Abstract 16453: Risk Factors for Obstructive Sleep Apnea in Patients With Pulmonary Hypertension

Introduction:Patients with pulmonary hypertension (PH) have been shown to have a high prevalence of obstructive sleep apnea (OSA). However, prevalence of OSA is partly dependent on the threshold of apnea hypopnea index (AHI) used. In addition, it is uncertain whether traditional risk factors for OSA apply to patients with PH. We examined the prevalence of OSA and associated risk factors in patients with PH using two different AHI thresholds.Methods:We reviewed consecutive patients diagnosed with PH who underwent in-lab sleep study at a single academic medical center. Patients without hemodynamic data or sleep study with supplemental oxygen use or significant central sleep apnea were excluded. Patients were classified into pre- vs. post-capillary PH based on a pulmonary artery wedge pressure < or ? 15 mmHg, respectively. OSA was defined by AHI ? 15/hr (significant OSA [higher threshold]) or 5/hr (any OSA [lower threshold]). Traditional risk factors of OSA including age, sex and body mass index (BMI), as well as pre- vs. post-capillary PH group were entered into a logistic regression model. Hypertension and left ventricular systolic dysfunction (EF <40%) are also established risk factors for OSA and were tested separately using the same model.Results:A total of 86 patients (Age: 62? 16 years, Female 51%) including those with pre- (n=27) and post-capillary PH (n=59) were analyzed. Patients in the post-capillary PH group tended to be male (55 vs. 33%, P=0.05) and more obese (BMI [kg/m2]: 37 vs. 31, p=0.02) compared to the pre-capillary PH group. As a group, OSA was highly prevalent (significant OSA: 68%; any OSA: 89% with no significant difference in prevalence between the two PH groups). Age (per 10 years) (OR: 1.7 [1.2, 2.7], p=0.007), male sex (9.7 [2.8, 42.3, p=0.0008] and higher BMI (? 30 vs. <30) (10.0 [2.7-45], p=0.001), but not pre- vs. post, were associated with significant OSA. Results were similar for ?any OSA?. No associations were found between hypertension or left ventricular systolic dysfunction and OSA.Conclusions:OSA is highly prevalent regardless of using two different AHI thresholds and shares traditional risk factors for OSA in general population (but more exaggerated). A larger study is needed to better define risk factors of OSA based on PH subtypes.