Dixon Yang, MD
Though non-vitamin K antagonist oral anticoagulants (NOACs) are the gold standard for stroke prevention in atrial fibrillation (AF) in absence of significant mitral stenosis or mechanical heart valve, a small proportion (1.4% per year) of residual risk for ischemic stroke remains based on pooled statistics from randomized control trials. Some observation studies report a residual annual risk as high as 10%. Therefore, Dr. Maurizio Paciaroni and authors sought to estimate the rates of ischemic and bleeding events after an acute ischemic stroke in patient with AF while on NOAC treatment and investigate predictive factors in those events, especially in the context of antiplatelet/anticoagulation strategies.
The study involved a multicenter observational cohort, which included consecutive acute ischemic stroke patients with AF who were taking NOAC at the time of event and adherent to NOAC therapy. Patients were enrolled and data collected prospectively from 43 centers across the United States and Europe. Investigators recorded NOAC doses and any reasons for reduced dosing. The study considered reduced NOAC dosing off-label in the absence of the recommended clinical and laboratory criteria for dose reduction. Participants were followed for at least 1 year. The study’s primary outcome was composite recurrent ischemic stroke, systemic embolism, intracranial bleeding, and major extracranial bleeding. The authors compared risk of outcome event between those who had their anticoagulant changed versus those who kept the same anticoagulation after index event.
A total of 1240 participants were included in the analysis, of whom 491 (40%) were on reduced NOAC dosing. About 30% of those on reduced dosing were treated with an off-label low dose. Two-thirds (n=920) of the index strokes were considered cardioembolic. After index event, among those who had been treated with an appropriate NOAC dose, 490 (40%) had their NOACs switched to another, 527 (43%) continued with the same dose, 83 (7%) continued with the same NOAC but at increased dosing, and the remaining minority were either switched to warfarin or low molecular weight heparin.
After an average follow-up time of 1550 patient-years, 192 (16%) participants had 207 outcome events (111 ischemic strokes, 15 systemic embolisms, 24 intracranial bleedings, 57 major extracranial bleeds) corresponding to a yearly rate of 13.4%. CHA2DS2-VASc score and hypertension were predictive factors of recurrent ischemic events, while age, history of major bleeding, and addition of an antiplatelet to an NOAC were predictive of a future hemorrhagic event. When comparing those who had their anticoagulant change against those who retained the same anticoagulation after index event, the authors found no difference in the rate of primary outcome, or rate specific to ischemic or hemorrhagic events. Those who switched to low molecular weight heparin had the highest rate of primary outcome when compared to other anticoagulation strategies. When evaluating only those with cardioembolic index stroke, the rate of primary outcomes was the same regardless of if a participant had their NOAC changed or not.
The authors find a higher rate of recurrent event (13.4% per year) in those on NOAC therapy for AF than reported in randomized control trials, though in line with prior observational studies. The reason for recurrent ischemic stroke is important to consider in order to target treatment. Some providers may feel compelled to add an antiplatelet agent to NOAC therapy in those who suffer a recurrent stroke due to atherothrombotic mechanism, but evidence in the present study clearly shows a significant added hemorrhage risk without justifiable benefit. This is supported by prior randomized control trial evidence. Therefore, the authors suggest avoidance of prescribing antiplatelets in combination with NOAC. For those who suffer a recurrent cardioembolic stroke despite NOAC, it does not appear switching an NOAC will add significant risk reduction overall. The authors speculate a possible role for intervention in certain high-risk patients, such as left atrial appendage closure, that would need to be borne out by further studies.
Together, the evidence would suggest the answer in preventing recurrent stroke for the majority of those adherent to appropriate NOAC therapy for AF often lies outside of antiplatelet/anticoagulation modification. At its most basic form, this could mean simply better addressing lifestyle modification or traditional vascular risk factors. For others, it may mean identifying untreated factors that increase risk for thromboembolism, such as occult malignancy. Further quantification of the high residual stroke risk among those on NOAC therapy for AF is needed.
