Arthur J. Vaught
•Pulmonary embolism accounts for 10% of maternal deaths
•The gold standard for diagnosis of pulmonary embolism in pregnancy is computed tomographic angiography of the pulmonary arteries
•Pulmonary embolism is considered massive (10%) if it is accompanied by systemic arterial hypotension resulting in shock
•Treatment considerations for massive pulmonary embolism include systemic thrombolytic therapy, catheter-directed thrombolysis, or emergent thrombectomy
•Amniotic fluid embolism occurs in 2–6 per 100,000 deliveries characterized by cardiovascular collapse, disseminated intravascular coagulopathy, and refractory hypoxemia resulting from acute respiratory distress syndrome
Both pulmonary embolism (PE) and deep venous thrombosis (DVT) occur more frequently in pregnancy. In a pooled meta-analysis, the overall prevalence of DVT and PE in pregnancy were 1.1% and 0.2%, respectively . Although PE occurs less frequently, its associated morbidity and mortality is higher than DVT . Despite the significant decrease in maternal mortality secondary to thromboembolic events, PE still remains a leader of maternal hospital cost, severe maternal morbidity, and prolonged hospital stay, and accounts for 10% of maternal deaths . See Figure 17.1.
Obesity, hypertensive disorders, postpartum hemorrhage, and inherited thrombophilia are all risk factors for PE in pregnancy . In fact, the physiologic and anatomic changes of pregnancy further augment these risks by increasing clotting factors and compression of the inferior vena cava (IVC) causing decreased venous flow and stasis . Although all of these factors increase the risk of a PE or thromboembolic event, the most important risk factor is a history of prior venous thromboembolic event (VTE). The risk of recurrent VTE during pregnancy increases by fourfold, and one-quarter of all VTEs in pregnancy are recurrences . Many women with VTE do not get a diagnosis of thrombophilia during pregnancy; however, these mutations certainly increase the risk and likelihood of having a clot. Routine screening for thrombophilia is not recommended; however, it is reasonable to screen if there is history of multiple VTE in a given woman, a strong family history, or if the VTE was severe (i.e., massive pulmonary embolism) .
Diagnosis of pulmonary embolism may be challenging in pregnancy, as shortness of breath and tachycardia are common features of normal pregnancy. (See Chapter 3 for more details of physiologic changes.)
Classic clinical signs and symptoms of PE include but are not limited to hypoxemia, tachypnea, changes in cardiac imaging studies, mild elevations of troponin levels, and tachyarrhythmia. Common presenting symptoms are listed in Table 17.1. It is important to note that symptoms of PE are neither sensitive nor specific, and therefore the obstetric practitioner should continue to have a wide differential when diagnosis of PE is being entertained.
Clinical Features and Diagnostic Testing in Pulmonary Embolism
Signs and Symptoms
Arterial blood gas
Chest CT angiography
Pleuritic chest pain (66%)
Lower extremity Doppler
Pro-brain natriuretic peptide or B-type natriuretic peptide
Includes pneumonia, myocardial infarction, preeclampsia, cardiac failure, and non-cardiogenic pulmonary edema.
Oddly enough, significant hypoxemia was a rare finding in PE, with many patients with oxygenation >95% . Cardiac arrhythmias are present in approximately 10% of patients diagnosed with a PE, usually a tachyarrhythmia. Although sinus tachycardia is the most common tachyarrhythmia, others can present with atrial fibrillation, atrial flutter, and multifocal atrial tachycardia .
Arterial blood gas is also clinically useful. Often blood gases will show a respiratory alkalosis and hypoxemia or normal oxygenation. Other lab parameters abnormally elevated in PE are D-dimer, cardiac troponins, and brain natriuretic peptide (BNP) levels. It should be noted that D-dimer levels increase in pregnancy and are not considered reliable in diagnosing PE during pregnancy [4,7–10]. Likewise, false negative results for D-dimer have been reported, and therefore D-dimer test is not used as part of the workup for PE .
Although not used to diagnose PE, cardiac troponins and BNP can be helpful in the risk stratification of patients with PE. Overall, approximately 50% of patients with PE will have a “positive” or elevated cardiac troponins . Further, approximately 30% of the patients exhibit right ventricular dysfunction that is associated with worsened short- and long-term mortality . BNP has also been shown to help stratify patients with PE into lower and higher risk categories. In particular, nonpregnant patients with pro-BNP levels >600 pg/mL were at increased risk of short-term mortality secondary to PE . Therefore, it is reasonable to check cardiac troponins and BNP levels at diagnosis of PE. High levels can help stratify patients into low- and high-risk categories which could aid practitioners in obtaining additional imaging (i.e., echo), consultation (cardiology, hematology, cardiac surgery), and transfer to higher levels of care (critical care, hospital transfer).
