Objective: We sought to clarify the association between the severity of obstructive sleep apnea (OSA) and hemoglobin A1c (HbA1c) levels.
Methods: HbA1c levels were measured in 330 subjects who were suspected of having OSA and who underwent polysomnography. A 75-g oral glucose tolerance test was performed in 308 subjects who had not been diagnosed to have diabetes mellitus. Patients with use of hypoglycemic agents were excluded. The 330 subjects were divided into three subgroups: 164 with normal glucose tolerance, 111 with impaired glucose tolerance, and 55 with diabetes mellitus (DM).
Results: Univariate regression analyses showed that natural logarithm(Ln)age, Lnbody mass index, hypertension, DM, Lnapnea-hypopnea index, Lnarousal index, and Lnlowest arterial oxyhemoglobin saturation (L-SpO(2)) were associated with LnHbA1c. A multiple regression analysis showed that Lnage (β=0.16, P=0.003), DM (β=0.39, P<0.001), and Ln L-SpO(2) (β=-0.29, P<0.001) were associated with LnHbA1c. Subgroup analyses showed that Ln L-SpO(2) was independently associated with LnHbA1c in each subgroup.
Conclusions: L-SpO(2) was independently associated with HbA1c regardless of the glucose tolerance status. This suggests that OSA-induced hypoxia could increase an individual's average blood glucose level.
Copyright © 2012 Elsevier B.V. All rights reserved.