Abstract
Rationale
Particulate matter (PM) exposure exacerbates health outcomes by causing lung damage.
Objectives
To investigate whether prior exposure to particulate matter ⩽10 μm and ⩽2.5 μm in aerodynamic diameter (PM10 and PM2.5) was associated with clinical outcomes among patients with coronavirus disease (COVID-19).
Methods
Data from the nationwide registration database of the National Health Insurance and Korea Disease Control and Prevention Agency in South Korea were used. The study included adult patients who were admitted to monitoring centers or hospitals between October 8, 2020 and December 31, 2021, after COVID-19 confirmation. AirKOREA database, which compiles air pollutant data from 642 stations in 162 cities and counties across South Korea, was used to extract data on PM levels. Average values of monthly exposure to PM10 and PM2.5 from the year previous to hospital admission because of COVID-19 to the date of confirmation of COVID-19 were calculated and used to define PM exposures of patients with COVID-19.
Results
In total, 322,289 patients with COVID-19 were included, and 4,633 (1.4%) died during hospitalization. After adjusting for covariates, a 1-μg/m3 increase in PM10 and PM2.5 exposure was associated with 4% (odds ratio [OR], 1.04; 95% confidence interval [CI], 1.03–1.05; P < 0.001) and 6% (OR, 1.06; 95% CI, 1.04–1.07; P < 0.001) increase in the risk of in-hospital mortality, respectively. In addition, a 1-μg/m3 increase in PM10 and PM2.5 was associated with 5% (OR, 1.05; 95% CI, 1.04–1.07; P < 0.001) and 8% (OR, 1.08; 95% CI, 1.06–1.10; P < 0.001) increase in the risks of requiring intensive care unit (ICU) admission and mechanical ventilation, respectively.
Conclusions
PM10 and PM2.5 exposure was associated with increased in-hospital mortality and the need for ICU admission and mechanical ventilation among patients with COVID-19 in South Korea.
Keywords: COVID-19, lung damage, particulate matter
Coronavirus disease (COVID-19) was declared a pandemic by the World Health Organization in 2020 (1). As of May 7, 2023, more than 765 million confirmed cases of COVID-19 and more than 6.9 million related deaths globally have been reported (2). With the widespread availability of COVID-19 vaccines and increased immunity from previous infection, COVID-19 is transitioning from a pandemic to an endemic disease, resulting in a lower mortality rate (3–5). As a new pandemic due to other viruses may occur in the future, identifying factors that can affect these diseases and reduce the associated risk of mortality is important.
Particulate matter (PM) contains microscopic solids or liquid droplets that can be inhaled into the human body. PM10 and PM2.5 measure ⩽10 μm and ⩽2.5 μm in diameter, respectively (6). PM exposure has been recognized as an important health issue in environmental medicine, owing to its adverse effects on human health (7, 8). Global epidemiological studies have reported PM10 and PM2.5 exposure as a leading risk factor for mortality (9, 10). Given the potential of PM to cause or mediate lung damage through oxidative stress and inflammatory responses, there might be a relationship between PM exposure and COVID-19 outcomes (11–13). Long-term exposure to PM was associated with severe COVID-19 after infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in the United Kingdom and China (14, 15). Moreover, short- and long-term exposures to PM2.5 were associated with increased risk of acute respiratory distress syndrome in patients with sepsis (16). However, information regarding the association between PM exposure and clinical outcomes among hospitalized patients with COVID-19 is still lacking.
Therefore, we aimed to examine whether prior exposure to PM10 and PM2.5 is associated with clinical outcomes among patients with COVID-19. We hypothesized that PM exposure is associated with increased in-hospital mortality after hospital admission owing to COVID-19.
Methods
Ethical Statement
The study protocol was approved by the institutional review board of Seoul National University Bundang Hospital (number: X-2205-758-901), who waived the requirement for informed consent because anonymized data were extracted and analyzed retrospectively.
Data Source of Study Population: National Health Insurance Service–Korea Disease Control and Prevention Agency Database
This study was part of public–private joint research on COVID-19 involving the Korea Disease Control and Prevention Agency (KDCA) and the National Health Insurance Service (NHIS) (research number: KDCA-NHIS-2022-1-489). The databases of the KDCA and the NHIS were used. The KDCA initially provided the NHIS with data on patients diagnosed with COVID-19 through polymerase chain reaction (PCR) testing from October 8, 2020, to December 31, 2021. The data included age, sex, date of COVID-19 diagnosis through PCR testing, death date if applicable, vaccination date, and infection route. The infection route was categorized into six groups: 1) inflow from foreign countries; 2) contact with person–related inflow from foreign countries; 3) outbreak in hospitals or nursing care centers; 4) outbreak in local communities; 5) contact with a patient with confirmed COVID-19; and 6) unknown. Using data from the KDCA, the NHIS supplemented information regarding socioeconomic status; use of medical institutions for any disease; diagnoses of all diseases using International Classification of Diseases, 10 Revision codes; and prescribed drugs and/or procedures.
Study Population
We initially screened all patients who were admitted to monitoring centers or hospitals after COVID-19 confirmation through PCR testing from October 8, 2020 to December 31, 2021. To focus on the latest episode of hospital or monitoring center admission for each patient, cases with multiple (two or more) admissions were excluded, including transfers from one hospital to another. This exclusion criterion ensured that only patients admitted to a hospital or monitoring center with COVID-19 were enrolled into the study, and only the most severe cases were included for survival analyses. Patients aged <18 years with COVID-19 or whose COVID-19 transmission route was inflow from foreign countries were excluded, as postal codes of home addresses of the patients with COVID-19 in South Korea would be used to evaluate PM exposure. In addition, patients with uncertain places of residence, those who did not live in areas where PM was measured, or those who changed their place of residence during the measurement period were excluded.
