In 2010, the World Health Organization (WHO) estimated maternal mortality rate to be approximately 17/100,000 pregnancies in the United States (1). The maternal mortality rate for the triennium, 2006 to 2008, was 11.39/100,000 maternities compared with the 13.95/100,000 maternities reported for the previous triennium, 2003 to 2005 in the United Kingdom (2). The Centre for Maternal and Child Enquiries (CMACE) reported maternal mortality rate from thromboembolism, the leading cause of death in the United Kingdom since 1985, was 32/100,000 maternities in the triennium 1985 to 1988 compared to 1.94 and 0.79 per 100,000 maternities in the trienniums, 2003 to 2005 and 2006 to 2008, respectively. In the summary section of the UK Obstetric Surveillance System (UKOSS) report on near-miss studies, Knight reported an estimated incidence of antenatal pulmonary embolism of 1.3/10,000 maternities (3). The main risk factors identified for pulmonary embolism were multiparity, increasing maternal age, and obesity. The overall increased prevalence of obesity among pregnant women reflects the increased prevalence in the general population. The prevalence of obesity in the UK women 16 years old or more was estimated 24% in 2007 compared to 16% in 1993 (2). Maternal weight was found to most significantly impact mortality from thromboembolism compared to any other cause of death and 78% of the mothers who died from thromboembolism were overweight or obese. Despite a significant decrease in the maternal mortality rate from thromboembolism following the publication of the 2004 Royal College of Obstetricians and Gynecologists’ guideline “Thromboprophylaxis during pregnancy, labour, and after normal vaginal delivery” risk assessment during all the three trimesters and the postpartum period continues to be a key factor in reducing maternal mortality (4,5,6). In addition, the guideline for thromboprophylaxis was updated and revised in 2009 to include weight specific dosage recommendations in morbidly obese patients (7). In 2012, Schoenbeck reported on the successful use of a scoring system of individual risk factors for the development of venous thromboembolism (VTE) in pregnancy to guide the administration of dalteparin for thromboprophylaxis in high-risk pregnant patients (8). The scoring system improved clinical management of women at risk for VTE without an increase in obstetric or anesthetic morbidity.

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, pulmonary embolism was a leading cause of maternal deaths (9). In 2008, a retrospective study of medical records from all maternal deaths in a series of 1.5 million deliveries within 124 Hospital Corporation of America (HCA) hospitals in the previous 6 years reviewed individual causes of maternal deaths (10). Clark et al. confirmed pulmonary thromboembolism as a leading cause of maternal death in addition to preeclampsia, AFE, obstetric hemorrhage, and cardiac disease, accounting for 10% of maternal deaths. See Table 23-1. Clark et al. concluded that a nationwide systematic and universal adoption of VTE prophylaxis measures would essentially eliminate thrombus as a preventable cause of maternal death from pulmonary embolism. In 2010, The Joint Commission released Sentinel Event Alert, Issue 44, which recommended the use of pneumatic compression devices for women at high risk for pulmonary embolism undergoing cesarean delivery (11). In addition, their National Patient Safety Goal 16 was elevated to one of the 2010 standards for hospitals, which requires the development of written criteria describing early warning signs of a change or deterioration in a patient’s condition and when to seek further assistance. Furthermore, The Joint Commission established a call to action for hospitals to develop effective strategies for preventing pregnancy-related mortality and severe morbidity. Despite increased awareness of the risks of VTE and the use of prophylaxis, pulmonary embolism continues to be a leading cause of maternal mortality. Therefore, a thorough understanding of the anesthetic implications of the risk factors, diagnosis, prevention, and management of pulmonary thromboembolism during pregnancy and anticoagulation during neuraxial anesthesia for vaginal and cesarean delivery is essential.

