476 Sleep-Disordered Breathing is More Predictive than Obesity of Increased Left Ventricular Mass Index in Bariatric Surgery Patients

Abstract

Abstract Introduction Obesity and obstructive sleep apnea (OSA) are associated with left ventricular hypertrophy and increased cardiovascular risk. Alternatively, the “obesity paradox” describes an improved prognosis from heart failure in obesity, though potentially attributable to confounding/bias. We sought to determine the contributions of obesity and sleep-disordered breathing (SDB) to left ventricular function and morphology in bariatric surgery candidates. Methods Patients undergoing polysomnography prior to bariatric surgery from 2011–2018 had demographic (age, gender, race), anthropometric (body mass index [BMI], neck circumference), comorbidities (atrial fibrillation, coronary artery disease, diabetes, hypertension, hyperlipidemia), echocardiographic and sleep-disordered breathing (apnea-hypopnea index [AHI], peak end-tidal CO2 [etCO2]) variables retrospectively examined. The echocardiographic visit closest to polysomnogram within two years was selected with missing values filled by available values within 6 months. Linear regression assessed the relationship of BMI, AHI, and etCO2 with left ventricular mass index (LVMI) after adjustment of demographics and comorbidities. Echocardiographic measures were logarithm transformed before regression analysis. Coefficients and 95% confidence intervals (CI) were calculated by exponential transformation. The analysis was performed based on an overall significance level of 0.05 using SAS software (version 9.4, Cary, NC). Results The total of 832 patients had 24% males, mean age 48.8±12, 60% white, and BMI:49.4±9.5kg/m2. Ejection fraction (%) was 60.0±7.0, and LVMI (g/m2): 80.9±23.7. In adjusted models, LVMI decreased by 2.1% for each 5kg/m2 increase in BMI (coefficient=0.979, 95%CI 0.961–0.997, p=0.022) and increased by 4.3% for each 5 mmHg increase in etCO2 (coefficient=1.043, 95%CI 1.013–1.073, p=0.005). Without adjustment, patients with AHI ≥ 5 had 15.3% higher LVMI than non-OSA group (coefficient=1.153, 95%CI 1.034–1.286, p=0.011) and moderate/severe OSA was associated with a 7.6% higher LVMI than those with AHI<15 (coefficient 1.076, 95%CI 1.003–1.153, p=0.040), but not statistically significant after adjustment. Conclusion In obese patients, nocturnal hypoventilation rather than obesity may have adverse influences on left ventricular morphology. Future studies should focus on clarifying whether obesity is truly protective in terms of LV mass, i.e. reflective of paradox versus a product of bias. The potential benefit of identifying/treating SDB in bariatric surgery candidates to mitigate cardiovascular risk also deserves further investigation. Support (if any) Cleveland Clinic Transformative Resource Neuroscience Award