https://orcid.org/0009-0006-0273-8879
Figures
Abstract
Objective
To investigate the association between long-term exposure to air pollution and the risk of microvascular complications in individuals with diabetes, and provide the evidence to inform environmental strategies for prevention.
Methods
The data were collected from UK Biobank cohort (n = 9,671). Kaplan-Meier curves and log-rank tests were used to compare cumulative incidence across exposure groups. Cox proportional hazards models were employed to evaluate the associations between air pollutant exposure and the development of microvascular complications, with adjusting for demographic, clinical, and biochemical covariates. Stratified and sensitivity analyses were conducted to assess the robustness of the findings. Additionally, restricted cubic splines were used to explore potential nonlinear relationships between pollutant concentrations and complication risk.
Results
With a median follow-up time of 12.4 years, 2,104 participants (21.8%) developed microvascular complications. The Kaplan-Meier analysis demonstrated early divergence in cumulative incidence curves, suggesting an early impact of air pollution exposure on vascular outcomes. After multivariable adjustment, higher levels of air pollution exposure were significantly associated with increased risk of microvascular complications in patients with diabetes. Compared to the lowest quartile, participants in the highest quartile of exposure had elevated risks (NO2: HR = 1.27, 95% CI: 1.12–1.45; PM10: HR = 1.44, 95% CI: 1.27–1.64). These associations remained consistent across stratified and sensitivity analyses.
Conclusion
Long-term exposure to air pollutants, particularly NO2 and PM10, is associated with an increased risk of microvascular complications among individuals with diabetes. The observed risk appears to be persistent and may begin at relatively low exposure levels, underscoring the need for preventive strategies targeting environmental risk factors.
Citation: Nong Y, Huang H, Xu X, Tan X, Xu S, Wang C, et al. (2026) Association of air pollution with diabetic microvascular complications: A prospective cohort study. PLoS One 21(8): e0354711. https://doi.org/10.1371/journal.pone.0354711
Editor: Mahendra Prasad Bhatt, Nepalese Army Institute of Health Sciences, NEPAL
Received: May 21, 2026; Accepted: July 10, 2026; Published: August 19, 2026
Copyright: © 2026 Nong et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data used in this study were obtained from the UK Biobank under approved application 68136. In accordance with UK Biobank’s Material Transfer Agreement and data access policies, we cannot share participant-level data directly. Access is managed solely by UK Biobank via their official application process (https://www.ukbiobank.ac.uk/about-us/how-we-work/access-to-uk-biobank-data/; https://www.ukbiobank.ac.uk/use-our-data/apply-for-access/). Interested researchers may apply through the UK Biobank Access Management System or contact the Access Team at ukbiobank@ukbiobank.ac.uk. The authors had no special access privileges.
Funding: This work was supported by grants from National Natural Science Foundation of China (81873486, 82403624), Natural Science Foundation of Jiangsu Province (BK20240437), Science and Technology Development Program of Jiangsu Province-Clinical Frontier Technology (BE2022754), Clinical Medicine Expert Team (Class A) of Jinji Lake Health Talents Program of Suzhou Industrial Park (SZYQTD202102), Suzhou Key Discipline for Medicine (SZXK202129), Demonstration of Scientific and Technological Innovation Project (SKY2021002), Suzhou Dedicated Project on Diagnosis and Treatment Technology of Major Diseases (LCZX202132), Research on Collaborative Innovation of medical engineering combination (SZM2021014, SZM2022003), Suzhou Key Laboratory of Diagnosis and Treatment of Panvascular Diseases (SZS2023021), Suzhou Industrial Park Healthcare Talent Support Initiative ((2024)117), Suzhou Science and Education Youth Science and Technology Project (KJXW2023086), Gusu Talent Program (GSWS2022119), Postgraduate Research & Practice Innovation Program of Jiangsu Province (26CXJH5974). The funders had no roles in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Diabetes has emerged as one of the most pressing global public health challenges of the 21st century. Currently, more than 580 million individuals worldwide are living with diabetes, and approximately 20% of the type 2 diabetes burden can be attributed to air pollution exposure [1]. This striking figure underscores the critical role of environmental factors in the diabetes epidemic. Over the past two decades, the health threat posed by air pollution has intensified alongside rapid global industrialization, particularly in developing countries where accelerated urbanization and industrial growth have been accompanied by severe pollution problems [2].
