Uses of different D-dimer levels can reduce the need for venous duplex scanning to rule out deep vein thrombosis in patients with symptomatic pulmonary embolism

Abstract

Objective: This study investigated the prevalence and distribution of deep vein thrombosis (DVT) in patients with symptomatic pulmonary embolism (PE) to establish a screening protocol to reduce unnecessary venous duplex scanning using different D-dimer level rather than single cutoff point of 0.5 microg/mL in patients with low and moderate pretest clinical probability (PTP).

Methods: The PTP score and D-dimer testing were used to evaluate 85 consecutive patients with symptomatically proven PE before venous duplex scanning. After calculating the PTP score, patients were divided into low (<or=0 points), moderate (1 to 2 points), and high (>or=3 points) PTP groups. The receiver operating characteristic (ROC) curves analysis was used to determine the appropriate D-dimer cutoff point in low and moderate PTP, with a negative predictive value of >98%.

Results: The study enrolled 81 patients. The prevalence of DVT was 63%, with 27 patients (33%) classified as low, 38 (47%) as moderate, and 16 (20%) as high PTP. DVT was detected in nine patients (33%) in the low PTP group, in 27 (71%) in the moderate group, and in 15 (94%) in the high group. In the low PTP patients, the difference in the value of D-dimer assay between positive-scan and negative-scan patients was statistically significant (9.99 +/- 7.33 vs 3.46 +/- 4.20, respectively; P = .008). Conversely, no significant difference in the D-dimer assay value between positive and negative scan results was found in the moderate PTP patients. ROC curves analysis were used to select D-dimer cutoff points of 2.0 microg/mL for the low PTP group and 0.7 microg/mL for the moderate PTP groups. For both groups, D-dimer testing provided 100% sensitivity and 100% negative predictive value in the diagnosis of DVT. In the low PTP group, specificity increased from 33% to 67% (P = .046). In the moderate PTP group, however, the determined D-dimer level did not improve the specificity. Overall, venous duplex scanning could have been reduced by 17% (14/81) by using different D-dimer cutoff points.

Conclusions: A combination of specific D-dimer level and clinical probability score is most effective in the low PTP patients in excluding DVT. In the moderate PTP group, however, the recommended cutoff point of 0.5 microg/mL may be preferable. These results show that a different D-dimer level is more useful than single cutoff point of 0.5 microg/mL in excluding DVT in established PE patients.