Sleep problems have been linked to atrial fibrillation (AF) in prior work, but many studies assessed sleep at only one time point or focused on specific sleep disorders. This secondary analysis of the Systolic Blood Pressure Intervention Trial (SPRINT) aimed to determine whether changes in sleep disturbance over time predict incident AF and whether AF predicts subsequent clinically significant sleep disturbance within the same hypertensive population.
The authors constructed two prospective analytic cohorts from SPRINT, a randomized trial of adults with hypertension and elevated cardiovascular risk without diabetes. For the analysis of sleep disturbance predicting AF, 7,232 participants without baseline AF were included. For the reciprocal analysis (AF predicting sleep disturbance), 6,079 participants without clinically significant sleep disturbance at baseline were analyzed. All participants were drawn from the original SPRINT trial cohort; this work is a secondary analysis of that trial.
Sleep disturbance was assessed repeatedly during follow-up using a patient-reported symptom score ranging from 0 to 3. Clinically significant sleep disturbance was prespecified as a score ≥2. AF was ascertained from protocol 12-lead electrocardiograms (ECGs) performed during study visits. The authors used these repeated measures to create time-updated exposure variables for each direction of the association.
Time-dependent Cox proportional hazards models were used to evaluate the associations. Models incorporated time-updated sleep disturbance to predict incident AF and time-updated AF to predict incident clinically significant sleep disturbance. Analyses adjusted for demographic factors, lifestyle characteristics, and cardiovascular risk factors. The authors report hazard ratios (HRs) with 95% confidence intervals (CIs). Prespecified subgroup analyses were performed to assess effect modification.
Over a median follow-up of 3.3 years in the cohort without baseline AF, 189 incident AF events occurred. Time-updated clinically significant sleep disturbance (score ≥2) was associated with a 78% higher adjusted risk of incident AF (HR 1.78; 95% CI, 1.27–2.51). When treating the sleep score continuously, 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).
When compared with a sleep score of 0, the adjusted HRs for incident AF were reported as follows: 1.20 (95% CI, 0.85–1.70) for score 1, 2.05 (95% CI, 1.33–3.15) for score 2, and 1.72 (95% CI, 1.03–2.88) for score 3. These categorical results indicate increased AF risk particularly at higher symptom scores.
In the reciprocal cohort (participants without clinically significant sleep disturbance at baseline), there were 1,422 incident clinically significant sleep-disturbance events over a median follow-up of 3.2 years. Time-updated AF was associated with a 59% higher adjusted risk of developing clinically significant sleep disturbance (HR 1.59; 95% CI, 1.16–2.16).
The authors report that associations in both directions were consistent across prespecified subgroups and that there was no evidence of significant effect modification. The source did not report additional sensitivity analyses or subgroup-specific point estimates beyond stating consistency across prespecified groups.
Among adults with hypertension enrolled in SPRINT, longitudinal, time-updated measures demonstrated a reciprocal association between sleep disturbance and atrial fibrillation. The findings support a bidirectional relationship and underscore the potential clinical relevance of assessing sleep health in patients with or at risk for AF. The authors suggest that routine attention to sleep symptoms may be important in the cardiovascular risk evaluation of hypertensive patients, although specific clinical recommendations were not provided in the source.
The SPRINT data used for this secondary analysis are publicly available through the NHLBI Biologic Specimen and Data Repository Information Coordinating Center (BioLINCC). This analysis is of SPRINT (Clinical Trial NCT01206062). The authors declared no competing interests and stated that institutional review boards approved the original SPRINT protocol and that participants provided written informed consent.