In South Korea, patients with COVID-19 were classified based on their symptoms and disease severity, which were key to determining whether the patient would stay at home under self-monitoring or be sent to government-monitored centers or a ward in hospital. Patients with no symptoms or mild symptoms, young people, and those with no underlying diseases were mainly assigned to wait at home under self-monitoring (17). Patients waiting at home were admitted to the hospital if their COVID-19 symptoms worsened.
Measuring PM from the AirKOREA Database
In 2004, the Korean Ministry of the Environment established the National Ambient Air Information System to provide the public with daily provincial-, city-, county-, and district-specific air pollution information. In total, 642 stations in 162 cities and counties across South Korea measure the air pollutant levels four times per day at 05:00, 11:00, 17:00, and 23:00 and provide up-to-date information on the AirKOREA website (https://www.airkorea.or.kr/eng). Measuring stations were constructed while considering various factors, such as population, region, and population density. All the stations use the same measurement method for air pollutant level assessment. AirKOREA provides daily, monthly, and annual average air pollutant levels based on all the four daily measurements during the exposure period. Data on six air pollutants of interest (sulfur dioxide [SO2] [ppm], ozone [O3] [ppm], carbon monoxide [CO] [ppm], nitrogen dioxide [NO2] [ppm], PM10 [μg/m3], and PM2.5 [μg/m3]) were collected from the AirKOREA database.
PM Exposure before Admission Owing to COVID-19
To ensure that all patients had been exposed to air pollution, including PM, for at least 1 year, we used data on monthly exposure to PM in the year before their hospital admission for COVID-19. In addition, we used data on monthly PM exposure until hospital admission in the year of hospitalization for COVID-19. For example, if a patient was admitted to the hospital because of COVID-19 in March 10, 2021, prior PM exposure was calculated using 14 months of average monthly data from January 2020 to February 2021. To define PM exposure of patients admitted to hospitals or monitoring centers because of COVID-19, postal codes were used to match monitoring stations measuring air pollution with the registered home addresses of the patients.
Study Endpoints
The primary endpoint in this study was in-hospital mortality, defined as any death during hospitalization after admission to a hospital or monitoring center owing to COVID-19. The NHIS database contains all death dates regardless of hospice discharge, and the patients who died after hospice discharge were also captured as in-hospital mortality data. The secondary endpoint was intensive care unit (ICU) admission and mechanical ventilation, to reflect the aggravation of COVID-19 during hospitalization.
Collected Variables
Several covariates were collected for adjustment. Age and sex were collected as demographic variables. To reflect the socioeconomic status of patients with COVID-19, the national household income level and employment status at the time of COVID-19 diagnosis were collected. Employment status included both employed and self-employed individuals. In South Korea’s NHIS system, individuals who have difficulty paying insurance premiums and meet certain criteria qualify for the medical aid program, where most of their treatment expenses are covered by the NHIS. Patients who had benefited from the medical aid program were categorized into the medical aid program group. In addition, all other patients were classified into four groups using quartile ratios (Q1 [lowest], Q2, Q3, and Q4 [highest] group). Comorbidity information, such as the Charlson Comorbidity Index and underlying disability, was also collected. Charlson Comorbidity Index scores were calculated using International Classification of Diseases, 10th Revision codes for 17 individual underlying diseases (see Table E1 in the data supplement), which were recorded from 1 year before diagnosis of COVID-19 to the date of COVID-19 diagnosis. Patients were classified into three groups based on the type of hospital where they were admitted for COVID-19: tertiary general hospitals, general hospitals, and long-term facility care centers. Information on the first and second vaccinations before COVID-19 diagnosis was also collected, as prior vaccination is known to affect the prognosis of COVID-19 (18). The confirmation date of COVID-19 was classified into five groups: 1) October 2020 to December 2020; 2) January 2021 to March 2021; 3) April 2021 to June 2021; 4) July 2021 to September 2021; and 5) October 2021 to December 2021. This confirmation date of COVID-19 was adjusted in the statistical analysis, because it could affect the duration of PM exposure.
Treatment information for patients with COVID-19 after hospitalization, including receipt of noninvasive ventilation, high-flow nasal cannula, extracorporeal membrane oxygenation support, continuous renal replacement therapy, and experience with cardiopulmonary resuscitation, was also collected.
Statistical Methodology
Mean values with standard deviations (SDs) and numbers (percentage) are used to describe the characteristics of the continuous and categorical variables, respectively. A multivariable logistic regression model was constructed to assess in-hospital mortality among patients with COVID-19, and the results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). All covariates were included in the multivariable model for adjustment, except for treatment information after hospitalization, as the main independent variable (exposure to PM) corresponded to exposure during the period before hospitalization for COVID-19. In the model, PM2.5 was included separately from PM10 because of multicollinearity between the two variables. There were no multicollinearity issues between the other variables, as assessed using a variance inflation factor threshold <2.0. A separate multivariable model was fitted for the secondary endpoint, ICU admission and mechanical ventilation, among patients with COVID-19. Subgroup analyses were performed to examine whether the impact of PM exposure differed according to the type of patient with COVID-19, such as those >60 years of age. Moreover, subgroup analyses were performed according to vaccination and type of hospitals in which patients with COVID-19 were admitted. Hosmer–Lemeshow statistics were used to assess the goodness of fit in the multivariable models. All statistical analyses were performed using the IBM SPSS Statistics for Windows (version 25.0; IBM Corp.). Statistical significance was set at P < 0.05.
