P323 Predictors of insulin resistance in obstructive sleep apnea syndrome

Abstract

Objectives: Insulin resistance has been reported to be high in obstructive sleep apnea syndrome (OSAS) patients compared to non-OSAS controls. The aim of this study was to determine the most significant independent factors that could predict insulin resistance in a cohort of OSAS patients. Design: Prospective cohort study in a tertiary hospital. Methods and measurements: Forty-four (44) patients referred to our sleep laboratory were involved in the study. No patient had a known history of diabetes. The mean age was 44.9 ± 11.3 years, the mean BMI was 31.8 ± 6.5 kg/m, the mean apnea–hypopnea index (AHI) was 29 ± 23.6 events/h, and the mean Epworth sleepiness scale (ESS) score was 12 ± 4.3. All patients underwent standard polysomnography (PSG). Fasting blood samples were taken for measurement of glucose and insulin the morning after the PSG. We estimated insulin resistance by the homeostasis model assessment method (HOMA-IR). The mean HOMA-IR was 5.04 ± 6.37. Age, BMI, anthropometric parameters, PSG findings, smoking, arterial blood hypertension, the Epworth sleepiness scale scoring, and the frequency of nocturia were investigated statistically as possible predictors of insulin resistance. Spearman’s coefficient correlation was used to estimate significant correlations and stepwise multiple regression analysis was performed to identify the independent determinant factors. A value of p < 0.05 was considered significant. Results: Insulin resistance was best correlated with BMI (r = 0.475, p = 0.001), the neck circumference (r = 0.548, p < 0.001), the AHI (r = 0.324, p = 0.031), the lowest SaO2 during the night (r = 0.0379, p = 0.011), the desaturation index (r = 0.375, p = 0.012), the percentage of total sleep time with SaO2 less than 90% (r = 0.427, p = 0.003), the average value of SaO2 < 90% (r = 0.380, p = 0.01) and the presence of nocturia one or more times every night (r = 0.468, p = 0.001). There was no significant correlation with age, waist circumference, the arousal index, the ESS score, smoking or the presence of arterial hypertension. The forward multiple regression analysis with HOMA-IR as the dependent variable showed that neck circumference (b = 0.325, p = 0.025) and nocturia (b = 0.438, p < 0.01) and the desaturation index (b = 0.175, p = 0.275) were the best independent predictors of HOMA-IR (adjusted R = 0.242, p < 0.003). Conclusions: The degree of obesity and the severity of apnea increase the risk for the development of insulin resistance but not independently. Neck circumference, nocturia and the frequency of intermittent apnea-related nocturnal hypoxemia were found to be the best predictors of the presence of insulin resistance in OSAS patients. The prediction of the presence of insulin resistance in OSAS patients may have important clinical implications.