Bidirectional relationship between Sleep Disturbance and Atrial Fibrillation: a SPRINT-based analysis

Background: Sleep disturbance has been associated with atrial fibrillation (AF), but prior studies have largely relied on sleep assessed at a single time point or examined specific sleep disorders. Whether changes in sleep disturbance over time predict AF and, conversely, whether AF predicts subsequent sleep disturbance within the same population is not well established. Methods: We conducted a secondary analysis of the Systolic Blood Pressure Intervention Trial (SPRINT), which enrolled adults with hypertension and increased cardiovascular risk without diabetes. Two prospective analytic cohorts were constructed to examine each direction of the association: 7,232 participants without baseline AF for the analysis of sleep disturbance and incident AF, and 6,079 participants without clinically significant sleep disturbance at baseline for the reciprocal analysis. Sleep disturbance was assessed repeatedly using a 0-3 patient-reported symptom score, with clinically significant sleep disturbance defined as a score [≥]2. AF was ascertained from protocol 12-lead ECGs. Time-dependent Cox models evaluated time-updated sleep disturbance in relation to incident AF and time-updated AF in relation to incident clinically significant sleep disturbance, with adjustment for demographic, lifestyle, and cardiovascular risk factors. Results: During a median follow-up of 3.3 years, 189 incident AF events occurred. Time-updated clinically significant sleep disturbance was associated with a 78% higher adjusted risk of incident AF (HR, 1.78; 95% CI, 1.27-2.51). Each 1-point increase in the time-updated sleep score was associated with a 26% higher risk of AF (HR, 1.26; 95% CI, 1.09-1.46). Compared with a sleep score of 0, the adjusted HRs were 1.20 (95% CI, 0.85-1.70), 2.05 (95% CI, 1.33-3.15), and 1.72 (95% CI, 1.03-2.88) for scores of 1, 2, and 3, respectively. In the reciprocal analysis, 1,422 incident clinically significant sleep-disturbance events occurred during a median follow-up of 3.2 years. Time-updated AF was associated with a 59% higher adjusted risk of incident clinically significant sleep disturbance (HR, 1.59; 95% CI, 1.16-2.16). Associations in both directions were consistent across prespecified subgroups, with no significant effect modification. Conclusion: In adults with hypertension, sleep disturbance and AF were prospectively and reciprocally associated when evaluated longitudinally. These findings support a bidirectional relationship between sleep disturbance and AF and highlight the potential clinical relevance of assessing sleep health in patients with or at risk for AF.