Pregnant women are at increased risk of thromboembolism compared with nonpregnant women especially during cesarean delivery. Pregnant women have a two- to fivefold increase of deep venous thrombosis (DVT) and PE compared with nonpregnant women of childbearing age (12). In the retrospective review in 2008, Clark et al. confirmed pulmonary thromboembolism was associated with a maternal mortality rate of 2/100,000 deliveries following cesarean delivery which was 10 times the maternal mortality rate following vaginal delivery of 0.2/100,000 deliveries (10). The majority of these deaths, 77% (7 of 9) patients died from pulmonary thromboembolism following primary and repeat cesarean delivery compared to 22% (2 of 9) patients following vaginal delivery. None of the nine women had received peripartum thromboembolism prophylaxis with either fractionated or unfractionated heparin or pneumatic compression devices. See Table 23-2. Regardless of the mode of delivery and risk
of major surgery associated with thromboembolism, maternal mortality rate of women undergoing vaginal and cesarean delivery exceeded expectations. Furthermore, one expects to also see a similar 70% to 80% reduction in VTE in pregnant women undergoing cesarean delivery as seen in surgical patients if a universal use of medical or mechanical thromboprophylaxis been in place. This universal practice alone would reduce the maternal mortality rate attributed causally to cesarean delivery from 2 to 0.9 per 100,000 (10).

In a 5-year retrospective, population-based, case-control study from 1996 to 2000 in France, Deneux-Tharaux et al. also confirmed an increased risk of postpartum death of 3.6 times following cesarean compared to after vaginal delivery (13) (Table 23-3). Deneux-Tharaux et al. found that cesarean delivery was associated with a significantly increased risk of maternal death from complications of general anesthesia, infection, hemorrhage, and VTE.

VTE, which refers to both DVT and pulmonary embolism (PE), is associated with significant morbidity and mortality. Berg et al. found that 11% of maternal deaths during pregnancy were related to PE (9). DVT accounts for 75% to 80% of pregnancy-related VTE and pulmonary embolism accounts for 20% to 25%. The incidence of VTE is 0.025% to 0.1% or 0.5 to 3 per 1,000 deliveries. Approximately, 66% of all DVTs occur antepartum and 50% of these occur before the third trimester. In comparison, PE occurs less frequently during pregnancy but more often than DVT postpartum. Approximately 25% of patients with DVTs will develop PE if untreated. Therefore, the goal of treatment of DVT is the prevention of PE. When DVT occurs during pregnancy, it is more likely to be proximal, massive, and in the left iliac vein of the lower extremity. Distal thrombosis can occur either on the right or the left. Pelvic vein thrombosis is usually associated with pregnancy or pelvic surgery, and accounts for 10% to 12% of DVT during pregnancy and the postpartum period. DVT of the upper extremities or neck is very rare and usually associated with pregnancies with assisted reproductive technologies or complicated by the ovarian hyperstimulation syndrome. DVT and PE are often preventable and usually treatable. Approximately 30% of patients will develop a recurrent DVT. When untreated, approximately 25% of patients with DVT will develop PE. Acute PE is a common and often fatal disease. As the cesarean rate now exceeds 30% in the United States, and maternal mortality rate due to pulmonary thromboembolism is higher with cesarean deliveries, it is the responsibility of the anesthesia care providers to understand the additional risk factors, anesthetic implications, and management of pulmonary thromboembolism.

The mortality rate of PE is approximately 15%. Mortality can be reduced by prompt diagnosis and treatment. The risk of pulmonary thromboembolism is decreased to 4.5% and mortality is decreased to less than 1% in pregnant patients with DVT who receive anticoagulation.

Pregnancy is a hypercoagulable state which is associated with changes in the coagulation pathways. Factors II, VII, VIII, and X, and fibrinogen are increased and protein S levels are decreased which may lead to clot formation. Also, fibrinolysis is inhibited during the third trimester of pregnancy. Venous stasis from swelling of the lower extremity veins due to edema formation and bed rest during pregnancy leads to VTE. Venous stasis due to the gravid uterus compressing the inferior vena cava (IVC) and venous outflow of the internal iliac, femoral, and popliteal veins from the lower extremities
and pelvis, also leads to venous thrombosis. This occurs on the left side to a greater extent. During pregnancy, the site of thrombosis is usually proximal in the iliofemoral vein. However, the popliteal area is more often affected in the postpartum period. Furthermore, hormonally induced, increased venous capacitance and vascular injury increase the risk of pregnancy-associated VTE. Pregnant patients who require prolonged bed rest, smoke, are obese, or have a thrombophilia are at increased risk of DVT and PE. Maternal age, multiparity, sepsis, instrumented or cesarean delivery, hemorrhage, and cardiac disease, including the presence of atrial fibrillation and mechanical valves also increase the risk of VTE. Although the risk of VTE may be higher during the third trimester than in the first or second trimester, the risk of VTE is still increased during the first 12 weeks, clearly before the anatomic changes of pregnancy produce increased venous stasis. However, the postpartum period is also associated with an increased risk of thrombosis of the ovarian veins, which is typically related to infection and cesarean delivery. Approximately 80% of thromboembolic events are venous and 20% are arterial. Compared to pregnancy, the risk of VTE is 20 to 80 times higher during the first 6 weeks postpartum and 100 times higher during the first week postpartum.

