Atrial fibrillation (AF) and heart failure (HF) commonly coexist and share similar risk factors such as hypertension, increasing age, valvular heart disease, previous ischemic cardiac event, and diabetes mellitus. Individually, AF and HF increase the risk of stroke and death, and the synergistic combination of AF and HF creates a prothrombotic state that results in worse stroke morbidity and mortality than the mere presence of either condition separately.
The 2 most widely used stroke risk stratification schemes in clinical practice, the CHADS 2 and CHA 2 DS 2 -VASc scores, use the letter “C” in each acronym to identify HF as a risk factor for stroke (and death) in patients with AF. However, the definition of HF varies slightly between these 2 risk scores: in CHADS 2 , C represents hospitalization for congestive HF, whereas in CHA 2 DS 2 -VASc, the C denotes the presence of signs or symptoms of either left ventricular (LV) dysfunction, or a recent acute decompensated HF episode irrespective of ejection fraction (EF).
In The American Journal of Cardiology , Taillandier et al describe the prognosis and clinical outcomes of patients with permanent or nonpermanent AF, in the presence of HF. First, their study reveals that in patients with permanent AF, the presence of HF (regardless of EF documented by echocardiography) was not significantly associated with a higher risk of stroke or thromboembolism (relative risk 1.27, 95% confidence interval 0.82 to 1.97, p = 0.29). Although initially surprising, these results are consistent with findings from a previous systemic review by the Stroke in AF Working Group and with the large Swedish Atrial Fibrillation cohort study by Friberg et al. Although moderate to severe cardiac dysfunction is clearly associated with thromboembolism in HF, a simple clinical code of “heart failure” is imprecise, and many such patients do not even have actual HF.
However, several other studies ( Table 1 ) have demonstrated otherwise, showing that congestive HF, recent decompensation of HF, and LV dysfunction are all associated with an increased risk of stroke, with the magnitude of risk ranging from 1.26 to 2.5 (see Table 1 ).
| Study | n | Age, Yrs (Mean ± SD) | % Female | HF Features | No. of Strokes | RR/OR/HR (95% CI) | Univariate/Multivariate Analysis | p Value |
|---|---|---|---|---|---|---|---|---|
| Banerjee et al | 7,156 | 70 | 38 | CHF ∗ | 553 | HR 1.20 (1.00–1.44) | Multivariate | † |
| Lin et al | 7,920 | † | 45.9 | CHF | † | OR 1.611 (1.299–1.999) | Univariate | <0.001 |
| van Staa et al | 79,844 | 73.3 ± 12.5 | 49.7 | CHF | 1,233 | RR 1.26 (1.11–1.42) | Multivariate | <0.05 |
| AFI Echo | 1,066 | 67 ± 10 | 22 | CHF | 78 | RR 1.7 (1.11–2.7) | Multivariate | 0.03 |
| 1,010 | ‡ | ‡ | LV dysfunction § | ‡ | RR 2.5 (1.5–4.4) | Multivariate | <0.001 | |
| Aronow et al | 312 | 84 ± 7 | 67 | LV dysfunction ‖ | 162 | RR 1.8 (1.2–2.6) | Univariate | 0.003 |
| SPAF study | 854 | 69 ± 11 | 31 | LV dysfunction ¶ | 73 | RR 1.8 (1.2–3.0) | Multivariate | 0.02 |
∗ CHF defined as history of HF at baseline.
§ LV dysfunction: moderate to severe LV impairment on 2-dimensional echocardiography.
‖ LV dysfunction: LV EF <50% on echocardiogram.
¶ LV dysfunction: fractional shortening <25% on echocardiogram.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
