ABSTRACT
Perfluoroalkyl substances (PFAS) have been reported to be harmful to multiple organs in the human body. Based on a previous study suggesting that hemodialysis (HD) may be a means of eliminating PFAS from the human body, we aimed to compare the serum PFAS concentrations of patients undergoing regular HD, patients with chronic kidney disease (CKD) and controls. Additionally, we also investigated the correlation between PFAS and biochemical data, as well as concurrent comorbidities. We recruited 301 participants who had been on maintenance dialysis for >90 days, 20 participants with stage 5 non-dialysis CKD, and 55 control participants who did not have a diagnosis of kidney disease, with a mean creatinine level of 0.77 mg/dl. Eight different PFAS, namely perfluorooctanoic acid (PFOA), total and linear perfluorooctanesulfonic acid (PFOS), perfluoroheptanoic acid (PFHpA), perfluorohexanesulfonic acid (PFHxS), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), and perfluoroundecanoic acid (PFUnDA), were measured using ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). Spearman correlation and multivariable linear regression with 5 % false discovery rate were used to evaluate the relationships between PFAS and clinical parameters in HD patients and controls. Circulating concentrations of seven PFAS, including total and linear PFOS (T-PFOS and L-PFOS) PFDA, PFNA, PFHxS, PFOA, and PFUnDA, were significantly lower in the HD group compared to the CKD and control group. For the interplay between biochemical data and PFAS, all of the studied PFAS were positively correlated with aspartate aminotransferase, alanine aminotransferase, glucose, blood urea nitrogen, ferritin, and vitamin D in the controls, while in HD patients, the PFAS were all positively correlated with albumin, uric acid, iron, and vitamin D. These findings may offer valuable insights for future studies seeking to eliminate PFAS.
