For all we know, some articles have discussed the correlation between air pollutants and autoimmune diseases, such as systemic lupus erythematosus (SLE) (Zhao et al. 2019; Jung et al.2019), rheumatoid arthritis (RA) (Alsaber et al. 2020; Di et al. 2019). Some studies have also conducted systematic review and meta-analysis on air pollution and neurological diseases (Cheng et al. 2019; Fu et al. 2019). However, there was no literature meta-analysis the association of air pollution exposure and MS. Our paper attempts to include 6 articles to analyze the effects of air pollutants on MS. Results from our meta-analysis suggest that except that PM10 is related to MS, other pollutants are not related to the MS. The reason may be that the air pollutant level, measurement time and exposure assessment methods were different during the study process, as well as the differences in related measurement methods and established model differences.
The literature on MS included in the analysis were all from developed countries in Europe (Jeanjean et al. 2018; Angelici et al. 2016) and America (Bai L et al. 2018; Yuchi et al. 2020; Chen et al. 2017; Palacios et al. 2017), which may bias the analysis to some extent due to differences in geography, climate and air quality (Witkowska et al. 2016). With the support of more literature, the inclusion of more reliable analysis data from Asian countries and underdeveloped regions will reduce the bias of poor results.
Except for the study region need to be discussed, the time lag effect has been taken into account in the extensive literature on the connection between ambient pollutant exposure and disease. It suggested that there may be a time lag on association between exposure to pollution and causing reactions in the body that affect the progression of the disease (Ritz et al. 2016; Hao et al. 2019). Among the literature on MS summarized in present research, two studies (Jeanjean et al. 2018; Angelici et al. 2016) adopted time-lag model of pollutant exposure, and the rest included literature did not study it. So, in our research, we did not analyze the time lag effect due to the limited number of studies, but we can not deny its potential role.
Considering the connection between air pollutants and autoimmune diseases, we realized that this could be affected by seasonal differences. In a study included in our meta-analysis, it separately analyzed hot season and cold season pollutant model, and found the influence of pollutants to autoimmune diseases varied. In hot season, there is no connection between the pollutants in MS, but in cold season, conversely. The result is similar with a study on COPD (Tsai et al. 2014). It could be observed that the effects of pollutants vary with the seasons. It may be related to the interaction of vegetation coverage, air temperature and humidity in different seasons. From the perspective of seasonal changes, the overall order winter > autumn > spring > summer is observed for atmospheric pollution, and the temporal heterogeneity was significant (Sun et al. 2019; Vojinović et al. 2015). Previous studies also shown that seasonal differences in MS were related to seasonal differences in ultraviolet radiation and vitamin D. Worse air pollution may affect the spread of UVB in the surface layer and caused a sharp decline in the body's production of vitamin D (Watad et al. 2017; Gallagher et al. 2019; Pan et al. 2019). All in all, these suggested that when explore the association of MS and ambient air pollution, more attention need to pay to seasons and meteorological factors in the future research. Several other risk factors, such as less sun exposure and viral infections also need to take into account. This provides guidance for future research to study the risk factors and mechanism of MS.
In terms of data sources, some studies come from large cohorts, while others come from national insurance systems with a nationwide file, etc. Bias may arise from differences in the sources of study population. Meanwhile, differences in the acquisition of pollutant exposure data and adoption of pollutant models may also produce bias to some extent. Limited by methods using in original research, we used single-pollutant models to analyze the effect size which could not consider adjustments for collinearity between pollutants and could not fully illustrate the connection of the disease and the overall mixed pollution exposure. However, as the studies we included were all high-quality cohort or case-control studies, the analysis results were of certain reference value in general.
Many studies found that worse air quality was associated with MS, but were not included in our current analysis due to the difference of statistical processing. Cross-sectional studies conducted in Turkey found that the incidence of MS was higher in the city with an iron-and-steel factory than a cleaner city in the same region (Börü et al. 2018; Börü et al. 2020). Similarly, a study also revealed mean annual PM2.5 levels was associated with MS (Corona-Vázquez et al. 2019). A significant correlation was also found in a case-control study of factory exhaust emissions and MS (Lavery et al. 2018). When considering the pediatric MS, a study found Ecological Quality Index (EQI) contributed to pediatric MS (Lavery et al. 2017). The air quality index (AQI) data was found significantly associated with higher expanded disability status scale (EDSS) of MS patients (Ashtari et al. 2018). Earlier studies also suggested that air pollutants may participate in the development of MS (Oikonen et al. 2003; Heydarpour et al. 2014; Gregory et al. 2008). On the other hand, a study explored the relationship between environmental pollution and the aggravation of MS disease, using the increase in the dosage of cortisol as an indicator to illustrate the possible role of air pollution to MS disease and was a new research approach (Faustini et al. 2018).
In our research, PM10 was found associated with the risk of MS. Many studies have attempted to study this correlation from the perspective of molecular and immune mechanism. Particulate pollutant is a complex mixture of several harmful materials. For example, metal aluminum can induce the production of inflammatory signals, disrupt the cellular metabolism, damage mitochondrial function and lead to oxidative stress. In addition, aluminum have neurotoxicity which can lead to neurological impairment and promote the development of MS. Known as aryl hydrocarbon receptor (AHR), PAHs and dioxins regulated pathological T cells of the central nervous system, and atmospheric particles may stimulate T cell differentiation, aggravated the inflammatory immune response and autoimmune diseases (Segal Y et al. 2018; O'Driscoll et al. 2019). Exposure to particulate matter could produce a dose-dependent proinflammatory cytokines, such a TNF-α, IL-6 and IL-12p40, and it could lead to a highly inflammatory response by contacting to lipopolysaccharide (LPS) by activated macrophage (Gawda et al. 2018). And exposure to particulate matter had a strong effect on redox biochemistry (Mikrut et al. 2018). Studies of the mechanism suggest that air pollutants have a potential role in the progression of MS disease, and epidemiological studies are needed to obtain evidence.
Several shortcomings in our study need to be acknowledged. Firstly, the studies included in our analysis were all observational studies, thus, there may be some bias existing. Secondly, due to the limitation of literature quantity, the amount of studies included in our analysis was small. It needs to update when more research results about this title published articles.
Despite these limitations, this study has several advantages. Firstly, the study year span and sample size in our meta-analysis were both large, which makes conclusions reliable. Secondly, we did not detect any publication bias, that is to say, the results we included may be unbiased, and the results of the sensitivity analysis indicated that our findings were stable. Thirdly, this is the first meta-analysis of air pollutants with MS, which provide a relatively credible conclusion and a direction for further research.