Approximately 15% to 25% of thromboembolic events during pregnancy are recurrent during pregnancy and subsequent pregnancies, and VTE is the cause of 10% of all maternal deaths. The most important risk factor for VTE during pregnancy is a history of thrombosis. The risk of recurrence VTE during pregnancy is increased 3 to 4 times. The rate of recurrence of VTE in women not receiving prophylaxis anticoagulation was reported to be 2.4% to 12.2%. However, in women receiving prophylaxis anticoagulation, the rate of recurrence of VTE was reported to be below 2.4%. Also, patients with a history of VTE during pregnancy have a recurrence rate 4% to 15% of VTE in a subsequent pregnancy. Women with a known thrombophilia associated with a hypercoagulable state are also at increased risk of VTE during pregnancy and should receive prophylaxis. Effective prophylaxis with unfractionated heparin is usually achieved with a twice-daily average daily dose of 16,400 IU/day or 225 IU/kg of body weight per 24 hours. It is not unusual for the prophylactic daily dose to increase through the second and third trimesters.

Risk factors include a history of prior VTE, inherited or acquired thrombophilia, a maternal age greater than 35, pregnancy related and delivery complications, multiparity, obesity, surgical procedures during pregnancy including cesarean delivery, and previous history of pulmonary thromboembolism (Table 23-4).

However, most women do not require anticoagulation despite the increased risk of VTE during pregnancy and the postpartum period. Indications for anticoagulation include women with current VTE, a history of VTE, thrombophilia and a history of poor outcome during pregnancy, or presence of risk factors during the postpartum period. The indications for prophylactic or level of therapeutic anticoagulation will depend on the risk factors and require a VTE risk assessment (14). Pregnant and postpartum patients are considered low risk if less than 35 years and following vaginal delivery. The risk of DVT without thromboprophylaxis is less than 10% in nonpregnant patients. Early ambulation and hydration are recommended. Patients at bed rest and following cesarean delivery and having one to two risk factors are at moderate risk. SCDs are recommended and prophylactic anticoagulation should be considered. However, pregnant or postpartum patients with three or more risk factors in addition to one or more of the following: Prior history of VTE, or thrombophilia are at high risk. These patients have a risk of DVT without thromboprophylaxis between 40% and 80%. Prophylactic anticoagulation is recommended and SCDs may be considered. The goal at delivery is to weigh the benefit of anticoagulation while continuing to minimize the risk of VTE and bleeding.

Approximately 30% of patients develop recurrent DVTs and approximately 20% to 30% will develop long-term complications, including venous insufficiency, right-sided heart failure, post-thrombotic syndrome, and pulmonary hypertension. Post-thrombotic syndrome is a long-term complication of DVT characterized by venous stasis chronic swelling, redness, ulcerations, persistent leg pain, and increased risk of future VTE (15,16). Hospitals have taken the initiative and adapted the use of performance measures and compliance with the Joint Commission’s 2012 Hospital National Patient Safety Goal 03.05.01 to decrease the likelihood of patient harm and complications associated with the use of anticoagulant therapy and treatment of VTE, and to improve the maternal mortality rates.

