Value of thrombin-antithrombin III complexes in major orthopedic surgery: relation to the onset of venous thromboembolism.

This study evaluated (a) the possible changes of plasma levels of thrombin-antithrombin III complexes during hospitalization to predict venous thromboembolism in patients undergoing elective total hip replacement and (b) the sensitivity and specificity of thrombin-antithrombin III complexes in the late incidence of deep vein thrombosis when these patients are discharged from the hospital. In 50 consecutive patients (18 men, mean age = 63 +/- 8 years) a venous blood sample was obtained from each patient before surgery and postsurgery on days 5 +/- 2, 9 +/- 2, and 45 to evaluate the thrombin-antithrombin III complexes by the enzyme-linked immunosorbent assay as a part of a larger surveillance program. Six of 50 patients developed deep vein thrombosis, diagnosed by phlebography on the 45th day postsurgery. From the day before until the ninth day after surgery, mean values of the thrombin-antithrombin III complexes increased to a greater extent in patients with deep vein thrombosis than in those without, although the differences were not significant (from 14.8 +/- 11.2 ng/mL to 36.2 +/- 19.1 ng/mL in the former group and from 13.6 +/- 3.3 ng/mL to 22.4 +/- 5.1 ng/mL in the latter, p = NS). On the 45th day after surgery the mean value of the thrombin-antithrombin III complexes reduced less in patients with deep vein thrombosis (up to 9.9 +/- 1.9 ng/mL and to 25.2 +/- 17.2 ng/mL, respectively, p = NS). In addition, thrombin-antithrombin III complexes remained over the level reached on the fifth day only in the patients who developed deep vein thrombosis. On the 45th day after surgery, thrombin-antithrombin III complexes exhibited a sensitivity of 17%, a specificity of 86%, and an accuracy of 78% in differentiating the presence and absence of deep vein thrombosis as compared with phlebography. We conclude that after total hip replacement (a) serial measurement of the thrombin-antithrombin III complexes does not appear helpful in predicting venous thromboembolism during hospitalization, and (b) measurement of thrombin-antithrombin III complexes has a low diagnostic accuracy in diagnosing delayed deep vein thrombosis. However, the greater and persistent increase of thrombin-antithrombin III complexes level in patients who developed deep vein thrombosis may deserve further investigations.