Perfluorinated compounds in adults and their association with fasting glucose and incident diabetes: a prospective cohort study

Abstract Background The endocrine disruption of perfluorinated compounds is an emerging issue. We aimed to examine the association of serum perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) levels with incident diabetes and fasting serum glucose concentration. Methods This prospective cohort study was based on an urban-based cohort subpopulation from the Korean Genome and Epidemiology Study. Serum samples (600 µL) were received from 100 participants in the normoglycemic baseline survey (2004–2013), and concentrations of PFOA and PFOS were measured using mass spectrometry. The incidence of diabetes was tracked in the follow-up survey (2012–2016). Results The mean age was 56.4 years (men, 59%). The median serum PFOA and PFOS concentrations were 4.29 ng/mL and 9.44 ng/mL, respectively. PFOA and PFOS concentrations differed according to age, sex, and residential area. After 60 months, 23 patients had diabetes. Log-transformed PFOA (lnPFOA) and log-transformed PFOS (lnPFOS) were significantly higher in those who transitioned to diabetes than in those who did not (both p < 0.05). After multivariate adjustment, lnPFOA (coefficient = 6.98, 95% CI -0.04–14 , p = 0.054) and lnPFOS (coefficient = 7.06, 95% CI -0.96–15.08 , p = 0.088) predicted increased fasting glucose without statistical significance. In addition, lnPFOA, but not lnPFOS, significantly predicted incident diabetes (HR = 3.98 , 95% CI 1.42–11.1, p < 0.01). Conclusion Exposure to PFOA and PFOS may have a potential dysglycemic effect. In particular, exposure to PFOA increased the risk of diabetes. Further research with larger sample size is warranted.