Air pollution and disease progression among patients with primary glomerular disease: an expanded study with extended follow-up Authors: Jonathan P. Troost, Abhijit V. Kshirsagar, William E. Smoyer, Jon Klein, Michael L. Merchant, Margaret Helmuth, Laura H. Mariani, Matthias Kretzler, Abigail R. Smith, Forrest Lacey, Lawrence S. Engel, Howard Trachtman, Cassandra R. O’Lenick Published: 2026 ABSTRACT BackgroundAir pollution is increasingly recognized as a risk factor for progression of kidney disease; however, few studies have examined its impact among patients with primary glomerular disorders. To address this knowledge gap, we previously reported positive associations between fine particulate matter ≤ 2.5μm in aerodynamic diameter (PM2.5) and black carbon with kidney disease progression in an observational cohort of children and adults (n = 925) with primary glomerular diseases, namely minimal change disease, focal segmental glomerulosclerosis, membranous nephropathy, and IgA nephropathy. In the current study, we leverage the same cohort to (1) identify additional air pollutants that may be associated with kidney disease progression using data from the National Center for Atmospheric Research (NCAR); (2) determine whether the association between baseline air pollution exposure and kidney disease progression is maintained over a longer follow-up period; and (3) assess whether associations identified in our previously published findings remain when using NCAR pollutant data.MethodsIn this retrospective cohort study, we obtained air pollutant concentration data from NCAR based on participant residential census tract at enrollment. For each census tract and pollutant, we aggregated daily pollutant concentrations to annual averages. We used Cox proportional hazards models to estimate associations between average baseline pollutant exposure and time to kidney disease progression, defined as a 40% decline in estimated glomerular filtration rate (eGFR) or occurrence of kidney failure (eGFR) < 15 ml/min/1.73 m2) during follow-up. Hazard ratios (HR) represented a doubling of exposure.ResultsUse of NCAR data supported our previous findings of adverse effects of PM2.5 and black carbon on the progression of glomerular disease. Moreover, using the NCAR data we identified novel associations with NO2 exposure, HR 1.12 [1.01, 1.26] and specific components of PM2.5, including organic matter, HR 1.17 [1.01, 1.37] and an increased risk of disease progression. When considering a longer follow-up period, associations between baseline exposures and kidney outcomes persisted, but were attenuated suggesting a need for recent, interval-specific risk assessment, inclusion of acute exposures, and enhanced spatial resolution of exposures.ConclusionOur findings highlight the importance of the systematic assessment including spatial and temporal variation of a broad range of air pollution components to determine the impact of exposure on short- and long-term outcomes in patients with primary glomerular disease (word count: 368).Supplementary InformationThe online version contains supplementary material available at 10.1186/s12940-026-01312-9. Background Air pollution, through induction of inflammation and oxidative stress, is a major contributor to the global burden of disease and mortality [1]. Most clinical research in this area has focused on the cardiopulmonary consequences of this environmental hazard [2, 3]. Over the last 10 years, there is increasing evidence that air pollution exposure is associated with an increased incidence and rate of progression of chronic kidney disease (CKD) in low-middle and high-income countries [4]. Importantly, the majority of published studies that include patients with CKD treat the condition as a single, homogenous disease entity. There are no reports that link the impact of air pollution on clinical outcomes in distinct, clearly specified subgroups based on underlying cause such as diabetes mellitus, hypertension, glomerular disease, or cystic kidney diseases [5–8]. Primary glomerular diseases, namely minimal change disease (MCD), focal segmental glomerulosclerosis (FSGS), membranous nephropathy (MN), and IgA nephropathy (IgAN), are characterized by immune-mediated injury to the glomerulus and are characterized clinically by the presence of hematuria, proteinuria, hypertension, and reduced kidney function [9]. These disorders may be particularly susceptible to the effects of air pollution given their links to inflammation, oxidative stress, and autoimmunity [10]. Emerging evidence, primarily from studies of