Comparison with Historical Control Group

A comparison of the composite and individual efficacy endpoints, as well as the major bleeding endpoint, for the lepirudin and control patients groups (categorical data), was performed from time of HIT diagnosis to end-of-observation period (Table 4a), as well as from time of start of treatment (Table 4b). (The latter time frame was also used for the analysis of the argatroban treatment studies (see Chapter 15). No patients were excluded from the categorical analysis.

The composite and single endpoints were also compared with the historical controls from start of treatment using time-to-event analyses (Fig. 5). The composite endpoint occurred less often in the lepirudin-treated patients as compared to controls (p = 0.04), primarily due to a reduction in new thrombotic events (p < 0.001), while the risk for limb amputation (p = 0.79) and death (p = 0.43) did not differ significantly (however, the studies were not powered to detect differences in these endpoints). However, the risk for major bleeding was increased in the lepirudin-treated patients (p = 0.015), even when the 22 patients who underwent CPB were excluded. Furthermore, 42 patients (11 from HAT-1,17 from HAT-2, and 14 from HAT-3) were excluded from the time-to-event comparison because the date of HIT confirmation was >21 days before start of treatment (n = 29), or because of missing HIT confirmation date or missing date of stopping of lepirudin therapy (n = 13).

0 0

Post a comment