key: cord-1013698-27t5b1ra authors: Li, Zhongqi; Tao, Bilin; Hu, Zhiliang; Yi, Yongxiang; Wang, Jianming title: Effects of short-term ambient particulate matter exposure on the risk of severe COVID-19 date: 2022-02-01 journal: J Infect DOI: 10.1016/j.jinf.2022.01.037 sha: 13fe256d6e77f023b1466b08cb929dfa1e09d468 doc_id: 1013698 cord_uid: 27t5b1ra OBJECTIVES: : Previous studies have suggested a relationship between outdoor air pollution and the risk of coronavirus disease 2019 (COVID-19). However, there is a lack of data related to the severity of disease, especially in China. This study aimed to explore the association between short-term exposure to outdoor particulate matter (PM) and the risk of severe COVID-19. METHODS: : We recruited patients diagnosed with COVID-19 during a recent large-scale outbreak in eastern China caused by the Delta variant. We collected data on meteorological factors and ambient air pollution during the same time period and in the same region where the cases occurred and applied a generalized additive model (GAM) to analyze the effects of short-term ambient PM exposure on the risk of severe COVID-19. RESULTS: : A total of 476 adult patients with confirmed COVID-19 were recruited, of which 42 (8.82%) had severe disease. With a unit increase in PM(10), the risk of severe COVID-19 increased by 81.70% (95% confidence interval [CI]: 35.45, 143.76) at a lag of 0-7 days, 86.04% (95% CI: 38.71, 149.53) at a lag of 0-14 days, 76.26% (95% CI: 33.68, 132.42) at a lag of 0-21 days, and 72.15% (95% CI: 21.02, 144.88) at a lag of 0-28 days. The associations remained significant at lags of 0-7 days, 0-14 days, and 0-28 days in the multipollutant models. With a unit increase in PM(2.5), the risk of severe COVID-19 increased by 299.08% (95% CI: 92.94, 725.46) at a lag of 0-7 days, 289.23% (95% CI: 85.62, 716.20) at a lag of 0-14 days, 234.34% (95% CI: 63.81, 582.40) at a lag of 0-21 days, and 204.04% (95% CI: 39.28, 563.71) at a lag of 0-28 days. The associations were still significant at lags of 0-7 days, 0-14 days, and 0-28 days in the multipollutant models. CONCLUSIONS: : Our results indicated that short-term exposure to outdoor PM was positively related to the risk of severe COVID-19, and that reducing air pollution may contribute to the control of COVID-19. In December 2019, coronavirus disease 2019 , caused by SARS-CoV-2, was first reported in Wuhan and escalated into a global pandemic (Novelli et al. 2020 . There is no doubt that COVID-19 poses a severe threat to global health. Patients with COVID-19 can be categorized as mild, moderate, severe, or critical based on their condition. The prognosis of severe and critical patients is poor , and the risk factors for these categories include older age and hypertension ). Levels of particulate matter (PM), which are produced by the combustion of biomass, diesel and spark-ignited vehicle emissions, is highly correlated with environmental pollution (Croft et al. 2020) . In particular, PM with an aerodynamic diameter of ≤10 μm (PM 10 ) or ≤2.5 μm (PM 2.5 ) has attracted widespread public attention, as particles of this size are inhalable (Prabhakaran et al. 2020 . PM was found to be positively associated with the risk of communicable and noncommunicable diseases (Landguth et al. 2020 , Matsuo et al. 2016 , Yao et al. 2019 . A global time-series study showed that short-term PM exposure contributed to increased mortality due to cardiovascular and respiratory disease (Liu et al. 2019) . Recently, several studies have revealed the possible links between PM exposure and the risk of developing COVID-19 (Tung et al. 2021 , Zoran et al. 2020 . For example, a multicity study in China found that for each 10 μg/m 3 increase in PM 10 and PM 2.5 , the risk of COVID-19 increased by 5% and 6%, respectively . Another study in Germany showed that every 1 μg/m 3 increase in PM 10 and PM 2.5 was associated with 52.38 and 199.46 more confirmed COVID-19 cases per 100,000 inhabitants, respectively (Prinz and Richter 2022) . The associations between ambient PM exposure and the risk of COVID-19 have been ardently discussed. In Italy, cases in the most polluted areas had higher rates of intensive care unit (ICU) admissions and mortality rates, indicating a possible link between air pollution and severe COVID-19 (Frontera et al. 