Results
Study Population
A total of 581,500 confirmed cases of COVID-19 were reported in South Korea between October 8, 2020, and December 31, 2021. Among them, 446,136 cases resulted in hospital or monitoring center admissions because of COVID-19. The following cases were excluded: 1) 65,111 cases with multiple admissions for a single patient; 2) 46,969 patients <18 years of age; 3) 5,306 patients with an infection route from foreign countries; and 4) 5,921 patients with uncertain place of residence or who did not reside where PM was measured or changed their place of residence during the measurement period. Finally, this study included 322,289 patients with COVID-19, of whom 4,633 (1.4%) died during hospitalization, as shown in Figure 1. Table 1 presents the demographic information of the study population. The mean age was 49.9 years (SD, 18.4 yr), and 163,793 (50.7%) were male. Among the patients, 2.3% (7,442/322,289) were admitted to the ICU, and 0.8% (2,600/322,289) received mechanical ventilatory support after ICU admission. Table E2 displays the exposure to PM and other air pollutants among patients with COVID-19 in this study. The mean exposure to PM10 and PM2.5 among patients with COVID-19 was 38.7 μg/m3 (SD, 3.9 μg/m3) and 21.2 μg/m3 (SD, 2.5 μg/m3), respectively. The range of PM10 was 22.6–49.3 μg/m3, and for PM2.5, it was 11.3–27.3 μg/m3.
Figure 1.
Flow chart depicting coronavirus disease (COVID-19) patient selection process. PM = particulate matter.
Table 1.
Demographic information of the study population
| Variable | Mean (SD) | n (%) |
|---|---|---|
| Age | 49.9 (18.4) | |
| Male sex | 163,793 (50.7) | |
| Having a job | 212,940 (66.0) | |
| Household income level | ||
| Medical aid group | 12,192 (3.8) | |
| Q1 (lowest) | 62,731 (19.4) | |
| Q2 | 71,554 (22.2) | |
| Q3 | 76,962 (23.8) | |
| Q4 (highest) | 94,545 (29.3) | |
| Unknown | 4,845 (1.5) | |
| Cause type | ||
| Outbreak in hospitals or nursing care centers | 19,295 (6.0) | |
| Outbreak in local communities | 68,624 (21.3) | |
| Contact with a patient with confirmed COVID-19 | 133,952 (41.5) | |
| Unknown | 100,958 (31.3) | |
| CCI, points | 25.0 (2.7) | |
| Underlying disability | ||
| Mild to moderate disability | 13,068 (4.0) | |
| Severe disability | 8,331 (2.6) | |
| LOS, d | 10.0 (4.4) | |
| Type of hospital | ||
| Tertiary general hospital | 33,928 (10.5) | |
| General hospital | 278,928 (86.4) | |
| Long-term facility care center | 9,973 (3.1) | |
| First vaccination | 282,569 (87.5) | |
| Second vaccination | 266,011 (82.4) | |
| ICU admission | 7,442 (2.3) | |
| Nasal or mask oxygen therapy | 31,683 (9.8) | |
| ICU admission and mechanical ventilation | 2,600 (0.8) | |
| NIV or HFNC | 6,071 (1.9) | |
| ECMO support | 283 (0.1) | |
| Experience of CPR | 720 (0.2) | |
| CRRT use | 136 (0.0) | |
| COVID-19 confirmed date | ||
| October 1–December 31, 2020 | 27,966 (8.7) | |
| January 1–March 31, 2021 | 30,869 (9.6) | |
| April 1–June 30, 2021 | 40,607 (12.6) | |
| July 1–September 30, 2021 | 105,265 (32.6) | |
| October 1–December 31, 2021 | 118,122 (36.6) | |
Definition of abbreviations: CCI = Charlson Comorbidity Index; COVID-19 = coronavirus disease; CPR = cardiopulmonary resuscitation; CRRT = continuous renal replacement therapy; ECMO = extracorporeal membrane oxygenation; HFNC = high-flow nasal cannula; ICU = intensive care unit; LOS = length of hospital stays; NIV = noninvasive ventilation; SD = standard deviation.
In-Hospital Mortality and Mechanical Ventilator Support after ICU Admission
Table 2 presents the results of the multivariable logistic regression model for in-hospital mortality among patients with COVID-19. After adjusting for covariates, including other ambient pollutants, a 1-μg/m3 increase in PM10 and PM2.5 exposures was associated with a 4% increased risk (OR, 1.04; 95% CI, 1.03–1.05; P < 0.001) and a 6% increased risk (OR, 1.06; 95% CI, 1.04–1.07; P < 0.001) of in-hospital mortality, respectively. Table 3 displays the results of the multivariable logistic regression model for ICU admission and mechanical ventilation among patients with COVID-19. After adjusting for covariates, including other ambient pollutants, a 1-μg/m3 increase in PM10 and PM2.5 was associated with 5% (OR, 1.05; 95% CI, 1.04–1.07; P < 0.001) and 8% (OR, 1.08; 95% CI, 1.06–1.10; P < 0.001) increased risks of requiring ICU admission and mechanical ventilation, respectively.
Table 2.