Early diagnosis utilizing imaging techniques and treatment decrease the risk of pulmonary embolism. However, diagnosis in pregnant patients can be challenging since symptoms associated with pulmonary embolism can also be seen throughout normal pregnancy. Shortness of breath, tachypnea, tachycardia, palpitations, and leg swelling are seen during normal pregnancy. Patients with acute pulmonary embolism typically develop symptoms and signs immediately after obstruction of the pulmonary artery or one of its branches by a thrombus usually originating in the lower extremity. Although symptoms may be absent in 70% of patients with documented PE, symptoms may include sudden onset of dyspnea, tachypnea, pleuritic chest pain, a nonproductive cough, hemoptysis, and tachycardia. Due to the nonspecificity of these symptoms, radiographic evaluation is important to the prompt diagnosis of PE.

The diagnosis of VTE is made by clinical examination and correlation with clinical presentation in addition to laboratory studies and radiographic studies. Accurate diagnostic testing to confirm or exclude DVT or PE is necessary because the diagnosis of either requires prolonged treatment during pregnancy, prophylaxis during future pregnancies, and the use of oral contraceptives should be avoided. Unfortunately, clinical presentation of pulmonary embolism is variable and nonspecific especially during pregnancy, making accurate diagnosis difficult. When pulmonary embolism is suspected and the clinical suspicion is high for PE, anticoagulation should be considered until the evaluation is completed and the diagnosis of PE is excluded.

Each diagnostic test presents advantages and disadvantages for the pregnant patient. Chest x-ray may be abnormal in 80% of patients with PE; however, the findings are usually nonspecific. In addition, ECG may also have nonspecific findings. In 70% of patients with PE, arterial oxygen tension is low. The efficacy of compression ultrasound, ventilation/perfusion (V/Q) scanning, pulmonary angiography, and spiral computerized tomography (CT) scanning in pregnant patients has not been thoroughly evaluated.

Contrast venography requires injection of radiopaque dye into the vein below the site of the suspected thrombus. A filling defect seen during imaging has been considered the gold standard in the diagnosis of DVT in nonpregnant patients. However, compression ultrasound and impedance plethysmography have replaced contrast venography in pregnant patients (17). Compression ultrasonography utilizes color flow Doppler imaging while applying firm compression to the ultrasound transducer to detect intraluminal filling defects of the major venous systems of the legs, including the common femoral, superficial femoral, greater saphenous, and popliteal veins (16). Noncompressibility of the venous lumen is the most accurate ultrasound criteria for thrombosis. Compression ultrasound is the least invasive test, can be repeated if necessary, and does not expose the mother or fetus to radiation. Therefore, ACOG recently made a level 1A recommendation for compression ultrasound as the initial diagnostic test for new onset of DVT during pregnancy (18). If a DVT is identified, then PE can be assumed, and treatment should be started without further testing. Sensitivity and specificity of compression ultrasound are 95% and 96%, respectively in the detection of proximal DVT. However, compression ultrasound is not effective in the diagnosis of isolated calf DVT or isolated iliac vein thrombosis. Additional testing is warranted when compression ultrasound scans are negative because these proximal DVTs are associated with a high risk for embolization. However, compression ultrasound is the diagnostic test of choice in women with clinical suspicion of DVT (19). Impedance plethysmography is noninvasive and safe during pregnancy and measures the impedance to blood flow. Impedance plethysmography utilizes the application of high-frequency continuous current to the affected lower extremity and a decrease in impedance of blood flow corresponds to an increased venous outflow resistance in the deep veins of the proximal lower extremity. It is not as useful as compression ultrasound in diagnosis of DVT in the femoral, superficial femoral, or popliteal veins, most isolated calf thrombi or nonobstructive proximal thrombi.

