Abstract

BORSINI, Eduardo et al. Clinical predictors of apnea hypopnea index (AHI) > 15 / hour in the respiratory polygraphy. Rev. amer. med. respiratoria [online]. 2012, vol.12, n.3, pp.90-97. ISSN 1852-236X.

Objectives: To identify the predictive value of variable risk factors for the diagnosis of clinically significant obstructive sleep apnea (apnea-hypopnea index =15 / hour) in a population referred for respiratory polygraphy (RP). Material and methods: Between January 2010 and March 2012 we studied prospectively 614 adult patients referred to our sleep clinic for the study of sleep respiratory disorders. The patients completed Berlin and Epworth questionnaires and were monitored by a level III respiratory polygraphy using a self placement method at home during one night. Results were correlated with demographic and anthropometric data. Obesity was defined as a body mass index (BMI) > 30. Clinically significant obstructive sleep apnea (OSA) was diagnosed if the patient was found to have an apnea - hypopnea index (AHI) greater than = 15 per hour. The analysis included BMI (> 30), age (> 50 years), educational level (primary school or more), Berlin questionnaire (high or low risk) and Epworth Sleepness Scale (> 10). Data were analyzed through logistic regression for factors predicting AHI =15 / hour. Results: 614 patients (392 men (63.8%), 222 women) with a mean age of 54.9 years old were included. High risk OSA according to Berlin questionnaire was identified in 536 subjects (87.3%) while only 207 (33.7%) had an Epworth score > 10 points. 330 subjects (53.74%) had a body mass index > 30. IAH > 15 / hour was diagnosed in 235 respiratory poligraphies (38.27%). Logistic regression analysis showed that three independent predictors were significant for the diagnosis of clinically significant OSA: male sex: OR 3.63 (CI 95%: 2.43 to 5.43) p = 0.0001; BMI > 30: OR 2.45 (CI 95% 1.69 to 3.56) p = 0.0001), and age > 50 years: OR 2.05 (CI 95% 1.39 to 3.02) p = 0.0001). We did not find significance for Berlin's high risk questionnaire: OR 1.17(CI 95% 0.63 to 2.17) p = 0.605, nor for the Epworth score > 10: OR 1.35 (CI 95% 0.93 to 1.97) p = 0.113. The differential analysis between both genders showed significant differences in age (women; p = 0.015 and men; p = 0.007) and in the BMI (women: 0.027 and men: 0.0001). The logistic regression indicated that there were three independent predictors for AHI >15 / hour, namely the male sex (OR: 3.6; CI 95%: 2.43-5.43) p = 0.0001; obesity (OR: 2.45; CI 95%: 1.69-3.56) p = 0.0001, and age more than 50 (OR: 2.05; CI 95%: 1.39-3.02) p = 0.0001. Using this model, a patient referred for respiratory polygraphy because of clinical suspicion of OSA who presents the three significant clinical predictors has a 70% chance of having an AHI > 15/hour. On the contrary, when the three predictors are absent the probability of finding an AHI > 15/hour is 7%. Conclusions: In a population referred for respiratory polygraphy in a general hospital, the male sex, obesity and age older than 50 years were predictors of AHI >15 / hour. These three clinical variables can help to predict the disease risk before the diagnostic test and therefore to determine priorities for respiratory sleep studies.

Keywords : Respiratory polygraphy; Risk factors; Clinical predictors.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )