Nine RCTs (n=13,756) were included in the meta-analysis: seven studies were double-blind and had a Jadad score of 5 with adequate concealment; one study scored 3 with unclear concealment; and one study scored 2 with unclear concealment.
All studies (total hip plus knee replacements): There was no significant difference between ximelagatran and comparators in occurrence of major venous thromboembolism. Ximelagatran was associated with a significant increase in major bleeding (OR 1.58, 95% CI 1.04 to 2.39; nine studies, n=6,361). High heterogeneity was found for venous thromboembolism (I2=74.3%). Moderate heterogeneity was found for major bleeding (I2=34.8%).
Type of surgery: Among patients who underwent total knee replacement, ximelagatran was associated with significantly fewer major venous thromboembolism events than comparators (OR 0.68, 95% CI 0.53 to 0.89). There was no significant difference in occurrence of major bleeding. No heterogeneity was found.
Initial timing of administration: Among total hip replacement patients started on melagatran/ximelagatran before surgery, ximelagatran was associated with significantly fewer major venous thromboembolism events compared with low molecular weight heparin (OR 0.32, 95% CI 0.19 to 0.53) but had a significant increase in major bleeding events (OR 3.33, 95% CI 1.88 to 5.89).
Dose: Among patients (all underwent knee replacement) who received ximelagatran 36mg, ximelagatran was associated with significantly fewer total venous thromboembolism events than comparators (OR 0.64, 95% CI 0.55 to 0.75). There was no difference between treatments in major venous thromboembolism or major bleeding events.
Publication bias was not detected.