The air pollutants, particularly fine particulate matter (PMX) and nitrogen oxides (NOX), have been widely documented. These pollutants have been associated with diabetes in previous studies, with proposed mechanisms including insulin resistance and impaired β-cell function [3]. In addition to directly increasing the risk of diabetes, air pollutants may also be associated with the progression of diabetic microvascular complications, a group of chronic disorders that includes diabetic retinopathy, nephropathy, and peripheral neuropathy. These complications are not only highly prevalent among people with diabetes but also represent leading causes of disability and mortality in this population [4].
Epidemiological data suggest that in patients with type 2 diabetes, the prevalence of nephropathy and retinopathy can reach 25%, while peripheral neuropathy may affect over 50% of individuals [5]. These complications significantly increase the risk of serious clinical outcomes such as coronary heart disease, stroke, renal failure, and blindness [6], thereby compromising quality of life and imposing a substantial burden on healthcare systems.
From a pathophysiological perspective, air pollution may aggravate microvascular damage in diabetes through multiple mechanisms. Fine particulate matter in ambient air can accumulate in the vascular endothelium, accelerating vascular aging and injury [7]. Chronic exposure has also been linked to endothelial dysfunction and systemic inflammation, which are key processes in atherosclerosis and acute coronary events [8].
Although the potential link between air pollution and diabetic microvascular complications has begun to attract attention, current evidence remains limited. Most existing studies have focused on short-term exposures and lack comprehensive evaluations of long-term cumulative effects. More importantly, prospective data from large-scale population cohorts are scarce. To address these gaps, we utilized study data from the UK Biobank, a large-scale prospective cohort of over 500,000 participants, to systematically investigate the association between long-term air pollution exposure and the risk of diabetic microvascular complications. Our findings aim to provide a scientific basis for individualized environmental interventions in diabetes care and offer critical parameters for estimating the disease burden attributable to air pollution.
2. Methods
2.1. Study design and population
This study is based on data from the UK Biobank, a large prospective longitudinal cohort that has collected biological and medical information from over 500,000 participants. The dataset includes demographic characteristics, laboratory biomarkers, physical measurements, lifestyle factors, self-reported questionnaires, and healthcare records. The UK Biobank was established to support the prevention, diagnosis, and treatment of a wide range of life-threatening and disabling diseases. All participants provided written informed consent, and the study was approved by the NHS North West Multi-centre Research Ethics Committee (Approval number: 16/NW/0274). The present analysis was conducted using data from participants recruited between 2006 and 2010 (Application ID: 68136).
2.2. Air pollution exposure
Air pollution data were derived from the European-wide regulatory air quality mapping network, with a spatial resolution of 100 × 100 meters. Individual long-term exposure levels were estimated by mapping residential coordinates to these grids and applying land use regression models incorporating satellite-derived data [9]. Participants were categorized into four exposure groups (Q1–Q4) based on pollutant concentration quartiles. This exposure assessment approach—using annual average concentrations at residential addresses combined with land use regression modeling—is well-established and has been widely validated in previous UK Biobank studies [10]. To capture long-term exposure, we used the annual average concentration of NO2 during 2005–2007 and PM10 in 2007 as primary exposure indicators. We selected NO2 and PM10 as primary exposure indicators because they are the most widely measured and consistently reported pollutants in relation to cardiovascular and metabolic outcomes in the UK Biobank, and their exposure data have been extensively validated in previous studies using the same assessment framework. Other pollutants such as PM2.5, SO2, and O3 were not included, as they were not available or had insufficient spatial resolution in the exposure dataset for the study period. While individual time-activity patterns and residential mobility were not directly captured, such non-differential misclassification generally biases effect estimates toward the null [11], suggesting that our findings likely represent conservative estimates of the true associations.
2.3. Study outcome
The primary outcome of interest was the incidence of diabetic microvascular complications. Diabetes diagnoses and dates were obtained through UK Biobank’s linkage with hospital inpatient records, using character string mapping and subsequent conversion to ICD-10 codes (ICD-10: E10–E15). Microvascular complications were defined as the first recorded diagnosis of any diabetes-related renal, neurological, ophthalmic, or peripheral circulatory condition. Detailed diagnostic codes are provided in Supplementary Table 1 in S1 Appendix. We identified 46,848 participants with diabetes. Those with incomplete or missing data, or who developed microvascular complications within two years of recruitment, were excluded. Follow-up time was calculated from the date of recruitment to the first occurrence of a microvascular complication, death, or the censoring date. The flow chart of participant selection is shown in Fig 1.
