Author + information
- Gregory Piazza, MD∗ (, )
- Neelima Karipineni, MD,
- Howard S. Goldberg, MD,
- Kathryn L. Jenkins, BA and
- Samuel Z. Goldhaber, MD
- ↵∗Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115
Failure to prevent stroke in patients with atrial fibrillation (AF) has become increasingly important for patient safety and quality of care. Therefore, we conducted a multicenter, outpatient clinic, retrospective cohort analysis using data abstracted through our electronic health record between March 2013 and March 2014.
The patient cohort was identified by a query to the Partners Quality Data Warehouse. Patients with modifiers to the problem list entry of AF, such as “history of,” were excluded. Patients with valvular heart disease were included. Clinical characteristics, CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke/transient ischemic attack, vascular disease, age 65 to 74 years, sex category), and HAS-BLED (Hypertension, Abnormal renal and liver function, Stroke, Bleeding, Labile INRs, Elderly, Drugs or alcohol) scores, guideline adherence, and 90-day outcomes (all-cause mortality, stroke, systemic embolic events, and bleeding) were determined by a web-based decision support tool.
A total of 5,062 patients were captured by our electronic data query and satisfied the inclusion criteria. Their mean age was 70 years. The study population was overweight, with a mean body mass index of 29 kg/m2. Most (78%) received their outpatient care at Brigham and Women’s Hospital.
The median CHA2DS2-VASc score was 3 points, with the majority (72.6%) of patients falling into a high stroke risk category (CHA2DS2-VASc score >1). The most frequently observed components of the CHA2DS2-VASc score were hypertension (57%), age ≥75 years (39.9%), and female sex (39.4%). The median HAS-BLED score was 2 points. The most frequently observed components of the HAS-BLED score were age ≥65 years (69.6%), hypertension (57%), and concomitant antiplatelet or nonsteroidal anti-inflammatory drug therapy (34.1%).
Any thromboprophylaxis, including aspirin or anticoagulation, for stroke prevention in AF was prescribed in 66.4% of the patient cohort (Figure 1A). The frequency of anticoagulation increased with CHA2DS2-VASc score until a score of 7 points, after which it declined (Figure 1B). AF patients age >75 years with an additional risk factor for stroke received anticoagulation more frequently than those age ≤75 years with an additional risk factor for stroke (55.3% vs. 51%; p = 0.01).
All-cause mortality at 90 days in the overall patient cohort was 1.2%. Stroke occurred in 1.6% of patients at 90 days. Bleeding events occurred in 3.7% of patients at 90 days. Gastrointestinal was the most frequent type of bleeding, occurring in 1.2% of the cohort. All-cause mortality (1.6% vs. 0.8%; p = 0.007), stroke (2.1% vs. 1.2%; p = 0.01), and bleeding events (4.3% vs. 2.8%; p = 0.004) at 90 days were more common in AF patients receiving anticoagulation. However, with multivariable regression analysis, prescription of anticoagulation was not independently associated with all-cause mortality, stroke, or bleeding events at 90 days. CHA2DS2-VASc score (adjusted odds ratio: 1.45; 95% confidence interval: 1.2 to 1.74) and HAS-BLED score (adjusted odds ratio: 1.39; 95% confidence interval: 1.07 to 1.82) were independently associated with stroke at 90 days.
Prescription of anticoagulation for stroke prevention in AF was lower in our study (46.9%) than in other large U.S. and international observational studies (1–3). Adherence to the 2011 American College of Cardiology Foundation/American Heart Association/Heart Rhythm Society Focused Update on Management of Patients With AF (4) and the 2012 Focused Update of the European Society of Cardiology Guidelines for the Management of AF (5) was low (53.3% and 49.8%, respectively) within our multicenter outpatient clinic network.
Our electronic health record data query strategy was not able to capture data on prior history of anticoagulation prescription or the rationale for why anticoagulation was not prescribed or discontinued. We were unable to discern whether patients were offered thromboprophylaxis to prevent stroke in AF and refused. Furthermore, our analysis is subject to the limitations of coding accuracy and completeness. The generalizability of our results may be limited because the majority of patients (78%) received care at Brigham and Women’s Hospital. Our short 90-day follow-up period may have limited our ability to discern a difference in clinical outcomes between patients who were anticoagulated and those who were not. Deaths, strokes, and bleeding events that occurred at outside facilities may have been missed.
Our findings suggest the presence of an important knowledge gap among providers regarding stroke prevention in AF patients.
Please note: This investigator-initiated study was funded, in part, by an unrestricted research grant from Janssen. The study was designed and led by an Executive Committee (Drs. Piazza, Karipineni, Goldberg, and Goldhaber). The Executive Committee designed the data query strategy and participated in the statistical analysis. Dr. Piazza has received research grant support from Bristol-Myers Squibb, Daiichi-Sankyo, the Thrombosis Research Institute, and Janssen; and has received consulting fees from Merck. Dr. Goldberg has served on the scientific advisory board of Clearsense. Dr. Goldhaber has received research grant support from Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi-Sankyo, the Thrombosis Research Institute, and Janssen; and has received consulting fees from Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi-Sankyo, Eisai, Merck, and Portola. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2016 American College of Cardiology Foundation
- Wann L.S.,
- Curtis A.B.,
- January C.T.,
- et al.
- Camm A.J.,
- Lip G.Y.,
- De Caterina R.,
- et al.