Multivariable logistic regression model for in-hospital mortality among patients with coronavirus disease
| Variable | OR (95% CI) | P Value |
|---|---|---|
| Exposure to air pollution | ||
| SO2, 0.001 | 1.07 (1.00–1.15) | 0.054 |
| NO2, 0.01 | 1.06 (0.94–1.19) | 0.361 |
| O3, 0.01 | 0.75 (0.50–1.09) | 0.135 |
| CO, 0.01 | 1.01 (1.00–1.01) | 0.339 |
| PM10 | 1.04 (1.03–1.05) | <0.001 |
| PM2.5 | 1.06 (1.04–1.07) | <0.001 |
| Age, yr | 1.11 (1.10–1.11) | <0.001 |
| Male sex | 1.79 (1.67–1.93) | <0.001 |
| Having a job | 0.95 (0.88–1.03) | 0.215 |
| Household income level | ||
| Medical aid group | 1.05 (0.92–1.21) | 0.466 |
| Q1 (lowest) | 1 | |
| Q2 | 0.90 (0.80–1.02) | 0.086 |
| Q3 | 1.06 (0.94–1.18) | 0.341 |
| Q4 (highest) | 0.97 (0.88–1.08) | 0.569 |
| Unknown | 0.74 (0.50–1.09) | 0.129 |
| Cause type | ||
| Outbreak in hospitals or nursing care centers | 1 | |
| Outbreak in local communities | 0.42 (0.37–0.49) | <0.001 |
| Contact with a patient with confirmed COVID-19 | 0.47 (0.42–0.52) | <0.001 |
| Unknown | 0.76 (0.69–0.84) | <0.001 |
| CCI, points | 1.11 (1.10–1.12) | <0.001 |
| Underlying disability | ||
| No disability | 1 | |
| Mild to moderate disability | 1.13 (1.02–1.26) | 0.017 |
| Severe disability | 1.69 (1.50–1.90) | <0.001 |
| Type of hospital | ||
| Tertiary general hospital | 1 | |
| General hospital | 0.24 (0.22–1.26) | <0.001 |
| Long-term facility care center | 0.20 (0.17–0.23) | <0.001 |
| First vaccination | 0.27 (0.23–0.31) | <0.001 |
| Second vaccination | 0.24 (0.21–0.28) | <0.001 |
| COVID-19 confirmed date | ||
| October 1–December 31, 2020 | 1 | |
| January 1–March 31, 2021 | 0.53 (0.46–0.62) | <0.001 |
| April 1–June 30, 2021 | 0.57 (0.48–0.68) | <0.001 |
| July 1–September 30, 2021 | 0.66 (0.57–0.76) | <0.001 |
| October 1–December 31, 2021 | 0.84 (0.75–0.93) | 0.001 |
Definition of abbreviations: CCI = Charlson Comorbidity Index; CI = confidence interval; CO = carbon monoxide; COVID-19 = coronavirus disease; NO2 = nitrogen dioxide; O3 = ozone; OR = odds ratio; PM = particulate matter; PM2.5 = PM ⩽ 2.5 μm in aerodynamic diameter; PM10 = PM ⩽10 μm in aerodynamic diameter; SO2 = sulfur dioxide.
Table 3.
Multivariable logistic regression model for intensive care unit admission and mechanical ventilation among patients with coronavirus disease
| Variable | OR (95% CI) | P Value |
|---|---|---|
| Exposure to air pollution | ||
| SO2, 0.001 | 1.06 (0.99–1.13) | 0.067 |
| NO2, 0.01 | 1.05 (0.90–1.15) | 0.250 |
| O3, 0.01 | 0.85 (0.65–1.11) | 0.352 |
| CO, 0.01 | 1.38 (0.39–4.83) | 0.614 |
| PM10 | 1.05 (1.04–1.07) | <0.001 |
| PM2.5 | 1.08 (1.06–1.10) | <0.001 |
| Age, yr | 1.07 (1.06–1.07) | <0.001 |
| Male sex | 1.75 (1.60–1.91) | <0.001 |
| Having a job | 0.96 (0.87–1.05) | 0.332 |
| Household income level | ||
| Medical aid group | 1.07 (0.90–1.28) | 0.432 |
| Q1 (lowest) | 1 | |
| Q2 | 1.06 (0.92–1.22) | 0.418 |
| Q3 | 1.07 (0.93–1.22) | 0.346 |
| Q4 (highest) | 0.97 (0.85–1.10) | 0.579 |
| Unknown | 0.90 (0.58–1.41) | 0.655 |
| Cause type | ||
| Outbreak in hospitals or nursing care centers | 1 | |
| Outbreak in local communities | 1.37 (1.14–1.65) | 0.001 |
| Contact with a patient with confirmed COVID-19 | 1.51 (1.30–1.76) | <0.001 |
| Unknown | 2.58 (2.23–2.99) | <0.001 |
| CCI, points | 1.13 (1.11–1.14) | <0.001 |
| Underlying disability | ||
| No disability | 1 | |
| Mild to moderate disability | 1.12 (0.98–1.28) | 0.110 |
| Severe disability | 1.68 (1.44–1.97) | <0.001 |
| Type of hospital | ||
| Tertiary general hospital | 1 | |
| General hospital | 0.15 (0.14–0.17) | <0.001 |
| Long-term facility care center | 0.01 (0.01–0.02) | <0.001 |
| First vaccination | 0.39 (0.33–0.47) | <0.001 |
| Second vaccination | 0.27 (0.22–0.32) | <0.001 |
| COVID-19 confirmed date | ||
| October 1–December 31, 2020 | 1 | |
| January 1–March 31, 2021 | 0.69 (0.57–0.83) | <0.001 |
| April 1–June 30, 2021 | 0.79 (0.65–0.96) | 0.016 |
| July 1–September 30, 2021 | 0.98 (0.83–0.15) | 0.791 |
| October 1–December 31, 2021 | 0.83 (0.72–0.95) | 0.007 |
Definition of abbreviations: CCI = Charlson Comorbidity Index; CI = confidence interval; CO = carbon monoxide; COVID-19 = coronavirus disease; NO2 = nitrogen dioxide; O3 = ozone; OR = odds ratio; PM = particulate matter; PM2.5 = PM < 2.5 mm in aerodynamic diameter; PM10 = PM < 10 mm in aerodynamic diameter; SO2 = sulfur dioxide.
