Comparison of four rapid D-Dimer tests for diagnosis of pulmonary embolism

Aim of this study was to evaluate rapid D-Dimer tests for their utility in diagnosis of acute pulmonary embolism (PE). Tests were performed in 183 consecutive pats referred for lung scanning because of clinically suspected PE. According to lung scans and the clinical course of disease 19 pats were classified to have PE with high probability and 164 with low probability. An ELISA (Agen) was used as the D-Dimer reference, and results compared with those of a turbidimetric (Behring), an immunofiltration (Nycomed), latex plasma and whole blood agglutination test (both Agen). There was a poor correlation between the turbidimetric test and either the ELISA (R = 0.38) and immunofiltration test (R = 0.49). The correlation between the ELISA and immunofiltration test was better (R = 0.73). The qualitative latex and whole blood agglutination tests were better fitted to ELISA since positive and negative samples were overlapped only in their 1st and 9th percentiles of ELISA values. The whole blood agglutination test was positive at lower ELISA values than the latex test. The highest sensitivity test for PE was the immunofiltration test (95%) (500ng/mL cut-off), followed by the turbidimetric method (89%) (66ng/mL), the ELISA (89%) (300ng/mL), the whole blood test (88%) and the latex test (68%). Specificity was lowest for the immunofiltration test (33%), intermediate (57-65%) for the turbidimetric and whole blood agglutination tests, and highest for the ELISA and the most insensitive latex test (76/77%). The whole blood assay was found to be the fastest and most suitable for bed site testing but weak positives were difficult to read. The immunofiltration test required plasma preparation but allowed objective semiquantitation of results. The less rapid turbidimetric assay was fully quantitative and objective.