The present study followed a panel of 37 young healthy college students in four various study periods, and assessed the effects of daily average indoor/outdoor PM2.5 and ambient temperature on morning/evening PEF simultaneously. Our results revealed that increased PM2.5 was associated with reduced PEF. Of note, the limitation in the two-pollutant models was that SO2 and NO2 data was available only at a site 3 km away, hence errors arising from the heterogeneous and differential exposure measurements could weaken the inferred results from co-pollutant models. Also, in the global analysis, higher ambient temperature was found to be associated with lower PEF. However, higher outdoor temperature showed a significant protective effect on evening PEF in winter. In addition, we also observed a very small antagonistic effect between temperature and PM2.5, which attenuated the harmful impacts of temperature and PM2.5. These findings may have helpful implications for us to better understand the adverse pulmonary impact of exposure to PM2.5 and the association between temperature and lung function among healthy adults.
4.1. Effects of PM2.5 on Lung Function
Previous studies have demonstrated that an elevated concentration of PM2.5
was associated with decreased lung function in susceptible populations, including children, the elderly, and those with respiratory conditions such as asthma and chronic obstructive pulmonary disease (COPD) [10
]. Subjects of the present study were young and healthy college students who might be more resistant to atmospheric pollution than the susceptible populations [27
]. The negative association between PM2.5
and PEF was significant, and so were the lagged and cumulative effects.
The present study revealed a stronger association between indoor/outdoor PM2.5
and evening PEF than morning PEF on the current day. Increased indoor/outdoor PM2.5
concentration was significantly associated with a decrement in current evening PEF but not the morning PEF, and larger effect estimates were also observed in the evening rather than in the morning (Table 4
). Several previous studies have examined the associations between PM and morning/evening PEF and showed inconsistent results. A panel study in children with chronic respiratory symptoms in Finland revealed that the changes in morning and evening PEF for the inter-quartile range (14 μg/m3
) of PM2.5
on the previous day were −1.06 L/min (p
< 0.05) and −0.43 L/min (not significant), respectively [28
]. Another study in European countries found that an increase of 10 μg/m3
was associated with PEF changes of 0.01 L/min (N.S.) in the morning and −0.06 L/min (p
< 0.05) in the evening, respectively [29
]. In 2011, Yamazaki et al.
] studied the effect of hourly concentration of PM on PEF in hospitalized children in Japan and no difference was found in the effect of PM on morning/evening PEF. Thus, it was still not clear whether exposure to PM was more strongly associated with morning or with evening lung function.
Both indoor and outdoor PM2.5
were associated with PEF among the college students (Table 4
& Figure 2
), which might result from the high correlation and strong permeation from the outdoor environment. In two previous studies conducted in Japan [30
] and America [31
], exposure to indoor PM2.5
showed a stronger association with decreased lung function when compared with outdoor PM2.5
. Notably, mean concentration of indoor PM in these two studies was higher than that of outdoor, and indoor and outdoor PM2.5
were weakly correlated. There was evidence showing that indoor-generated particles might be more bioactive than outdoor particles by assessing the in vitro
toxicity of indoor and outdoor PM2.5
collected in Boston-area homes [32
]. Even in the absence of obvious pollutant sources, some allergens, such as house dust mites, might also contaminate the indoor environment. In our study, college students spent most of their time in the dormitories and classrooms on campus. Therefore, assessment of the indoor micro-environment was more appropriate to evaluate the association between PM and PEF of subjects for the present study. Important strength of this study is that ambient PM2.5
exposure was monitored outdoors and indoors where study participants lived and attended school, resulting in more accurate exposure estimates than that based on PM2.5
measured at central site monitors.
