Clinical predictors of obesity hypoventilation syndrome in morbidly obese subjects with obstructive sleep apnoea

Introduction: Obesity hypoventilation syndrome (OHS) is associated with significant morbidity and mortality but is largely underdiagnosed. The prevalence of OHS increases significantly with raising body mass index (BMI). The objective of this prospective observational study was to determine the prevalence and predictors of OHS in patients with a BMI ≥45 kg/m2 referred for suspected obstructive sleep apnoea (OSA) to St. Vincent's University Hospital, Dublin, Ireland. Methods: Arterial blood gas (ABG) was performed at baseline. Demographic and anthropometric features, history of hypertension and chronic obstructive pulmonary disease (COPD), Epworth sleepiness scale (ESS) and polysomnographic and spirometry data were recorded. Results: Of 98 subjects (51 male), 24 (24%) had OHS (pCO2>6 kPa) with 5 (5%) having a pCO2>7 kPa. Subjects with OHS presented with lower daytime oxygen saturation (SaO2), forced expiratory volume in 1 second (FEV1), higher ESS and more severe OSA than subjects without OHS. Using a multivariate logistic regression model, the percentage of the total sleep time with SaO2 30% was associated with a sensitivity of 92% and a specificity of 62% for OHS with a negative predictive value of 96%. Conclusion: OHS is common in OSA patients with a BMI≥45 but hypercapnia is mild in the majority of cases. A TST90>30% might be a useful cut-off value to determine patients who require further evaluation with ABG.