2020) . Nevertheless, studies on the effects of outdoor PM exposure on the severity of COVID-19 are insufficient, notably in China. On July 20, 2021, Nanjing Lukou International Airport identified nine domestic COVID-19 cases through regular screening. Subsequently, the disease spread rapidly to surrounding cities, resulting in hundreds of confirmed cases in four cities in Jiangsu Province, including Nanjing, Yangzhou, Huaian, and Suqian. The scale of this epidemic in China was second only to the Wuhan epidemic in 2020. Genome sequencing confirmed that the pathogen causing this epidemic was the SARS-CoV-2 B.1.617.2 (Delta) variant, which initially appeared in India (Cherian et al. 2021 ). Thus, we collected data from COVID-19 patients identified in this outbreak and data on meteorological factors and air pollutant concentrations during the same time period and in the same region where the cases occurred, aiming to evaluate the relationships between short-term ambient PM exposure and COVID-19 severity. We collected data from COVID-19 patients admitted to Nanjing Public Health Medical Center from July 20, 2021 to August 17, 2021. All of the cases came from Nanjing, Yangzhou, Huaian, or Suqian and were related to the recent outbreak of COVID-19 that originated in Nanjing Lukou International Airport. The locations of the four cities are shown in Figure 1 . The inclusion criteria were as follows: (1) patients aged ≥18 years and (2) patients infected with the Delta variant. We collected data on the general characteristics of patients, including city, sex, age, current or past hypertension (yes or no), current or past diabetes (yes or no), current or past heart disease (yes or no), current or past carcinoma (yes or no), current or past COPD (yes or no), current or past asthma (yes or no), current or past autoimmune disease (yes or no), vaccination status (unvaccinated, partially vaccinated, or fully vaccinated), and number of days between onset and hospitalization. Patients were defined as fully vaccinated if they had received two doses of a vaccine with an interval between the two doses of ≥21 days and a disease onset date of ≥14 days from the second dose (Kang et al. 2021) . The diagnosis and classification of COVID-19 was based on the "Guidelines for Diagnosis and Treatment of COVID-19 (Trial Eighth Edition)" issued by the National Health Commission (http://www.nhc.gov.cn/). Patients were categorized as mild, moderate, severe, and critical based on their symptoms. In the current study, we combined the severe and critical categories , referred to as "severe" (Hu et al. 2022 ). We extracted data on meteorological factors, including daily average temperature (°C) and daily average wind speed (m/s) in four cities between June 15, 2021 and August 15, 2021 from the China Meteorological Data Sharing Center (http://data.cma.cn/), as well as daily average concentrations of six ambient air pollutants, including PM 10 , PM 2.5 , SO 2 , NO 2 , CO, and O 3 (the concentration of O 3 was the maximum 8-hour moving average) in four cities during the same period from the National urban air quality real-time release platform (http://106. 37.208.233:20035/) . Except for the unit of CO concentration, which was mg/m 3 , the concentration of other pollutants was measured in μg/m 3 . We applied the generalized additive model (GAM) with the logit link function to estimate the effects of short-term PM exposure on the severity of COVID-19. The GAM has been widely used in exploring the associations between air pollutants and health outcomes (Liu et al. 2019, Xie and . Covariates adjusted for in the model included city, sex, age, current or past hypertension, current or past diabetes, current or past heart disease, current or past carcinoma, current or past COPD, current or past asthma, current or past autoimmune disease, vaccination status, days between disease onset and hospitalization, daily average ambient temperature, and daily average wind speed. The thin plate spline function was applied to control for the nonlinear effects of meteorological factors on COVID-19 (Xie and Zhu 2020) . Previous studies have shown that PM exposure might have lag effects on health (Chen et al. 2017 , You et al. 2016 ). Thus, we calculated the moving average concentration of PM to describe personal PM exposure levels