Inflammatory bowel disease and biomarkers of gut inflammation and permeability in a community with high exposure to perfluoroalkyl substances through drinking water

摘要

Perfluoroalkyl substances (PFAS) can act as surfactants and have been suggested to be capable of affecting gut mucosa integrity, a possible factor in the pathogenesis of inflammatory bowel disease (IBD). So far, only PFOA has been shown to have a positive association with ulcerative colitis. The present study aimed to investigate the association of PFAS and clinically diagnosed IBD in the Ronneby cohort, a population with high PFAS exposure (especially high PFOS and PFHxS) from Aqueous Film-Forming Foam through drinking water, using registry data. Additionally, to explore associations of PFAS with fecal zonulin and calprotectin, subclinical biomarkers of gut inflammation and permeability, in a sub-set of participants from Ronneby and Karlshamn (a nearby control municipality). The registry study included all people that ever resided in Ronneby municipality at least one year between 1980 and 2013. Yearly exposure to contaminated drinking water was assessed based on residential addresses and waterworks supply data, and the population classified by early, mid and late periods in ascending level of contamination. Diagnosed IBD cases were retrieved from the Swedish National Patient register and cause-of-death register. The Cox proportional hazards model was used to derive the hazard ratios (HRs) for diagnosed IBD. The biomarker study included 189 individuals who provided fecal samples. Serum PFAS were measured using LC-MS/MS. Fecal zonulin and calprotectin were measured using ELISA. Linear regression was used to assess the associations between measured PFAS and biomarker levels. In the registry study, no raised HRs for diagnosed IBD were found for cohort subjects with mid (1995-2004) or late period (2005-2013) exposure compared to never exposure. Early period exposure only (1985-1994) showed raised HRs for Crohn's disease (HR = 1.58, p = 0.048) and other non-specified IBD (HR = 1.38, p = 0.037). In the biomarker study, Karlshamn showed higher fecal calprotectin levels (median = 99.6 mg/kg in Karlshamn vs. 66.8 mg/kg in Ronneby, p = 0.04). A trend of decreased calprotectin with increased serum PFAS indicated higher PFAS was associated with lower degree of gut inflammation (p = 0.002). No association between serum PFAS and fecal zonulin was found. In conclusion, the present study found no consistent evidence to support PFAS exposure as a risk factor for IBD.