Observer dependency of the simplified D-dimer assay in 81 consecutive patients with suspected pulmonary embolism

The SimpliRed D-dimer assay is a whole blood agglutination test for the exclusion of pulmonary embolism (PE) and has been advocated as a reliable rapid bedside alternative to more time-consuming quantitative assays. However, widely differing negative predictive values (77-100%) are reported. This study assessed the intra- and interobserver variability as a possible cause for this wide accuracy range and evaluated the accuracy of the SimpliRed test in 81 consecutive patients with clinically suspected PE. Absolute D-dimer concentration was measured using the Tinaquant (Roche Diagnostics GmbH, Mannheim, Germany) assay. Every assay was immediately scored by one observer and then photographed. Photos were randomised and read twice by two other independent observers. The intraobserver reproducibility was good: kappa = 0.81 and 0.73. However, we found poor interobserver reproducibility with kappa 0.47. This seems due to the fact that the SimpliRed test proved difficult to interpret in the 0.3-1.3 mu g/mL plasma concentration D-dimer range, which contained 32/81 (40%) patients, 8 of whom had PE, Using pulmonary angiograms and ventilation perfusion lung scans with a concurrent spiral computed tomography scan as a gold standard (25 patients had PE), SimpliRed's accuracy to exclude PE was moderate, with negative predictive values of 0.78 and 0.84. In particular, the two observers scored the SimpliRed test as normal in 12 of 25 and 5 of 25 patients with proven PE. We conclude that SimpliRed is strongly observer-dependent. This explains the varying sensitivities for the detection of venous thromboembolism as reported in previous studies. (C) 1999 Elsevier Science Ltd. All rights reserved.