More Support for NOACs, from ESC 2016

As novel oral anticoagulants (NOACs) continue to gain popularity, in part because of their simple dosing schedule and lack of significant food-drug and drug-drug interactions, results of a study presented at the European Society of Cardiology Scientific Sessions in Rome, Italy and published as a research letter in JAMA, may offer one more reason to consider the non-vitamin K oral anticoagulants vs warfarin-perhaps even for patients already taking warfarin. It is probably true that in their patients with atrial fibrillation (AF) on warfarin with stable INRs (between 2.0 and 3.0), few providers would consider switching these individuals to a NOAC. The data recently presented at ESC, however, may challenge the “hands-off” practice.

The study, from the ORBIT outpatient registry of 3749 patients with AF, was presented by Dr Eric Peterson of the Duke Clinical Research Institute. Of note, this registry includes mostly stable outpatients with AF who have been on warfarin long-term and does not represent critically ill inpatients. There were ~43% women, ~90% white patients with a mean age of 75 years. When patients were followed for 3 years, only 26% had >80% of the INR levels taken in therapeutic range; in other words, 74% had <80% of their INR values between 2-3 in the first 6 months-very low overall time in therapeutic range. After therapeutic range was established, only 34% remained stable over the next year and there was little correlation between stable INRs in the past and future stable INRs (R²=11%, C index 0.61).  Even in the 10% of patients who had 100% of their INR values  stable during the first 6 months of the study, 64% of these did not have stable INRs in the following year and 33% had an INR that was well out of the therapeutic range.  In general, those with more stable INRs had been on warfarin longer (4.4 years vs. 3.8 years, p=0.003).  In addition, those with more stable INRs tended to have lower CHA₂DS₂-Vasc scores (4.0 vs 4.3, p<.001).

Limitations of this study:

 â–º Non-randomized, observational study so it is subject to confounding.

 â–º Under-representation of minorities, such as African Americans, Hispanics, etc.

 â–º No adjustment for prior bleeding events, presence of mechanical valves, etc.

 â–º All patients were “assumed” to have a target INR of 2.0-3.0 as “normal”

We have been under the misconception for quite some time that if a patient is doing well on warfarin, we don't need to: (1) keep checking INRs and (2) change them over to a NOAC.  This important study offers real world information on the challenges of long term warfarin therapy and the difficulty in maintaining time in therapeutic range, which is one of the most important predictors of thrombotic complications such as stroke as well as bleeding complications. So, the next time you have a patient in front of you that has been “doing well” on warfarin for years, consider checking a few series INR values and ask yourself whether a NOAC may be a better long-term option.

Source: Pokorney SD, Simon DN, Thomas L, et al. Stability of international normalized rations in patients taking long-term warfarin therapy.JAMA. 2016;316:661-663. doi:10.1001/jama.2016.9356.