绝经前女性的乳腺密度与空气污染物暴露：一项横断面研究。

Mammographic density and exposure to air pollutants in premenopausal women: a cross-sectional study.

机构信息

Department of Preventive Medicine, Public Health and Microbiology, Universidad Autónoma de Madrid (UAM).

HM CINAC (Centro Integral de Neurociencias AC), Hospital Universitario Puerta del Sur, Fundación HM Hospitales.

出版信息

Environ Health Prev Med. 2024;29:65. doi: 10.1265/ehpm.24-00209.

DOI:10.1265/ehpm.24-00209

PMID:39581598

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11604911/

Abstract

BACKGROUND

Mammographic density (MD) is a well-established risk factor for breast cancer. Air pollution is a major public health concern and a recognized carcinogen. We aim to investigate the association between MD and exposure to specific air pollutants (SO, CO, NO, NO, NO, PM, PM, and O) in premenopausal females.

METHODS

This cross-sectional study, carried out in Spain, included 769 participants who attended their gynecological examinations. Hourly concentrations of the pollutants were extracted from the Air Quality Monitoring System of Madrid City over a 3-year period. Individual long-term exposure to pollutants was assessed by geocoding residential addresses and monitoring stations, and applying ordinary kriging to the 3-year annual mean concentrations of each pollutant to interpolate the surface of Madrid. This exposure variable was categorized into quartiles. In a first analysis, we used multiple linear regression models with the log-transformed percent MD as a continuous variable. In a second analysis, we used MD as a dichotomous variable ("high" density (MD > 50%) vs. "low" density (MD ≤ 50%)) and applied multiple logistic regression models to estimate odds ratios (ORs). We also analyzed the correlation among the pollutants, and performed a principal component analysis (PCA) to reduce the dimensionality of this set of eight correlated pollutants into a smaller set of uncorrelated variables (principal components (PCs)). Finally, the initial analyses were applied to the PCs to detect underlying patterns of emission sources.

RESULTS

The first analysis detected no association between MD and exposure to any of the pollutants. The second analysis showed non-statistically significant increased risks (OR; IC95%) of high MD were detected in women with higher exposure to SO (1.50; 0.90-2.48), and PM (1.27; 0.77-2.10). In contrast, non-significant ORs < 1 were found in all exposure quartiles for NO (OR = 0.72, OR = 0.68, OR = 0.78), and PM (OR = 0.69, OR = 0.82, OR = 0.72). PCA identified two PCs (PC1: "traffic pollution" and PC2: "natural pollution"), and no association was detected between MD and proximity to these two PCs.

CONCLUSIONS

In general, our results show a lack of association between residential exposure to specific air pollutants and MD in premenopausal females. Future research is needed to confirm or refute these findings.

摘要

背景

乳腺密度（MD）是乳腺癌的一个既定风险因素。空气污染是一个主要的公共卫生问题，也是一种公认的致癌物质。我们旨在研究 MD 与特定空气污染物（SO、CO、NO、NO、NO、PM、PM 和 O）在绝经前女性中的暴露之间的关联。

方法

这是一项在西班牙进行的横断面研究，共纳入 769 名参加妇科检查的参与者。在 3 年期间，从马德里市空气质量监测系统中提取污染物的每小时浓度。通过地理编码居住地址和监测站，对污染物的个体长期暴露进行评估，并应用普通克里金插值法对每种污染物的 3 年年均浓度进行插值，以绘制马德里的表面图。该暴露变量分为四分之一。在第一次分析中，我们使用具有对数变换的百分比 MD 作为连续变量的多元线性回归模型。在第二次分析中，我们将 MD 作为二分类变量（“高”密度（MD > 50%）与“低”密度（MD ≤ 50%）），并应用多变量逻辑回归模型估计比值比（OR）。我们还分析了污染物之间的相关性，并进行了主成分分析（PCA），以将这组 8 个相关污染物的维度降低为一组较小的不相关变量（主成分（PCs））。最后，将初始分析应用于 PCs 以检测排放源的潜在模式。

结果

第一项分析未发现 MD 与任何一种污染物暴露之间存在关联。第二项分析显示，SO 暴露较高的女性中，高 MD 的风险（OR；IC95%）呈非统计学意义的增加（1.50；0.90-2.48），而 PM 暴露较高的女性中，高 MD 的风险（OR；IC95%）呈非统计学意义的增加（1.27；0.77-2.10）。相比之下，在所有暴露四分位数中，NO（OR = 0.72，OR = 0.68，OR = 0.78）和 PM（OR = 0.69，OR = 0.82，OR = 0.72）的 OR 均小于 1。PCA 确定了两个 PCs（PC1：“交通污染”和 PC2：“自然污染”），但未检测到 MD 与这两个 PC 的接近程度之间存在关联。