Clinical evaluation of a diagnostic strategy for deep venous thrombosis with exclusion by low plasma levels of fibrin degradation product D-dimer

Abstract

Clinical research studies have indicated the possibility of diagnostic strategies for deep venous thrombosis (DVT), strategies which include a step where the diagnosis is excluded by low or undetectable plasma levels of fibrin degradation product D-dimer. In collaboration with two local hospitals in Sweden, three implementations of such a strategy are evaluated in this study. Procedures 1, 2 and 3 differed in the method for D-dimer determination, i.e. latex agglutination, immunofiltration and both, respectively. The evaluated procedures were performed in parallel and compared with the current procedure in the different hospitals. At both hospitals, the current procedure stipulated mandatory phlebography and laboratory analysis of acute coagulation status and routine haematology with report-back time of 2 h. Within the 2 h the hospitals' clinical chemistry laboratories also determined plasma D-dimer by the two methods. Of 180 patients enrolled in the study, phlebography was successful in 155 and unsuccessful in 25. The phlebographies revealed 47 proximal DVT, 13 distal DVT and 95 no DVT. With Procedure 1, 53 patients (29%) were excluded in the D-dimer step. For these patients, 47 successful phlebographies revealed one proximal DVT and two distal DVT. With Procedure 2, 71 patients (39%) were excluded. For these patients, 65 successful phlebographies revealed two proximal DVT and four distal DVT. With Procedure 3, 44 patients (24%) were excluded. For these patients, 41 successful phlebographies revealed two distal DVT. The negative predictive values of the D-dimer exclusion step, with 95% confidence intervals given within parentheses, were 96% (88-100%), 91% (84-98%) and 95% (89-100%) for Procedures 1, 2 and 3, respectively. The evaluation demonstrated that the diagnostic potential of D-dimer revealed in research studies can be achieved in clinical practice. The study also indicated that the positive diagnostic value of high levels of D-dimer may be of use in finalizing the diagnosis in the 14% of patients for whom phlebography is unsuccessful.