Abstract
This study investigated the relationship between obstructive sleep apnea and haemoglobin A1c (HbA1c) among Hispanics/Latinos in the United States and assessed whether this relationship was moderated by glycaemic status. This was a cross-sectional analysis of the Hispanic Community Health Study/Study of Latinos cohort. The sample consisted of 13,394 participants with valid measures of obstructive sleep apnea, HbA1c, and study covariates. Obstructive sleep apnea was assessed with the apnea-hypopnea index and categorised as obstructive sleep apnea if the apnea-hypopnea index was ≥5 events/h. HbA1c measures were obtained through fasting blood samples. Fasting plasma glucose (FPG), 2-h post-load plasma glucose (2h-PG) and use of antihyperglycaemic medications were used to define glycaemic status (i.e., normoglycaemia [FPG < 5.6 mmol/L (< 100 mg/dL) and 2h-PG < 7.8 mmol/L (140 mg/dL)], prediabetes [FPG 5.6-6.9 mmol/L (100-125 mg/dL), and/or 2h-PG 7.8-11.0 mmol/L (140-199 mg/dL)], diabetes without treatment [FPG > 7.0 mmol/L (≥ 126 mg/dL) and/or 2h-PG ≥ 11.1 mmol/L (≥ 200 mg/dL)], and diabetes with treatment. Multivariable linear regression was used to calculate adjusted least square means. Overall, 25.9% of the sample had obstructive sleep apnea and 49.2% had normal glycaemic levels, 36.1% had prediabetes, 6.5% diabetes without receiving treatment, and 8.3% diabetes and undergoing treatment for it. Participants with obstructive sleep apnea had significantly higher adjusted mean HbA1c (adjusted mean [standard error] 5.85 [0.03)]) than those without (5.70 [0.02)]; p < 0.001). Models stratified by diabetes status showed that the association between obstructive sleep apnea (versus not) and higher HbA1c was only for participants with normal glycaemic status (adjusted mean [standard error] 5.27 [0.01] versus 5.30 [0.01]; p = 0.013) and prediabetes (5.59 [0.01] versus 5.66 [0.01]; p < 0.001). In conclusion, obstructive sleep apnea was associated with higher HbA1c in a diverse sample of Hispanic/Latino adults in the United States. This association was present only for participants with normal glycaemic status or with prediabetes. Studies are needed to further understand the clinical implications of the association between obstructive sleep apnea and HbA1c according to glycaemic status.