Relationship between smoking and the severity of OSA

Abstract

Objective:To explore the relationship between smoking and the severity of OSA. Method:There were 719 patients included in the study, who were accompanied by snoring, daytime sleepiness and other symptoms. Laboratory-based polysomnographic variables （including AHI, oxygen desaturation index and microarousal index, etc.）, and anthropometric measurements （including weight, neck circumference, waist circumference, hip circumference etc.） were collected for all participants. The severity of OSA was determined by AHI. No OSA was defined as AHI<5, mild OSA as AHI of 5 to 15，moderate OSA as AHI of >15 to 30, and severe OSA as AHI of >30. Smoking severity was determined by the smoking index （SI）. Light smoke was defined as SI<200, moderate smoke was as SI 200 to 400, and severe smoke as SI>400. Result:There were 138 cases of non-OSA and 581 cases of OSA. There were 381 non-smokers, 279 smokers and 59 quit smokers. The smoking rate of OSA group was significantly higher than that of non-OSA group （41.5% vs. 27.5%,P<0.01）. After excluding 59 quit smokers, the remaining 660 subjects were divided into four groups according to the severity of smoking, then each group was further divided into four groups according to OSA severity. Unadjusted analysis showed that OSA severity positively correlated with smoking severity （r=0.203,P<0.01）. The positive correlation remained significant after further adjustment for age, BMI and waist-hip ratio. In addition, logistic regression analysis showed that compared to non-smokers, the odd ratios for OSA in moderate smokers were 1.72 （95%CI 1.08-2.75） and in severe smokers were 2.68 （95%CI 1.61-4.46）, after adjustment for age, BMI and waist-hip ratio. Conclusion:The severity of smoking significantly correlated with the severity of OSA. There was increased risk of OSA in patients with severe smoke. The correlation was independent of some confounders such as age and obesity.