residents in China and Taiwan, suggests a link between air pollution exposure and the incidence of two specific glomerular diseases, MN, and IgAN [11–13]. However, there are no published studies that assess the impact of air pollution on the clinical course of disease among patients with existing glomerular disorders. We recently investigated whether air pollution was associated with an increased risk of kidney disease progression in North America-based patients with primary glomerular disorders — MCD, FSGS, MN, and IgAN. We leveraged the detailed characterization of these patients who were enrolled in the Nephrotic Syndrome Study Network (NEPTUNE) and Cure Glomerulonephropathy (CureGN) observational cohort studies [14, 15], by linking them with the publicly available air pollution dataset developed and maintained at Washington University in St. Louis (WUSTL). We demonstrated that higher exposure to total fine particulate matter ≤ 2.5μm in aerodynamic diameter (PM2.5) and black carbon (a species of PM2.5) were associated with an increased risk of disease progression. Higher levels of exposure to sulfate and total PM2.5 were positively correlated with systemic inflammation based on circulating levels of TNF-α and Il-1β, and sulfate exposure was correlated with activation of two intra-renal signaling pathways, namely TNF and JAK-STAT [15]. Contemporary air pollution estimates generated by other research centers with a wider range of components and/or longer observation periods offer an opportunity to identify additional components of air pollution that may impact glomerular disease progression over a longer time period and lay the groundwork for evaluating the effects of acute exposures. The National Center for Atmospheric Research (NCAR) has recently released a publicly available dataset that includes hourly concentrations of a comprehensive set of air pollutants through 2024. Therefore, the objectives of this study were to: (1) Determine the association between air pollution exposure from the NCAR dataset containing sub-daily concentrations of several health relevant gaseous (e.g., ozone, nitrogen dioxide, sulfur dioxide) and aerosol pollutants (e.g., PM2.5 and specific components) [16] with kidney disease progression in the cohort of patients with primary glomerular disease; (2) Determine whether the previously observed associations between baseline air pollution exposure and disease progression persist over a longer follow-up period; and (3) Assess the comparability of the associations with glomerular disease progression identified based on the NCAR and WUSTL estimates of exposure. Methods Patients NEPTUNE and CureGN are two NIDDK-funded multicenter observational cohort studies that enroll incident and prevalent patients, respectively, with primary glomerular disease [17, 18]. A description of these two studies, details of the eligibility criteria for inclusion in this assessment of the impact of air pollution on disease progression, and the clinical and laboratory data retrieved through 2022, are summarized in our previous report [15]. Briefly, pediatric and adult patients with MCD (n = 235), FSGS (n = 203), MN (n = 163), IgAN (n = 202), or other diagnoses (n = 16) who were enrolled in NEPTUNE and CureGN, and for whom residential census tract data were available, were included in the present study. There were 106 non-biopsied participants, namely pediatric participants who were enrolled in NEPTUNE at their initial presentation with nephrotic syndrome with available residential census tract data, who were included. For this retrospective cohort study, clinical outcomes were updated through 2024 for NEPTUNE and March 2025 for CureGN study participants to assess whether the associations between baseline air pollution exposure and disease progression are sustained over a more extended follow-up period, i.e., a minimum of 2 more years in all participants. Air pollution exposure Our previous analysis, was based on annual average estimates, at 1 km x 1 km resolution, of total PM2.5, black carbon, and sulfate from WUSTL [15]. Overall, aerosols and aerosol components of PM2.5, black carbon, and sulfate (SO4) were estimated by both WUSTL and NCAR datasets. Gaseous pollutants including ozone daily maximum 8-h average (MDA8), nitrogen dioxide (NO2), and sulfur dioxide (SO2), and other speciated aerosol components are reported solely from the NCAR dataset. In the current NCAR-based analyses, pollutant concentrations were estimated using the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) and the Community Multiscale Air Quality