Eleven RCTs (748 patients, range 8 to 256) were included in the review. Three RCTs reported using random number tables or computer generation. Five RCTs reported appropriate methods to conceal allocation. Three RCTs reported blinding patients and investigators. None of the RCTs reported losses to follow-up. The duration of follow-up ranged from 'in hospital' to 30 days.
Compared with heparin, thrombolytic therapy resulted in a non-statistically significant reduction in recurrent pulmonary embolism/death (OR 0.67, 95% CI 0.40 to 1.12; 11 RCTs); no publication bias was observed. When restricted to RCTs that included patients with haemodynamically unstable pulmonary embolism, the risk of recurrent pulmonary embolism/death was significantly reduced with thrombolytic therapy (OR 0.45, 95% CI 0.22 to 0.92; NNT=10; five RCTs), but there were no difference when restricted to RCTs that excluded these patients (six RCTs).
Compared with heparin, thrombolytic therapy did not produce any statistically significant difference for recurrent pulmonary embolism (OR 0.67, 95% CI 0.33 to 1.37; 11 RCTs), death (OR 0.70, 95% CI 0.37 to 1.30; 11 RCTs), or major bleeding (OR 1.42, 95% CI 0.81 to 2.46; 11 RCTs). There was a statistically significant increase in non-major bleeding (OR 2.63, 95% CI 1.53 to 4.54; eight RCTs; NNH=8).
Results for further subgroup and sensitivity analyses were reported. No statistically significant heterogeneity was observed for any outcome.