according to the onset date of COVID-19 and city of residence (Chen et al. 2017 . For instance, if the onset date was August 1, we extracted the daily average concentration of PM in the same city from July 4 to August 1. Then, we calculated the average concentration from July 25 to August 1 as the 8-day moving average, from July 18 to August 1 as the 15-day moving average, and from July 4 to August 1 as the 29-day moving average. We used four different lag times, including lags of 0-7 days (8-day moving average), 0-14 days (15-day moving average), 0-21 days (22-day moving average), and 0-28 days (29-day moving average) Zhu 2020, Zhu et al. 2020 ). The strength of associated evidence was expressed as the change in the risk of severe COVID-19 and its corresponding 95% confidence interval (CIs) for a unit increase in PM concentration. We conducted two sensitivity analyses to examine the robustness of the relationships between PM exposure and severe COVID-19. First, as mentioned above, we estimated the association at different lag times. Second, we included data for other air pollutants to construct multipollutant models. We used the Spearman rank correlation test to evaluate the correlation among air pollutants. Only pollutants with an |r| of <0.7 were entered in the multipollutant models to address the problem of multiple collinearity (Zhu et al. 2018) . Moreover, to ensure the comparability of the models, only pollutants with an |r| of <0.7 at each of the four lag times were entered in the multipollutant models. For example, if the |r| between SO 2 and PM 10 was <0.7 at all four lag times, while the |r| between NO 2 and PM 10 were <0.7 at only three pof the lag times, then SO 2 was included in the multipollutant models, while NO 2 was not. Additionally, we performed subgroup analyses to explore whether the effects of PM exposure on the risk of severe COVID-19 were modified by sex or age. The difference in effects between subgroups was examined by the following formula: where 1 and 2 are the estimated effects, and 1 and 2 are the standard errors of the estimates. The difference was considered to be statistically significant if the value was >1.96 (Zheng et al. 2020 ). Moreover, we plotted the exposure-response curve between average wind speed and the risk of severe COVID-19 based on the single-pollutant models of PM 10 and PM 2.5 . If the curve was linear or approximately linear, then we used a linear function to estimate the effects of wind speed on severe COVID-19. Otherwise, we used a piecewise linear function to assess the effects of wind speed on severe COVID-19. All analyses were performed using R software version 4.0.4 (https://www.r-project.org/). The significance level for testing was 0.05. (Table 2) . In the single-pollutant models, PM 10 was positively associated with the risk of severe COVID-19 at lags of 0-7 days, 0-14 days, 0-21 days, and 0-28 days. The maximum effect was at lag 0-14 days. For a unit increase in PM 10 , the risk of severe COVID-19 increased by 86.04% (95% CI: 38.71, 149.53). In the multipollutant models, the associations remained significant at lags of 0-7 days, 0-14 days, and 0-28 days (Table 3) . In the single-pollutant models, PM 2.5 was positively associated with the risk of severe COVID-19 at lags of 0-7 days, 0-14 days, 0-21 days, and 0-28 days. The maximum effect was at a lag of 0-7 days. For a unit increase in PM 2.5 , the risk of severe COVID-19 increased by 299.08% (95% CI: 92.94, 725.46). In the multipollutant models, the associations remained significant at lags of 0-7 days, 0-14 days, and 0-28 days (Table 3) . The associations between PM 10 and the risk of severe COVID-19 remained significant in different sex or age groups at lags of 0-14 days and 0-21 days. For a unit increase in PM 10 at lag 0-14 days, the risk of severe COVID-19 increased by 108.55% (95% As shown in Figure 2 , the exposure-response curves between average wind speed and the risk of severe COVID-19 were approximately linear at different lag times. Based on the single-pollutant models of PM 10 , the wind speed was negatively associated with severe COVID-19 at all lag times but was only significant at a lag of 0-7 days. For a unit increase in wind speed at a lag of 0-7 days, the risk of severe COVID-19 decreased by 62.44% (95% CI: -85.09, -5.37). Based on the