In this study, we examined the association between environmental temperature and sperm quality among 1780 men from one hospital in Wuhan. A threshold effect of ambient exposure temperature on sperm quality was found. A decreased percentage of normal sperm morphology was associated with the increase of ambient temperature above the thresholds at five exposure windows, whereas the effect of exposure temperature below the thresholds was significantly positive. However, a statistically significant association between ambient temperature and sperm concentration or progressive motility only appeared at the early stages of sperm development (lag 15–69 days or lag 70–90 days). We found that there was an interactive effect between ambient temperature and PM2.5 on sperm quality. Specifically, PM2.5 enhanced the effect of ambient temperature on sperm quality when exposure temperatures were above the thresholds.
Seasonal differences have been found for semen quality in previous studies [19, 27], with improved sperm quality during winter and spring. In our study, we found that the population in Wuhan had the best sperm quality in autumn and spring. As temperature change is a main feature of the four seasons, this led us to question whether ambient temperature is the main factor for this seasonal change pattern. Laboratory data from animal models indicates that physiological temperature is an important cause of poor sperm quality [14]. However, few studies have reported that environmental temperature has an influence on semen quality parameters; one study conducted in Italy found a significant relationship between environmental temperature and sperm concentration at lag 3 months, indicating that the environmental temperature had an adverse effect on sperm quality [15]. However, Momen et al. reported the opposite finding that semen parameters were within normozoospermic levels when under high environmental temperature [17]. Although our results found that daily mean temperature correlated with sperm concentration and progressive motility only at the early stage of sperm development (lag 15–69 days or lag 70–90 days), the correlations between ambient temperature and the percentage of normal sperm morphology at the five exposure windows were statistically significant, indicating that environmental temperature affected the percentage of normal sperm morphology over the entire period and at each key stage of spermatogenesis, including sperm storage in the epididymis, sperm motility development, and spermatogenesis.
The exposure–response curve of environmental factors and health outcomes is an important issue. An association between environmental temperature and health outcomes has been demonstrated in many districts around the world, and the relationships generally have been described as U-, V-, W-, or J-shapes [28–30]. In this study, a threshold effect of temperature on health outcomes was observed, and the exposure-response curves of ambient temperature and sperm quality had inverse U-shapes. Our results referred that the threshold values of five exposure windows were inconsistent, and the study population in ≤ threshold and > threshold group of five exposure windows were different. Therefore, the effect at lag 0–90 days in two groups were not simply equal to the sum of the former four estimates at lags 0–9, 10–14, 15–69 and 70–90 days.
In addition, the percentage of normal sperm morphology was found to be associated with ambient temperature over the entire period and at each key stage of spermatogenesis. However, ambient temperature only at 70–90 days prior to semen ejaculation was associated with sperm concentration, and ambient temperature at lag 15–69 days and lag 70–90 days had an effect on sperm progressive motility. Consistent with this finding, a study in Wuhan has reported that PM2.5 exposure only at lag 70–90 days is significantly associated with sperm concentration [4]. Both findings suggest that ambient factors tend to affect sperm concentration and sperm progressive motility at the early stages of sperm development, whereas ambient temperature affects the percentage of normal sperm morphology through all stages of spermatogenesis.
Many research groups have attempted to elucidate the underlying mechanisms of the association between ambient temperature and sperm quality, but they remain unclear. He and colleagues proposed that hot days could cause physiological stress [31], and Cheng et al. utilized an animal model to show that high temperature exposure may induces oxidative stress [32], which could cause an increase of reactive oxygen species (ROS). The excessive production of ROS can disrupt the integrity of DNA in sperm cells and in turn decrease sperm quality. Wang and colleagues have shown that high temperatures can cause apoptosis of spermatogenic cells by overexpression of heat shock proteins (HSP), which induces spermatogenic disorders [33]. Further investigations are required to clarify the detailed underlying biological mechanisms of the effects of ambient temperature on sperm quality.
Previous studies have demonstrated that ambient temperature and air pollution may interact synergistically to affect health outcomes [34–36]. Kim and colleagues examinated whether temperature and air pollution had an interactive effect on mortality in South Korea, and they found that temperature modified the effect of PM10 and increased the risk of daily mortality [34]. An influence of air pollution on sperm quality has been reported in several studies [37, 38], and it has been reported that air pollution is correlated with temperature [39]. Altogether, ambient temperature and air pollution may interact to affect sperm quality. However, no study has explored the modification effect of air pollution for temperature on sperm quality so far. In our study, we revealed a statistically interactive effect of temperature and PM2.5 on sperm quality. We found that PM2.5 enhanced the temperature effect on sperm quality when exposure temperatures reached high levels. Some laboratrory data suggest that PM2.5 inhalation can cause inflammation and oxidative stress [40, 41]. Since both PM2.5 and high ambient temperatures can cause oxidative stress, it is rational to think that there is an interactive effect between them.
One strength of this study is that we are the first to investigate the nonlinear relationship between ambient temperature and sperm quality, finding that daily mean temperature has a threshold effect on sperm quality. In addition, our analysis controlled for various potential confounders, including body mass index (BMI), age, education level, smoking status, season of semen sample collection, days of abstinence, and relative humidity. Moreover, we took the effect of PM2.5 into account and examined whether there was an interaction between daily mean temperature and PM2.5 on sperm quality.
However, there are several limitations in current study. First of all, the participants were recruited from the hospital. Despite the participants were selected through several inclusion and exclusion criteria, the subjects could not fully represent the general population, this might cause a misestimate of the association between ambient temperature and semen quality. The subgroup analysis yielded that ambient temperature was correlated with semen quality, but some results in “normal” group were inconsistent with whole group. It need to be explored in further study whether there are differences in the association between temperature and semen parameters in “normal” and “abnormal” people. Secondly, the city-wide outdoor average temperatures were used instead of individual temperature exposures, lacking of indoor temperature data and activity patterns of participants might lead to exposure measurement error. More advanced techniques and methods, such as land use regression models, are required for more accurate individual exposure assessment. Thirdly, the variation in temperature over the day was not taken into consideration in this study, a future study may focus on the association between variation in temperature over the day and sperm quality, and the comparison of the effects of daily environmental temperature and variation in temperature.