The Gap between Trial Data and Clinical Practice - An Analysis of Case Reports on Bleeding Complications Occurring under Dabigatran and Rivaroxaban Anticoagulation

BACKGROUND: The novel direct oral anticoagulants (NOA), dabigatran (a thrombin inhibitor), rivaroxaban and apixaban (factor Xa inhibitors) have shown at least noninferiority compared to warfarin concerning the prevention of stroke and systemic embolism as well as the risk of hemorrhagic complications in large phase III trials in patients with atrial fibrillation (AF). These results have been obtained under regular monitoring of side effects and reinforcement of medication adherence in carefully controlled clinical trials. To what extent they translate into clinical practice is a matter of ongoing research. While postmarketing registers, most prominently the GLORIA-AF registry, are currently recruiting and will not report data for several years, we aimed at extracting risk factors for hemorrhagic complications under NOA from all available case reports and single case series published to date. METHODS: To identify risk factors which increase the risk of hemorrhage under NOA, we performed a PubMed search for both dabigatran and rivaroxaban, as well as three search terms for hemorrhagic complications. The cases of hemorrhagic complications were analyzed for the presence of the following four factors: 'prescriber errors', 'unfavorable comedications', 'renal impairment' and 'prescription of NOA in the frail elderly'. RESULTS AND DISCUSSION: We found a discrepancy in the frequency of case reports on hemorrhagic complications to the disadvantage of dabigatran which can hardly be attributed to the earlier marketing time of dabigatran alone. As risk factors, we identified prescriber errors, impaired renal function, comedication with antiplatelet drugs or p-glycoprotein inhibitors, old age and low body weight. Strikingly, the majority of the bleeding complications reported in this compilation of case reports showed at least one and in most cases several risk factors. CONCLUSIONS: We should, therefore, carefully select our patients for treatment with the NOA with an emphasis on age, body weight, renal function and comedications and follow them faithfully concerning their medication adherence and eventual side effects.