Examining the Evidence
Q Does inducing labor raise the risk of amniotic fluid embolism?
<huc>A</huc> Yes. In this retrospective, population-based study involving more than 3 million hospital deliveries in Canada over 12 years, medical induction of labor was strongly associated with fatal amniotic fluid embolism (AFE) and a near doubling of the risk of overall AFE. Maternal age (≥35), grand multiparity, cesarean and instrumental vaginal delivery, polyhydramnios, cervical laceration or uterine rupture, placenta previa or abruption, eclampsia, and fetal distress were also associated with an increased risk of AFE.
Although earlier studies suggested an association between AFE and medical induction of labor, as well as an association between AFE and the risk factors listed above, no controlled trials had confirmed or refuted this assumption until now. A population-based study by Gilbert in 19991 looked at pregnancy complications and mortality associated with AFE, rather than risk factors per se.
Incidence of AFE is hard to define
Worldwide, the incidence of AFE is estimated to range from 1 in 8,000 to 1 in 83,000 live births.2,3 In the United States, the estimate is 1 in 20,000 to 1 in 30,000 live births.4,5 In the study by Kramer and colleagues, the incidence of AFE was approximately 1 case per 17,000 singleton pregnancies, or 6 cases per 100,000—slightly higher than the 4.8 cases per 100,000 reported by Gilbert.1
The fatality rate varies, too
Overall, the reported mortality rate of AFE ranges from 26% to 90%.2,6 AFE is the fifth most common cause of maternal death in the world.7,8 In Canada, maternal deaths from AFE rank third behind cerebrovascular and hypertensive disorders.
Kramer and colleagues found a comparatively low mortality rate of only 13%. They reasoned that higher mortality rates in earlier uncontrolled case series may have been caused by a bias arising from selective reporting of more severe cases of AFE, such as fatal cases. They also postulated that the relatively constant annual rate of fatal AFE in their population argues against significant misdiagnosis.
AFE is linked to risk factors
The association between AFE and specific risk factors such as labor induction, cesarean section, and operative vaginal delivery was strengthened considerably when Kramer and colleagues restricted their analysis to fatal cases. The authors acknowledged that the link between AFE and cesarean section, instrumental delivery, and the presence of fetal distress may reflect the difficult labors that led to operative delivery—that is, reverse causality. If that is the case, it would substantiate the current belief that AFE is propagated by the tearing and shearing of fetal membranes and uterine vessels, which occurs more frequently in difficult and augmented labors with strong contractions. This theory is consistent with the authors’ finding that dystocia (probably associated with weaker contractions and early rupture of membranes) resulted in a significant reduction in the risk of AFE.
Nonfatal cases were hard to identify
The identification and analysis of nonfatal AFE cases were more problematic. Kramer and colleagues pointed out that the lower mortality rate found in their study may be secondary to some degree of overdiagnosis of nonfatal AFE, which can be difficult to identify. The lack of an absolute gold standard and the lack of specificity of the signs and symptoms of AFE may contribute to both overreporting and underreporting of this syndrome.
Compounding the dilemma is the fact that many signs and symptoms associated with AFE are also clinical syndromes that can occur in its absence, such as anaphylactic shock, left ventricular failure, sepsis, adult respiratory distress syndrome, and diffuse intravascular coagulation. Most commonly, AFE is a diagnosis of exclusion, made during pregnancy or the immediate postpartum period. However, even this temporal association lends little support to the diagnosis of AFE, because any and all of these other clinical syndromes can and do occur in isolation during and after pregnancy.
The constellation of many of the clinical sequelae, rather than isolated signs and symptoms, is what many recognize as the sine qua non of AFE. However, this hypothesis is also slightly problematic because not all cases of AFE—especially nonfatal cases—present with multiorgan disease involving cardiac, respiratory, renal, neurologic, and hematologic failure.
Choose elective induction wisely
Despite the relative rarity of AFE, the phenomenon is worthy of attention because the incidence of elective labor induction has been increasing steadily. The absolute increase in the risk of AFE among women who undergo medical induction is very small: 4 or 5 cases of AFE and 1 or 2 cases of fatal AFE for every 100,000 labors induced. However, with nearly 4 million births annually and induction rates of 20% in the United States, the incidence of AFE could escalate to 30 to 40 cases a year and 10 to 15 deaths. This statistic is alarming, especially given the increasing use of elective medical induction. Extra caution in choosing elective induction is therefore justified.
1. Gilbert WM, Danielson B. Amniotic fluid embolism: decreased mortality in a population-based study. Obstet Gynecol. 1999;93:973-977.
2. Clark SL, Hankins GD, Dudley DA, et al. Amniotic fluid embolism: analysis of the national registry. Am J Obstet Gynecol. 1995;172:1158-1167.
3. Tuffnell DJ. Amniotic fluid embolism. Curr Opin Obstet Gynecol. 2003;15:119-122.
4. Killam A. Amniotic fluid embolism. Clin Obstet Gynecol. 1985;28:32-36.
5. Sperry K. Landmark perspective: amniotic fluid embolism. To understand an enigma. JAMA. 1986;255:2183-2186.
6. Tuffnell DJ. United Kingdom amniotic fluid embolism register. BJOG. 2005;112:1625-1629.
7. Moore J, Baldisseri MR. Amniotic fluid embolism. Crit Care Med. 2005;33:S279-285.
8. Turner LA, Kramer MS, Liu S. for the Maternal Mortality and Morbidity Study Group of the Canadian Perinatal Surveillance System. Cause-specific mortality during and after pregnancy and the definition of maternal death. Chronic Dis Can. 2005;23:31-36.