Subgroup Analysis
Table 4 presents the results of subgroup analyses. A 1-μg/m3 increase in PM10 exposure was associated with an increased risk of in-hospital mortality in both the >60-year-old group (OR, 1.04; 95% CI, 1.03–1.05; P < 0.001) and the ⩽60-year-old group (OR, 1.03; 95% CI, 1.01–1.07; P = 0.044). Similarly, a 1-μg/m3 increase in PM2.5 exposures was associated with an increased risk of in-hospital mortality in both the >60-year-old group (OR, 1.06; 95% CI, 1.04–1.07; P < 0.001) and the ⩽60-year-old group (OR, 1.07; 95% CI, 1.02–1.12; P = 0.006). Moreover, a 1-μg/m3 increase in PM10 exposure was associated with an increased risk of in-hospital mortality in both the vaccination (OR, 1.04; 95% CI, 1.02–1.05; P < 0.001) and no-vaccination groups (OR, 1.04; 95% CI, 1.02–1.05; P < 0.001). Similarly, a 1-μg/m3 increase in PM2.5 exposure was associated with an increased risk of in-hospital mortality in both the vaccination (OR, 1.06; 95% CI, 1.04–1.08; P < 0.001) and no-vaccination groups (OR, 1.05; 95% CI, 1.04–1.07; P < 0.001).
Table 4.
Subgroup analyses according to age
| Variable | OR (95% CI) | P Value |
|---|---|---|
| Exposure to air pollution and in-hospital mortality | ||
| >60 yr old | ||
| PM10 | 1.04 (1.03–1.05) | <0.001 |
| PM2.5 | 1.06 (1.04–1.07) | <0.001 |
| ⩽60 yr old | ||
| PM10 | 1.03 (1.01–1.07) | 0.044 |
| PM2.5 | 1.07 (1.02–1.12) | 0.006 |
| Vaccination group | ||
| PM10 | 1.04 (1.02–1.05) | <0.001 |
| PM2.5 | 1.06 (1.04–1.08) | <0.001 |
| No-vaccination group | ||
| PM10 | 1.04 (1.02–1.05) | <0.001 |
| PM2.5 | 1.05 (1.04–1.07) | <0.001 |
| Type of hospital: tertiary general hospital group | ||
| PM10 | 1.03 (1.01–1.05) | <0.001 |
| PM2.5 | 1.04 (1.02–1.07) | 0.003 |
| Type of hospital: General hospital group | ||
| PM10 | 1.04 (1.03–1.05) | <0.001 |
| PM2.5 | 1.06 (1.04–1.07) | <0.001 |
| Type of hospital: long-term facility care center group | ||
| PM10 | 1.03 (1.00–1.06) | 0.025 |
| PM2.5 | 1.07 (1.02–1.11) | 0.002 |
| Exposure to air pollution and MV support in ICU | ||
| >60 yr old | ||
| PM10 | 1.05 (1.04–1.07) | <0.001 |
| PM2.5 | 1.07 (1.05–1.09) | <0.001 |
| ⩽60 yr old | ||
| PM10 | 1.06 (1.03–1.09) | <0.001 |
| PM2.5 | 1.09 (1.05–1.14) | <0.001 |
| Vaccination group | ||
| PM10 | 1.06 (1.04–1.08) | <0.001 |
| PM2.5 | 1.09 (1.06–1.11) | <0.001 |
| No vaccination group | ||
| PM10 | 1.05 (1.03–1.07) | <0.001 |
| PM2.5 | 1.07 (1.05–1.10) | <0.001 |
| Type of hospital: tertiary general hospital group | ||
| PM10 | 1.07 (1.05–1.09) | <0.001 |
| PM2.5 | 1.12 (1.09–1.15) | <0.001 |
| Type of hospital: general hospital group | ||
| PM10 | 1.04 (1.03–1.06) | <0.001 |
| PM2.5 | 1.05 (1.02–1.07) | <0.001 |
| Type of hospital: long-term facility care center group | ||
| PM10 | 0.99 (0.86–1.15) | 0.934 |
| PM2.5 | 1.04 (0.82–1.32) | 0.746 |
Definition of abbreviations: CI = confidence interval; ICU = intensive care unit; MV = mechanical ventilator; OR = odds ratio; PM = particulate matter; PM2.5 = PM ⩽ 2.5 μm in aerodynamic diameter; PM10 = PM ⩽ 10 μm in aerodynamic diameter.
A 1-μg/m3 increase in PM10 exposure was associated with an increased risk of in-hospital mortality in the tertiary general hospital (OR, 1.03; 95% CI, 1.01–1.05; P < 0.001), general hospital (OR, 1.04; 95% CI, 1.03–1.05; P < 0.001), and long-term facility care center groups (OR, 1.03; 95% CI, 1.00–1.06; P = 0.025). A 1-μg/m3 increase in PM2.5 exposure was associated with an increased risk of in-hospital mortality in the tertiary general hospital (OR, 1.04; 95% CI, 1.02–1.07; P < 0.001), general hospital (OR, 1.06; 95% CI, 1.04–1.07; P < 0.001), and long-term facility care center groups (OR, 1.07; 95% CI, 1.02–1.11; P = 0.002).
