INR targets for mechanical valves: A systematic review and meta-analysis

Introduction: Guidelines recommend higher INR targets for patients with mechanical valves who are believed to be at higher risk for thromboembolism. Higher INR targets are associated with increased bleeding risk. We performed a systematic review and meta-analysis assessing effects of lower and higher INR targets on thromboembolism and bleeding risk in patients with mechanical heart valves. Methods: We searched Cochrane CENTRAL, MEDLINE and EMBASE for randomized controlled trials (RCTs) and observational studies evaluating lower versus higher INR targets for adults with bileaflet mechanical valves. We performed title and abstract screening, full-text review, risk of bias evaluation, and data collection independently and in duplicate. We evaluated overall quality of evidence with the GRADE framework, and pooled data using a random effects model. Results: We identified 6 RCTs (n = 5497) and 2 observational studies (n = 1195). In RCTs, lower INR targets were associated with significantly less bleeding - 22% versus 40% - (RR 0.54, 95%CI 0.31, 0.93, p=0.03, very low quality). There was no difference in thromboembolism - 2% in both groups (RR 1.28, 95%CI 0.88, 1.85, p=0.20, very low quality) or mortality - 5.5% with lower INR targets versus 8.5% (RR 1.00, 95%CI 0.82, 1.21, p=0.47, moderate quality). Observational studies showed no significant differences in bleeding or thromboembolism, whereas lower INR targets were associated with lower mortality (RR 0.56, 95%CI 0.49, 0.63, p<0.00001). Discussion: In patients with bi-leaflet mechanical valves, lower INR targets appear to be associated with lower bleeding risk when compared to higher INR targets without compromising efficacy.