Whereas the available evidence mostly supports chronic sleep disruption as a risk factor for incident AF, less is known about the near-term risk of a discrete atrial fibrillation (AF) episode following a night of disrupted or poor-quality sleep. This prospective, randomized, case-crossover study aims to examine the acute effects of sleep disruption on clinically relevant cardiac ectopy pertinent to AF.

One hundred patients aged 21 years or older who have a hypoglossal nerve stimulation (HGNS) device implanted for obstructive sleep apnea, without a history of permanent AF, not on Class 1 or 3 antiarrhythmic medications and willing to participate in the study, will be randomized to "off-on" versus "on-off" periods (where "off" is turning off the HGNS device for one night and "on" is turning on the HGNS device), assuring no more than two consecutive days of either device use or not, over the ensuing 14 days. Participants will be fitted with an automatically recording electrocardiographic monitor. The primary endpoint will be the daily number of cardiac ectopic beats, and secondary endpoints: daily number of PACs, daily number of PVCs, and daily presence versus absence of AF episodes ≥ 30 seconds. Primary analyses will employ intention-to-treat principles, examining randomization assignment as the primary predictor.

Inclusion Criteria:

• Are age 21 years or older
• Have a HGNS device implanted for obstructive sleep apnea and followed up at UCSF Health
• Are willing to abstain from using the upper airway stimulation device for no more than two consecutive days as instructed over the 14-day trial period
• Able and willing to provide written informed consent

Exclusion Criteria:

• Currently pregnant or trying to get pregnant
• Are currently taking class 1 or 3 anti-arrhythmic medications
• Have a history of permanent AF or expected to have continuous AF throughout the study period
• Have congenital heart disease
• Ventricular pacing >40%
• Are unable to read or sign to provide informed consent