Discussion
This population-based cohort study in South Korea demonstrated that prior exposure to both PM10 and PM2.5 was a potential risk factor for severe COVID-19 and increased in-hospital mortality. Moreover, this association was observed in both older and younger age groups of patients with COVID-19. Our results suggest that individuals with COVID-19 who reside in areas with high PM levels are a high-risk population that requires more careful monitoring, intensive treatment, and care.
Our methodology for calculating the duration of PM exposure could be an important issue, because different patients can have quite different durations of PM exposure window, based on COVID-19 admission month (i.e., 12–13 mo for a January admission and 23–24 mo for a December admission). PM exposure is known to have an increasingly harmful effect on the human body with the duration of exposure (19), and previous studies used an exposure time of at least 1 year to identify the harmful effects of long-term PM exposure (20–22). A recent study used a relatively long exposure length of 19 years (2000–2018) to find the relationship between PM2.5 exposure and COVID-19 fatalities in California (23). Against this background, we expected that an exposure duration of at least 1 year would be ensured and that a longer exposure time may be reflected, depending on the patient in this study design. Then, the possible bias owing to varied lengths of the PM exposure window in each patient was statistically controlled by incorporating the confirmation date of COVID-19 as an important covariate. However, because our approach has not been widely used, our results should be carefully interpreted considering this issue.
In this study, the rates of in-hospital mortality, ICU utilization, and mechanical ventilatory support after ICU admission were 1.4%, 2.3%, and 0.8%, respectively. The overall mortality rate among all patients with COVID-19 in South Korea was reported as 0.13%, which was the lowest among the rates of 30 countries with the highest case counts (24). High vaccine coverage rate, an effective healthcare system, and active cooperation between private sectors and the central government seem to have contributed to the low mortality rate (24).
The effect of air pollution on COVID-19 severity and mortality has been an important issue. A recent systematic review and meta-analysis analyzing data from Europe and North America (25) provided evidence of a relationship between long-term exposure to PM10 or PM2.5 and the severity of COVID-19, and reducing air pollution may alleviate COVID-19–related deaths. Another cohort study conducted in China reported an association between increased exposure to PM pollution and COVID-19 case fatality rate (26).
Several mechanisms have been identified regarding lung damage caused by PM exposure. First, exposure to PM leads to the production of free radicals, which are associated with oxidative stress and lung toxicity (27). Second, PM exposure triggers the activation of the immune system, resulting in increased systemic inflammation (28). Third, PM exposure may affect immunity by modulating the antiviral response of patients with respiratory tract (29) viral infections. In summary, exposure to PM can alter the immune response in lung cells, leading to oxidative and inflammatory stress that can result in acute lung injury in patients with viral respiratory infections. A previous study demonstrated that exposure to PM alters the inflammatory response to respiratory infections caused by the respiratory syncytial virus (30).
The relationships among PM, angiotensin-converting enzyme 2, and COVID-19 should also be considered. SARS-CoV-2, the virus causing COVID-19, possesses spike proteins that allow it to bind to angiotensin-converting enzyme receptors on cells in the lungs, blood vessels, heart, kidneys, and intestine. The role of angiotensin-converting enzyme 2 has been emphasized during the COVID-19 pandemic because of its antiinflammatory function (31). Angiotensin-converting enzyme 2 has been found to reduce severe lung damage caused by hyperoxia by regulating severe inflammatory (31) reactions and oxidative stress. Angiotensin-converting enzyme 2 can inhibit the intracellular signaling pathway that activates the inflammatory response and nuclear factor erythroid 2–related factor 2, which in turn activates the antiinflammatory response as a protective mechanism against oxidative stress (32). Lin and colleagues reported that exposure to PM2.5 can induce acute lung injury with increased inflammation (33), and this effect was more pronounced in angiotensin-converting enzyme 2 knockout mice. This suggests that angiotensin-converting enzyme 2 may play a role in defense against the proinflammatory actions of PM2.5. Angiotensin-converting enzyme 2 is crucial, as it activates the antiinflammatory pathway and suppresses the inflammatory pathway to prevent a dysregulated inflammatory response. However, long-term exposure to PM can disrupt the protective mechanism of angiotensin-converting enzyme 2 (33), potentially leading to the severity of COVID-19.
The results of subgroup analyses are important because they suggest that exposure to PM is associated with increased in-hospital mortality and severity of COVID-19 regardless of age. Furthermore, the ORs for exposure to PM10 or PM2.5 in relation to in-hospital mortality were higher in the ⩽60-yearold group than in the >60-year-old group. A previous study conducted in the United States reported a significant association between respiratory emergency department visits and PM exposure in adults younger than 65 years, but not in adults aged 65 years and older (34). As age can influence the morphology of the respiratory tract, immune response, and mechanical properties of the lungs (35), the effect of PM exposure on the lungs or immune system may differ according to age. Although the current evidence is limited, our results suggest that younger age groups may be more vulnerable to the effects of PM exposure on the prognosis of COVID-19.
Limitations
This study has several limitations. First, we did not accurately measure the exposure time of all patients with COVID-19 because we relied on the postal codes of their residential addresses. For example, some patients may have lived and worked in different cities, which could have influenced our results. Second, important information, such as smoking status, alcohol consumption history, and body mass index at the time of COVID-19 diagnosis, was not included for adjustment because the NHIS database did not contain this information. Third, as we did not include patients with COVID-19 who were staying at home under self-monitoring, the impact of PM exposure on hospital admission for this population could not be determined. Fourth, the work location was not considered in this study because of a lack of data. PM exposure could be influenced by the proportion of the day spent at home versus at work in commercial locations during the study period. Last, during the COVID-19 pandemic, there may have been challenges in the allocation of hospitals, ICU beds, and medical staff, and not all patients with COVID-19 may have received the same level of treatment, which could have influenced the results of this study.
Conclusions
Exposure to PM10 or PM2.5 was associated with increased in-hospital mortality and the need for ICU admission and mechanical ventilation among patients with COVID-19 in South Korea. Our results suggest that PM exposure may be a risk factor for higher COVID-19 mortality and morbidity.
Footnotes
Author Contributions: T.K.O. designed the study, analyzed the data, interpreted the data, and drafted the manuscript; S.K. and D.H.K. collected the data; I.-A.S. contributed to the study conceptualization, acquisition of data, and review of the manuscript. All authors approved the final version of the manuscript.
A data supplement for this article is available via the Supplements tab at the top of the online article.
Author disclosures are available with the text of this article at www.atsjournals.org.
References
- 1. Gulati A, Pomeranz C, Qamar Z, Thomas S, Frisch D, George G, et al. A comprehensive review of manifestations of novel coronaviruses in the context of deadly COVID-19 global pandemic. Am J Med Sci . 2020;360:5–34. doi: 10.1016/j.amjms.2020.05.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.World Health Organization. 2023. https://iris.who.int/bitstream/handle/10665/366534/nCoV-weekly-sitrep16Mar23-eng.pdf?sequence=1
- 3. Mahajan R, Lim WM, Kumar S, Sareen M. COVID-19 and management education: from pandemic to endemic. Int J Manag Educ . 2023;21:100801. [Google Scholar]
- 4. Wong MK, Brooks DJ, Ikejezie J, Gacic-Dobo M, Dumolard L, Nedelec Y, et al. COVID-19 mortality and progress toward vaccinating older adults: World Health Organization, Worldwide, 2020–2022. MMWR Morb Mortal Wkly Rep . 2023;72:113–118. doi: 10.15585/mmwr.mm7205a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. de Gier B, Huiberts AJ, Hoeve CE, den Hartog G, van Werkhoven H, van Binnendijk R, et al. Effects of COVID-19 vaccination and previous infection on Omicron SARS-CoV-2 infection and relation with serology. Nat Commun . 2023;14:4793. doi: 10.1038/s41467-023-40195-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Nezis I, Biskos G, Eleftheriadis K, Kalantzi O-I. Particulate matter and health effects in offices: a review. Build Environ . 2019;156:62–73. [Google Scholar]
- 7. Dominici F, Greenstone M, Sunstein CR. Science and regulation: particulate matter matters. Science . 2014;344:257–259. doi: 10.1126/science.1247348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Harrison RM, Yin J. Particulate matter in the atmosphere: which particle properties are important for its effects on health? Sci Total Environ . 2000;249:85–101. doi: 10.1016/s0048-9697(99)00513-6. [DOI] [PubMed] [Google Scholar]
- 9. Hammer MS, van Donkelaar A, Li C, Lyapustin A, Sayer AM, Hsu NC, et al. Global estimates and long-term trends of fine particulate matter concentrations (1998–2018) Environ Sci Technol . 2020;54:7879–7890. doi: 10.1021/acs.est.0c01764. [DOI] [PubMed] [Google Scholar]
- 10. Karagulian F, Belis CA, Dora CFC, Prüss-Ustün AM, Bonjour S, Adair-Rohani H, et al. Contributions to cities’ ambient particulate matter (PM): a systematic review of local source contributions at global level. Atmos Environ . 2015;120:475–483. [Google Scholar]
- 11. Comunian S, Dongo D, Milani C, Palestini P. Air pollution and Covid-19: the role of particulate matter in the spread and increase of Covid-19’s morbidity and mortality. Int J Environ Res Public Health . 2020;17:4487. doi: 10.3390/ijerph17124487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Pardo M, Porat Z, Rudich A, Schauer JJ, Rudich Y. Repeated exposures to roadside particulate matter extracts suppresses pulmonary defense mechanisms, resulting in lipid and protein oxidative damage. Environ Pollut . 2016;210:227–237. doi: 10.1016/j.envpol.2015.12.009. [DOI] [PubMed] [Google Scholar]
- 13. Li Y, Dong T, Jiang X, Wang C, Zhang Y, Li Y, et al. Chronic and low-level particulate matter exposure can sustainably mediate lung damage and alter CD4 T cells during acute lung injury. Mol Immunol . 2019;112:51–58. doi: 10.1016/j.molimm.2019.04.033. [DOI] [PubMed] [Google Scholar]
- 14. Hyman S, Zhang J, Andersen ZJ, Cruickshank S, Møller P, Daras K, et al. Long-term exposure to air pollution and COVID-19 severity: a cohort study in Greater Manchester, United Kingdom. Environ Pollut . 2023;327:121594. doi: 10.1016/j.envpol.2023.121594. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Wang B, Liu J, Li Y, Fu S, Xu X, Li L, et al. Airborne particulate matter, population mobility and COVID-19: a multi-city study in China. BMC Public Health . 2020;20:1585. doi: 10.1186/s12889-020-09669-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Reilly JP, Zhao Z, Shashaty MGS, Koyama T, Jones TK, Anderson BJ, et al. Exposure to ambient air pollutants and acute respiratory distress syndrome risk in sepsis. Intensive Care Med . 2023;49:957–965. doi: 10.1007/s00134-023-07148-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Kang J, Jang YY, Kim J, Han S-H, Lee KR, Kim M, et al. South Korea’s responses to stop the COVID-19 pandemic. Am J Infect Control . 2020;48:1080–1086. doi: 10.1016/j.ajic.2020.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Abu-Raddad LJ, Chemaitelly H, Butt AA, National Study Group for COVID-19 Vaccination Effectiveness of the BNT162b2 Covid-19 vaccine against the B.1.1.7 and B.1.351 variants. N Engl J Med . 2021;385:187–189. doi: 10.1056/NEJMc2104974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Schwartz J. Long-term effects of exposure to particulate air pollution. Clin Occup Environ Med . 2006;5:837–848. doi: 10.1016/j.coem.2006.07.008. [DOI] [PubMed] [Google Scholar]
