associated with improvement of hemostasis. Pathologic examination of peripheral blood revealed few fetal squames cells. Other possible differential diagnoses were ruled out. AFE following amniocentesis is very rare.

Atopy and Male Fetus

Clark showed that 41% of the patients had a history of atopy or known drug allergies (5). In the analysis of the national registry, Clark showed that 67% of the pregnant patients with AFE had male fetuses.

Blunt Abdominal Trauma

Trauma is a leading nonobstetric cause of maternal death in the United States. The primary causes of trauma in pregnancy include motor vehicle accidents, falls, assaults, homicides, domestic violence, and penetrating wounds. Not only is trauma associated with increased risk of abruption but also fatal hemorrhage and AFE (38). Rainio describes a case AFE in a patient at 38 weeks of gestation who sustained blunt abdominal trauma associated with improper use of a seat belt in a motor vehicle accident. Pathologic examination revealed hematoxylin–eosin staining, phloxine–tartrazine red staining of squamous material in the pulmonary blood vessels. In addition, there was positive immunohistochemical staining for cytokeratins and positive monoclonal antibody staining specific for tryptase-positive granules. The baby also did not survive and died later of anoxic brain damage and pneumonia (39). Ellingsen also describes a case of minor blunt abdominal trauma in a patient whose injuries would not have been associated with AFE had she not been pregnant, although there were no signs of abruption or uterine tear with presumed entry of amniotic fluid into the maternal circulation. Pathologic examination of the lungs revealed blood vessels with mucus and epithelial squames and the lower uterine segment revealed blood vessel with mucus (40). Pluymakers describes a case of blunt abdominal trauma due to a blow in the stomach requiring laparotomy for torsion of the left adnexa in a patient who developed tachypnea and hypotension requiring intubation and hemodynamic support. Pathologic examination of blood drawn from the pulmonary artery and bronchoalveolar lavage demonstrated fetal squames cells (41).

Cervical Suture Removal

Although AFE is most commonly seen in a laboring patient, AFE can have many presentations. Haines described a case of presumed nonfatal AFE associated with pulmonary edema, hypoxemia and oxygen desaturation, and hypotension in a multiparous patient with a history cerclage or cervical suture removal under general anesthesia complicated by placental abruption and required emergency cesarean delivery. Patient had a history of cerclage placement in three prior successful pregnancies followed by six previous recurrent spontaneous abortions (42). This patient had several possible risk factors: Multiparity, premature separation of placenta, and presumed cervical tear during removal of the cervical suture which allowed a mode of entry of the amniotic fluid into the maternal circulation, right side of the heart, and into the pulmonary vasculature. Pluymakers decribes a case of AFE in a patient with a history of cervicouterine suture at 14 weeks who was admitted with spontaneous rupture of membranes with meconium-stained fluid, fetal demise, and suspected intrauterine infection for removal of cervicouterine suture and curettage. Patient developed signs and symptoms of AFE requiring intubation, pressure-controlled ventilation with high PEEP, 100% oxygen, up to 20 ppm nitric oxide by inhalation, and hemodynamic support. Pathologic examination of blood aspirated from the pulmonary artery and bronchoalveolar lavage revealed fetal squamous cells and mucin (41).

Cesarean Delivery

McDougall describes a case in which a patient developed AFE during a cesarean delivery under general anesthesia following delivery of the placenta (43). Although resuscitation was initially successful, the patient developed sepsis, acute respiratory distress syndrome, and acute renal failure and died on the seventh day. Postmortem examination was suggestive of AFE with evidence of diffuse alveolar damage, DIC, foci of fetal squames in the pulmonary vessels. There was no evidence consistent with a thromboembolism (44).

Epidural and Spinal Blockade

Sprung describes a case of a 27-year-old patient who had a cesarean delivery of a healthy baby for previous cesarean and failed induction under an uneventful epidural anesthesia. The patient first complained about a funny sensation in her head, became unresponsive, and developed tonic–clonic seizures immediately after delivery of the placenta and exteriorization of the uterus. Patient became hemodynamically unstable with severe hypotension, bradycardia, DIC, and respiratory arrest and desaturation. Resuscitation was successful. The epidural catheter was removed atraumatically prior to the patient receiving blood products, 4 units of FFP, 4 units of PRBCs, and 8 units of cryoprecipitate to treat the coagulopathy. If there is bleeding around the epidural insertion site, correction of coagulopathy is recommended before catheter removal. Spontaneous epidural hematomas rarely develop, but may require surgical laminectomy (45).

Pang suggests that a favorable gradient for amniotic fluid to enter maternal circulation occurred following the sympathetic block during cesarean delivery under spinal anesthesia. This contributed to dilated uterine vessels and effective pooling of fetal debris which were mobilized into the maternal circulation during treatment with vasopressors and return of sympathetic tone. Therefore, avoidance of hypotension with effective coloading of fluids and prophylactic vasopressors during cesarean delivery is recommended to avoid the possibility of circulation of amniotic fluid and debris into the maternal circulation (46).