2.4. Covariate assessment
Baseline covariates were collected from assessment center records and questionnaires at enrollment, including demographic variables (age, sex, ethnicity, smoking, alcohol consumption, annual income), medical history (hypertension, chronic renal failure, diabetes duration), physical measurements, and laboratory tests. The laboratory variables included: body mass index (BMI), alanine aminotransferase (ALT), aspartate aminotransferase (AST), albumin, alkaline phosphatase (ALP), C-reactive protein (CRP), serum calcium, cholesterol, creatinine, γ-glutamyl transferase (γ-GGT), glycated hemoglobin (HbA1c), phosphate, triglycerides (TG), blood urate, and vitamin D levels. Hospital diagnoses were also extracted for baseline disease assessment.
2.5. Statistical analysis
Continuous variables were summarized as medians with interquartile ranges (IQR), while categorical variables were presented as counts and percentages. Group differences were assessed using the Kruskal-Wallis test for continuous variables and the chi-square test for categorical variables. Kaplan–Meier curves were plotted to estimate cumulative incidence of microvascular complications across exposure groups, and differences were tested using the log-rank test. Cox proportional hazards models were used to assess the association between air pollution exposure and risk of microvascular complications, with proportional hazards assumptions checked using Schoenfeld residuals. Covariates were selected based on a priori knowledge from the literature and clinical plausibility, rather than data-driven stepwise selection. We constructed 3 Cox models. Model 1 estimated the total effect. Model 2 included established confounders. Model 3 further added biochemical markers to assess independence from metabolic status. Some of these markers may act as mediators. Model 1: adjusted only for air pollution quartile; Model 2: additionally adjusted for age, sex, BMI, ethnicity, smoking, alcohol use, annual income, HbA1c, hypertension, chronic kidney disease, and diabetes duration; Model 3: further adjusted for biochemical and inflammatory markers including AST, ALT, ALP, CRP, serum calcium, cholesterol, creatinine, γ-GGT, phosphate, triglycerides, blood urate, and vitamin D. To examine effect modification, stratified analyses were performed by age, sex, BMI, and diabetes duration.
Potential nonlinear exposure-response relationships were explored using restricted cubic spline functions. Sensitivity analyses were conducted by excluding participants with less than three years of follow-up to minimize reverse causality and potential bias.
3. Results
3.1. Baseline characteristics of the study population
Among the included participants, 5,858 (60.6%) were men and 3,813 (39.4%) were women. During the follow-up period, 2,104 individuals (21.8%) developed microvascular complications. Baseline characteristics by complication status are shown in Table 1. The median age in both groups was 62 years (IQR: 56–66 in the non-complication group; 57–66 in the complication group). Participants who developed complications had significantly higher BMI (31.36 vs. 30.54 kg/m2, P < 0.001) and HbA1c levels (56.45% vs. 50.20%, P < 0.001). The prevalence of hypertension (75.0% vs. 70.4%, P = 0.021) and chronic renal failure (7.2% vs. 5.4%, P < 0.001) was also higher in the complication group. Baseline characteristics by quartile-based exposure groups of NO2 and PM10 are presented in Supplementary Table 2 and Supplementary Table 3 in S1 Appendix. A higher proportion of women was observed in the highest PM10 exposure group (P = 0.01), while no significant sex difference was found across NO2 quartiles (P = 0.412). Notably, the proportion of White participants decreased significantly in high exposure groups, with corresponding increases in Asian and Black populations (P < 0.001). Lifestyle factors such as smoking and drinking status also varied significantly across exposure categories (P < 0.001). Clinically, NO2 exposure was associated with elevated CRP levels (P = 0.014), and both pollutants were significantly associated with variations in serum creatinine (P < 0.05). Strikingly, vitamin D levels were significantly lower in the highest PM10 quartile (P < 0.001 for both NO2 and PM10), suggesting meaningful associations between air pollution exposure and multiple demographic and clinical indicators.