model (WRF-CMAQ) at gridded 12 X 12 km horizontal spacing for the period of 2005 through 2020 [19]. Physical parameterizations of the atmosphere follow those being used in the near-real-time operational forecasting system run at NCAR [20]. Anthropogenic emissions of trace gases and aerosols are based on the available gridded EPA National Emissions Inventory (NEI), with sector and county-level emissions assumed to be conserved in years for which NEI emissions data are not available. Fire emissions are based on the Fire Inventory from NCAR version 2.5 [21] that provides daily varying biomass burning emissions at a resolution of 1 × 1 km2 using fire location information from the NASA MODIS and VIIRS satellites. Fire emissions are vertically distributed online within the model following Frietas et al. [22]. Natural emissions (vegetation and the Great Lakes) are calculated online within the model using high-resolution meteorological information and the Model of Emissions of Gases and Aerosol from Nature (MEGAN) Version 3 [23]. The gas-phase chemistry is represented using the MOZART-T1 scheme [24] that includes updated isoprene chemistry. Aerosol processes are represented using the Model for Simulating Aerosol Interactions and Chemistry [25], which is the most sophisticated aerosol module in WRF-Chem and WRF-CMAQ. Model biases are further reduced by assimilating the MODIS AOD retrievals corresponding to the Multi-Angle Implementation of Atmospheric Correction (MAIAC) algorithm that has been providing AOD retrievals at 1 km resolution and Measurements of Pollution in the Troposphere (MOPITT) CO column values. The assimilation of satellite AOD and CO retrievals has been shown to significantly improve air quality and weather forecasts in our recent studies [26]. The community Gridpoint Statistical Interpolation (GSI) system was used for assimilation. To spatially link the air pollution data to the NEPTUNE and CureGN cohort baseline residential locations, zonal statistics for pollutant concentrations were calculated using Python and TIGER shapefiles for cohort members’ residential census tracts at study enrollment. The air pollution data were temporally aggregated to census tract-specific average annual concentrations for each pollutant species evaluated, as with the prior study using the WUSTL dataset [15]. For total and speciated particulate matter, we evaluated annual averages of the daily 24-h mean concentrations. For ozone, we evaluated annual averages of the daily 8-h max concentration, and for NO2 and SO2 we evaluated annual averages of the daily 1-h max and 24-h mean concentrations. Statistical methods We used Cox proportional hazards models to examine the relationships, estimated as hazard ratios (HR) and 95% confidence intervals (CI), between each baseline pollutant exposure and time to the composite endpoint of kidney failure (eGFR < 15 ml/min/1.73 m2) or ≥ 40% reduction in eGFR from baseline and the duration of follow-up observation (Tables 2 and 3). The eGFR was calculated using the CKD-EPI formulas for adult and the modified Schwartz or CKiD U25 formulas for pediatric patients. Air pollution exposures were log-2 transformed and HR estimates are scaled to represent a doubling of exposure. We performed unadjusted models as well as two adjusted models, namely, adjusted-1: age, race (Black vs. non-Black), maternal education as a proxy for socioeconomic status (< college, some college, ≥ 4 year degree), and adjusted-2: age, race, maternal education, and baseline eGFR. Separate models were performed for each pollutant. Analyses were performed using SAS v9.4 (SAS Institute Inc. Cary, NC). Ethics approval The study was approved by the University of Michigan Institutional Review Board as a combined NEPTUNE-CureGN Ancillary Study. Because the analyses involved use of already collected clinical and laboratory data from patients that were deidentified and the use of publicly available air pollution estimates, informed consent was not required from individual study participants. Results The demographic and clinical characteristics and the key laboratory data at baseline in the combined cohort of 925 patients are summarized in Table 1. The median age was 21 years (48% were in the pediatric age range [≤ 18 years], 52% were adults), median eGFR was 88 ml/min/1.73m2, and median urine protein:creatinine ratio (UPCR) was 1.8. The patients were approximately evenly divided among the four different primary glomerular disease subgroups and the median follow-up interval was 49 months.Table 1Demographic, clinical and laboratory characteristics