Long-term Exposure to Particulate Matter and Residential Greenness in Relation to Androgen and Progesterone Levels among Rural Chinese Adults
Dandan Wei,Shanshan Li,Xue Liu,Li Zhang,Pengling Liu,Keliang Fan,Luting Nie,Lulu Wang,Xiaotian Liu,Jian Hou,Wenqian Huo,Songcheng Yu,Linlin Li,Tao Jing,Xing Li,Wenjie Li,Yuming Guo,Chongjian Wang,Zhenxing Mao
DOI: https://doi.org/10.1016/j.envint.2021.106483
IF: 11.8
2021-01-01
Environment International
Abstract:Background: Population-based studies on the associations of long-term exposure to particulate matter (PM) with androgen and progesterone are still scant. Residential greenness is benefits health by promoting physical activity, reducing air pollution, and improving mental health, but it remains unclear whether it is related to androgen and progesterone levels among humans. Aims: This study aimed to explore the individual and interactive effects of PM and residential greenness on serum testosterone and progesterone levels among rural Chinese adults. Methods: A total of 6017 subjects were recruited from the baseline of the Henan Rural Cohort Study in 2016. Serum testosterone and progesterone were measured with liquid chromatography-tandem mass spectrometry. Particulate matters (PM) (PM1, PM2.5, and PM10) were assessed by machine learning algorithms. Residential greenness was assessed using the normalized difference vegetation index (NDVI) within 500-m, 1000-m, and 3000-m buffers around participants' residences. The effects of air pollutants and residential greenness and their interaction on serum testosterone and progesterone levels were assessed using linear mixed-effects models with township as a random intercept. Results: After adjusting for potential confounding factors, a 1 & micro;g/m3 increase in PM2.5 or PM10 was associated with a 0.037 or 0.030 ng/ml increase in serum testosterone, respectively, in females and with a 0.111 or 0.182 ng/ml decrease in serum progesterone, respectively, in males. A 1 & micro;g/m3 increase in PM1, PM2.5 or PM10 was associated with a 0.222, 0.306, or 0.295 ng/ml decrease in serum progesterone, respectively, among females. Moreover, a 0.1-unit increase in NDVI was associated with a 0.310 ng/ml increase in serum testosterone and a 0.170 ng/ml increased in serum progesterone in males, as well as with a 0.143 ng/ml increase in serum progesterone in females. Interaction effects of PM and residential greenness on serum testosterone and progesterone levels were observed, indicating that the effects of residential greenness on serum testosterone and progesterone were modified by high levels of PM. In addition, physical activity significantly mediated 2.92% of the estimated association between greenness and testosterone levels. Conclusions: Our study suggested that long-term exposure to PM was positively associated with serum testosterone in males but negatively associated with progesterone levels in both genderssin. In addition, positive associations of residential greenness with serum testosterone and progesterone levels were observed, but they were modified by high levels of PM. Furthermore, the estimated effects of residential greenness on testosterone levels were partly mediated by physical activity.