3.2. Cumulative risk of microvascular complications by air pollution exposure
Kaplan–Meier analyses (Fig 2a–b) demonstrated significant relationships between exposure to both NO2 and PM10 and the cumulative incidence of microvascular complications. For NO2, the highest exposure group (Q4) showed a consistently elevated cumulative risk compared to the lowest group (Q1) throughout the follow-up period (log-rank P < 0.001), with early divergence of risk curves suggesting early emergence of exposure-related risk differentials. A similar but more pronounced pattern was observed for PM10 exposure. The Q4 group had the steepest incidence curve (log-rank P < 0.001), with earlier and wider separation from lower exposure groups, suggesting a potentially stronger risk elevation. For both pollutants, significant divergence appeared within two years of follow-up, with risk differences widening over time.
(a) NO2 (b) PM10.
3.3. Association between pollutant exposure and risk of microvascular complications
Cox proportional hazards models showed significant associations between both NO2 and PM10 exposure and complication risk (Table 2). In the unadjusted Model 1, the highest NO2 quartile (Q4) was associated with increased risk (HR = 1.31, 95% CI: 1.16–1.48, P < 0.001), with a borderline significant increase in Q3 (HR = 1.13; 95% CI: 1.00–1.28; P = 0.05), but no significant risk in Q2 (HR = 1.03, 95% CI: 0.90–1.16, P = 0.695). PM10 exposure showed a clearer risk gradient, with significantly elevated risk beginning in Q2 (HR = 1.20, 95% CI: 1.06–1.36; P = 0.005) and peaking in Q4 (HR = 1.46, 95% CI: 1.29–1.65, P < 0.001). These associations persisted after adjusting for confounders in Models 2 and 3. In the fully adjusted Model 3, NO2 Q4 remained significantly associated with increased risk (HR = 1.27; 95% CI: 1.12–1.45; P < 0.001), while PM10 Q4 demonstrated a stronger association (HR = 1.44; 95% CI: 1.27–1.64; P < 0.001). PM10 exposure showed stronger and earlier risk elevations than NO2, suggesting broader health impacts even at lower levels. Further subtype analyses showed that the associations of NO2 and PM10 exposure with different outcomes (nephropathy, retinopathy, and neuropathy) were generally consistent in terms of direction and trend, although the effect estimates appeared somewhat stronger for nephropathy than for the other two endpoints (Supplementary Table 4 in S1 Appendix).
3.4. Subgroup analyses
To further assess the differences in the impact of air pollutants on different groups of people, stratified analyses were conducted by age (<60/ ≥ 60 years), sex, BMI (<30/ ≥ 30 kg/m2), and diabetes duration (<5/ ≥ 5 years). Although none of the interaction terms reached statistical significance (all P-interaction > 0.05), numeric differences in risk estimates were observed (Table 3). In age-stratified models, individuals ≥60 years had slightly higher risk in the NO2 Q4 group (HR = 1.31; 95% CI: 1.12–1.53) compared to those <60 years (HR = 1.26; 95% CI: 1.01–1.57), though P-interaction = 0.784. A similar trend was seen for PM10. By sex, adjusted risks for NO2 were higher in women (HR = 1.36; 95% CI: 1.10–1.67) than in men (HR = 1.24; 95% CI: 1.06–1.47), but interaction P = 0.332. PM10 showed comparable effects across sexes. For BMI, individuals with BMI ≥ 30 showed higher HRs for both pollutants, but interaction terms remained non-significant. Regarding diabetes duration, NO2 had a greater impact in those with <5 years since diagnosis, whereas PM10 had stronger effects in the ≥ 5-year group. However, interactions were marginal (P = 0.08 and 0.071, respectively).
3.5. Nonlinear associations between air pollution and microvascular complications
Restricted cubic spline models revealed nonlinear associations between pollutant exposure and complication risk, as shown in Fig 3. For NO2, risk remained flat below 20 μg/m3 but rose sharply above 25 μg/m3, suggesting a potential threshold effect. In contrast, PM10 exhibited increasing risk even at low concentrations (15–25 μg/m3), with sustained high risk at higher levels. These patterns were consistent across both unadjusted and fully adjusted models, although effect estimates were attenuated after adjustment, indicating partial confounding. These findings highlight a concentration-dependent relationship, especially for PM10, and underscore the need for tailored pollution control strategies in diabetes populations.
(a) NO2 (b) PM10.