Although practitioners can use physical exam and lab data to aid in the diagnosis of PE, the gold standard for diagnosis is computed tomographic angiography (CTA) of the pulmonary arteries, even in pregnancy . In clinical practice, practitioners may argue against CTA for fear of radiation exposure to the fetus; however, this exposure is very low and rarely if at all causes fetal harm from just one CTA . Further, the fetus can be shielded by placing the lead on the maternal abdomen if there is concern. In clinical scenarios of renal injury, failure, or severe contrast allergy, ventilation-perfusion scan is a reasonable alternative; however, this imaging modality also has radiation exposure . Fetal radiation exposure from CTA is lower than ventilation-perfusion scan (0.32–0.64 mGy vs. 0.003–0.1398 mGy); however, the maternal radiation exposure to the breast is lower with ventilation-perfusion scan . See Table 17.2.
Risks and Benefits of Imaging Modalities in Pulmonary Embolism
CT Pulmonary Arteries
Sensitivity of 90%–95%
Detects other pulmonary abnormalities, i.e., pneumonia, fibrotic disease
Sensitivity varies on probability
Low: 4% risk of PE
Intermediate: 15% risk of PE
High: 16% risk of PE
Scoring further changes based on clinical probability
Diagnosis right ventricular dysfunction
Not gold standard
Needs confirmatory test
After a PE is diagnosed it should be categorized appropriately to help delineate severity, treatment options, and reduce maternal morbidity.
Types of Pulmonary Embolism
The three types of PE are subsegmental, submassive, and massive. Many PEs experienced in obstetric practice are subsegmental and can be treated with careful therapeutic anticoagulation. However, massive and submassive PE account for 5–10% and 20%–25% of cases, respectively, in the general population and carry an increase in morbidity and mortality [17–19]. A PE is considered massive if it is accompanied by systemic arterial hypotension resulting in shock . A PE is considered submassive when systemic hemodynamics are preserved but there is evidence of right ventricular dysfunction on either electrocardiogram, echocardiography, or CTA. The 90-day mortality for a massive PE can reach as high as 50%, and up to 19% for submassive PE. Therefore, when a PE is diagnosed proper categorization should be undertaken.
Management of Pulmonary Embolism
One could argue that the mainstay of treatment of PE in pregnancy is prevention. Ideally, personal or family history of thromboembolism should be reviewed for each and every pregnant patient to identify those at high risk of complications. Patients without current PE but with risk factors should be considered for prophylactic, intermediate, or adjusted dose low molecular weight heparin (LMWH) or unfractionated heparin (UFH) per ACOG guidelines . When selecting an appropriate dose of LMWH or UFH for prevention, maternal weight, past medical history, severity of previous disease (i.e., massive PE), thrombophilia workup (high-risk thrombophilia vs. low-risk thrombophilia), and other hypercoagulable comorbidities (nephrotic syndrome, ulcerative colitis, antiphospholipid antibody) should be considered.
Women diagnosed with a PE should be treated with full anticoagulation for at least 3–6 months and for at least 6 weeks postpartum. Although many practitioners will keep patients on full anticoagulation, it is reasonable to decrease this dose to intermediate or prophylactic dosing after the initial 3–6 months of therapy into the postpartum period . Acceptable therapeutic dosing for LMWH and UFH are 1 mg/kg every 12 hours or 250 units/kg every 12 hours, respectively. Usually, patients receiving LMWH do not need anti-factor Xa levels tested. However, patients with renal injury, morbid obesity, and antithrombin III deficiency should be considered for level testing which is 4–6 hours after LMWH administration. An anti-factor Xa level of 0.6–1.0 u/mL is considered therapeutic range . When using UFH, a goal aPTT 1.5–2.5 times control 6 hours after injection is considered therapeutic (see Chapter 9 for more details).
One of the central issues the obstetrician faces is managing full anticoagulation around the timing of delivery. Full anticoagulation is not a contraindication to vaginal delivery or neuroaxial blockade but timing and coordination with obstetric anesthesia is paramount. Decisions regarding delivery timing should be based on usual obstetric and maternal indication; however, it is reasonable to time an induction for coordination purposes after 39 weeks. In general, for women receiving prophylactic LMWH, discontinuation is recommended 12 hours before neuroaxial blockade and 24 hours before an adjusted dose regimen . Concerning UFH, doses greater than 7500 units twice daily need a 12-hour interval from the most recent dose as well as coagulation studies before induction or neuroaxial blockade placement .
The optimal time to resume full anticoagulation therapy postpartum is unknown. Obstetricians must weigh severity of disease burden and risk of hemorrhage or bleeding. A reasonable approach to minimize bleeding is to resume therapy no sooner than 4–6 hours after vaginal delivery and 6–12 hours after cesarean section [4,22]. Although it is safe to use LMWH in the postpartum period, it is reasonable to use intravenous UFH protocol to avoid further hemorrhage.
