A Propensity Score Matched Comparison of Clinical Outcomes in Atrial Fibrillation Patients Taking Vitamin K Antagonists: Comparing the “Real-World” vs Clinical Trials

Abstract

ObjectiveTo investigate the incidence and risk of adverse clinical outcomes in a “real-world” cohort of patients with atrial fibrillation (AF) anticoagulated with vitamin K antagonists (VKAs) from the Murcia AF Project in comparison with the warfarin arm of the randomized clinical trial (RCT) AMADEUS (Evaluating the Use of SR34006 Compared to Warfarin or Acenocoumarol in Patients With Atrial Fibrillation). Patients and MethodsWe included 1361 patients with AF from the Murcia AF Project (recruitment from May 1, 2007, to December 1, 2007) and 2293 from the AMADEUS trial (started in September 2003 and primary completed in March 2006), all taking VKA treatment. After propensity score matching (PSM), we investigated differences in rates and risks of several events, including major bleeding, ischemic stroke, and all-cause mortality at 365 (interquartile range, 275-428) days of follow-up. ResultsAfter PSM there were 1324 patients for the comparative analysis, whereby annual event rates for most adverse events were significantly higher in the “real-world” population. Cox regression analyses demonstrated that the risk of primary outcomes was also increased in the “real-world” (vs RCT: hazard ratio [HR], 6.32; 95% CI, 2.84-14.03 for major bleeding; HR, 3.56, 95% CI, 1.22-10.42 for ischemic stroke; HR, 5.13, 95% CI, 3.02-8.69 for all-cause mortality). The risk of all other adverse events was higher in the real-world cohort, except for cardiovascular mortality. ConclusionThis study comparing the Murcia AF Project and the AMADEUS trial demonstrates that there is a great heterogeneity in both populations, which is translated into a higher risk of several adverse outcomes in the real-world cohort, including major bleeding, ischemic stroke, and mortality.