3.6. Sensitivity analyses
To assess model robustness, sensitivity analyses excluded participants with follow-up ≤3 years. Results were consistent with the primary analysis, as shown in Table 4. In unadjusted models, the highest NO2 and PM10 quartiles were associated with elevated risks (NO2 Q4: HR = 1.28, 95% CI: 1.13–1.45, P < 0.001, PM10 Q4: HR = 1.46, 95% CI: 1.28–1.65, P < 0.001). In fully adjusted models (Model 3), risk estimates remained stable (NO2 Q4: HR = 1.24, 95% CI: 1.09–1.42, P = 0.001, PM10 Q4: HR = 1.45, 95% CI: 1.27–1.65, P < 0.001). All effect size differences between main and sensitivity analyses were <5%, and directions of association remained unchanged. These consistent results suggest strong robustness to confounding adjustment and confirm the reliability of the observed associations between long-term air pollution exposure and diabetic microvascular complications.
4. Discussion
Diabetic microvascular complications—including nephropathy, retinopathy, and peripheral neuropathy—remain leading causes of disability and mortality among individuals with diabetes. In this study based on 9,671 diabetic participants from the UK Biobank, we systematically assessed the association between long-term exposure to air pollutants (NO2 and PM10) and the risk of microvascular complications over a follow-up period of up to 16.7 years. Our findings showed a significant positive association between long-term exposure to NO2 and PM10 and the risk of developing diabetic microvascular complications, even after adjustment for demographic characteristics, lifestyle factors, metabolic indicators, and inflammatory biomarkers. These results may have critical clinical and public health implications, as air pollution is a modifiable environmental factor that warrants consideration as a potential target for preventing diabetes-related complications.
Previous studies have indicated that NO2, a major traffic-related air pollutant, is independently associated with increased risks of all-cause, cardiovascular, and respiratory mortality, irrespective of other pollutants [12]. Mechanistically, based on previous experimental evidence, NO2 has been suggested to induce oxidative stress by generating free radicals, which may contribute to structural damage in cells and abnormal intracellular signaling, or by affecting mitochondrial function [13,14]. Similarly, particulate matter such as PM10 is a well-recognized environmental hazard. Our results suggest that PM10, like NO2, exerts considerable microvascular toxicity and elevates the risk of diabetic complications. Evidence from cohort studies has linked these pollutants with increased incidence of aortic dissection and cardiovascular mortality, potentially through mechanisms involving endothelial dysfunction and vascular smooth muscle remodeling. Notably, several recent large-scale UK Biobank-based studies have consistently reported positive associations between long-term air pollution exposure and diabetic microvascular complications, with hazard ratios ranging from 1.06 to 1.15 per interquartile range increase in PM2.5, PM10, and NO2 [15,16]. Our observed effect estimates (HRs up to 1.44 for PM10 Q4) are slightly higher than those reported in some previous studies, which may be attributed to our longer follow-up duration (up to 16.7 years). Furthermore, our study extends previous findings in several aspects: PM10 showed earlier (from Q2) and stronger effects than NO2; restricted cubic splines revealed a threshold effect for NO2 (~25 μg/m3) versus a nearly linear risk for PM10; the extended follow-up (up to 16.7 years) captured early divergence of risk curves within two years; and subgroup analyses consistently suggested higher susceptibility in older adults, females, and obese individuals. These distinctive features position our study as a meaningful complement to the existing evidence base for air pollution and diabetic microvascular complications.
The biological plausibility linking air pollutant exposure to diabetic microvascular complications is strongly supported by experimental and clinical evidence. Mechanistically, both NO2 and PM10 are potent inducers of systemic inflammation and oxidative stress. Once inhaled, these pollutants can trigger a cascade of inflammatory responses, reflected in our observed elevation of C-reactive protein (CRP) levels in higher exposure groups [17]. At the cellular level, these pollutants promote the overproduction of reactive oxygen species (ROS) and reactive nitrogen species (RNS), overwhelming endogenous antioxidant defenses. In diabetes, chronic hyperglycemia itself establishes a self-perpetuating vicious cycle with oxidative stress: hyperglycemia enhances ROS production through mitochondrial dysfunction, while excessive ROS further impairs insulin signaling and β-cell function, creating a feedback loop that amplifies tissue damage [18]. In particular, NO2 exposure can induce nitro-oxidative stress, a process that may synergize with the “metabolic memory” phenomenon in diabetes—the persistent vascular damage even after glycemic control is achieved—by perpetuating epigenetic modifications that sustain pro-inflammatory gene expression [19]. Furthermore, pollutant-induced mitochondrial dysfunction not only impairs cellular energy metabolism but also amplifies ROS generation, creating a vicious cycle of oxidative damage. Over time, these molecular insults—inflammation, oxidative/nitro-oxidative stress, and mitochondrial impairment—can drive epigenetic alterations (e.g., DNA methylation, histone modifications) in vascular endothelial cells, smooth muscle cells, and podocytes, ultimately predisposing to the development and progression of organ-specific microvascular injuries, including retinopathy, nephropathy, and neuropathy [20]. Our findings provide an epidemiological link that aligns with previously proposed mechanistic pathways.
