Identification of significant obstructive sleep apnoea in the obese patient: development of the novel DX-OSA score

Abstract

Background and objectives. There is a high prevalence of undiagnosed obstructive sleep apnoea (OSA) in obese surgical patients. We investigated the extent to which anthropometric measurements can be used to identify the presence of significant OSA (Apnoea/Hypopnoea Index (AHI) > 20) in adult patients.Materials and methods. We prospectively studied 1357 adult patients scheduled for elective laparoscopic bariatric surgery. Prior to surgery, body mass index (BMI), gender, neck circumference, STOP-Bang score, SpO 2 , neck and trunk fat (by dual X-ray absorptiometry) were recorded. All patients with a STOPBang score > 5 underwent polysomnography. Auto-titrated Positive Airway Pressure (APAP) therapy was instituted when AHI > 20/h. Predictors of OSA were identified and their cut-off values determined.Results: In total, 1357 patients were screened; 345 patients underwent preoperative polysomnography; 190 had AHI > 20/h and received APAP treatment. The novel Dual X-Ray-Obstructive Sleep Apnoea (DX-OSA) score was derived from the data. The score included 6 items: the STOP-Bang score, BMI, neck fat, trunk fat, baseline SpO 2 , and Expiratory Reserve Volume (ERV), and its sensitivity, specificity, positive-predictive values, negative-predictive values, likelihood ratios, and post-test probabilities determined. At a cut-off of 3, the DX-OSA score had the same sensitivity as the STOP-bang score, but better specificity. The lowest likelihood ratio was found for STOP-Bang and the highest for the DX-OSA score (OSA probability > 83%). Conclusion:The DX-OSA score may be useful for identifying obese patients with significant OSA who require CPAP (continuous positive airway pressure) treatment, and CPAP could be commenced without the need for polysomnography, therefore, without delaying surgery.