What is the recommended diagnostic approach for suspected pulmonary embolism (PE)?

Diagnostic Approach for Suspected Pulmonary Embolism

The recommended diagnostic approach for suspected pulmonary embolism (PE) should be based on clinical probability assessment followed by selective testing, with D-dimer testing for low and intermediate risk patients, and immediate imaging for high-risk patients. 1

Initial Risk Stratification

The first step in evaluating patients with suspected PE is to determine the pretest clinical probability using either:

• Validated clinical decision tools (Wells score or revised Geneva score)
• Clinical gestalt (physician's implicit assessment)

This risk stratification is essential as it determines the subsequent diagnostic pathway 1.

Risk-Based Diagnostic Algorithm

For Low Clinical Probability Patients:

1. Apply the Pulmonary Embolism Rule-Out Criteria (PERC):

   • Age < 50 years
   • Heart rate < 100 beats/minute
   • Oxygen saturation ≥ 95%
   • No history of venous thromboembolism
   • No recent trauma or surgery
   • No hemoptysis
   • No estrogen use
   • No unilateral leg swelling

2. If all PERC criteria are met (PERC negative):

   • No further testing is required - PE can be safely excluded 1, 2
   • This approach has a negative predictive value of 99.5% 3

3. If any PERC criterion is not met (PERC positive):

   • Proceed to D-dimer testing 1
   • If D-dimer is normal (< 500 ng/mL or age-adjusted threshold for patients >50 years): No further testing needed
   • If D-dimer is elevated: Proceed to imaging (CTPA preferred)

For Intermediate Clinical Probability Patients:

1. Perform D-dimer testing:

   • If normal (< 500 ng/mL or age-adjusted threshold): PE excluded
   • If elevated: Proceed to imaging 1

2. Age-adjusted D-dimer threshold:

   • For patients >50 years: age × 10 ng/mL
   • This maintains sensitivity >97% while significantly improving specificity 1

For High Clinical Probability Patients:

1. Proceed directly to imaging without D-dimer testing 1
   • CTPA is the preferred imaging modality when available and not contraindicated
   • V/Q scan should be used when CTPA is unavailable or contraindicated

Important note: Recent evidence suggests that D-dimer testing may be safe even in high-probability patients, with a failure rate of 0% in one study, though confidence intervals were wide (0.0%-6.5%) 4. However, current guidelines still recommend proceeding directly to imaging for these patients.

Imaging Modalities

Computed Tomographic Pulmonary Angiography (CTPA)

• First-line imaging when available and not contraindicated
• High sensitivity and specificity (95% and 98% respectively)
• Can identify alternative diagnoses

Ventilation-Perfusion (V/Q) Scan

• Alternative when CTPA is unavailable or contraindicated (contrast allergy, renal insufficiency)
• Normal perfusion scan reliably excludes PE (Level A evidence) 1
• High-probability V/Q scan confirms PE (Level A evidence) 1
• Non-diagnostic scans may require additional testing

Lower Limb Compression Ultrasound (CUS)

• May be considered in selected patients to obviate the need for further imaging if positive 1
• Finding of proximal DVT confirms PE (Level B evidence) 1
• Particularly useful when imaging of the pulmonary vasculature is contraindicated

Special Considerations

Hemodynamically Unstable Patients

• For patients with shock or hypotension (high-risk PE):
  • Emergency CT or bedside echocardiography is recommended depending on availability and clinical circumstances 1
  • Echocardiography can identify right ventricular dysfunction suggesting PE

Pregnant Patients

• Modified diagnostic approach may be needed
• V/Q scan may be preferred over CTPA in early pregnancy
• CTPA causes less fetal radiation exposure than V/Q scan except in third trimester

Common Pitfalls to Avoid

1. Skipping clinical probability assessment - This is the essential first step that determines the entire diagnostic pathway

2. Ordering D-dimer in high clinical probability patients - This is not recommended by guidelines as a negative result does not safely exclude PE 1

3. Failure to use age-adjusted D-dimer thresholds - Using age × 10 ng/mL for patients >50 years maintains sensitivity while improving specificity

4. Applying PERC criteria to all patients - PERC should only be applied to patients with low clinical probability, not as a universal screening tool 2

5. Overreliance on single test results - The diagnostic approach should follow the evidence-based algorithm based on pretest probability

By following this structured approach, clinicians can safely diagnose or exclude PE while minimizing unnecessary testing and radiation exposure.