In diabetic individuals, vascular injury is often exacerbated by glucolipotoxicity. Even in the absence of air pollution, the progression of microvascular complications is almost inevitable as the disease advances [21]. However, chronic exposure to air pollution may substantially accelerate this trajectory, further compromising patients’ quality of life and increasing healthcare burdens.
Although our stratified analyses did not detect statistically significant effect modification by age, sex, BMI, or diabetes duration, we observed trends indicating a potentially stronger adverse effect of air pollution in subgroups such as older adults (≥60 years), females, individuals with obesity (BMI ≥ 30), and those with longer disease duration (except for NO2). These findings may reflect increased vascular vulnerability in these populations, amplifying the detrimental effects of pollutant exposure.
Several limitations should be acknowledged. Selection bias may exist due to the UK Biobank’s volunteer-based recruitment, which tends to enroll healthier and more socioeconomically advantaged individuals, potentially limiting generalizability to broader populations. Moreover, which may limit the generalizability of our findings to other air pollutants such as PM2.5, SO2, and O3. However, NO2 and PM10 are recognized as major traffic-related pollutants with well-established health effects, and our findings provide a solid basis for future investigations incorporating a broader range of pollutants. Furthermore, despite rigorous adjustment for a wide range of confounders, residual confounding by unmeasured variables—including physical activity, dietary patterns, and occupational exposure—cannot be entirely excluded. In addition, the use of baseline residential addresses without accounting for residential mobility may introduce non-differential exposure misclassification, which typically biases effect estimates toward the null. Finally, some biomarkers adjusted in Model 3 (e.g., CRP, lipids) may lie on the causal pathway, and controlling for these potential mediators could represent over-adjustment, yielding conservative estimates. Nonetheless, the consistent positive associations across Models 1–3 support the robustness of our findings.
5. Conclusion
Our study provides robust evidence that long-term exposure to ambient air pollution—specifically NO2 and PM10—is associated with increased risk of microvascular complications in patients with diabetes. These associations were observable even among participants with shorter exposure duration and appeared to strengthen with longer follow-up time. Our findings provide information that may be useful for integrating air pollution control into diabetes management strategies and offer a scientific rationale for developing personalized environmental interventions targeting high-risk diabetic populations.
Supporting information
S1 Appendix. Supplementary tables including ICD-10 diagnostic codes for diabetic microvascular complications, participant baseline characteristics stratified by NO2 and PM10 exposure quartiles, and subtype-specific associations between air pollutants and nephropathy, retinopathy and peripheral neuropathy.
https://doi.org/10.1371/journal.pone.0354711.s001
(DOC)
References
- 1. GBD 2019 Diabetes and Air Pollution Collaborators. Estimates, trends, and drivers of the global burden of type 2 diabetes attributable to PM(2·5) air pollution, 1990-2019: an analysis of data from the Global Burden of Disease Study 2019. Lancet Planet Health. 2022;6:e586–600.
- 2. Clark SN, Anenberg SC, Brauer M. Global burden of disease from environmental factors. Annu Rev Public Health. 2025;46(1):233–51. pmid:39689276
- 3. Rajagopalan S, Brook RD. Air pollution and type 2 diabetes: mechanistic insights. Diabetes. 2012;61(12):3037–45. pmid:23172950
- 4. Beckman JA, Creager MA. Vascular complications of diabetes. Circ Res. 2016;118:1771–85.
