Eleven RCTs met the inclusion criteria with 3,674 participants.
Compared with UFH, LMWH reduced mortality rates over 3 to 6 months of patient follow-up (OR = 0.71, 95% CI: 0.53, 0.94; p = 0.02) which was statistically significant. The NNT was 61.
For major bleeding complications, LMWH was favoured over UFH (OR = 0.57, 95% CI: 0.33, 0.99; p = 0.047), but the absolute risk reduction was small and not statistically significant (risk reduction 0.61%, 95% CI: -0.04% to 1.26%; p = 0.07) using a fixed-effect model. Using a random=effects model produced an OR of 0.71, (95% CI: 0.40, 1.27) and absolute risk reduction of 0.66 (95% CI: -0.09, 1.41).
For preventing thromboembolic recurrences, LMWH seemed as effective as UFH (OR = 0.85, 95% CI: 0.63, 1.14; p > 0.2) however this was not statistically significant.
No statistical heterogeneity was found between studies for major bleeding, recurrent thromboembolism and mortality rates (p > 0.2 for all outcomes).
Sensitivity analysis showed that the review's findings were not altered by removing individual studies, excluding participants with distal thrombosis, including recent abstract results, or relaxing the study eligibility criteria to include studies that administered adjusted dosages. Cumulative meta-analysis showed a significant mortality advantage for LMWH by the time of the third published study.