Self-reported sleep quality of patients with atrial fibrillation and the effects of cardioversion on sleep quality.

Background: Atrial fibrillation (AF) is the most common sustained cardiac rhythm disturbance encountered in clinical practice and is associated with impaired quality of life. Data from the previous studies have shown that sleep quality (SQ), as a component of life quality, may also deteriorate in patients with AF. However, it remains unclear; we do not know whether SQ improves after sinus rhythm is maintained. Therefore, we aimed to examine the relationship between SQ and AF, as well as the effects of sinus rhythm restoration with direct current cardioversion (DCC) on SQ among patients with persistent AF.

Methods: One hundred fifty-three patients with a diagnosis of nonvalvular AF and 150 age-matched control subjects with sinus rhythm were recruited. SQ was assessed using the Pittsburgh Sleep Quality Index (PSQI). The study was designed with two stages. First, the difference in SQ between AF patients and age-matched controls was examined. Patients with global PSQI scores greater than 5 were defined as "poor sleepers." Thus, a higher global PSQI score indicated worsened SQ. Predictors of poor SQ were also analyzed using a regression model. Second, the effect of rhythm control on SQ was studied in patients with AF who were eligible for DCC. Of the 65 patients with persistent AF, 54 patients with successful cardioversion were followed for 6 months. The remaining 11 patients, whose cardioversion was unsuccessful, were not followed. After 6 months of follow-up, the PSQI scores of patients with sinus rhythm maintenance (n = 39) and patients with AF recurrence (n = 15) were reassessed. Changes in global PSQI scores (baseline vs after 6 months) were analyzed.

Results: The PSQI scores were significantly higher in the AF group compared to the control group (9.4 ± 4.6 vs 5.8 ± 4.1, P = 0.001, respectively). The prevalence of poor sleepers was significantly higher in the AF group (76%) than in the control group (45%) (P < 0.001 by the χ(2) test). Multivariate logistic regression analysis showed that AF (odds ratio [OR]: 3.36, 95% confidence interval [CI]: 2.00-5.55), age (OR: 1.02, 95% CI: 1.00-1.04), and diabetes mellitus (OR:1.79, 95% CI: 1.03-3.14) were independent predictors of poor SQ. In the second stage, the effect of rhythm control on the SQ of the 54 patients with successful DCC was analyzed. PSQI scores improved significantly between baseline and the 6 months in sinus rhythm maintenance group (8.7 ± 4.1 vs 7.2 ± 3.8, P < 0.001, respectively). However, in the AF recurrence group, the change in global PSQI scores between baseline and the sixth month was not statistically significant (9.8 ± 4.5 vs 9.2 ± 4.2, P = 0.56, respectively).

Conclusion: Patients with AF have shorter sleep duration and poor SQ. Maintenance of sinus rhythm after DCC may have a favorable effect on the SQ of patients with AF. Nevertheless, AF is an independent predictor of poor SQ.