Bastien describes a case of AFE with the presenting sign of DIC following a forceps-assisted vaginal delivery under epidural anesthesia. Patient developed epistaxis, bleeding from the epidural site, and postpartum hemorrhage associated with uterine atony. Despite aggressive resuscitation and transfusion of multiple blood products including 10 units of PRBCs, 6 packed platelets, 6 units of FFP, patient did not survive.
During coagulopathy and evidence of bleeding, removal of an in situ epidural catheter should not be attempted until the coagulopathy is treated and corrected and the patient survives (47).

Fetal Demise and Second Trimester Abortion

In 1995 Clark found three cases of AFE during second trimester abortions (5). Ray describes a case of AFE in a woman with an intrauterine fetal demise at 18 weeks of gestation. Patient also had a history of asthma and had several allergies to antibiotics. In addition, the fetus was male. Following placement of cervical osmotic dilators and aspiration of amniotic fluid, the patient had a dilation and extraction (D & E) with acute onset of paroxysmal coughing, restlessness, and peripheral cyanosis. The patient developed hypotension, shock, respiratory arrest, pulseless electrical activity, hemorrhage, and DIC. The patient was successfully resuscitated after cardiopulmonary resuscitation, fluid resuscitation, vasopressors, blood products, and uterotonics (48).

Price describes a case of a patient at 24.5 weeks with fetal demise who became profoundly hypotensive and developed severe peripheral cyanosis and DIC during dilation and suction curettage. She was successfully resuscitated and received packed red blood cells, fresh frozen plasma, and cryoprecipitate (49).

The risk of maternal mortality increases with increasing gestational age at time of abortion. The risk is 24 times greater when the abortion is at 20 weeks as compared to at 15 weeks of gestation.

Induction of Labor

Caughey, in a recent cochrane systematic review of randomized controlled trials comparing elective induction of labor versus expectant management of labor, found a decreased risk for cesarean delivery and meconium-stained amniotic fluid with elective induction of labor (50). Although Kramer found medical induction of labor nearly doubled the risk of overall cases of AFE, an association which was stronger for fatal cases, multiple factors, such as, maternal age of 35 years or older, cesarean or instrumented delivery, polyhydramnios, cervical laceration or uterine rupture, placenta previa or abruption, eclampsia, and fetal distress were also associated with an increased risk of AFE (24). Albeit a relative low risk of AFE and a need for standardization in epidemiologic studies of AFE, obstetricians should be aware of this risk when making decisions about elective induction of labor. Price describes a case of a morbidly obese patient who arrested approximately 15 hours after spontaneous rupture of membranes, 6.5 hours after a scalp electrode and an internal pressure catheter were placed and pitocin augmentation started. Resuscitation was not successful (49). Fletcher describes a case of a 41-year-old primigravida at 41 weeks of gestation who presented for induction of labor with prostin and oxytocin infusion and artificial rupture of membranes with meconium-stained fluid and developed fetal late decelerations requiring emergency cesarean delivery under general anesthesia. The patient developed respiratory and renal failure requiring mechanical ventilation and hemodialysis, hemodynamic instability, and coagulopathy requiring hysterectomy. Transthoracic echocardiography did not reveal any obstruction to blood flow and showed mild dilation of the right ventricle and atriums and normal-sized left ventricle (51). Knight also suggests that induction of labor is associated with an increased risk factor of 35%, in addition to cesarean section, multiple pregnancies, and in older ethnic-minority groups in the occurrence of AFE (25). However, many of these cases of AFE are not associated with only one potential risk factor but are multifactorial in their presentation and a direct causal relationship between induction of labor and AFE has not been established.

Multiple Gestations

In a retrospective, Canadian population-based cohort study, Kramer estimated the incidence of AFE was 6/100,000 deliveries for singleton deliveries and 14.8/100,000 for multiple-birth delivery (24). Mortality rate for women with singleton deliveries who had AFE was 13%. Papaioannou reports a possible case of nonfatal AFE in a twin pregnancy associated with premature rupture of membranes and acute respiratory failure following tocolysis with ritodrine. Patient developed severe hypoxemia, hypotension, fever, and coagulopathy 2 hours following a cesarean delivery with spinal anesthesia which required intubation and intensive care (52).

Meconium Staining of Amniotic Fluid

In experimental animal studies, amniotic fluid containing meconium was associated with a higher risk of developing AFE syndrome than filtered amniotic fluid. It was thought that the particulate matter seen in meconium was responsible for the emboli in the maternal lung. In cases of intra-amniotic infection, meconium-stained amniotic fluid is seen more frequently than clear amniotic fluid. Romero reported that immunogenic endotoxins can be detected more frequently in meconium-stained amniotic fluid as compared to clear amniotic fluid (22).

Preeclampsia, Placental Abruption, and Placenta Previa

Ratten describes two cases of AFE associated with cesarean delivery under general for placenta previa. Resuscitation was successful in one patient despite the use of epinephrine, hydrocortisone, defibrillation, and internal cardiac massage. Pathologic examination revealed the presence of amniotic fluid debris in pulmonary arterioles (53).

Only gold members can continue reading. Log In or Register to continue