Household air pollution exposure in adult women is associated with increased carotid intima-media thickness: A cross-sectional study of the Household Air Pollution Intervention Network trial.

Lindsay J. Underhill, Lisa de las Fuentes, Laura Nicolaou, Shakir Hossen, Anaité Díaz-Artiga, Ajay Pillarisetti, Aris T. Papageorghiou, Florien Ndagijimana, Ghislaine Rosa, Gurusamy Thangavel, John P. McCracken, Kalpana Balakrishnan, Krishnendu Mukhopadhyay, Kyle Steenland, Lisa M. Thompson, Lance A. Waller, Maggie L. Clark, Michael Johnson, Sarada S. Garg, Sankar Sambandam, Suzanne Simkovich, Vigneswari Aravindalochanan, Kendra N. Williams, Wenlu Ye, Jennifer L. Peel, Thomas Clasen, William Checkley, Víctor G. Dávila‐RománInternational Journal of Hygiene and Environmental Health 2025 · DOI

TL;DR

This study found that higher personal exposure to fine particulate matter (PM2.5) and carbon monoxide (CO) from household air pollution (HAP) is associated with increased carotid intima-media thickness (CIMT), a key indicator of atherosclerosis. This research provides crucial evidence linking HAP, prevalent in low- and middle-income countries, to cardiovascular disease risk, even in the absence of traditional risk factors.

What Problem Does This Paper Address?

Cardiovascular disease (CVD) is a leading global cause of death and disability, disproportionately affecting individuals in low- and middle-income countries (LMICs). Atherosclerosis, the underlying process of most CVDs, involves the buildup of plaque in artery walls. While traditional risk factors like high cholesterol and hypertension are well-known, the role of environmental factors, particularly household air pollution (HAP), remains less understood, especially in LMICs where HAP exposure is widespread. HAP, generated from burning biomass fuels for cooking, contains pollutants like PM2.5 and carbon monoxide (CO). Previous studies have often relied on indirect measures of HAP exposure. This paper aims to bridge this knowledge gap by directly measuring personal HAP exposure and its association with CIMT, a well-established imaging biomarker for atherosclerosis.

Methods

The researchers conducted a cross-sectional study using baseline data from the Household Air Pollution Intervention Network (HAPIN) trial. This trial involved adult women aged 40-79 years from rural communities in Guatemala, India, Peru, and Rwanda.

Key methodological aspects include:

  • Personal Exposure Monitoring: Participants wore lightweight monitors for 24 hours to capture personal exposure levels to PM2.5, black carbon (BC), and CO. This direct measurement approach is more precise than relying on modeled or area-based estimates.
  • Carotid Artery Ultrasound: Certified sonographers performed ultrasound imaging of the carotid arteries to measure intima-media thickness (CIMT) – both mean and maximal – and to identify atherosclerotic plaques. CIMT is a non-invasive measure of arterial wall thickness, and increased CIMT is a strong predictor of CVD risk.
  • Statistical Analysis: Single- and multi-variable linear regression models were used to assess the association between personal HAP exposures and CIMT. Covariates included study site, socioeconomic status (SES) index, secondhand smoke exposure, and age. Both linear and log-linear models were tested, and pollutant exposures were natural log-transformed. Non-linear associations were also explored using generalized additive models.

Datasets

The study utilized baseline data from the Household Air Pollution Intervention Network (HAPIN) trial. This dataset comprises information from 397 adult women aged 40-79 years across four LMICs: Guatemala, India, Peru, and Rwanda. The data includes:

  • 24-hour personal exposure measurements for PM2.5, BC, and CO.
  • Carotid artery ultrasound measurements (mean and maximal CIMT, and atherosclerotic plaque presence).
  • Sociodemographic information, health history, and SES data.

This dataset is unique in its use of direct personal exposure monitoring for HAP pollutants in conjunction with CIMT measurements across multiple LMIC settings.

