Preventing Stroke in Patients With Atrial Fibrillation—A Steep Climb Away From Achieving Peak Performance

There are many ways of going forward, but only one way of standing still.

Franklin D. Roosevelt

Atrial fibrillation (AF) increases the risk of stroke by 5-fold, and more than 40% of all strokes in patients older than 80 years are attributable to AF.¹ Worse yet, AF-related thromboembolic events lead to larger infarcts and poorer outcomes, such that patients with AF are more likely to develop significant disability after a stroke.² Moreover, AF-related strokes are 2.5-fold more likely to be fatal.¹ Despite the significant morbidity and mortality due to AF, AF-related strokes can be prevented. Oral anticoagulation has been shown to reduce the risk of stroke by 64% and the risk of death by 26% in patients with AF.³ In fact, oral anticoagulation has been the only intervention to improve survival in randomized clinical trials of AF therapy. For decades, warfarin sodium was the sole oral anticoagulant available for the prevention of AF-related stroke and systemic embolization. However, in the past 6 years, there have been 4 randomized clinical trials of non–vitamin K antagonist oral anticoagulants (NOACs), including more than 71 000 patients.⁴ These trials revealed that NOACs provide better protection than warfarin, reducing the risk of ischemic stroke by 19% and the risk of intracranial hemorrhage by approximately 50%.⁴ The medical community now has several highly effective therapies to prevent stroke.

Unfortunately, studies in a variety of clinical settings have demonstrated that a substantial proportion of patients with AF who are eligible for oral anticoagulation (with moderate to high risk of stroke and no contraindications) do not receive it. Less than half of the hospitalized Medicare patients with chronic AF, only 65% of patients hospitalized with AF and heart failure, and as few as 1 in 5 patients with AF with acute coronary syndromes receive oral anticoagulation at hospital discharge.⁵ However, many have questioned whether underutilization of oral anticoagulation in hospitalized patients is due to the presence of acute illness and other mitigating factors. Therefore, it remains speculative whether or not underutilization of oral anticoagulation is as much of a problem in stable outpatient AF populations, particularly if cared for by cardiologists.

In this issue of JAMA Cardiology, Hsu and colleagues⁶ sought to determine how frequently patients with AF receive oral anticoagulation and whether or not the use of anticoagulation increased in patients with higher risk for stroke. Using the National Cardiovascular Data Registry’s Practice Innovation and Clinical Excellence (PINNACLE) prospective, office-based quality improvement registry, they analyzed the use of oral anticoagulation in more than 429 000 patients with AF from 144 cardiology practices in 38 states. After adjusting for patient, physician, and practice characteristics, the authors found that increasing stroke risk was associated with an increased likelihood of oral anticoagulation. However, prescription of oral anticoagulation by cardiovascular specialists did not surpass 50% even in higher-risk patients with a CHA₂DS₂VASc score exceeding 4. So, based on the data from the PINNACLE Registry, it seems that the more things have changed, the more they have stayed the same, despite recent advances in pharmacotherapy for stroke prevention.

While the PINNACLE Registry included a very large, representative outpatient cohort, there are some limitations to the analysis that we should keep in mind when we think about the results. First, Hsu and colleagues⁶ only considered treatment at the first clinic encounter. Management of chronic medical illnesses is a process rather than a decision at a single point in time, and therapy decisions often evolve over several health care encounters. Would the results be the same after multiple follow-up clinic visits and more opportunities for intervention? The authors performed a sensitivity analysis that suggested there was little change in the use of anticoagulation therapy with additional follow-up out to 1 year. Second, while the authors excluded 28 088 of 457 505 patients (6.1%) with known contraindications to oral anticoagulation, the prevalence of contraindications in clinical practice is much higher. So, some (but not all) of the underutilization may have reflected appropriate avoidance of therapy that was not adequately captured.

What are the clinical implications of this PINNACLE Registry report? In untreated patients with AF, the overall risk of stroke is 2 events per 100 patient-years.⁷ Oral anticoagulation decreases the risk of stroke by approximately two-thirds.³ Assuming these figures, the use of oral anticoagulation in only half of the 380 093 patients with a CHA₂DS₂-VASc score of 2 or greater and no contraindications may lead to as many as 2500 strokes each year—strokes that should be prevented with higher-quality care. To echo President Franklin D. Roosevelt’s observation, there are many ways to change practice and improve stroke prevention, but only one way to “stand still.” In particular, it is critical to move beyond just documenting gaps in quality to actually improving AF care. If we consider some of the high-yield opportunities available in most hospitals and outpatient practices, several potential evidence-based interventions should be implemented. First, electronic medical record systems should be configured to remind health care professionals when a patient has an indication for oral anticoagulation and to provide clinical decision support tools. Second, while health care professionals have a key role in improving treatment, better patient education and shared decision making are also of paramount importance. Survey data indicate that only 64% of patients with AF understand that they are at increased risk for stroke.⁸ If patients do not understand the importance of drug therapy, they may be less likely to assent or adhere to oral anticoagulation. Third, organized quality improvement efforts are a vital component in improving quality of care. For example, the Get With The Guidelines program has achieved anticoagulation treatment rates for eligible patients with AF exceeding 95%,⁹ and high levels of performance (80%) have also been reported among practices participating in the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF).¹⁰ One of the potential reasons for the lag in utilization of oral anticoagulation in AF relative to other cardiac drug therapies—for example, β-blockers in heart failure or aspirin in acute coronary syndromes—is the relative paucity of formally endorsed AF quality measures being used for accountability and public reporting. Universal performance measures make health systems more likely to monitor treatment and implement system-based changes to improve utilization.

Another interesting (and perhaps vexing) question raised by the data from Hsu and colleagues⁶ is why cardiology practices participating in the PINNACLE Registry did not rapidly improve their utilization of oral anticoagulation, as has been observed in other quality improvement registries. Registries should provide health care professionals and health systems with timely feedback about treatment targets and performance that is actionable. Furthermore, performance improvement programs should be interactive, such that hospitals and outpatient practices use the data to change practice and achieve public recognition. Health systems that attain quality improvement should be identifiable by the public, so patients and payers can make informed decisions about their care.

There is overwhelming evidence for the net clinical benefit of oral anticoagulation in patients with moderate to high risk of stroke. The data by Hsu and colleagues⁶ make an important contribution and help highlight that significant opportunity to improve care exists among many outpatient practices, such as those participating in the PINNACLE Registry. However, it is now critical to move beyond documenting these quality gaps to implementing meaningful and effective quality improvement. Doing so will require ensuring that every setting where patients with AF are cared for has programs to provide health care professionals with real-time clinical decision support, useful reminders based on the guidelines, and continuous assessment of success in delivering therapies to patients who can benefit from them. The challenge going forward will be to ensure that the progress is universal and consistent.

Corresponding Author: Gregg C. Fonarow, MD, Division of Cardiology, Department of Medicine, David Geffen School of Medicine, University of California, 10833 LeConte Ave, Room A2-237, Center for Health Sciences, Los Angeles, CA 90095 (gfonarow@mednet.ucla.edu).

Published Online: March 16, 2016. doi:10.1001/jamacardio.2015.0382.

Conflict of Interest Disclosures: Both authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Piccini reported receiving grants for clinical research from ARCA Biopharma, Boston Scientific, Johnson & Johnson, Gilead, St Jude Medical, and ResMed and reported serving as a consultant to Bristol-Myers Squibb, GlaxoSmithKline, Janssen Pharmaceuticals, Medtronic, and Spectranetics. Dr Fonarow reported serving as a consultant to Janssen Pharmaceuticals and Medtronic. No other disclosures were reported.