American Heart Association

Monthly Archives: April 2014

Warfarin no better than aspirin at preventing cognitive decline in A-Fib patients

Matthew Edwardson, MD

Mavaddat N, Roalfe A, Fletcher K, Lip GYH, Hobbs FDR, Fitzmaurice D, and Mant J. Warfarin Versus Aspirin for Prevention of Cognitive Decline in Atrial Fibrillation: Randomized Controlled Trial (Birmingham Atrial Fibrillation Treatment of the Aged Study). Stroke. 2014


Atrial fibrillation (AF) is associated with impaired cognition. Knowing this association has led many to speculate that treatment with warfarin may reduce the accumulation of silent infarcts, thereby preventing or at least slowing cognitive decline. In this article, Mavvadat and colleagues analyzed the protective effects of warfarin by measuring cognitive function over time in warfarin vs. aspirin treated patients.


The authors performed a retrospective analysis using data from the Birmingham Atrial Fibrillation Treatment of the Elderly (BAFTA) study. 973 patients age > 75 with AF were randomized to aspirin (ASA) vs. warfarin if their primary care physician was undecided on the most appropriate treatment. Subjects received mini-mental status examinations (MMSE) at 9, 21, and 33 months after randomization. Of note, subjects were excluded from analysis if they suffered a symptomatic stroke. The authors found a non-significant effect favoring warfarin over ASA at 33 months (mean MMSE 26.9 vs. 26.4 respectively, P = 0.16) that was likely too small to be clinically meaningful.

This is the first study to test the ability of warfarin to prevent cognitive decline over time related to silent infarcts in patients with AF. Prior studies were confounded by inclusion of subjects with clinically symptomatic stroke during follow up. Unfortunately neuroimaging was not obtained in this study. It would be fascinating, for example, to determine whether there was a significant accumulation of silent infarcts on FLAIR despite little change in MMSE over time. This study suggests that undertreated AF has little impact on cognitive decline in the absence of new symptomatic infarcts. Despite these findings, the evidence for warfarin in terms of stroke prevention is overwhelming. Whenever possible we should push our primary care colleagues to choose warfarin (or one of the novel anticoagulant drugs) over aspirin for patients with AF.

Validity of CHA2DS2-VASc score in young patients

Vivek Rai, MD

Melgaard L, Rasmussen LH, Skjøth F, Lip GYH, and Bjerregaard Larsen TB. Age Dependence of Risk Factors for Stroke and Death in Young Patients With Atrial Fibrillation: A Nationwide Study. Stroke. 2014


Atrial fibrillation (AF) is the most common arrhythmia and prevalence of AF of increases with age. Patients with AF are at high risk for stroke and thromboembolism but the risk is also dependent on presence of various stroke risk factors in the individual patient. CHADS2 score is the most commonly used risk stratification score and has been refined to produce CHA2DS2-VASc score. The importance of CHA2DS2-VASc score lies in its ability to identify “truly low-risk” patient who may not need anti-thrombotic treatment. The risk stratification scores have primarily been evaluated in patients >65 years of age and there is little data, if any, about utility of these scores in younger age group. 



Melgaard et al analyzed the data of patients aged 30 to 65 years diagnosed with AF between 2000 and 2011 utilizing The Danish National Patient Registry. The authors identified 73799 non-valvular AF, of which about half (51.2%) were aged <65 years. The overall incidence of stroke per year for 1 year (5 years) follow-up was 1.2% (0.6%), 3.5% (1.6%), and 5.6% (2.8%) respectively, for the age groups of 30-50, 50-65, and 65-75. A higher CHA2DS2-VASc score was associated with higher mortality. Prior stroke, heart failure, vascular disease, diabetes, and hypertension remained independent predictors of stroke and death.

Of note, the risk stratification score used in the analysis is modified to exclude scoring for age and sex and it was not possible to determine the subtype of stroke. Also, this is a retrospective analysis of data obtained using diagnostic codes which may have resulted in over or underestimation of risk. Nevertheless, the authors have reliably shown that CHA2DS2-VASc score is a valid tool for risk stratification of younger patients as well. The study will not change my clinical practice but reinforces the, so far presumed, validity of CHA2DS2-VASc score in identifying young patients who are at very low risk of stroke due to AF and may not need anti-thrombotic treatment.

Declining stroke rates after MI: A testament to improved coronary care

Seby John, MD

Kajermo U, Ulvenstam A, Modica A, Jernberg T, and Mooe T. Incidence, Trends, and Predictors of Ischemic Stroke 30 Days After an Acute Myocardial Infarction. Stroke. 2014


We have long known that ischemic stroke is a devastating complication following myocardial infarction (MI).  Stroke incidence has been previously reported by numerous clinical trials, but is limited by the narrow group of patients selected for the trial, which precludes extrapolation to the population seen in day-to-day practice. Besides, treatment of acute MI has undergone paradigm shifts with the introduction of in-hospital interventions and drugs for secondary prevention. The influence of these modern treatments on the risk of stroke was unknown.



Kajermo and colleagues examined trends of stroke incidence after an MI, and its predictors. Using a national registry, patients admitted with a first MI to cardiac ICUs in Sweden from 1998-2008 were included. They then identified patients who subsequently developed an ischemic stroke. 2.1% of acute MI patients developed an ischemic stroke within 30 days.  Over time, the incidence decreased significantly with rates during 2007-2008 being 2%, compared to 2.2% during 1998-2000. This decrease was attributed to the increased utilization of PCI, statins, ASA and P2Y12-inhibitors. The study confirmed other variables that have previously been shown to increase stroke risk such as prior stroke, age, female sex, diabetes and atrial fibrillation.

This is the largest study (n=173,233) that has examined this question in an unselected population. Although you would suspect that the stroke etiology was likely cardioembolic from an LV thrombus, there was no association between in-hospital anticoagulation and stroke occurrence. In fact, the association of MI secondary prevention treatments with stroke reduction would suggest that stroke etiology is related to pathophysiological players common to both MI and stroke, such as atherosclerosis and platelet activation. Similarly, one would have thought that an invasive interventional procedure (PCI) would be associated with higher stroke risk. However, PCI decreased this risk and is possibly related to early revascularization reducing myocardial infarct volume and the subsequent development of heart failure/atrial fibrillation, thereby reducing stroke.  

Congratulations are due to the cardiology community for rigorous, effective implementation of early revascularization and secondary prevention.