of the combined NEPTUNE-CureGN cohort at study enrollmentCharacteristicOverall(n = 925)Age at baseline (years), median (IQR)21.0 (9.0 to 48.0)Sex, n (%) Male516 (56) Female409 (44)Race, n (%) Asian/Asian American57 (6) Black/African American204 (22) Multi-Racial41 (4) Native Hawaiian/Other Pacific Islander6 (1) White/Caucasian584 (63) Unknown33 (4)Ethnicity, n (%) Hispanic or Latino157 (17) Not Hispanic or Latino762 (82) Unknown6 (1)Maternal education, n (%) Unknown92 (10) < High school204 (22) High school401 (43) ≥ 4-year college228 (25)Diagnosis, n (%) FSGS203 (22) IgA nephropathy202 (22) MCD235 (25) MN163 (18) Non-biopsied106 (11) Other16 (2)eGFR at baseline, median (IQR)87.9 (57.3 to 111.7)UPCR at baseline, median (IQR)1.8 (0.2 to 6.6)Follow-up time (months), median (IQR)49.0 (27.1 to 67.1)Abbreviations: eGFR estimated glomerular filtration rate, FSGS focal segmental glomerulosclerosis, MCD minimal change disease, MN membranous nephropathy, PM particulate matter, UPCR urine protein:creatinine ratio Table 1Demographic, clinical and laboratory characteristics of the combined NEPTUNE-CureGN cohort at study enrollmentCharacteristicOverall(n = 925)Age at baseline (years), median (IQR)21.0 (9.0 to 48.0)Sex, n (%) Male516 (56) Female409 (44)Race, n (%) Asian/Asian American57 (6) Black/African American204 (22) Multi-Racial41 (4) Native Hawaiian/Other Pacific Islander6 (1) White/Caucasian584 (63) Unknown33 (4)Ethnicity, n (%) Hispanic or Latino157 (17) Not Hispanic or Latino762 (82) Unknown6 (1)Maternal education, n (%) Unknown92 (10) < High school204 (22) High school401 (43) ≥ 4-year college228 (25)Diagnosis, n (%) FSGS203 (22) IgA nephropathy202 (22) MCD235 (25) MN163 (18) Non-biopsied106 (11) Other16 (2)eGFR at baseline, median (IQR)87.9 (57.3 to 111.7)UPCR at baseline, median (IQR)1.8 (0.2 to 6.6)Follow-up time (months), median (IQR)49.0 (27.1 to 67.1)Abbreviations: eGFR estimated glomerular filtration rate, FSGS focal segmental glomerulosclerosis, MCD minimal change disease, MN membranous nephropathy, PM particulate matter, UPCR urine protein:creatinine ratio Demographic, clinical and laboratory characteristics of the combined NEPTUNE-CureGN cohort at study enrollment Abbreviations: eGFR estimated glomerular filtration rate, FSGS focal segmental glomerulosclerosis, MCD minimal change disease, MN membranous nephropathy, PM particulate matter, UPCR urine protein:creatinine ratio We have previously reported on the interrelationships between the air pollution components measured in the WUSTL database [15]. Similarly, there was a high degree of correlation between the individual particulate air pollution components that were measured in the NCAR data set (Supplemental Table 1). Table 2 summarizes the HR for disease progression defined as a 40% decline in eGFR or occurrence of kidney failure (eGFR < 15 ml/min/1.73 m2), during follow-up through 2022, per two-fold difference at baseline concentration of each air pollutant, unadjusted, adjusted for demographic factors, and adjusted for demographic factors and baseline eGFR. During this period, 226 (24.4%) patients met the criteria for kidney disease progression [15]. The estimated effect size in the fully adjusted models for the air pollution components based on the NCAR estimates with kidney disease progression consistently demonstrated a HR > 1, although most confidence intervals included the null. Thus, the NCAR-based findings show an increased HR for kidney disease progression per unit increase in total PM2.5 (HR 1.28 [0.90–1.82]). The results also suggest that PM2.5 composed of elemental carbon (HR 1.16 [0.99–1.37]) and organic material (HR 1.43 [0.99–2.07]) and black carbon (HR 1.26 [0.99, 1.37]) may be contributors to the adverse effects of the particulate matter, with potential effects of other components of PM2.5 such as ammonium (NH4+) and nitrate (NO3−). Recognizing the inclusion of 1 in the confidence interval, these findings should be viewed as informative trends that require confirmation in additional independent cohorts with sufficient follow-up duration. In addition, we observed significant associations with nitrogen dioxide (NO2) daily average (HR 1.13 [1.01–1.27]) and 1-h daily maximum concentrations (HR 1.14 [1.01–1.28]), as well as suggestive associations for SO2.Table 2Hazard ratios for progression of kidney disease in association with NCAR-measured air pollution exposure based on renal function data through 2022UnadjustedAdjusted-1 (age, race, maternal education)Adjusted-2 (age, race, maternal education, and baseline eGFR)HRp-valueHRp-valueHRp-valuePM2.51.43 [1.03, 1.98]0.03061.28 [0.90, 1.82]0.16441.28 [0.90, 1.82]0.1634Black carbon1.17 [1.01, 1.37]0.04141.14 [0.97, 1.35]0.11221.16 [0.99, 1.37]0.0697SO4 (sulfate)1.23 [0.82, 1.83]0.31031.04 [0.68, 1.59]0.84490.96 [0.63, 1.48]0.8558Ozone MDA80.67 [0.16, 2.87]0.58731.05 [0.21, 5.16]0.95491.48 [0.28, 7.91]0.6498NO21.12 [1.01, 1.26]0.03861.12 [0.99, 1.26]0.0671.13 [1.01, 1.27]0.0376NO2 (1 h daily max)1.12 [1.00, 1.26]0.05431.12 [0.99, 1.27]0.08111.14 [1.01, 1.28]0.0381PM11.32 [1.02, 1.70]0.03241.23 [0.94, 1.62]0.13821.24 [0.94, 1.63]0.1211PM101.20 [0.87, 1.63]0.26331.07 [0.76, 1.51]0.69131.06 [0.75, 1.50]0.7303PM2.5 (ammonium)1.25 [1.00, 1.56]0.0481.27 [1.00, 1.60]0.04841.20 [0.95, 1.51]0.1298PM2.5 (nitrate)1.12 [0.97, 1.29]0.11481.14 [0.98, 1.33]0.09911.12 [0.97, 1.30]0.1308PM2.5 (organic matter)1.58 [1.12, 2.22]0.00881.37 [0.96, 1.97]0.08271.43 [0.99, 2.07]0.0558SO21.18 [1.03, 1.36]0.01951.14 [0.98, 1.32]0.09071.11 [0.96, 1.29]0.1729SO2 (1 h daily max)1.17 [1.02, 1.35]0.02571.13 [0.97, 1.31]0.10781.10 [0.95, 1.27]0.2162The hazard ratios were determined using the WRF-CMAQ reanalysis-based air quality metricsAbbreviations: eGFR estimated glomerular filtration rate, MDA8 daily maximum 8-h average, PM particulate matter Table 2Hazard ratios for progression of kidney disease in association with NCAR-measured air pollution exposure based on renal function data through 2022UnadjustedAdjusted-1 (age, race, maternal education)Adjusted-2 (age, race, maternal education, and baseline eGFR)HRp-valueHRp-valueHRp-valuePM2.51.43 [1.03, 1.98]0.03061.28 [0.90, 1.82]0.16441.28 [0.90, 1.82]0.1634Black carbon1.17 [1.01, 1.37]0.04141.14 [0.97, 1.35]0.11221.16 [0.99, 1.37]0.0697SO4 (sulfate)1.23 [0.82, 1.83]0.31031.04 [0.68, 1.59]0.84490.96 [0.63, 1.48]0.8558Ozone MDA80.67 [0.16, 2.87]0.58731.05 [0.21, 5.16]0.95491.48 [0.28, 7.91]0.6498NO21.12 [1.01, 1.26]0.03861.12 [0.99, 1.26]0.0671.13 [1.01, 1.27]0.0376NO2 (1 h daily max)1.12 [1.00, 1.26]0.05431.12 [0.99, 1.27]0.08111.14 [1.01, 1.28]0.0381PM11.32 [1.02, 1.70]0.03241.23 [0.94, 1.62]0.13821.24 [0.94, 1.63]0.1211PM101.20 [0.87, 1.63]0.26331.07 [0.76, 1.51]0.69131.06 [0.75, 1.50]0.7303PM2.5 (ammonium)1.25 [1.00, 1.56]0.0481.27 [1.00, 1.60]0.04841.20 [0.95, 1.51]0.1298PM2.5 (nitrate)1.12 [0.97, 1.29]0.11481.14 [0.98, 1.33]0.09911.12 [0.97, 1.30]0.1308PM2.5 (organic matter)1.58 [1.12, 2.22]0.00881.37 [0.96, 1.97]0.08271.43 [0.99, 2.07]0.0558SO21.18 [1.03, 1.36]0.01951.14 [0.98, 1.32]0.09071.11 [0.96, 1.29]0.1729SO2 (1 h daily max)1.17 [1.02, 1.35]0.02571.13 [0.97, 1.31]0.10781.10 [0.95, 1.27]0.2162The hazard ratios were determined using the WRF-CMAQ reanalysis-based air quality metricsAbbreviations: eGFR estimated glomerular filtration rate, MDA8 daily maximum 8-h average, PM particulate matter Hazard ratios for progression of kidney disease in association with NCAR-measured air pollution exposure based on renal function data through 2022 The hazard ratios were determined using the WRF-CMAQ reanalysis-based air quality metrics Abbreviations: eGFR estimated glomerular filtration rate, MDA8 daily maximum 8-h average, PM particulate matter Table 3 replicates the analyses detailed in Table 2 based on the NCAR estimates of air pollution exposure with follow-up extended through 2024 in the NEPTUNE participants and March 2025 in the CureGN participants. During this 2–3-year longer follow-up period, 275 (29.7%) patients developed a 40% decline in eGFR or occurrence of kidney failure (eGFR < 15 ml/min/1.73 m2), an increment of 49 events compared to our previous report [15]. While HRs in relation to particulate matter are all > 1 and the lower bound of the 95% confidence interval is generally > 0.90, the strength of the associations with disease progression was lessened when the impact of the baseline exposures were assessed over a longer period of observation. This trend was observed for all exposures that were associated with disease progression documented in our original report using the WUSTL exposure estimates and similar to the additional associations identified using the NCAR exposure estimates (Supplemental Table 2). To summarize, Tables 2 and 3 represent the associations between air pollution exposure measured by NCAR and disease progression during the initial and more prolonged follow-up periods, respectively. For comparison to the WUSTL dataset, the associations over the initial observation period are available in our previous report [15] and over the extended follow up period in Supplemental Table 2 in this report,Table 3Hazard ratios for progression of kidney disease associated with air pollution