• Sylvie Mazière, Jean-Louis Pépin, Natalia Siyanko, Catherine Bioteau, Sandrine Launois, Renaud Tamisier, Nathalie Arnol, Patrick Lévy, Pascal Couturier, Jean-Luc Bosson, and Gaëtan Gavazzi.
• Geriatric Department, Grenoble University Hospital, Grenoble, France. Electronic address: SMaziere@chu-grenoble.fr.
• J Am Med Dir Assoc. 2014 Jun 1; 15 (6): 447.e9-14.

BackgroundSleep Apnea Syndrome (SAS) prevalence increases with age. In the elderly, symptoms are less specific (falls, cognitive or functional decline, polymedication). Polysomnography, the gold standard technique to diagnose SAS, is challenged by sleep laboratories' waiting lists and high associated costs. Nocturnal oximetry is an easy-to-use tool widely available outside the sleep medicine field identifying intermittent hypoxia, the landmark of SAS. It might be an interesting and easy way to screen for SAS in the functionally and cognitively impaired elderly living in long-term care settings.ObjectivesThe primary goal of this study was to assess the accuracy of the variability index of nocturnal pulse oximetry to detect moderate to severe SAS in patients older than 75 hospitalized in stable condition. The secondary goals were to assess the accuracy of the other indices of pulse oximetry (oxygen desaturation index [ODI]), and to determine the prevalence of moderate to severe SAS in our population.MethodsIn-hospital sleep studies with simultaneous respiratory polygraphy and nocturnal pulse oximetry were performed. Comorbidities were assessed by the Cumulative Illness Rating Scale for Geriatrics (CIRS-G) in association with a comprehensive geriatric assessment.ResultsEighty patients (mean age 85.3 ± 5.3 years) were included. Seventy-two percent of the patients exhibited moderate to severe SAS (95% CI 58.9-82.9), including 59.5% of severe SAS (apnea + hypopnea index >30/hour). SaO2 variability index using a threshold of 0.51, the sensitivity and negative predictive value (NPV) were 100%. With a value above 0.88, positive predictive value and specificity were high (respectively 96.6% and 93.8%). ODI of 3% or higher and 4% or higher were highly specific but less sensitive.ConclusionPrevalence of moderate to severe SAS in multimorbid hospitalized elderly patients is high. Automatic analysis of the variability of nocturnal SaO2 is a reliable tool for geriatricians to screen and rule out moderate to severe SAS. Our study suggests an important role of pulse oximetry as the first step in the diagnostic strategy for moderate to severe SAS in this population.Copyright © 2014 American Medical Directors Association, Inc. Published by Elsevier Inc. All rights reserved.

20 keywords...

hide…