Sleep Stage Monitoring in Congenital Heart Disease (CHD) Using a Digital Health Application Programming Interface (API).

Background: Adults with congenital heart disease (CHD) are living longer but face increasing comorbidities. Sleep is a key health determinant, yet objective data in CHD remain limited. This study compared sleep characteristics of adults with CHD and controls using wearable technology and a Health Application Programming Interface (API). Methods: A total of 175 CHD patients (33.1 ± 10.3 years, 49.2% women) and 52 controls (34.4 ± 12.4 years, 40.4% women) completed seven continuous days of wrist-worn Garmin Vivosmart® 5 during routine follow-up at the TUM Klinikum Deutsches Herzzentrum. Sleep duration, phases, Sleep Scores, and weekday-weekend differences were analyzed, and multivariate models examined clinical and demographic predictors. Results: Total sleep duration and rapid eye movement (REM) sleep did not differ between groups. CHD patients had more deep sleep (83 ± 19 vs. 75 ± 16 min, p = 0.004) but lower Sleep Scores (74 ± 9 vs. 77 ± 9, p = 0.041). Within CHDs, deep sleep was higher on weekends than on weekdays (p = 0.033). Multivariate analyses showed no overall group effect, but age (p = 0.016), sex (p = 0.013), and body mass index (BMI; p < 0.001) significantly predicted sleep outcomes. Regression analyses in CHDs revealed female sex associated with longer REM sleep (p < 0.001), while higher BMI consistently predicted poorer outcomes. Disease severity was linked to lower Sleep Scores. Conclusions: Sleep in CHDs is broadly comparable to controls, but BMI, sex, and disease severity significantly shape outcomes. The additional variability between weekends and weekdays and a higher risk of sleep-disordered breathing, according to the literature, underscores that sleep is an underestimated target for prevention and clinical care in CHD.