- 20. Alexeeff SE, Deosaransingh K, Van Den Eeden S, Schwartz J, Liao NS, Sidney S. Association of long-term exposure to particulate air pollution with cardiovascular events in California. JAMA Netw Open . 2023;6:e230561. doi: 10.1001/jamanetworkopen.2023.0561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Alexeeff SE, Liao NS, Liu X, Van Den Eeden SK, Sidney S. Long-term PM2.5 exposure and risks of ischemic heart disease and stroke events: review and meta-analysis. J Am Heart Assoc . 2021;10:e016890. doi: 10.1161/JAHA.120.016890. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Hoek G, Krishnan RM, Beelen R, Peters A, Ostro B, Brunekreef B, et al. Long-term air pollution exposure and cardio-respiratory mortality: a review. Environ Health . 2013;12:43. doi: 10.1186/1476-069X-12-43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. English PB, Von Behren J, Balmes JR, Boscardin J, Carpenter C, Goldberg DE, et al. Association between long-term exposure to particulate air pollution with SARS-CoV-2 infections and COVID-19 deaths in California, U.S.A. Environ Adv . 2022;9:100270. doi: 10.1016/j.envadv.2022.100270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Lim S, Sohn M. How to cope with emerging viral diseases: lessons from South Korea’s strategy for COVID-19, and collateral damage to cardiometabolic health. Lancet Reg Health West Pac . 2023;30:100581. doi: 10.1016/j.lanwpc.2022.100581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Hernandez Carballo I, Bakola M, Stuckler D. The impact of air pollution on COVID-19 incidence, severity, and mortality: a systematic review of studies in Europe and North America. Environ Res . 2022;215:114155. doi: 10.1016/j.envres.2022.114155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Yao Y, Pan J, Wang W, Liu Z, Kan H, Qiu Y, et al. Association of particulate matter pollution and case fatality rate of COVID-19 in 49 Chinese cities. Sci Total Environ . 2020;741:140396. doi: 10.1016/j.scitotenv.2020.140396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Mantecca P, Sancini G, Moschini E, Farina F, Gualtieri M, Rohr A, et al. Lung toxicity induced by intratracheal instillation of size-fractionated tire particles. Toxicol Lett . 2009;189:206–214. doi: 10.1016/j.toxlet.2009.05.023. [DOI] [PubMed] [Google Scholar]
- 28. Bengalli R, Zerboni A, Marchetti S, Longhin E, Priola M, Camatini M, et al. In vitro pulmonary and vascular effects induced by different diesel exhaust particles. Toxicol Lett . 2019;306:13–24. doi: 10.1016/j.toxlet.2019.01.017. [DOI] [PubMed] [Google Scholar]
- 29. Farina F, Sancini G, Longhin E, Mantecca P, Camatini M, Palestini P. Milan PM1 induces adverse effects on mice lungs and cardiovascular system. BioMed Res Int . 2013;2013:583513. doi: 10.1155/2013/583513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Becker S, Soukup JM. Exposure to urban air particulates alters the macrophage-mediated inflammatory response to respiratory viral infection. J Toxicol Environ Health A . 1999;57:445–457. doi: 10.1080/009841099157539. [DOI] [PubMed] [Google Scholar]
- 31. Fang Y, Gao F, Liu Z. Angiotensin-converting enzyme 2 attenuates inflammatory response and oxidative stress in hyperoxic lung injury by regulating NF-κB and Nrf2 pathways. QJM . 2019;112:914–924. doi: 10.1093/qjmed/hcz206. [DOI] [PubMed] [Google Scholar]
- 32. Rodrigues Prestes TR, Rocha NP, Miranda AS, Teixeira AL, Simoes-E-Silva AC. The anti-inflammatory potential of ACE2/Angiotensin-(1-7)/Mas receptor axis: evidence from basic and clinical research. Curr Drug Targets . 2017;18:1301–1313. doi: 10.2174/1389450117666160727142401. [DOI] [PubMed] [Google Scholar]
- 33. Lin CI, Tsai CH, Sun YL, Hsieh WY, Lin YC, Chen CY, et al. Instillation of particulate matter 2.5 induced acute lung injury and attenuated the injury recovery in ACE2 knockout mice. Int J Biol Sci . 2018;14:253–265. doi: 10.7150/ijbs.23489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Strosnider HM, Chang HH, Darrow LA, Liu Y, Vaidyanathan A, Strickland MJ. Age-specific associations of ozone and fine particulate matter with respiratory emergency department visits in the United States. Am J Respir Crit Care Med . 2019;199:882–890. doi: 10.1164/rccm.201806-1147OC. [DOI] [PubMed] [Google Scholar]
- 35. Postma DS, Reddel HK, ten Hacken NH, van den Berge M. Asthma and chronic obstructive pulmonary disease: similarities and differences. Clin Chest Med . 2014;35:143–156. doi: 10.1016/j.ccm.2013.09.010. [DOI] [PubMed] [Google Scholar]