What are the indications for a D-dimer (Deep vein thrombosis/pulmonary embolism marker) test in adults with suspected deep vein thrombosis (DVT) or pulmonary embolism (PE)?

D-Dimer Test Indications in Adults with Suspected DVT or PE

D-dimer testing should only be ordered in patients with suspected venous thromboembolism who have LOW or MODERATE pretest probability of disease—never in high-risk patients who should proceed directly to imaging. 1

Mandatory First Step: Clinical Pretest Probability Assessment

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

• Wells criteria for DVT or PE - assigns points based on active cancer, immobilization, localized tenderness, leg swelling, previous VTE, heart rate >100, hemoptysis, and likelihood of alternative diagnosis 1
• Geneva score - alternative validated tool for PE risk stratification 1
• Categorizes patients into low (5% DVT prevalence), moderate (17% prevalence), or high (53% prevalence) risk groups 1

D-Dimer Testing Algorithm by Risk Category

Low Pretest Probability Patients

• Highly sensitive D-dimer assay (ELISA or turbidimetric) is the first-line test 1
• If D-dimer is negative (<500 ng/mL for age ≤50 years), VTE is excluded with 99% negative predictive value—no imaging needed 1, 2
• For patients >50 years old, use age-adjusted cutoff (age × 10 ng/mL) to maintain sensitivity >97% while improving specificity from 34% to 46% 3, 4
• If D-dimer is positive, proceed to imaging (compression ultrasound for DVT, CT pulmonary angiography for PE) 1

Moderate Pretest Probability Patients

• Highly sensitive D-dimer is recommended as initial test (Grade 2C) 1
• Negative D-dimer safely excludes VTE without further testing 1, 2
• Positive D-dimer requires imaging confirmation 1
• The number needed to test is approximately 3 patients to exclude one PE in this population 3

High Pretest Probability Patients

• Do NOT order D-dimer—proceed directly to imaging 1
• Even highly sensitive D-dimer assays have insufficient negative predictive value in high-risk patients 1, 5
• Two of nine patients with confirmed PE and negative D-dimer in one study had high pretest probability 5

Critical Assay Selection Requirements

Only highly sensitive quantitative assays (ELISA, enzyme-linked, or latex platforms) should be used for low-to-moderate risk patients 1:

• Sensitivity must be ≥95% with negative likelihood ratio of 0.07 3, 4
• Moderate sensitivity qualitative (point-of-care) assays are acceptable only for low pretest probability patients (Grade 2C) 1
• Point-of-care assays have lower sensitivity (88% vs 95%) and should be avoided when laboratory-based tests are available 4

Populations Where D-Dimer Has Limited Utility

Avoid D-dimer testing or interpret with extreme caution in 1, 3:

• Recent surgery or major trauma (likely positive regardless of VTE presence)
• Active malignancy (29% of patients with D-dimer >5000 ng/mL have cancer) 3
• Hospitalized patients (specificity drops dramatically but sensitivity remains high) 3, 4
• Pregnancy (normal third-trimester levels range 0.16-1.3 μg/mL, up to 2.0 μg/mL may be normal) 3
• Elderly patients >80 years (specificity drops to 10%, though age-adjusted cutoffs help) 3, 4
• Sepsis or severe infection (causes elevation independent of thrombosis) 3

Common Pitfalls to Avoid

• Never order D-dimer without first assessing clinical probability—this is the most common error 3, 4
• Never use standard 500 ng/mL cutoff in patients >50 years—leads to unnecessary imaging due to poor specificity 3, 4
• Never rely on positive D-dimer alone for diagnosis—positive predictive value is low and requires imaging confirmation 1
• Never apply PERC criteria to patients >50 years—age <50 is one of the eight required criteria 4
• Never order D-dimer in high pretest probability patients—wastes time and resources while delaying definitive imaging 1, 4

Diagnostic Outcomes and Safety Thresholds

• The acceptable miss rate for VTE diagnostic strategies is ≤2% during initial evaluation and subsequent 3-6 months 1
• Negative D-dimer combined with low/moderate pretest probability yields 0% 3-month thromboembolic risk (95% CI: 0.0-5.6%) 2
• Negative predictive value reaches 94.4-100% when D-dimer is properly combined with appropriate pretest probability assessment 6, 5