ELUCIDATION OF OBSTRUCTIVE SLEEP APNEA-RELATED HEART RATE RESPONSE USING A NOVEL CONTINUOUS BEAT TO BEAT BLOOD PRESSURE MONITORING TECHNOLOGY

Abstract Introduction Obstructive sleep apnea (OSA) episode related sympathetic surge may mediate the association of OSA with cardiovascular disease. Heart rate (HR) response (HRR) to OSA events may reflect magnitude of sympathetic surge but is expected to be variable between individuals. We investigated the variability of HRR to OSA events and its relation to degree of oxygen desaturation. Methods We included patients undergoing clinically-indicated polysomnography (PSG). We calculated HRR by deriving inter-beat interval (IBI) from a novel continuous beat-to-beat (b-b) blood pressure (BP) monitoring technology (Caretaker™) that was concurrently recorded during PSG.HRR from respiratory (apnea, hypopnea and desaturation alone events) and non-respiratory events (spontaneous or leg movement-related arousals) were compared. We also examined the association of degree of oxygen desaturation with HRR in a given respiratory event combining all events accounting for the counts of the number of events. In the sub-cohort who underwent split night sleep study, we compared the hourly HR surge events (HRR> 20 pm) between diagnostic and CPAP phases. Results A total of 17 patients(12 men, mean 52 years old, 9 diagnostic and 8 split night PSGs) were included after excluding one patient with poor signal quality due to excessive movement. The device was well tolerated by patients and IBI and BP data were successfully aligned with PSG data. Mean respiratory HRR ranged from 1.0 to 44.0 (Median [IQR]= 11.00[2.0, 20.0]) mmHg. Mean HRR was more pronounced during the non-respiratory events than respiratory events (12.5[7.2] vs. 11.9[6.6] mmHg, p=0.034).Accounting for the count distribution of desaturation/HRR data pair events, there was a moderately positive correlation between the degree of oxygen desaturation and HRR(R=0.63). Hourly HR surge events were significantly reduced during CPAP phase compared to diagnostic phase (11 events/hr vs 126 events/hr, p=0.032). Conclusion We demonstrated highly variable OSA-related HRR patterns between individuals with OSA. There was moderate correlation between degree of oxygen desaturation and HRR to respiratory events. HRR to non- respiratory events was more pronounced than to respiratory events. Future studies should evaluate the clinical implications of the OSA specific HRR. Support (If Any)