single-pollutant models of PM 2.5 , the wind speed was negatively associated with severe COVID-19 at all lag times and was significant at lags of 0-7 days, 0-21 days, and 0-28 days. For a unit increase in wind speed at lags of 0-7 days, 0-21 days, and 0-28 days, the risk of severe In this study, we conducted a time-series analysis of 476 patients with COVID-19 caused by the Delta SARS-CoV-2 variant to explore the effects of short-term PM exposure on the risk of severe COVID-19. We observed that short-term exposure to PM was positively associated with the risk of severe COVID-19. To our knowledge, this is the first individual-level study to evaluate the relationship between short-term PM exposure and the risk of severe COVID-19 in China. A multicenter study in 33 European countries found that PM 2.5 was positively related to the number of COVID-19 deaths (Lembo et al. 2021 hospitalization rates in Cincinnati and found that a unit increase in 10-year average PM 2.5 concentration was correlated with an 18% higher hospitalization rate (Mendy et al. 2021) . Although most of the previous studies were epidemiological and focused on COVID-19 mortality or hospitalization rates, their findings implied that PM exposure was related to the severity of COVID-19. Moreover, our results showed that the effect of PM 2.5 seemed to be stronger than that of PM 10 . This may be attributed to the smaller particle size of PM 2.5 , which can penetrate more deeply into the alveoli and bronchioles and thus has more potent biological toxicity (Liu et al. 2019) . We also found an inverse relationship between average wind speed and severe COVID-19, although this association was only significant at some lag times. One possible explanation is that higher wind speeds can dilute the concentration of PM in the environment, thereby indirectly reducing the risk of severe COVID-19. The following reasons may explain the potential links between PM and severe COVID-19. First, PM suspended in the air, especially PM 2.5 , may not only carry SARS-CoV-2 but also enhance the attachment and replication of the virus in the bronchus by damaging the integrity of bronchial epithelial cells (Mendy et al. 2021) . Second, as pointed out by Domingo et al., SARS-CoV-2 attached to PM may survive longer and have a stronger effect on the immune system, which is triggered by exposure to high concentrations of air pollutants (Domingo and Rovira 2020, Zhan et al. 2021) . Third, as mentioned before, PM 2.5 can reach the alveoli, thereby delivering SARS-CoV-2 to target type II alveolar cells (Mendy et al. 2021) . Previous studies have shown that PM, especially PM 2.5 , can stimulate activated alveolar macrophages and then induce proinflammatory cytokine production and release, thus triggering allergic inflammation in the lungs (Manivannan and Sundaresan 2021). Fourth, the metals and polycyclic aromatic hydrocarbons that make up PM 2.5 facilitate the production of free radicals, which may oxidize alveolar cells. Excessive free radicals weaken the cellular antioxidant capacity, leading to lipid peroxidation and increased intracellular calcium concentrations, further inducing cellular damage (Xing et al. 2016 ). Finally, SARS-CoV-2 enters the cell through binding to the angiotensin-converting enzyme 2 (ACE2) receptor, and this process can be enhanced by PM exposure (Tung et al. 2021) . The binding of SARS-CoV-2 and the ACE2 receptor resulted in the downregulation of the latter. ACE2 mediated the transformation of angiotensin II to angiotensin 1-7 through the G protein-coupled receptor pathway and worked with angiotensin 1-7 by anti-inflammatory and antioxidant activities to protect the body. Downregulation of ACE2 decreased its protective effect and lessened the effect of angiotensin II (Zhu et al. 2021 ). Frontera et al. also postulated that long-term exposure to PM 2.5 resulted in overexpression of alveolar ACE2 receptors. This may increase the viral load in PM-exposed patients, weakening the defenses of the host (Frontera et al. 2020) . Our study had several limitations. First, we estimated the PM exposure level of each patient based on the monitoring data from the fixed sites, which may not accurately reflect individual exposure. Second, other factors related to the severity of COVID-19, which were not considered in the analysis, may