What is the recommended approach for diagnosing and managing pulmonary embolism in pregnant women, according to the American College of Obstetricians and Gynecologists (ACOG)?

Diagnostic Approach for Pulmonary Embolism in Pregnant Women

The recommended approach for diagnosing pulmonary embolism (PE) in pregnant women follows a specific algorithm that begins with clinical assessment, followed by compression ultrasound for those with DVT symptoms, and then appropriate imaging based on chest radiography findings. 1

Clinical Assessment and Initial Considerations

• Maintain a high index of suspicion for PE in pregnant women, as PE is a leading cause of pregnancy-related mortality, accounting for 20% of maternal deaths in the United States 2
• Normal pregnancy can mimic PE symptoms, making clinical diagnosis difficult, with less than 10% of pregnant women with concerning features ultimately having confirmed VTE 1
• D-dimer testing should not be used to exclude PE during pregnancy due to poor specificity and inadequate sensitivity (weak recommendation, very-low-quality evidence) 2, 1
• D-dimer levels are physiologically elevated during pregnancy and increase with gestational age, limiting their utility as a screening tool 3, 4

Diagnostic Algorithm

Step 1: Assess for DVT Symptoms

• For pregnant women with suspected PE and signs/symptoms of DVT:

  • Perform bilateral venous compression ultrasound (CUS) of lower extremities 2
  • If CUS is positive, initiate anticoagulation treatment 2
  • If CUS is negative, proceed with further testing 2

• For pregnant women with suspected PE but no signs/symptoms of DVT:

  • Proceed directly to pulmonary vasculature studies rather than CUS (weak recommendation, very-low-quality evidence) 2

Step 2: Chest Radiography

• Perform chest radiography (CXR) as the first radiation-associated procedure for all pregnant women with suspected PE (strong recommendation, low-quality evidence) 2, 1
• CXR results will guide the selection between lung scintigraphy and CTPA 2

Step 3: Select Appropriate Imaging Based on CXR Results

• If CXR is normal:

  • Perform lung scintigraphy (V/Q scan) rather than CTPA (strong recommendation, low-quality evidence) 2, 1
  • V/Q scanning delivers lower radiation dose to maternal breast tissue (0.98-1.07 mGy vs. 10-60 mGy with CTPA) 2
  • Diagnostic yield of V/Q scans in pregnant women with normal CXR ranges from 75-94% 2

• If CXR is abnormal:

  • Perform CTPA rather than lung scintigraphy (weak recommendation, very-low-quality evidence) 2, 1
  • CTPA provides better diagnostic yield and can identify alternative diagnoses when CXR is abnormal 2

Step 4: Management of Non-Diagnostic Results

• If V/Q scan is non-diagnostic:
  • Proceed with further diagnostic testing rather than clinical management alone (weak recommendation, low-quality evidence) 2
  • Use CTPA rather than digital subtraction angiography (DSA) (strong recommendation, very-low-quality evidence) 2
  • CTPA has been shown to be more sensitive than DSA in detecting emboli in the general population 2

Important Considerations and Pitfalls

• Technically inadequate CTPA studies occur in 6-36% of pregnant women due to:

  • Suboptimal vascular opacification
  • Respiratory motion artifacts 1
  • Repeat CTPA after a non-diagnostic study should be undertaken cautiously 1

• For unstable patients or when studies are not available promptly:

  • Consider empiric anticoagulation while awaiting diagnostic results 1
  • The diagnostic algorithm assumes patients are stable and all studies are equally available 2

• Radiation concerns:

  • CTPA delivers higher radiation dose to maternal breast tissue compared to V/Q scanning 2
  • The lifetime relative risk of radiation-induced breast cancer in a 25-year-old woman undergoing a single CTPA is estimated at 1.011 1
  • V/Q scanning is preferred when CXR is normal to minimize maternal radiation exposure 2

• MR pulmonary angiography (MRPA) is relatively contraindicated due to uncertain long-term effects of gadolinium on the fetus 2