Effects of filtered fresh air ventilation on classroom indoor air and biomarkers in saliva and nasal samples: A randomized crossover intervention study in preschool children

To evaluate the short-term effects of filtered fresh air ventilation system on classroom indoor air and biomarkers in saliva and nasal samples in preschool children, a randomized crossover study was conducted in a kindergarten in Shanghai, China in 2016. Two classrooms at the same grade (n = 43) were selected and fresh air ventilation systems (FAVS) with high efficiency particulate air filter (HEPA) were installed. In the first week, FAVS-HEPA was run in one classroom for 2 continuous school days and the other classroom was remained as usual with no use of FAVS-HEPA. After one week of wash-out, the ventilation modes exchanged between two classrooms and another 2 days of intervention were repeated. Real-time indoor and outdoor air pollution and climate factors (PM2.5, Temp and relative humidity (R.H.)) were measured. Saliva and nasal internal mucosa samples were collected immediately at the end of each intervention scenario. Linear mixed-effect regression model was applied to evaluate the effects of intervention on children's health indicators controlling for age, gender, height, BMI and temperature. The results showed, with FAVS-HEPA, the classroom indoor fine particulate matter (PM2.5) (29.1 +/- 17.9 mu g/m(3)) was on average significantly lower than that without FAVS-HEPA (85.7 +/- 43.2 mu g/m(3)). By regression analysis, each 10 mu g/m(3) decrease of indoor PM2.5 during the 8 school hours in the first intervention day was associated with an average of 1.76% (95% confidence interval (CI) 0.43-3.08%) increase in saliva lysozyme. This percentage increased to 2.41% (95%CI 0.52-4.26%) if related to the PM2.5 level in 16 school hours over 2 days of intervention. A total of 19 nasal bacterial taxa were lower in subjects exposed to FAVS-HEPA, compared to that with no use of FAVS-HEPA, despite the general bacteria diversity levels in nasal samples were not statistically different. Among others, Providencia species showed significant effects in mediating the associations between higher PM2.5 and lower lysozyme. In conclusion, using FAVS-HEPA was effective in decreasing the classroom indoor PM2.5. Saliva lysozyme, as a non-specific immune biomarker, was significantly inversely associated with indoor PM2.5. Certain nasal bacteria might play key roles in mediating PM2.5 exposure and children's lysozyme levels.