In patients with atrial fibrillation (AF), the prevalence of moderate-to-severe sleep-disordered breathing (SDB) ranges between 21% and 72% and observational studies have demonstrated that SDB reduces the efficacy of rhythm control strategies, while treatment with continuous positive airway pressure lowers the rate of AF recurrence. Currently, the number of apneas and hypopneas per hour (apnea-hypopnea-index, AHI) determined during a single overnight sleep study is clinically used to assess the severity of SDB. However, recent studies suggest that SDB-severity in an individual patient is not stable over time but exhibits a considerable night-to-night variability which cannot be detected by only one overnight sleep assessment. Nightly SDB-severity assessment rather than the single-night diagnosis by one overnight sleep study may better reflect the exposure to SDB-related factors and yield a superior metric to determine SDB-severity in the management of AF. In this review we discuss mechanisms of night-to-night SDB variability, arrhythmogenic consequences of night-to-night SDB variability, strategies for longitudinal assessment of nightly SDB-severity and clinical implications for screening and management of SDB in AF patients.