affect the results. Third, since the epidemic was under control for a short time, the sample size of this study was relatively small, especially in terms of the number of severe patients. The association between PM and severe COVID-19 needs to be further confirmed in future studies. Our results showed that short-term PM exposure was positively correlated with the risk of severe COVID-19. Curbing outdoor PM pollution will help decrease the burden of COVID-19 and improve patient prognosis. The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper. This study was funded by the National Natural Science Foundation of China All data generated or analyzed during this study are included in this published article. This study was approved by the Ethics Committee of Nanjing Public Health Medical Center. Not required. past heart disease, current or past carcinoma, current or past COPD, current or past asthma, current or past autoimmune disease, vaccination status, days between onset and hospitalization, and average temperature and average wind speed at the same lag time. b : Based on single-pollutant models, additionally adjusted for other air pollutants at the same lag time. Fine Particulate Air Pollution and Daily Mortality. A Nationwide Analysis in 272 Chinese Cities SARS-CoV-2 Spike Mutations, L452R, T478K, E484Q and P681R, in the Second Wave of COVID-19 in Maharashtra Associations between Source-Specific Particulate Matter and Respiratory Infections in New York State Adults Effects of air pollutants on the transmission and severity of respiratory viral infections Severe air pollution links to higher mortality in COVID-19 patients: The "double-hit" hypothesis Effectiveness of inactivated COVID-19 vaccines against severe illness in B.1.617.2 (Delta) variant-infected patients in Jiangsu, China Effectiveness of inactivated COVID-19 vaccines against COVID-19 pneumonia and severe illness caused by the B. 1.617. 2 (Delta) variant: evidence from an outbreak in Guangdong The delayed effect of wildfire season particulate matter on subsequent influenza season in a mountain west region of the USA Air pollutants and SARS-CoV-2 in 33 European countries Risk factors for severity and mortality in adult COVID-19 inpatients in Wuhan Ambient Particulate Air Pollution and Daily Mortality in 652 Cities Systems level insights into the impact of airborne exposure on SARS-CoV-2 pathogenesis and COVID-19 outcome -A multi-omics big data study Short-Term Exposure to Fine Particulate Matter and Risk of Ischemic Stroke Effect of environmental pollutants PM-2.5, carbon monoxide, and ozone on the incidence and mortality of SARS-COV-2 infection in ten wildfire affected counties in California COVID-19 update: the first 6 months of the pandemic Regional and global contributions of air pollution to risk of death from COVID-19 Exposure to Particulate Matter Is Associated With Elevated Blood Pressure and Incident Hypertension in Urban India Long-term exposure to fine particulate matter air pollution: An ecological study of its effect on COVID-19 cases and fatality in Germany Particulate matter and SARS-CoV-2: A possible model of COVID-19 transmission Airborne particulate matter, population mobility and COVID-19: a multi-city study in China Inflammatory health effects of indoor and outdoor particulate matter Association between ambient temperature and COVID-19 infection in 122 cities from China The impact of PM2.5 on the human respiratory system Ambient air pollution exposures and risk of drug-resistant tuberculosis On the association between outdoor PM2.5 concentration and the seasonality of tuberculosis for Beijing Effect of short-term ambient PM2.5 exposure on fasting blood glucose levels: A longitudinal study among 47,471 people in eastern China Risk factors for disease severity, unimprovement, and mortality in COVID-19 patients in Wuhan, China Role of atmospheric particulate matter exposure in COVID-19 and other health risks in human: A review A Novel Coronavirus from Patients with Pneumonia in China Ambient air pollutants are associated with newly diagnosed tuberculosis: A time-series study in Chengdu Association between short-term exposure to air 26 There was no significant difference in the effects between the subgroups (P >0.05) The effects between the subgroups were significantly different (P <0.05)