Association between nickel exposure and body compositions in the United States: a population-based cross-sectional study

Background Increasing body fat or decreasing muscle and bone mass were associated with worse health outcomes in the adult population. The effects of nickel exposure on body composition are not known. The aim of the current study was to investigate the relationship between urinary nickel levels and body compositions. Materials and methods Two thousand seven hundred sixty-two participants were included in the analysis from the National Health and Nutrition Examination Surveys of 2017–2018 after excluding participants who have missing data on urinary nickel and those with missing all body mass component data. We used weighted generalized linear models to explore the relationship between urinary nickel and body mass components under interpolating missing covariable values. Simultaneously, sensitivity analyses and subgroup analysis were conducted to verify stability of analysis result. Curve fitting and saturation effect analysis were used to explore the possible nonlinear relationship between urine nickel and body compositions. Results Among the 2,762 participants, the average urinary nickel level was 1.58 ug/L. The weighted generalized linear models, the sensitivity analyses and subgroup analyses found no significant linear relationship between urinary nickel and body compositions. For body weight, BMI, TLM, ALM, TRF, TOF and BMC, the urine nickel saturation effect values were 0.76, 0.74, 0.5, 0.67, 0.64, 0.48, and 0.45 ug/L, respectively. For each 1 ug/L rise in urinary nickel levels at levels below the turning point, body weight increases (β = 9.06, 95% CI = 2.75, 15.36, p = 0.01), BMI increases (β = 3.20, 95% CI = 1.36, 5.05, p = < 0.001), TLM decreases (β = -47.39, 95% CI = -97.38, 2.59, p = 0.06), ALM decreases (β = -37.25, 95% CI = -63.25, -11.24, p = 0.01), TRF increases (β = 20.68, 95% CI = 1.50, 39.86, p = 0.03), TOF increases (β = 57.92, 95% CI = -0.12, 115.95, p = 0.05), and BMC decreases (β = -6.84, 95% CI = -12.64, -1.04, p = 0.02). Conclusions In summary, our study demonstrated that a dose–response relationship exists between urinary nickel and body compositions, with a low inflection point level of urinary nickel for the saturation effect.