Long-term exposure to ambient air pollution and prevalence of lower urinary tract symptoms suggestive of benign prostatic hyperplasia: A nationwide study in China

Technical Summary: Long-term Exposure to Ambient Air Pollution and Prevalence of LUTS/BPH in China

Problem Statement
Benign prostatic hyperplasia (BPH) and its associated lower urinary tract symptoms (LUTS) represent a significant public health burden among aging males, with prevalence rates in China rising from 36.6% in men over 40 to 69.2% in those over 80. While established risk factors include age, lifestyle, and comorbidities, the role of environmental exposures remains underexplored. Although emerging evidence suggests air pollution may contribute to urological disorders via mechanisms like oxidative stress and systemic inflammation, epidemiological data linking specific ambient pollutants to LUTS/BPH in China is limited. Existing studies are often geographically restricted, lack comprehensive pollutant panels (e.g., missing PM1 or specific gaseous pollutants), or fail to adequately address confounding variables and co-exposure effects.

Methodology
This nationwide cross-sectional study utilized data from the fourth wave (2018) of the China Health and Retirement Longitudinal Study (CHARLS), analyzing 7,066 males aged 45 and older.

• Outcome Definition: LUTS/BPH was defined based on self-reported physician diagnosis, as symptom-based instruments (e.g., IPSS) were unavailable in the dataset.
• Exposure Assessment: Individual-level exposure to seven major air pollutants (CO, NO2, O3, SO2, PM1, PM2.5, PM10) was estimated using the CHAP dataset. This dataset employs artificial intelligence (Space-Time Extremely Randomized Trees model) to integrate ground measurements, remote sensing, and atmospheric reanalysis, providing city-level annual average concentrations from 2013 to 2018 at a 0.1° spatial resolution.
• Statistical Analysis:
  • Logistic Regression: Multivariable logistic regression models (crude, base, and fully adjusted) assessed the association between per-unit increases in individual pollutants and LUTS/BPH prevalence. The fully adjusted model controlled for age, BMI, education, marital status, residence, smoking, drinking, and various comorbidities (hypertension, diabetes, dyslipidemia, kidney disease, etc.).
  • Mixture Analysis: Weighted Quantile Sum (WQS) regression was employed to evaluate the joint effect of the pollutant mixture and to determine the relative contribution of each pollutant, addressing high inter-pollutant correlations.
  • Subgroup Analysis: Stratified analyses by age (45–59 vs. ≥60 years) and spatial autocorrelation (Global Moran's I) were conducted.

Key Results

• Prevalence: 11.8% of the study population reported LUTS/BPH.
• Individual Pollutant Associations: In fully adjusted models, long-term exposure to Carbon Monoxide (CO) and Particulate Matter (PM1, PM2.5, PM10) showed significant positive associations with LUTS/BPH:
  • CO: Each 1 mg/m³ increase was associated with a 35% higher odds of LUTS/BPH (OR = 1.35, 95% CI: 1.00–1.82).
  • PM1: Each 1 µg/m³ increase associated with OR = 1.012 (95% CI: 1.00–1.02).
  • PM2.5: Each 1 µg/m³ increase associated with OR = 1.007 (95% CI: 1.00–1.01).
  • PM10: Each 1 µg/m³ increase associated with OR = 1.004 (95% CI: 1.00–1.01).
  • Non-Significant Pollutants: NO2, O3, and SO2 did not show statistically significant associations in the fully adjusted models.
• Mixture Effects: WQS regression confirmed a significant joint effect of the pollution mixture (OR = 1.12 per quartile increase). CO (weight = 0.41) and PM10 (weight = 0.36) were identified as the primary contributors to this mixture effect, collectively accounting for over 77% of the weight.
• Age Stratification: Significant associations between pollutants (CO, NO2, PMs) and LUTS/BPH were observed primarily in the older subgroup (≥60 years), whereas no significant associations were found in the 45–59 age group, though interaction terms were not statistically significant.

Key Contributions

1. First Nationwide Evidence in China: This is the first study to examine the association between a comprehensive panel of ambient air pollutants and LUTS/BPH prevalence across a nationally representative Chinese cohort.
2. Advanced Exposure Modeling: The study utilized high-resolution AI-driven exposure estimates (2013–2018) rather than relying on coarse monitoring station data, allowing for more precise individual-level exposure assessment.
3. Mixture Analysis: By applying WQS regression, the study addressed the challenge of multicollinearity among pollutants (e.g., the near-collinearity between PM1 and PM2.5), identifying CO and PM10 as the dominant drivers of the observed mixture effect.
4. Specific Pollutant Differentiation: The study distinguishes between the effects of particulate matter and gaseous pollutants, finding that while PMs and CO are significantly associated with LUTS/BPH, other gases (NO2, O3, SO2) are not, suggesting specific pathophysiological pathways related to combustion-derived pollutants and hypoxia.

Significance and Claims
The authors claim that their findings provide new insights into potentially modifiable environmental drivers of LUTS/BPH in aging societies. The study suggests that long-term exposure to ambient air pollution, particularly CO and particulate matter, is associated with a higher prevalence of LUTS/BPH among Chinese men.

The paper posits that these associations may be biologically plausible through mechanisms such as:

• Systemic Hypoxia: CO's ability to reduce oxygen-carrying capacity may activate proliferative pathways (e.g., HIF-1α/VEGF) in prostatic tissue.
• Oxidative Stress and Inflammation: Fine particulate matter (PM1, PM2.5) may induce systemic inflammation and oxidative stress, contributing to prostate tissue remodeling.
• Endocrine and Neurogenic Disruption: Potential impacts on hormonal balance and autonomic nervous system function.

The authors conclude that these findings highlight the importance of air quality regulations targeting combustion-derived pollutants and particulate matter, especially in regions with aging populations. However, they explicitly acknowledge limitations, including the cross-sectional design (precluding causal inference), potential outcome misclassification due to self-reporting, and the inability to fully distinguish independent contributions of highly correlated particulate pollutants.