Effect of tranexamic acid in vascular surgery- a sub-study of the POISE-3 randomized clinical trial

Abstract Background/Introduction In the POISE-3 trial involving patients undergoing noncardiac surgery, tranexamic acid (TXA) reduced the incidence of the composite outcome of life-threatening bleeding, major bleeding, or bleeding into a critical organ at 30 days. Bleeding is a major issue in vascular surgery. However, TXA use may be limited due to concerns about its safety in this high-risk population. With 1,399 participants in POISE-3 undergoing vascular surgery, this is the largest trial assessing TXA's efficacy in this population. Purpose In this post hoc subgroup analysis, we aimed to investigate the interaction between surgery type (vascular/non-vascular) and the effects of TXA vs placebo on bleeding and cardiovascular (CV) safety outcomes in the POISE-3 population. Methods: Eligible patients were ≥45 years, undergoing inpatient noncardiac surgery, and at risk for bleeding and CV complications. Patients were randomised (1:1) to TXA 1 gm IV bolus at start and end of surgery (total 2 gm) or placebo. We used Cox proportional hazards models incorporating tests for interaction to analyse time to the first occurrence of the composite bleeding outcome, and the CV safety outcome (composite of myocardial injury after noncardiac surgery, non-hemorrhagic stroke, peripheral arterial thrombosis, or symptomatic proximal venous thromboembolism) at 30 days. Estimates of hazard ratios (HR) with 95% confidence intervals (CI) were calculated for each subgroup with the interaction P value. Results 9,468 of 9535 (99.3%) POISE-3 patients are included in this analysis; 1399 of whom underwent vascular surgery (699 TXA, 700 placebo). Majority (89.8%) of the vascular surgeries were aorto-iliac, peripheral vascular and extracranial surgeries. In the vascular surgery group, 12.6% of patients receiving TXA experienced the composite bleeding outcome versus 14.7% with placebo (HR 0.85, 95% CI 0.64-1.13). For non-vascular surgery patients, these rates were 8.6% with TXA and 11.3% with placebo (HR 0.74, 95% CI 0.65-0.86), with an interaction P value of 0.42. In vascular surgery patients, the CV safety outcome occurred in 20.2% with TXA versus 18.1% with placebo (HR 1.12, 95% CI 0.88-1.42). In the non-vascular surgery group, both TXA and placebo arms showed a 13.1% incidence rate (HR 1.00, 95% CI 0.89-1.13), with an interaction P value of 0.43. Conclusion In vascular surgery patients, there were higher rates of bleeding and cardiovascular events compared to those undergoing non-vascular surgeries; however, there was no significant interaction between the effects of TXA and the type of surgery. The overall POISE-3 trial provides the best estimate of effect with TXA for all patients including those undergoing vascular surgery. The main limitation is that this subgroup analysis was not pre-specified.