- 5. Faselis C, Katsimardou A, Imprialos K, Deligkaris P, Kallistratos M, Dimitriadis K. Microvascular complications of type 2 diabetes mellitus. Curr Vasc Pharmacol. 2020;18(2):117–24. pmid:31057114
- 6. Cole JB, Florez JC. Genetics of diabetes mellitus and diabetes complications. Nat Rev Nephrol. 2020;16(7):377–90. pmid:32398868
- 7. Münzel T, Gori T, Al-Kindi S, Deanfield J, Lelieveld J, Daiber A, et al. Effects of gaseous and solid constituents of air pollution on endothelial function. Eur Heart J. 2018;39(38):3543–50. pmid:30124840
- 8. Pope CA 3rd, Bhatnagar A, McCracken JP, Abplanalp W, Conklin DJ, O’Toole T. Exposure to fine particulate air pollution is associated with endothelial injury and systemic inflammation. Circ Res. 2016;119(11):1204–14. pmid:27780829
- 9. Vienneau D, de Hoogh K, Bechle MJ, Beelen R, van Donkelaar A, Martin RV, et al. Western European land use regression incorporating satellite- and ground-based measurements of NO2 and PM10. Environ Sci Technol. 2013;47(23):13555–64. pmid:24156783
- 10. Ji Q, Liu Q, Xu Y, Xu M, Zhan Y. Long-term exposure to residential greenspace, bluespace, traffic, and air pollutants with dementia: a prospective cohort study with exploratory mediation by plasma metabolites and telomere length. Ecotoxicol Environ Saf. 2026;310:119750. pmid:41547191
- 11. Allen NE, Lacey B, Lawlor DA, Pell JP, Gallacher J, Smeeth L, et al. Prospective study design and data analysis in UK Biobank. Sci Transl Med. 2024;16(729):eadf4428. pmid:38198570
- 12. Huang S, Li H, Wang M, Qian Y, Steenland K, Caudle WM, et al. Long-term exposure to nitrogen dioxide and mortality: a systematic review and meta-analysis. Sci Total Environ. 2021;776:145968. pmid:33640547
- 13. Yan W, Ji X, Shi J, Li G, Sang N. Acute nitrogen dioxide inhalation induces mitochondrial dysfunction in rat brain. Environ Res. 2015;138:416–24. pmid:25791864
- 14. Persinger RL, Poynter ME, Ckless K, Janssen-Heininger YMW. Molecular mechanisms of nitrogen dioxide induced epithelial injury in the lung. Mol Cell Biochem. 2002;234–235(1–2):71–80. pmid:12162462
- 15. Wang B, Sun Y, Zhang K, Wang Y, Tan X, Wang N, et al. Long-term exposure to ambient air pollution and risk of microvascular complications among patients with type 2 diabetes: a prospective study. Int J Epidemiol. 2024;53(3):dyae056. pmid:38632038
- 16. Wang Z, Yang J, Huang W, Li D. Long-term exposure to ambient PM2.5 constituents and risk of microvascular complications among individuals with type 2 diabetes mellitus: a nationwide cohort study in UK. Environ Res. 2026;302:124589. pmid:42055387
- 17. Liu Q, Gu X, Deng F, Mu L, Baccarelli AA, Guo X, et al. Ambient particulate air pollution and circulating C-reactive protein level: a systematic review and meta-analysis. Int J Hyg Environ Health. 2019;222(5):756–64. pmid:31103472
- 18. Yurdgulu EE, Ay YA, Bayir Y, Halici Z. The vicious cycle of hyperglycemia and oxidative stress: novel mechanistic insights into a pathogenic alliance. Int J Biochem Cell Biol. 2026;199:106990. pmid:42303171
- 19. Yao Y, Song Q, Hu C, Da X, Yu Y, He Z, et al. Endothelial cell metabolic memory causes cardiovascular dysfunction in diabetes. Cardiovasc Res. 2022;118(1):196–211. pmid:33483741
- 20. Bhatt MP, Lee Y-J, Jung S-H, Kim YH, Hwang JY, Han E-T, et al. C-peptide protects against hyperglycemic memory and vascular endothelial cell apoptosis. J Endocrinol. 2016;231(1):97–108. pmid:27554111
- 21. Crasto W, Patel V, Davies MJ, Khunti K. Prevention of microvascular complications of diabetes. Endocrinol Metab Clin North Am. 2021;50:431–55.