Thirteen randomised studies. Analysis of the incidence of recurrent venous thromboembolism used 6 studies with 1,228 patients. Comparisons of bleeding complications assessed 10 studies including 1,684 patients, whilst mortality (overall and from pulmonary embolism) was assessed through 12 studies with 1,723 patients. Mortality for cancer patients was assessed in 4 studies (155 patients).
In level 1 studies comparing LMWH with UFH, the RR of recurrent venous thromboembolism varied between the different treatment periods. During the first 15 days, and over the entire period of anticoagulant therapy, the RR of recurrent venous thromboembolism for LMWH compared to UFH was 0.24 (95% confidence interval, CI: 0.06, 0.80, P=0.02) and 0.39 (95% CI: 0.30, 0.80, P=0.006), respectively. In contrast, treatment between 16 and 90 days had a RR of 0.60 (95% CI: 0.2, 1.5, P=0.3), showing no significant difference between the treatments. Level 2 studies showed no significant difference in the rates of recurrence of venous thromboembolism for any treatment period.
The effects of LMWH compared to UFH on bleeding complications varied between level 1 and 2 studies, and types of bleed. For major bleeding, the RR in level 1 studies was 0.42 (95% CI: 0.2, 0.9, P=0.01) for LMWH, whilst in level 2 studies there was no significant effect with an RR of 0.85 (95% CI: 0.3,1.9, P=0.7). The risk of minor bleeding was higher for LMWH than UFH, as reflected in the RR of 1.16 (95% CI: 0.7,1.9, P=0.5) for level 1 studies and 1.03 (95% CI: 0.5, 1.8, P=0.8) for level 2 studies, although this difference is not statistically significant.
Comparisons of mortality were undertaken by pooling level 1 and 2 studies. The RR for overall mortality and mortality in cancer patients during the entire treatment period, 0.51 (95% CI: 0.2, 0.9, P=0.01) and 0.33 (95% CI: 0.1, 0.8, P=0.01) respectively, were significantly lower for LMWH than UFH. While similar relationships were evident for the two periods of treatment (1 to 15 days and 16 to 90 days) for overall mortality, there was no significant difference between LMWH and UFH for cancer patients in the first 15 days of treatment (RR=0.5, 95% CI: 0.01, 7.7, P=0.52). Mortality from pulmonary embolism during the entire treatment period was not significantly different between LMWH and UFH, as reflected in the RR of 0.59 (95% CI: 0.1, 2.5, P=0.4).