Failure of soluble fibrin polymers in the diagnosis of clinically suspected deep venous thrombosis.

ABSTRACT

A new diagnostic test has recently become available which is highly specific for plasma soluble fibrin polymers, the thrombus precursor protein (TpP) enzyme immunoassay. In order to evaluate the diagnostic accuracy of this test and that of a new rapid and automated test for the determination of D-dimers, the BC D-dimer test, in patients with clinically suspected deep vein thrombosis (DVT), 70 consecutive symptomatic patients underwent laboratory analysis with both tests and with the classic enzyme-linked immunosorbent assay (ELISA) D-dimer test, followed by the execution of a compression ultrasound (CUS) test of the affected limb. Patients with a positive CUS test were considered to have DVT (20 of 70), whereas those with a serially negative test and an uneventful 3-month follow-up test were regarded as not having DVT (50 of 70). The sensitivity of the TpP test (45.0%) was significantly lower than that of both the BC D-dimer test (80.0%; P = 0.02) and the classic ELISA test (90.0%; P = 0.002). The specificity of the TpP test (66.0%) did not differ from that of either D-dimer test (60.0 and 64.0%, respectively). The negative predictive value of the TpP test (75.0%) was significantly lower than that of the classic ELISA D-dimer test (94.1%; P = 0.02), which in turn did not differ from that of the BC D-dimer test (88.2%). The positive predictive value was similarly low for each investigated test (34.6, 44.4, and 50.0%, respectively). In conclusion, the TpP test can neither be used to detect a DVT nor to exclude its development in patients with the clinical suspicion of this disease. By contrast, the BC D-dimer might safely replace the classic ELISA test for excluding DVT in symptomatic patients.