Key Insights and Ideas

  • Direct Measurement Matters: Using personal monitors for HAP exposure provides a more accurate picture of individual risk compared to relying on self-reports or area monitoring, which can be prone to misclassification.
  • PM2.5 and CO as Key Culprits: The study highlights PM2.5 and CO as significant contributors to subclinical atherosclerosis, even at exposure levels common in LMICs.
  • Biomarker Validation: The findings reinforce CIMT's role as a valuable biomarker for assessing the vascular impact of environmental exposures like HAP.
  • Pollutant-Specific Effects: The observed associations suggest that different HAP components may have distinct impacts on vascular health, with PM2.5 and CO showing significant links to CIMT, while BC did not in this study.

Key Results

  • PM2.5 Association: Each 10 µg/m³ increase in PM2.5 exposure was associated with a 0.001 mm increase in mean CIMT (p=0.008) and a 0.002 mm increase in maximal CIMT (p=0.005).
  • CO Association: Each 10 ppm increase in CO exposure was associated with a 0.04 mm increase in maximal CIMT (p=0.048). Women in the highest quartile of CO exposure had significantly greater mean (0.036 mm, p=0.04) and maximal CIMT (0.05 mm, p=0.02) compared to those in the lowest quartile.
  • No BC or Plaque Association: No significant associations were found between black carbon (BC) exposure and CIMT, nor between any of the measured pollutants and the presence of carotid atherosclerotic plaques.
  • High Exposure Levels: A significant majority (81%) of participants experienced personal PM2.5 exposures exceeding the World Health Organization's annual interim target-1 guideline.

Limitations and Future Work

The authors acknowledge several limitations:

  • Cross-Sectional Design: The study design limits causal inference; it can only show associations, not prove that HAP causes the increased CIMT.
  • Single 24-Hour Exposure: Baseline personal exposure was measured over a single 24-hour period, which may not fully capture long-term or seasonal variability in exposure. However, the authors argue this provides a more precise individual estimate than other methods.
  • Higher Exposure Distribution: The study population was concentrated at the higher end of the exposure distribution, which might have limited the ability to detect associations at lower exposure levels.
  • Unmeasured Heterogeneity: Differences in fuel types, cooking practices, and underlying cardiovascular risk profiles across study sites could influence results, even with site adjustments.

Future work could involve longitudinal studies to establish causality, explore the impact of interventions on CIMT, and investigate the specific mechanisms by which PM2.5 and CO affect vascular health. Further research into the role of BC and its interaction with other pollutants is also warranted.

Summary updated

Abstract

In this cross-sectional study, higher personal exposures to PM<sub>2.5</sub> and CO were associated with greater mean and maximal CIMT, a well-established biomarker of atherosclerosis, further supporting the association between HAP and cardiovascular disease.
BibTeX
@article{underhill2025household,
  author = {Lindsay J. Underhill and Lisa de las Fuentes and Laura Nicolaou and Shakir Hossen and Anaité Díaz-Artiga and Ajay Pillarisetti and Aris T. Papageorghiou and Florien Ndagijimana and Ghislaine Rosa and Gurusamy Thangavel and John P. McCracken and Kalpana Balakrishnan and Krishnendu Mukhopadhyay and Kyle Steenland and Lisa M. Thompson and Lance A. Waller and Maggie L. Clark and Michael Johnson and Sarada S. Garg and Sankar Sambandam and Suzanne Simkovich and Vigneswari Aravindalochanan and Kendra N. Williams and Wenlu Ye and Jennifer L. Peel and Thomas Clasen and William Checkley and Víctor G. Dávila‐Román},
  title = {{Household air pollution exposure in adult women is associated with increased carotid intima-media thickness: A cross-sectional study of the Household Air Pollution Intervention Network trial.}},
  journal = {International Journal of Hygiene and Environmental Health},
  year = {2025},
  doi = {10.1016/j.ijheh.2025.114649},
}
Figure