The direct oral anticoagulants (DOACs) offer distinct advantages over warfarin – among them, dose adjustment is only needed in cases of impaired renal function, advanced age, low body weight, or potential for drug-drug interactions. Incidentally, these factors (ie, older age or chronic kidney disease) can also increase the risk of ischemic stroke and of bleeding. Thus, the safety and efficacy profile of DOACs vs warfarin in the subgroup of patients on a reduced dose of the former could be different from that seen in the general population. This situation has not been previously well described as these patients are mostly under-represented in the landmark large randomized trials that led to the approval of these drugs.
Looking at a large Danish cohort group (from three nationwide registries), a recent study delved into this issue using propensity matching to compare outcomes in patients on dose-reduced DOACs (apixaban 2.5 mg, dabigatran 110 mg, rivaroxaban 15 mg) compared to warfarin. There were 55,664 patients total with ~8% taking apixaban, ~16% dabigatran, ~6% rivaroxaban; the majority received warfarin (~70%). Average follow-up was 2.3 years with the apixaban group having only a 1 year follow-up. There was wide variability in age groups (overall, mean age 73.9 years; mean age 83.9 years for apixaban, 79.9 for dabigatran, 77.9 for rivaroxaban and 71.0 for warfarin) and in the number of women represented (overall 44.9%, range 40.4% [warfarin] to 60.6% [apixaban]). Mean CHA2DS2--vasc score was 3.3 overall, with the lowest score in the warfarin group (3) and highest in the apixaban group (4.3).
While there was no statistically significant difference in the rate of ischemic stroke and systemic embolism in an inverse probability treatment-weighted analysis for all three DOACs vs warfarin, there was a trend towards a higher rate of the endpoint for apixaban (4.8%) over dabigatran (3.3%), rivaroxaban (3.5%) and warfarin (3.7%). Similarly, there was no statistically significant difference in weighted bleeding outcomes but there was a trend towards lower bleeding in the dabigatran group (4.1%) vs warfarin (5.1%), compared to apixaban (5.1%) and rivaroxaban (5.6%), which were similar to warfarin.
Interesting insights from this study:
1) In this high-risk population, with a high CHA2DS2-Vasc score the rate of ischemic stroke and bleeding over study follow-up was relatively high compared to standard population (~4-5%).
2) There is significant variability among the dose-adjusted DOACs, with dabigatran being the most frequently used dose-adjusted DOAC in patients who require lower doses.
3) There is no statistically significant difference in efficacy or safety of dose-reduced DOAC vs warfarin although apixaban seemed to consistently have a (statistically nonsignificant) trend towards a higher rate of ischemic stroke (even in sensitivity analyses of elderly patients or those with AF diagnosed in hospital).
4) Rates for bleeding were not statistically significantly different for apixaban and rivaroxaban compared with warfarin but there was a trend towards lower rates with dabigatran.
When interpreting these results, keep in mind that although the study is large, the majority of the patients were treated with warfarin, the minority with a dose-reduced DOAC, which may limit the study’s power to detect statistically significant differences. In addition, the significant variability in the baseline characteristics (age, sex, CHA2DS2-Vasc score) within both groups, even with the use of propensity matching, subjects these results to residual confounding (as is the case with all observational studies). Pending randomized controlled trials, this study does offer preliminary insights into the safety and efficacy of dose-reduced DOACs. Additional studies with larger numbers of patients on DOACs are needed, however, to validate these findings.
Brønnum NP, Flemming S, Mette S, et al. Effectiveness and safety of reduced dose non-vitamin K antagonist oral anticoagulants and warfarin in patients with atrial fibrillation: propensity weighted nationwide cohort study BMJ 2017; 356 :j510.