Balance of symptomatic pulmonary embolism and symptomatic intracerebral hemorrhage with low-dose anticoagulation in recent ischemic stroke: a systematic review and meta-analysis of randomized controlled trials.

Background The current consensus is that anticoagulation therapy has no role acutely in the management of ischemic stroke, although there is still debate for specific conditions, such as cerebral venous thrombosis and cervical dissection. In addition, anticoagulation is used in the prevention of venous thromboembolic events. We assess the balance between preventing symptomatic pulmonary embolism (sPE) and causing symptomatic intracerebral hemorrhage (sICH) in patients with recent stroke who were randomized to low-dose subcutaneous anticoagulation in trials. Methods We systematically searched the Cochrane Library, Medline, Embase, and Science Citation Index for prospective randomized controlled trials assessing the effect of heparin and other antithrombotic therapies in patients with acute/early ischemic stroke. Included trials had to record information on pulmonary embolism and sICH. Risk ratios (RRs) were calculated for sICH per sPE for each trial using a random effects model. Results We identified 15 trials of low-dose subcutaneous anticoagulation. The trials studied low molecular weight heparin, heparinoids, and unfractionated heparin. The ratio of sICH to sPE was increased with low molecular weight heparin (RR 2.1; 95% confidence interval [CI] 1.03-4.28), but was in approximated unity with heparinoids (RR 1.27; 95% CI 0.31-5.17) and unfractionated heparin (RR 0.99; 95% CI 0.65-1.52). Conclusions Prophylactic/low-dose heparin increased sICH by more than they reduced sPE in patients with recent ischemic stroke. Therefore, their routine acute use cannot be recommended, but they may still be relevant in patients at very high risk of PE (eg, morbid obesity, previous venous thromboembolism, and inherited thrombophilia) or if started later, although trials have not assessed these issues.