2026-01-14 マウントサイナイ医療システム(MSHS)
<関連情報>
- https://www.mountsinai.org/about/newsroom/2026/exposure-to-forever-chemicals-linked-to-higher-risk-of-gestational-diabetes-major-review-finds
- https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(25)00682-0/fulltext
パーフルオロアルキルおよびポリフルオロアルキル化合物と血糖コントロール、インスリン分泌および感受性、糖尿病リスクのマーカーとの関連:系統的レビューおよびメタアナリシス Associations of perfluoroalkyl and polyfluoroalkyl substances with markers of glycaemic control, insulin secretion and sensitivity, and diabetes risk: a systematic review and meta-analyses
Sandra India Aldana ∙ Xin Yu ∙ Meizhen Yao ∙ Nathan Cohen ∙ Eftychia Markopoulou ∙ Maanal Chowdhury ∙ et al.
eClinicalMedicine Published: January 12, 2026
DOI:https://doi.org/10.1016/j.eclinm.2025.103747
Summary
Background
Growing literature examines the impact of per- and polyfluoroalkyl substances (PFAS) on diabetes risk. We aimed to conduct a comprehensive systematic review and meta-analysis of epidemiological studies to characterize the associations of exposures to PFAS with markers of glycemic control, insulin resistance, pancreatic β-cell function, and diabetes risk.
Methods
A systematic search of epidemiological articles published through July 21, 2025 was conducted by two researchers in PubMed/MEDLINE and Ovid/EMBASE following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Experimental studies were excluded from our review. Reported findings were extracted from published articles. Risk of bias was evaluated using the Navigation Guide. Random-effects meta-analyses stratified by study design estimated PFAS associations with gestational diabetes mellitus (GDM), type 2 diabetes (T2D), and continuous measures of Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), HOMA-β, fasting insulin, fasting glucose, and hemoglobin A1c (HbA1c). This study was registered in PROSPERO (CRD42022369711).
Findings
Out of 738 records retrieved, we identified 129 eligible studies. Most studies focused on GDM (n = 25) and/or T2D (n = 36), while three focused on type 1 diabetes (T1D). Participant numbers ranged from n = 40 to n = 1,331,541 in the systematic review and from n = 399 to n = 111,544 in the meta-analyses. We found consistent associations between 8 different PFAS and higher odds of GDM across prospective and other study designs, including PFOS [n = 8, OR (95%CI) per doubling PFOS increase: 1.13 (1.01, 1.26), I2 = 0.0%] among other PFAS. We also found positive associations between several legacy PFAS such as PFOS with HOMA-IR [(n = 8), β (95%CI): 0.06 (0.01, 0.12), I2 = 0.0%] and fasting insulin [n = 5, β (95% CI) in μU/mL: 0.23 (0.06, 0.40), I2 = 0.0%] in prospective studies, and HOMA-β in cross-sectional studies [(n = 6), β (95% CI): 5.93 (1.72, 10.2), I2 = 67.0%], among other. Less consistent or null associations were with T2D, fasting glucose, and HbA1c. The evidence was of low-moderate quality and limited strength. Most studies were categorized as low risk of bias for other criteria, except for study design (cross-sectional).
Interpretation
Evidence from observational studies supports PFAS associations with higher odds of GDM and increased markers of insulin resistance and secretion. PFAS associations with established T2D or T1D remain to be elucidated, as evidence is still limited and effect sizes for some continuous diabetes markers were small and should be interpreted with caution. Larger life-course prospective studies with greater representation of well-characterized cases and evaluating emerging PFAS and mixtures are needed to fully capture the potential PFAS impacts on diabetes.
Funding
National Institutes of Health (NIH), National Institute of Environmental Health Sciences (NIEHS).