V/Q scanning, spiral CT scanning, and pulmonary angiography are associated with dose-dependent radiation exposure, generally 10 to 37 mrad and 6 mrad for the first two, respectively. The fetus is exposed to 30% of the maternal radiation dose. Shielding reduces fetal radiation dose by 30%. However, radiation exposure greater than 5 rads are thought to potentially cause radiation-induced central nervous system fetal damage, especially during the 8 to 15 weeks of organogenesis. Although fetal exposure to radiation is minimal during V/Q scans, an initial diagnostic perfusion scan further reduces fetal exposure (20). V/Q scans are categorized into low, intermediate, high, normal, and indeterminate diagnostic probability categories according to comparative images produced by inhaled radioactive aerosol gases and radiolabeled markers injected intravenously, such as technetium-99m (21,22). A low probability refers to either no perfusion defects or nonsegmental defects, matched V/A defects of subsegmental perfusion defects. A high probability refers to either two or more mismatched segmental defects, or defects much larger than chest radiograph abnormality. See Figure 23-1. A perfusion scan is considered diagnostic if a defect of pulmonary arterial blood flow is seen in symptomatic patients. A ventilation scan will differentiate matched defects from unmatched defects in perfusion scans with defects not attributable to PE. A low, intermediate, or indeterminate result may require additional testing. Prior V/Q scan results should be reviewed since defects from prior thromboembolism may not have resolved completely. Chest radiographs should also be reviewed for atelectasis, effusions, and consolidations. Technetium-99m, the intravenous contrast media used during perfusion scan, is excreted by the kidneys and in breast milk, and fetal exposure may be decreased by increased fluid intake for 4 to 6 hours in the pregnant patient and substituting breast milk with formula for 2 days following a test. A perfusion scan should be ordered before a ventilation scan since a normal perfusion scan rules out a PE. Xe-133 is the radioactive agent utilized for ventilation scan in addition to technetium-99m. Therefore, maternal and fetal radiation exposure may be decreased by utilizing a ventilation scan when the perfusion scan is abnormal. V/S scan is the preferred diagnostic testing in pregnant women with suspected PE who have a normal chest x-ray.

If the V/Q scan is normal, no additional testing is required. However, in a large study of pregnant patients, only 3.3% of the V/Q scans were interpreted as high probability as compared to nonpregnant patients (20). In pregnant patients with a moderate to high clinical suspicion for PE, treatment should be started without additional testing when the V/Q scan has a high probability. In the same study, only 25% of V/Q scans were considered nondiagnostic in pregnant patients as compared to 47% to 57% in nonpregnant patients and required additional testing. Pregnant patients tend to be both younger and healthier than most patients requiring chest imaging and have normal ventilation and perfusion scans more frequently as compared to nonpregnant patients, 72.5% versus 27% to 36%, respectively.

D-dimers are produced from the breakdown of fibrin and levels are increased with pulmonary embolism. Therefore, a negative D-dimer test reliably excludes the diagnosis of PE in nonpregnant patients with a low clinical probability of PE. During pregnancy, D-dimers increase during gestation, after surgery, during preterm labor, preeclampsia, and placental abruption. Therefore, pregnancy decreases the accuracy of D-dimer testing and limits the efficacy in the diagnosis of VTE during pregnancy. Therefore, D-dimer testing is not recommended for the diagnostic testing of suspected DVT or PE in pregnancy or early postpartum period (19).

In a recent review of the literature, spiral CT was reported to be equivalent to pulmonary angiography in excluding PE in nonpregnant patients and has been shown to be cost effective and safe in all trimesters during the evaluation of suspected PE in pregnant patients (23). Fetal exposure to radiation with the use of intravenous contrast is minimal during spiral CT and less than during V/Q scanning (20). The use of contrast may lead to allergic reactions and renal dysfunction, and patients should be well-hydrated postspiral CT using contrast. Spiral CT scan is highly predictive of PE with a sensitivity and specificity range between 57% to
94% and 64% to 100%, respectively. However, during pregnancy, spiral CT scans have a higher nondiagnostic rate of 36% as compared to a nondiagnostic rate of 25% associated with V/Q scan. Other differential diagnoses of PE such as pulmonary infiltrate, pneumonia, and effusions may also be excluded by a spiral CT scan. Spiral CT scans are considered safe during pregnancy and have greater diagnostic value of pulmonary embolism in main and lobar arteries compared to segmental pulmonary arteries. Therefore, spiral CT scan is just as highly diagnostic of a PE as is pulmonary angiography. Spiral CT is recommended in women with an abnormal chest x-ray or when V/Q scan is inconclusive (19).