exposure using the NCAR air quality dataset based on renal function data through 2024 (NEPTUNE) or 2025 (CureGN)UnadjustedAdjusted-1 (age, race, maternal education)Adjusted-2 (age, race, maternal education, and baseline eGFR)HRp-valueHRp-valueHRp-valuePM2.51.30 [0.97, 1.75]0.07821.15 [0.83, 1.57]0.40451.14 [0.83, 1.57]0.4132Black carbon1.11 [0.96, 1.28]0.1441.07 [0.92, 1.25]0.36821.09 [0.94, 1.27]0.2567SO4 (sulfate)1.19 [0.83, 1.72]0.34591.03 [0.70, 1.51]0.88280.98 [0.67, 1.45]0.9378Ozone MDA80.82 [0.21, 3.17]0.77491.26 [0.29, 5.40]0.75841.18 [0.25, 5.53]0.8342NO21.08 [0.98, 1.20]0.12131.07 [0.96, 1.20]0.2131.08 [0.97, 1.21]0.146NO2 (1 h daily max)1.08 [0.97, 1.20]0.1591.07 [0.95, 1.20]0.24941.08 [0.97, 1.21]0.1605PM11.23 [0.97, 1.55]0.0821.13 [0.88, 1.44]0.35051.13 [0.88, 1.45]0.3207PM101.16 [0.87, 1.54]0.32311.03 [0.75, 1.40]0.87291.02 [0.75, 1.39]0.8968PM2.5 (ammonium)1.15 [0.95, 1.40]0.16091.17 [0.94, 1.44]0.15411.15 [0.93, 1.42]0.1966PM2.5 (nitrate)1.08 [0.95, 1.23]0.22921.09 [0.95, 1.25]0.2121.10 [0.96, 1.26]0.185PM2.5 (organic matter)1.43 [1.05, 1.93]0.02211.21 [0.88, 1.67]0.23721.21 [0.88, 1.67]0.2489SO21.12 [0.99, 1.27]0.07611.08 [0.95, 1.24]0.23511.07 [0.94, 1.23]0.3218SO2 (1 h daily max)1.11 [0.98, 1.26]0.10011.08 [0.94, 1.23]0.27271.06 [0.93, 1.22]0.3741The hazard ratios were determined using the WRF-CMAQ reanalysis-based air quality metricsAbbreviations: eGFR estimated glomerular filtration rate, MDA8 daily maximum 8-h average ozone concentration, PM particulate matter Table 3Hazard ratios for progression of kidney disease associated with air pollution exposure using the NCAR air quality dataset based on renal function data through 2024 (NEPTUNE) or 2025 (CureGN)UnadjustedAdjusted-1 (age, race, maternal education)Adjusted-2 (age, race, maternal education, and baseline eGFR)HRp-valueHRp-valueHRp-valuePM2.51.30 [0.97, 1.75]0.07821.15 [0.83, 1.57]0.40451.14 [0.83, 1.57]0.4132Black carbon1.11 [0.96, 1.28]0.1441.07 [0.92, 1.25]0.36821.09 [0.94, 1.27]0.2567SO4 (sulfate)1.19 [0.83, 1.72]0.34591.03 [0.70, 1.51]0.88280.98 [0.67, 1.45]0.9378Ozone MDA80.82 [0.21, 3.17]0.77491.26 [0.29, 5.40]0.75841.18 [0.25, 5.53]0.8342NO21.08 [0.98, 1.20]0.12131.07 [0.96, 1.20]0.2131.08 [0.97, 1.21]0.146NO2 (1 h daily max)1.08 [0.97, 1.20]0.1591.07 [0.95, 1.20]0.24941.08 [0.97, 1.21]0.1605PM11.23 [0.97, 1.55]0.0821.13 [0.88, 1.44]0.35051.13 [0.88, 1.45]0.3207PM101.16 [0.87, 1.54]0.32311.03 [0.75, 1.40]0.87291.02 [0.75, 1.39]0.8968PM2.5 (ammonium)1.15 [0.95, 1.40]0.16091.17 [0.94, 1.44]0.15411.15 [0.93, 1.42]0.1966PM2.5 (nitrate)1.08 [0.95, 1.23]0.22921.09 [0.95, 1.25]0.2121.10 [0.96, 1.26]0.185PM2.5 (organic matter)1.43 [1.05, 1.93]0.02211.21 [0.88, 1.67]0.23721.21 [0.88, 1.67]0.2489SO21.12 [0.99, 1.27]0.07611.08 [0.95, 1.24]0.23511.07 [0.94, 1.23]0.3218SO2 (1 h daily max)1.11 [0.98, 1.26]0.10011.08 [0.94, 1.23]0.27271.06 [0.93, 1.22]0.3741The hazard ratios were determined using the WRF-CMAQ reanalysis-based air quality metricsAbbreviations: eGFR estimated glomerular filtration rate, MDA8 daily maximum 8-h average ozone concentration, PM particulate matter Hazard ratios for progression of kidney disease associated with air pollution exposure using the NCAR air quality dataset based on renal function data through 2024 (NEPTUNE) or 2025 (CureGN) The hazard ratios were determined using the WRF-CMAQ reanalysis-based air quality metrics Abbreviations: eGFR estimated glomerular filtration rate, MDA8 daily maximum 8-h average ozone concentration, PM particulate matter Discussion The key findings of this report are that: (1) Higher levels of specific components of particulate matter, including elemental carbon and organic material, and the gases NO2 and SO2 are associated with an increased risk of disease progression among pediatric and adult patients with primary glomerular disorders. The PM2.5 effects are slightly stronger than those seen for SO2 and NO2 but similar to those observed for other forms of particulate matter such as particulate matter with ≤ 1.0μm in aerodynamic diameter (PM1). The adverse effects associated with PM2.5 exposure are consistent with our previous findings based on WUSTL air quality exposure estimates; and (2) The association of baseline air pollution exposures with kidney disease progression becomes attenuated with longer duration of follow-up. Overall, the findings are established in a population that combines pediatric and adult patients, an approach that is supported by the comparable disease course in children and adults with primary glomerular diseases [27, 28]. The NCAR database, with a wider array of PM2.5 