4.2. Effects of Temperature on Lung Function
Several previous studies have reported that seasonal differences in temperature were associated with lung function, and warmer temperature led to lower FEV1
in cystic fibrosis patients [19
] and lower PEF in subjects with chronic pulmonary diseases [33
]. A recent panel study of 270 asthmatic children in five cities in Australia revealed that higher ambient temperature was significantly associated with lower PEF and FEV1
even after controlling for children’s respiratory symptoms and air pollutants (PM2.5
and other gaseous pollutants), and the effects of temperature on children’s lung function varied by cities [21
It was also demonstrated in our research that higher indoor/outdoor temperature was associated with lower morning/evening PEF among college students in the global analysis. Results of seasonally stratified analyses suggested that, however, in the winter with general low average temperatures (7.8 ± 2.5 °C), a 1 °C increase of outdoor temperature was associated with an increase of PEF by 2.07 L/min (95%CI
: 0.34–3.80 L/min), which was contradictory to the global analysis and the studies referred to above. Moreover, significant positive associations between temperature and lung function were found in several epidemiological studies. A previous long-term study of 76 elderly COPD patients conducted in east London found that a fall in outdoor or bedroom temperature was associated with increased frequency of exacerbation and decline in lung function [20
]. Belli et al.
] reported that lower outdoor temperatures were associated with increased symptom severity and reduced lung function in former smokers with COPD during cold season as well. As for children with asthma, adverse effects of low indoor temperatures on lung function were also found in the Heating House and Health Study conducted in New Zealand [35
As many studies indicated, both extremely low and high temperatures were robustly associated with adverse cardiopulmonary and cardiovascular events, including increased morbidity and mortality [13
]. The authors further speculated that ambient temperature could be a potential and important confounding factor for lung function. Nevertheless, health effects of ambient temperature might vary by seasonal and climatic factors, study populations and study locations [21
Very few studies have discussed the possible mechanisms underlying the inverse association between both high/low temperature and lung function, which remained unclear. For susceptible populations, such as asthmatic children, one possible explanation is that higher temperature is associated with higher allergen exposure such as pollen loads that may potentially lead to trigger of asthma [21
], and infectious agents (such as through P. aeruginosa
) that may mediate the association between temperature and lung function among patients with cystic fibrosis lung disease [19
]. Another possibility might be that higher temperature is associated with airway drying, which may result in bronchoconstriction and decrement of lung function [43
]. Besides, the reduction of lung function can be caused by increased airway inflammation in cryogenic environment among both asthmatic and COPD patients. On the other hand, cold temperatures can induce peripheral vasoconstriction and shunt blood centrally, and inhalation of cold air can also cause post-exertional bronchoconstriction in asthmatics, both of which thus will result in decreased lung function [20
4.3. Interactive Effects between PM2.5 and Temperature on Lung Function
Seasonal variations have been reported in a few previous studies about the effects of ambient PM on non-accidental mortality of the entire population [12
]. Another two studies in Europe revealed that season and temperature levels strongly modified the PM10
-mortality association [14
]. Synergistic effects of PM10
and high temperature on daily non-accidental, cardiovascular, and cardiopulmonary mortality were also found in a Chinese study conducted in Wuhan [17
]. However, very few studies have explored the interactive effects of PM and ambient temperature on human lung function.
In our research, both ambient PM2.5
and temperature demonstrated a certain seasonal variation in the effects on lung function of healthy college students. Moreover, PM2.5
and temperature were found to have a significant antagonistic interactive effect in reducing PEF (Table 4
), which was not consistent with a recent study conducted in Beijing of China [22
]. However, the effect size of interaction was very small compared to the main effects, so it is hard to say whether the interactive effect between PM2.5
and temperature is of clinical significance for young and healthy adults in the present study.
The seasonal variation of effects of PM on PEF might be explained by changes in PM sources and constituents with different toxicological characteristics in each season [45
], and various chemical components of ambient PM2.5
may play different and complicated roles in affecting lung function of young healthy adults [9
]. However, to date, the mechanisms underlying the interactive effects of ambient PM pollution and temperature on lung function are still unclear. It was biologically assumed that changes of environmental temperature act on the thermoregulatory system, activation of which has direct or indirect effects on the entry of toxicants into the body, thus enhancing or attenuating total intake of airborne pollutants [14
]. Another possibility might be that through effects on reaction kinetics, higher daily ambient temperatures lead to increased ozone levels [48
], which can also result in impaired lung function. The confounding effect of ozone associated with temperature may finally have modified the observed adverse impact of PM on lung function. In addition, some other potential factors may affect both air pollutants and temperature [49
], thus resulting in interactive effects between pollutants and temperature.