Association of traffic air pollution with severity of obstructive sleep apnea in urban areas of Northern Taiwan: A cross-sectional study

Background: Although recent studies have indicated an association between obstructive sleep apnea (OSA) and air pollution, they have reported inconsistent results. Moreover, few studies investigated the effects of short-term air pollution exposure. Objective: To estimate the health effects of short- and long-term exposure to traffic air pollution on mild OSA in Taipei. Methods: We collected participants' data from Taipei Sleep Center and air pollution data from Taiwan Environmental Protection Administration. A spatiotemporal model was used to estimate the individual exposure level. Generalized linear models were used to assess the percent change of overall apnea-hypopnea index (AHI), AHI in rapid eye movement period (AHI-REM), AHI in non-REM (AHI-NREM), and oxygen desaturation index (ODI) associated with an interquartile (IQR) increase in personal pollution exposure. A generalized logistic model was used to estimate the ORs of different severities of OSA compared with the reference group. Results: In the patients with AHI of <15, both short- and long-term exposure to NO2 were significantly associated with AHI and ODI increases: an IQR increase in 2-year mean NO2 increased 7.3% of AHI and 8.4% of ODI; these values were the highest among all exposure windows. The effects of NO2 on AHI increase were stronger in the men and younger patients. Moreover, the association between AHI and NO2 in the patients with AHI of <15 was mediated by the REM stage. NO2 exposure was associated with an increased risk of mild OSA that reached up to 24.8% per IQR increase in NO2 averaged over 2 years. PM2.5 exerted no effects on AHI, but an IQR increase in 1-year and 2-year mean PM2.5 was associated with 6.8% and 8.8% increases in ODI, respectively. Conclusions: Both short- and long-term exposure to traffic air pollution were associated with the risk of mild OSA, which was modified by REM stage.