components than available in the WUSTL database, extends our previous research by identifying the adverse effects of specific components of PM2.5, namely, elemental carbon and organic material particulate matter. In addition, our findings suggest an association of glomerular disease progression with exposure to NO2 and SO2, gaseous pollutants commonly associated with traffic and industrial sources. The concentrations of these gases can fluctuate and are dependent on the local dominant sources. These air pollutants, which are not included in the WUSTL data set, had similar effect sizes. Importantly, both of these pollutants have adverse acute effects on kidney function that may be mediated by increased oxidant stress and inflammation, mitochondrial dysfunction, and epigenetic changes [29–31]. These disease mechanisms are activated in all forms of chronic kidney disease including primary glomerulopathies and cause injury to glomeruli and tubules leading to glomerulosclerosis, interstitial fibrosis, and ultimately kidney failure [32]. It is important to note that these associations between exposure and kidney disease progression were documented in a modest-sized cohort that is considerably smaller than most studies that have investigated the impact of air pollution on CKD in general. It is important to note that there are reports that detail markedly contrasting findings. In a study of 30,154 Swedish adults with levels of air pollution comparable to the United States and with preserved kidney function, higher PM2.5 exposure was associated with a higher level of kidney function, albeit with an elevated serum concentration of matrix metalloproteinase-2 [33]. This highlights the importance of conducting future studies that incorporate measurements of NO2 and SO2 along with other components of air pollution in order to obtain a comprehensive assessment of the impact of air pollution on kidney disease progression in the overall population and in patients with primary glomerulopathies in particular. Our study noted some differences in effect size when comparing HRs based on the NCAR and WUSTL air pollutant estimates. These differences could be attributed to differences in spatial resolution of the air pollution estimates (NCAR data are estimated at 12 × 12 km versus 1 × 1 km with WUSTL), which generally has a greater impact on more spatially heterogenous pollutants like PM2.5. The importance of the degree of spatial resolution on the effect of air pollution exposure on health outcomes has been well documented on mortality attributable to PM2. levels in Finland in 2015 [34]. The impact of the larger measurement area in the NCAR database could misidentify the level of exposure for a given individual patient and bias the conclusion to the null hypothesis. Additionally, in contrast to the WUSTL data, the NCAR-based pollutant estimates are not influenced by sparse placement of Environmental Protection Agency Air Quality System sites because they do not rely on surface observations except for model validation. Thus, the NCAR method, which primarily uses satellite observations to estimate concentrations, likely offers improved air quality estimates in regions with sparse ground monitoring networks [19]. In contrast to the lower degree of spatial resolution compared to WUSTL measurements, NCAR datasets include hourly estimates of air pollution. This appreciably higher temporal resolution may improve the power to detect meaningful changes in exposures across populations. It can be aggregated to any exposure metric of interest across daily, weekly, monthly, and annual temporal scales allowing for comparability with other studies and exploration of acute and chronic health effects. Notably, this temporal granularity enables assessment of the impact of acute exposures on acute health outcomes, which can improve quantification of risk for many exposure-disease associations, but for which we lacked the necessary outcome data in the present study. Importantly, temporally high resolution air quality data may enable better assessment of the biological mechanisms through which the observed associations between air pollution exposure and glomerular disease progression may be occurring [35]. The increased occurrence of massive wildfires over the last 3 years represents a prominent example of a rapidly changing exposure that has been linked to short-term cardiovascular morbidity and mortality [36]. These catastrophic events highlight the heterogeneity of air pollution in space and in time and the need to accumulate granular data in both dimensions. Utilization