The association between short-term exposure to wood smoke and salivary cortisol – a panel study

Abstract

Background: Ambient air pollution, has been described as a big contributor for increasing mortality rates. In the Netherlands approximately one million woodstoves and fireplaces are actively being used, resulting in roughly 1.100 million kilograms of wood being burned yearly. This contributes to a significant amount of wood smoke. Currently, the exposure to wood smoke in developing countries has been associated with high indoor air pollution concentrations and (respiratory) health effects. Several studies showed that exposure to wood smoke may be involved as a factor inducing changes in secretion of stress hormones, such as cortisol. However, in the Netherlands no studies have been conducted on this relationship. Aim: This study will investigate the short-term variation of cortisol (stress) in association with outdoor wood smoke exposures in adults with and without asthma and/or COPD. Methods: Our study is a panel study with repeated observations of the stress marker cortisol. Participants completed daily symptom reporting in a diary and provided three weekly saliva samples on one day (two in the morning and one evening sample) between February 13th 2021 and May 10th 2021 for cortisol determination. Exposure to wood smoke was measured daily at central monitoring sites in the four study areas (Bergen, IJburg, Zutphen and De Meern) where the study participants lived during the study period. Measurements include real-time measurements of 24-hour average concentrations of levoglucosan (the most specific wood smoke marker), fine particulate matter (PM2.5), black carbon (BC) and ultrafine particles (UFP). Mixed linear models were used to evaluate associations between measured wood smoke exposure and cortisol. Results: We included 46 participants in our study. The total amount of cortisol samples was 1519, evenly divided between the first morning sample (34%), second morning sample (33,4%) and evening sample (32,6%). More than 67% of all evening samples were below the detection limit of 1 nmol/l (332/495). For cortisol morning 1 and cortisol morning 2 there were only 6 samples below the detection limit. Levoglucosan exposure shows a median and 25th – 75th percentile of 8,3 and 4,3 to 19,2 ng/m3. Black Carbon (Ch 1-6) has a median and 25th – 75th percentile of 0,11 and 0,03 to 0,52 µg/m3. Wood smoke exposure had highly non-significant associations with cortisol in the morning, with no pattern of direction of association. In the evening and cortisol awake response (CAR) associations were negative but highly non-significant. Conclusion: No consistent evidence for an association between salivary cortisol and wood smoke was found.