BACKGROUND: Obstructive sleep apnea (OSA) is a common sleep disorder that causes daytime dysfunction and cardiovascular diseases. Nocturnal polysomnography (NPSG) is the standard method of evaluating OSA; however, it is time-consuming, inconvenient, and expensive. Selective performance of NPSG would be possible if we could better predict those who are more likely to have clinically significant OSA. The aim of this study is to define clinical and anthropometric predictors of OSA. METHODS: We included 100 consecutive patients in whom OSA was clinically suspected. Structured sleep interview, anthropometric measurement, and NPSG were performed in all subjects. Presence of OSA was defined when the apnea-hypopnea index was five or more. Parameters from sleep interview and anthropometric data were investigated with multiple logistic regression using the SAS program (ver 8.1, USA) to identify independent predictors of OSA. RESULTS: OSA was diagnosed in seventy-six patients after NPSG. Univariate analysis showed that the male sex, co-existing diabetes, overweight (BMI>or=25), habitual alcohol drinking, large neck circumference (>or=40 cm), high waist circumference/hip circumference (WC/HC) ratio (>or=0.94), and observed apnea were significantly more frequent in OSA patients (p<0.05). Using multivariate analysis, large neck circumference (>or=40 cm) (adjusted OR=4.43, 95% CI: 1.05~18.61) and high WC/HC ratio (adjusted OR=3.48, 95% CI: 1.12~10.80) were found to be the independent predictors of OSA on the NPSG. CONCLUSIONS: We report the predictors of OSA that could be easily identified by clinical and anthropometric evaluations before performing NPSG. This might aid the clinical decision whether or not to perform NPSG in subjects with clinically suspected OSA syndrome.
Alcohol Drinking ; Apnea ; Cardiovascular Diseases ; Humans ; Logistic Models ; Male ; Multivariate Analysis ; Neck ; Overweight ; Polysomnography ; Sleep Apnea, Obstructive* ; Waist-Hip Ratio