When should a D-dimer be ordered in the evaluation of a suspected pulmonary embolism?

When to Order D-Dimer

D-dimer should only be ordered after clinical probability assessment using validated tools (Wells score or Geneva score), and only in patients with low or intermediate probability of pulmonary embolism—never in high probability patients who should proceed directly to imaging. 1, 2, 3

Essential First Step: Clinical Probability Assessment

Before ordering any D-dimer test, you must stratify patients using validated prediction rules 1, 2:

• Low probability patients (Wells score <2): ~3-13% PE prevalence 3
• Intermediate probability patients (Wells score 2-6): ~16-26% PE prevalence 3
• High probability patients (Wells score >6): ~36-50% PE prevalence 3

The British Thoracic Society emphasizes that clinical probability assessment must be documented before proceeding with any testing 1. This prevents physicians from being inappropriately influenced by D-dimer results when evaluating clinical likelihood 4.

When D-Dimer Should Be Ordered

Low Probability Patients

• First apply PERC criteria (8 objective criteria including age <50, HR <100, O2 sat ≥95%, no leg swelling, no hemoptysis, no recent trauma/surgery, no prior VTE, no hormone use) 2, 3
• If all PERC criteria are met: no D-dimer needed—PE is safely excluded 2, 3
• If PERC criteria not met: order D-dimer 1, 2

Intermediate Probability Patients

• Always order D-dimer as the initial diagnostic test 1, 3
• These patients cannot be managed with PERC alone 3

High Probability Patients

• Never order D-dimer—proceed directly to CT pulmonary angiography 1, 2, 3
• A normal D-dimer does not safely exclude PE in this population, even with highly sensitive assays 3, 4
• The 3-month VTE rate in high probability patients with normal D-dimer is 9.3% (95% CI 4.8-17.3%), compared to only 1.1% in low probability patients 4

Age-Adjusted D-Dimer Interpretation

The standard 500 ng/mL cutoff performs poorly in older patients due to dramatically reduced specificity 2, 3:

• Patients ≤50 years: Use standard cutoff of <500 ng/mL 2, 3
• Patients >50 years: Use age-adjusted cutoff of age × 10 ng/mL 2, 3, 5

The age-adjusted approach maintains sensitivity >97% while significantly improving specificity across all older age groups 3, 5:

• Ages 51-60: specificity increases from 57.6% to 62.3% 3
• Ages 71-80: specificity increases from 24.5% to 44.2% 3
• Ages >80: specificity increases from 14.7% to 35.2% 3

In the landmark ADJUST-PE study of 3,346 patients, age-adjusted D-dimer increased the proportion of elderly patients (≥75 years) in whom PE could be excluded from 6.4% to 29.7%, with a failure rate of only 0.3% 5.

D-Dimer Result Interpretation

Negative D-Dimer (Below Appropriate Threshold)

• PE is safely excluded—no imaging needed 1, 2, 6
• Negative predictive value is 99.5% when combined with low clinical probability 3
• The 3-month thromboembolic risk is 0% (95% CI 0.0-5.6%) 6

Positive D-Dimer (Above Threshold)

• Proceed immediately to CT pulmonary angiography 2, 3
• Remember that D-dimer has high negative predictive value but poor positive predictive value 2
• Elevated D-dimer alone is not diagnostic of PE 2

Assay Requirements

• Only use high-sensitivity D-dimer assays (ELISA or turbidimetric methods) with sensitivity ≥97% 3
• Point-of-care assays have lower sensitivity (88% vs. 95%) and should be avoided when laboratory-based tests are available 3
• Each hospital should provide information on the sensitivity and specificity of its D-dimer test 1

Special Populations

Hospitalized Patients

• D-dimer specificity is lower in inpatients due to comorbidities and systemic inflammation 3
• However, sensitivity remains high, so testing remains appropriate after proper pretest stratification 3
• A normal D-dimer with appropriate clinical probability assessment still prevents unnecessary imaging 3

Markedly Elevated D-Dimer

• D-dimer levels ≥5,000 ng/mL (approximately 10× upper limit of normal) have high positive predictive value 3
• Consider proceeding directly to imaging without formal probability scoring in this scenario 3

Critical Pitfalls to Avoid

• Never order D-dimer before clinical probability assessment—this leads to misinterpretation and unnecessary imaging 1, 2, 4
• Never order D-dimer in high probability patients—it wastes time and resources while potentially delaying definitive diagnosis 1, 3, 4
• Never use standard 500 ng/mL cutoff in patients >50 years—this leads to unnecessary imaging due to poor specificity 2, 3
• Never proceed directly to imaging in low/intermediate probability patients without checking D-dimer first—this exposes patients to unnecessary radiation and contrast risks 2, 3
• Never ignore clinical assessment when D-dimer is elevated—elevated D-dimer with low clinical suspicion does not mandate imaging 2

The British Thoracic Society data confirms that PE is present in only 9% of low clinical probability patients, with negative predictive values of 89-96% 1. This underscores why proper clinical assessment before D-dimer testing is essential to avoid both missed diagnoses and unnecessary testing 1, 4.