BACKGROUND: The relationship between the severity of obstructive sleep apnea (OSA) and impaired glucose metabolism (IGM) has not yet been fully elucidated in patients with OSA. Accordingly, we sought to clarify this relationship in Japanese patients with OSA. METHODS: The study population consisted of 129 Japanese patients with OSA (apnea-hypopnea index [AHI] > or = 5). A 75-g oral glucose tolerance test was performed in all patients who had not been diagnosed as diabetes mellitus (DM). IGM was defined as either diabetes mellitus (DM) or impaired glucose tolerance (IGT). RESULTS: IGM was observed in 78 (60.5%) patients: DM in 39 (30.2%) and IGT in 39 (30.2%). The frequency of IGM was significantly different among patients with AHI > or = 30, those with 15 < or = AHI < 30, and those with AHI < 15 (72.1%; 53.7%; 35.0%; respectively, p=0.001). Univariate logistic regression analyses showed male sex, the BMI, the AHI, and the lowest SpO(2) to be significantly associated with IGM. A stepwise multivariate logistic regression analysis showed a male sex and the AHI to be independently associated with IGM. CONCLUSION: IGM was observed in 60.5% of Japanese patients with OSA (AHI > or = 5), and the prevalence of IGM increased according to the severity of OSA. Furthermore, the AHI was independently associated with IGM, thus suggesting that OSA may contribute to the development of IGM.
BACKGROUND: The relationship between the severity of obstructive sleep apnea (OSA) and impaired glucose metabolism (IGM) has not yet been fully elucidated in patients with OSA. Accordingly, we sought to clarify this relationship in Japanese patients with OSA. METHODS: The study population consisted of 129 Japanese patients with OSA (apnea-hypopnea index [AHI] > or = 5). A 75-g oral glucose tolerance test was performed in all patients who had not been diagnosed as diabetes mellitus (DM). IGM was defined as either diabetes mellitus (DM) or impaired glucose tolerance (IGT). RESULTS: IGM was observed in 78 (60.5%) patients: DM in 39 (30.2%) and IGT in 39 (30.2%). The frequency of IGM was significantly different among patients with AHI > or = 30, those with 15 < or = AHI < 30, and those with AHI < 15 (72.1%; 53.7%; 35.0%; respectively, p=0.001). Univariate logistic regression analyses showed male sex, the BMI, the AHI, and the lowest SpO(2) to be significantly associated with IGM. A stepwise multivariate logistic regression analysis showed a male sex and the AHI to be independently associated with IGM. CONCLUSION: IGM was observed in 60.5% of Japanese patients with OSA (AHI > or = 5), and the prevalence of IGM increased according to the severity of OSA. Furthermore, the AHI was independently associated with IGM, thus suggesting that OSA may contribute to the development of IGM.