Warfarin is a vitamin K antagonist that easily crosses the placenta. In the setting of acute VTE and PE in pregnancy, warfarin is seldom used secondary to increased fetal risks which include fetal embryopathy (nasal bone hypoplasia and stippled epiphyses), fetal bleeding (intracranial, intraabdominal hemorrhage), and fetal loss [23,24]. Although warfarin is shown to be teratogenic, it is dose dependent and a dose of less than 5 mg/daily is rarely associated with poor fetal outcome. Although warfarin is infrequently used antenatally, it can be used in the postpartum period as it safe in breastfeeding [25,26].
New oral anticoagulants (NOAC) have shown great promise in the nonpregnant populations. They are preferred over LMWH in the prevention and treatment of VTE and PE, as there is no transition therapy needed for many NOACs; they are also superior in reducing the recurrence in VTE and PE, and the risk of bleeding appears less when compared to warfarin [27,28]. Despite these advances, LMWH is still the preferred anticoagulant in pregnancy and postpartum. There are no randomized trials on NOAC versus LMWH in pregnancy or in the immediate postpartum period. Further, fetal effects are unknown, and the medication is in breast milk . There are also treatment failures noted in case reports with use postpartum. The pharmacologic failure is thought to be secondary to a shared hypermetabolic renal and hepatic system in the pregnant and immediate postpartum woman .
Treatment of Massive and Submassive Pulmonary Embolism
As stated earlier, submassive and massive PE are associated with high mortality; this is increasingly complex and morbid in the setting of a third trimester pregnancy. Although systemic thrombolysis with tissue plasminogen activator (tPA) improves pulmonary perfusion and improves right-sided heart failure, it is not without risk . In fact, systemic tPA use has a major bleeding risk of 10% and a 3%–5% risk of hemorrhagic stroke . Therefore, use of systemic tPA is FDA approved for massive PE, but use for submassive PE remains controversial. In addition, tPA is “relatively” contraindicated in the setting of pregnancy and recent surgery (cesarean section) because of risk of urgent delivery and hemorrhage . Therefore, other targeted options in the setting of pregnancy may be preferred if available.
In a systematic review of pregnant patients, 83 severe PE in pregnancy received systemic thrombolytic therapy; 61 were antenatal and 9 were treated with catheter-directed thrombolysis (CDT) or emergent thrombectomy. Of the cohort, 80% were massive PEs and 20% underwent cardiac arrest. Among all massive PEs, maternal survival was 93%, with 96% survival antepartum and 85% postpartum . Systemic thrombolysis was efficacious in hemodynamic improvement, and the five women who died received thrombolysis during or after cardiac arrest . Major maternal bleeding occurred in approximately 30% of pregnant cases treated with systemic thrombolysis, with the greatest risk in the immediate postpartum period. In antenatal massive PE, fetal and neonatal death was 12%, and spontaneous preterm labor occurred in 14% of pregnancies .
In the review, 36 pregnancies underwent surgical thrombectomy, of which 13 had cardiac arrest. In this cohort, maternal survival was 84%, and major bleeding was prevalent at 20%. When thrombectomy occurred in the antepartum period without combined delivery, fetal loss reached 20%.
CDT represents an alternative to systemic thrombolysis and surgical thrombectomy as it acutely preserves right ventricular function as well as immediate pulmonary perfusion . CDT with thrombectomy is reported in the literature, with seven cases reported. In these seven cases, two had suboptimal results and required use of extracorporeal membranous oxygenation (ECMO) . However, maternal survival in this group was 100%, with only one episode of major bleeding, or 20%; fetal loss was 25%.
In the setting of massive PE, practitioners should consider systemic thrombolysis, CDT thrombolysis, and thrombectomy as well as surgical thrombectomy and ECMO.
All of these therapeutic options carry high risk of major hemorrhage, stroke, and fetal loss. However, overall outcomes are promising, with survival reaching 90% in the pregnant population. The obstetric specialist must balance risk and benefits in these settings optimally with a multidisciplinary team, including intensivist, cardiac surgery, and skilled interventional radiologist if available.
Amniotic Fluid Embolism
Amniotic fluid embolism (AFE) is a rare and usually lethal condition characterized by cardiovascular collapse, disseminated intravascular coagulopathy (DIC), and refractory hypoxemia resulting from acute respiratory distress syndrome (ARDS). The exact incidence of AFE is unknown but is estimated to affect 2–6 women per 100,000 deliveries . In the past, diagnostic criteria were lacking; however, the Society of Maternal Fetal Medicine and the Amniotic Fluid Embolism Foundation created criteria. The diagnostic criteria are intended for research purposes but can also aid in clinical diagnosis (Table 17.3) .
Diagnostic Criteria for Amniotic Fluid Embolism
Sudden onset of cardiorespiratory arrest or refractory hypotension
Overt disseminated intravascular coagulopathy (Score ≥3)
PT or INR
Clinical onset during labor or within 30 minutes of placental delivery
Absence of fever (38°C) during labor