LAUSR

Background It is known that perchlorate, nitrate and thiocyanate have the property of inhibiting sodium iodide symporter. Animal studies have suggested that these compounds, especially perchlorate, might also interfere with insulin secretion. However, the association between their exposure and diabetes risk is largely unknown in humans. Methods Among 11 443 participants (mean age 42.3 years) from the National Health and Nutritional Examination Survey 2001-14, urinary perchlorate, nitrate and thiocyanate were measured by using ion chromatography coupled with electrospray tandem mass spectrometry. Diabetes was defined as self-reported doctor diagnosis, use of oral hypoglycaemic medication or insulin, fasting plasma glucose ≥ 126 mg/dl or glycated haemoglobin A1c (HbA1c) ≥ 6.5%. Results The median (interquartile range) levels of urinary perchlorate, nitrate and thiocyanate were 3.32 (1.84, 5.70) μg/l, 46.4 (27.9, 72.0) mg/l and 1.23 (0.59, 2.78) mg/l, respectively. Higher levels of urinary perchlorate were associated with elevated levels of fasting glucose, HbA1c, insulin and homeostatic model assessment of insulin resistance (all Ptrend < 0.001). After multivariate adjustment including urinary creatinine, smoking status and body mass index (BMI), higher urinary perchlorate, but not nitrate or thiocyanate, was associated with an increased prevalence of diabetes mellitus. Comparing extreme quintiles, the odds ratio (95% confidence interval) of diabetes was 1.53 (1.21, 1.93; Ptrend < 0.001) for perchlorate, 1.01 (0.77, 1.32; Ptrend = 0.44) for nitrate and 0.98 (0.73, 1.31; Ptrend = 0.64) for thiocyanate. When urinary perchlorate, nitrate and thiocyanate were further mutually adjusted, the results did not materially change. Similar results were observed when analyses were stratified by smoking status, as well as by age, gender, kidney function and BMI. Conclusions Higher urinary perchlorate levels are associated with an increased prevalence of diabetes mellitus, independent of traditional risk factors. Future prospective studies are needed to confirm these findings.