Case Study
A healthy nulliparous 36-year-old woman presented at 41 weeks’ gestation for postdates induction of labor. The progress of labor was slow, whereupon the patient became increasingly distressed and agitated and could not cooperate in order to receive epidural analgesia. Eventually, a Category I cesarean delivery for persistent fetal bradycardia was performed.
Standard general anesthesia with a rapid-sequence induction using propofol and suxamethonium was administered without complication. The patient was intubated and ventilated, and anesthesia was maintained with a mixture of oxygen, nitrous oxide, and sevoflurane. Lightly stained meconium liquor was noted at uterine incision, and a pediatrician was present to resuscitate the neonate. A 5-unit oxytocin bolus was given as standard at delivery. Following delivery, the patient’s blood pressure decreased acutely to 72/40 mmHg with a heart rate of 52 beats/min and an oxygen saturation of 88 percent on 40% oxygen. There was equal air entry with no audible wheeze on auscultation, and ventilatory pressures were normal. Despite aggressive resuscitation with 1,000 ml crystalloid, 1000 ml colloid, 100% oxygen, and lung recruitment maneuvers, there was no significant clinical improvement. Initial estimated blood loss was 700 ml, and an oxytocin infusion (10 IU/h) was started. A bedside Hemocue was 7.8 g/dl. An arterial line was inserted, and a right internal jugular central venous catheter was placed under ultrasound guidance. The central venous pressure was elevated at 18 mmHg, and initial arterial blood gases were p. 7.28, PaO2 9.6 kPa, PaCO2 6.2 kPa, base excess –10 mmol/L, and lactate concentration 5 mmol/L.
Surgery continued, during which escalating inotropic support with a norepinephrine infusion of up to 15 µg/min was required. The obstetricians noted a well-contracted uterus but ongoing abnormal oozing and suspected defective coagulation. A total of 4 units of packed red cells (PRCs), 4 units of fresh frozen plasma (FFP), and 4 units of platelets were transfused. The total estimated blood loss was 1.6 liters. The abdomen was packed and closed with an abdominal drain, and the patient was transferred to the ICU sedated and ventilated.
Immediate postoperative blood tests performed in the ICU were consistent with disseminated intravascular coagulation (DIC), with a hemoglobin concentration of 8.2 g/dl, platelet count of 63 × 109/L, white cell count of 10 × 109/L, international normalized ratio (INR) of 2.8, activated partial thromboplastin time (aPTT) of 59 s, and fibrinogen concentration of 0.6 g/liter. The chest x-ray showed bilateral interstitial infiltrates. Bedside transthoracic echocardiography showed left ventricular impairment with an ejection fraction of 30 percent. Right ventricular and pulmonary arterial pressures were normal.
The patient further received a transfusion of 2 units of PRCs, 4 units of FFP, 4 units of platelets, and 6 units of cryoprecipitate. Her coagulopathy was corrected and stable by postoperative day 1. Inotropic support was weaned off on day 3. However, the patient developed acute respiratory distress syndrome (ARDS) and required mechanical ventilation for 5 days. She was eventually extubated and discharged from the ICU on day 6.
Key Points
This patient developed fulminant cardiovascular collapse during delivery.
She exhibited premonitory symptoms, and the cardinal features of amniotic fluid embolism (AFE) syndrome: hypoxia, hypotension, and coagulopathy.
Early aggressive resuscitation and postoperative ICU support were key to her survival.
Discussion
AFE is a rare but potentially devastating obstetric complication. Owing to rarity, variation in diagnostic criteria, and a range of symptoms and signs, it can be difficult to diagnose, and diagnosis is usually one of exclusion. The prospective UK Obstetric Surveillance System (UKOSS) estimated an incidence of AFE of 2.0 per 100,000 maternities.1 It is among the top five leading causes of direct maternal death in the United Kingdom, with case-fatality estimates ranging from 11 to 61 percent.2 High-quality supportive care is the mainstay for meaningful maternal outcome.
Risk Factors
There has been a plethora of associations with AFE, including advanced maternal age, multiparity, male fetus, induction of labor, assisted or cesarean delivery, placenta previa, placental rupture, and ethnic minority. However, no consistent data have yet proven any significant causation or, moreover, justified any prospective alteration of standard obstetric practice to reduce the risk of AFE.1, 3
Pathophysiology
In 1941, Stiener and Lauschburg initially proposed the traditional concept of a simple obstructive mechanism of injury based on the finding of fetal debris in the pulmonary circulation of postmortem women who died from “obstetrical shock.”4 This has since been challenged; the close similarities between the manifestations of AFE and those of anaphylaxis and the systemic inflammatory response syndrome (SIRS) suggest a much more dynamic and complex picture. It has been suggested that the term anaphylactoid syndrome of pregnancy may better describe the pathophysiologic process and may be more suitable.
There first must be a breach of the maternal-fetal barrier and a favorable pressure gradient permitting transfer of amniotic fluid into the maternal circulation, perhaps via endocervical veins or a site of uterine trauma. Amniotic fluid contains many vasoactive and procoagulant substances, including arachidonic acid metabolites and cytokines. These components, however, have been demonstrated in pregnant women without clinical evidence of AFE. Therefore, it has been proposed that a secondary idiosyncratic humoral response must exist. This would subsequently trigger a pro-inflammatory cascade, a coagulation cascade, and potential deterioration to multiorgan failure in susceptible maternal-fetal pairs.3, 5
Clinical Presentation
AFE classically presents during labor and delivery or in the immediate postpartum period. Hallmark findings are the triad of hypoxia, hypotension with cardiovascular collapse, and DIC. Outside of this triad, there also may be confusion and agitation, seizures, breathlessness, and evidence of fetal distress.
Both hypoxia and hypotension appear to follow a temporal pattern, representing the evolving mechanism of injury, from obstructive shock arising from embolic debris in the pulmonary vasculature to inflammatory shock. Early hypoxia is likely due to severe ventilation-perfusion mismatch. During this initial phase, severe pulmonary hypertension from pulmonary vasospasm can lead to acute right ventricular failure, impairment of left ventricular filling, severe left ventricular dysfunction, obstructive shock, and cardiogenic pulmonary edema.
In those who survive the initial insult, late hypoxia occurs primarily from noncardiogenic exudative pulmonary edema secondary to alveolar-capillary membrane damage. In this later phase, pulmonary hypertension is often not evident, and the ongoing left ventricular failure that is seen may be a result of myocardial ischemia or the presence of inflammatory myocardial depressants such as endothelin and other cytokines that are implicated in SIRS. During the later phase, although obstructive and cardiogenic shock may persist, a distributive pattern of shock predominates.
In both phases, severe hypoxic encephalopathy has been implicated in brain death and long-term neurologic sequelae. Cardiac arrhythmias including bradycardia, ventricular fibrillation, pulseless electrical activity, and asystole may also present and further complicate management5 (Figure 37.1). Coagulopathy occurs in over 80 percent of cases and can itself lead to hemorrhagic shock and death. Although uncertain, it may be the result of a consumptive process or massive fibrinolysis.6