of the temporally high-resolution NCAR exposure data would be of great value by enabling assessment of the impact of intermittent, severe environmental events like wildfires or seasonal variation in air pollution levels on kidney outcomes including disease flare, hospitalization, and mortality in patients with glomerular disease. Ultimately, we concur with the suggestion that analytic approaches that combine the WUSTL with its high level spatial resolution and NCAR that captures short-term temporal variations in air pollution may provide the best estimate of the impact of the environmental exposure on the clinical course in patients with glomerular disorders [37]. We observed an attenuation of the association between baseline exposures and disease progression with increasing follow-up time. A similar phenomenon has recently been reported in a case–control study from Taiwan in which the HR’s for 30% and 40% decline in eGFR were stronger based on average exposures over the 1–2 years versus 3–5 years prior to outcome occurrence [38]. There are several possible explanations of the weakening of the association between baseline levels of air pollution exposure and glomerular kidney disease progression. The biological changes induced by air pollution exposure assessed at a single time period may wane during extended follow-up. A single annual average estimate of exposure at a residential census tract at baseline fails to capture potentially important peaks in exposure that may be particularly detrimental. It also fails to account for any changes in ambient levels at a given location over time or for changes in ambient levels due to residential movement, which we were unable to assess in the present study. This suggests that there is a need for updated or time-integrated measures of air pollution exposure in order to determine the contribution of this environmental factor on the clinical course of glomerular disease more precisely. Use of the NCAR database, with its appreciably higher temporal resolution of air pollution estimates, would facilitate an assessment that accounts for temporal variations in exposures. We acknowledge several limitations in this analysis. The large area, 12 × 12 km, as a unit of exposure may have limited the accuracy of the associations. In addition, the impact of the exposures was assessed based on a single measurement that may not be representative of the chronic ambient levels of each component. The sample is also modest in size, and the patients enrolled in NEPTUNE and CureGN may not be typical for the specific glomerular disease entities. Finally, we were unable to fully correct for a number of potential confounders including occupation, level of physical activity, diet, and smoking status. In contrast, a key strength of the study is that the population was well characterized and likely broadly representative of patients with primary glomerular disease. In addition, the standardized follow-up protocol and centralized laboratory measurements lent credence to the impact of the exposures on disease progression. Conclusion In conclusion, we observed adverse effects of PM2.5 species elemental carbon and organic matter, as well as of the criteria pollutants NO2 and SO2, in relation to kidney function among children and adults with primary glomerular diseases, using air pollution estimates from the NCAR. These results confirm—and expand—our previous findings of associations between air pollution exposures and the risk of disease progression in patents with primary glomerular disorders based on air pollution estimates from WUSTL. These findings also underscore some relative advantages of the NCAR database to address the heterogeneity and rapid changes in exposures in space and time in assessing the impact of air pollution on the clinical course of kidney disease. There is a need to account for cumulative exposure and for peaks in exposure, especially those caused by extreme events like wildfires, to better quantify the risk of kidney disease progression associated with poor air quality. The NCAR database, with its high temporal resolution, would enable a more rigorous evaluation of the relationship between air pollution exposures, especially transient ones, and acute health outcomes. Taken together, our findings provide a strong rationale for a systematic assessment of the effects of air pollution exposure on short- and long-term outcomes in a large population of patients with primary glomerular disease (word count: 3544). Supplementary Information Supplementary Material 1 Supplementary Material 1