SESSION TITLE: Sleep Disorders Posters II: Consequences of OSA and Treatment
SESSION TYPE: Original Investigation Poster
PRESENTED ON: Wednesday, October 28, 2015 at 01:30 PM - 02:30 PM
PURPOSE: The ability of questionnaires to estimate the probability of obstructive sleep apneas (OSA) is varying, and it is challenging to know the performance of STOP-BANG with simplified methods such as home respiratory polygraphy (RP). The aim of this study was to assess the performance of STOP-BANG and its capacity to predict sleep apnea in patients with high pre-test likelihood to present OSA referred for a home RP.
METHODS: We conducted a cross-sectional study of patients recruited over 26 months. They were asked to complete the STOP-BANG questionnaire during evaluation prior to RP and were evaluated according to different apnea hypopnea index (AHI) cut-offs. Areas under ROC curves were calculated for multiple logistic regression models.
RESULTS: 869 patients were studied. 557 were male (64.1%) with a median age of 52.82±14.43 years, a BMI of 32.88±8.51, and Epworth Sleepiness Scale (ESS) score of 7.95±5.17. The performance for AHI ≥5/hour (ROC area) was: STOP: 0.62, BANG:0.66, and STOPBANG 0.69. The best sensitivity (S) - specificity (Sp) relationship for AHI ≥5/hour was found with 5 components in any combination (S: 56.02%; Sp: 70%). For AHI ≥30/h, STOP was 0.68, BANG: 0.66 and STOPBANG: 0.73 and the best S-Sp relationship was obtained with 5 components (S: 68% - Sp: 63.6%). Six variables (snoring, observed apneas, high blood pressure (HBP), BMI>35, neck perimeter>40 cm, and male gender) showed the best performance for AHI>30/h; ROC area: 0.76.
CONCLUSIONS: STOP-BANG shows moderate discrimination for AHI >5 and ≥30/hour using RP. Five components in any combination have acceptable diagnostic S to identify patients with severe OSA. SOP-BNG performed best to identify AHI≥30/h.
CLINICAL IMPLICATIONS: This work contributes both to the understanding of the performance of a multi-component scale and to the better interpretation of different combinations of clinical and anthropometric variables when simplified methods are used for the ambulatory diagnosis of OSA.
DISCLOSURE: The following authors have nothing to disclose: Eduardo Borsini, Glenda Ernst, Martín Bosio, Fernando Grassi, Tamara Decima, Jeronimo Campos, Fernando Ditullio, Felipe Chertcoff, Alejandro Salvado
No Product/Research Disclosure Information