0614 Diurnal Patterning of Autonomic Measures in Sleep Apnea and Paroxysmal Atrial Fibrillation and Response to Continuous Positive Airway Pressure Therapy

Introduction

Diurnal patterning of autonomic function in paroxysmal atrial fibrillation (PAF) and sleep disordered breathing (SDB) is unknown. We hypothesize heart rate variability (HRV) as surrogates of autonomic function, exhibit diurnal differences in PAF relative to SDB severity and treatment.

Methods

We leveraged the Sleep Apnea and Atrial Fibrillation Biomarkers and Electrophysiologic Atrial Triggers (SAFEBEAT,NCT02576587) study focused on participants with PAF and SDB (apnea hypopnea index,AHI≥15,3% oxygen desaturation hypopnea). Attended 16-channel polysomnography (PSG) and continuous ECG monitoring (Heartrak Telemetry®) for 7-21 days was performed at baseline and after 3-months of continuous positive airway pressure (CPAP). Linear mixed-effects models (least square means,95%CI) were used to assess relationships between daily average HRV measures (frequency domain:LF,HF,LF/HF;time domain:MNN,RMSSD,SD1,SD ratio and novel non-linear:DFA-alpha measures) with SDB (AHI),%sleep time with SaO2<90%(TRT<90): per 5-unit increase),effect of 3-month CPAP relative to sleep-wake and statistical interaction of sleep-wake. Analyses were conducted using SAS version v.9.4, Cary, NC.

Results

The analytic sample was comprised of 33 cases with PAF and SDB:61.1±11.7 years,62.5% male, BMI:33.9±7.2kg/m2,75% Caucasian,AHI 15.1 (IQR: 4.4,29.4) and 68.8% on atrioventricular nodal blocking medications. AHI was associated with frequency (HF:0.08[0.01,0.16] and LF/HF:-0.11[-0.20, -0.01]), time (SD1:0.08[0.02,0.14] and SD ratio: 0.09[0.04,0.14]) and non-linear (DFA-alpha1: -0.02[-0.036,-0.003]) domain measures during wake, but not sleep. Significant sleep-wake and AHI as well as TRT<90 interactions relative to HRV measures were observed (p≤0.001). Only SD ratio was associated with TRT<90 (0.12[0.03,0.24]). Baseline to follow-up CPAP time domain measures were altered mainly during wake versus sleep with MNN increased 0.13: [0.08,0.19],p<0.001; RMSSD increased 0.13 [0.08,0.19], p<0.001; SD1 increased 30% [0.09,0.55], p=0.004; SD ratio increased 20% [0.01,0.43], p=0.033,and also frequency domain: HF increased 33%[0.03,0.72], p=0.028.

Conclusion

SDB defined by AHI--more so than nocturnal hypoxia--was associated with surrogate autonomic measures impacted by CPAP intervention during wake and not sleep in PAF. SDB-related autonomic influences in PAF appear to be more pronounced during wakefulness suggesting long-term potentiation-like influences.

Support

This study was supported by the National Heart, Lung and Blood Institute (NHLBI) [Grant R01 HL108493] and National Institutes of Health (NIH) National Center for Research Resources [Grant UL1 RR024989]

0558 Sleep Disordered Breathing and Right Ventricular Electrocardiographic and Functional Characteristics in Group 1 Pulmonary Arterial Hypertension

Introduction

Right ventricular (RV) electrophysiologic and functional alterations related to sleep disordered breathing (SDB) in pulmonary arterial hypertension (PAH) are not well understood. We hypothesize an association between SDB and RV electrophysiological/functional measures in World Symposium of Pulmonary Hypertension (WSPH) Group 1 PAH.

Methods

The NHLBI multicenter PVDOMICS study (NCT02980887) enrolls patients with PAH undergoing a battery of assessments including home sleep apnea testing(NOX-T3, Carefusion®) or with historical sleep study data. Logistic(OR,95%CI) and linear(beta coefficients,95%CI) regression models adjusted for age, sex, race, body mass index (BMI, kg/m2), PAH medications, supplemental oxygen(O2), positive airway pressure(PAP) were used to assess associations of SDB(apnea hypopnea index,(AHI), ≥3% desaturations(hypopnea), percentage recording time with SaO2<90% (TRT<90%) with electrocardiographic measures: RV hypertrophy(RVH), right bundle branch block(RBBB), and right axis deviation(RAD), echocardiographically-derived RV systolic pressure(RVSP) and RV ejection fraction(RVEF) from cardiac MRI. Analyses were performed based on an overall significance level of 0.05, using SAS software (version 9.4, Cary, NC).

Results

The analysis consisted of 182 PAH participants with age: 52.5±13.9 years, 71.4% female, 88.9% Caucasian, BMI:30.3±7.8 kg/m2, RVEF: 37.3±11.6, and RVSP: 67.0±23.4. None of the electrocardiographic measures were associated with AHI and only RVH was significantly associated with TRT<90% (1.25:1.09,1.43),p=0.001. Although AHI was not associated with RVSP, a 10% increase in TRT<90% was associated with a 2.60mmHg increase in RVSP (2.60:1.44,3.76),p<0.001. Each 10-unit increase in AHI was associated with a 2.72% reduction of RVEF (-2.72:-4.89,-0.56),p=0.014, and each 10-unit increase in TRT<90% was associated with a 0.72% reduction of RVEF (-0.72:-1.38,-0.06),p=0.033.

Conclusion

We identify nocturnal hypoxia as a predictor of RV electrophysiological and functional alterations even after consideration of confounding factors. SDB as determined by AHI was also more so associated with reduced RVEF than hypoxia. Future mechanistic studies should focus on further elucidation of SDB and nocturnal hypoxia on pathogenesis of RV dysfunction in PAH.

Support

U01HL125218/U01HL125205/U01HL125212/U01HL125208/U01HL125175/U01HL125215, U01HL125177/Pulmonary Hypertension Association