Objective To investigate the association between urinary metal concentrations and hyperuricemia in elderly men.

Methods In this cross-sectional study, elderly men were selected from a place in Guangxi from August 2016 to July 2017, and physical examination, urinary sample collection, and a questionnaire survey were performed. Inductively coupled plasma mass spectrometry was used to measure the concentrations of 17 metal elements in morning urinary samples, i.e., manganese, titanium, vanadium, iron, cobalt, nickel, copper, zinc, arsenic, molybdenum, cadmium, lead, selenium, rubidium, strontium, barium, and thallium. The Wilcoxon rank-sum test, the multivariate logistic regression model, and restricted cubic spline (RCS) functions were used to explore the association of urinary metal concentrations with hyperuricemia.

Results A total of 375 elderly male subjects were enrolled, among whom 86 (22.9%) were diagnosed with hyperuricemia. Compared with the non-hyperuricemia group, the hyperuricemia group had significantly lower urinary concentrations of vanadium, iron, arsenic, lead, rubidium, strontium, and thallium and a significantly higher urinary concentration of zinc (P < 0.05). RCS analysis showed a U-shaped curve for the correlation between urinary copper and serum uric acid and an inverted U-shaped curve for the correlation between urinary zinc and serum uric acid. Urinary iron, cobalt, rubidium, and strontium were negatively correlated with serum uric acid (P < 0.05), and the curve for the correlation between urinary rubidium and serum uric acid showed a downward trend, with a rapid reduction followed by a slow reduction (P_nonlinear < 0.05). The participants were divided into three groups according to the tertiles of urinary metal concentrations: T1 (low concentration), T2 (medium concentration), and T3 (high concentration) groups. The single-metal multivariate logistic regression model showed that the risk of hyperuricemia in the groups with T3 concentrations of zinc, strontium, and rubidium were 3.74 times (95% confidence intervalCI: 1.64-8.50), 0.42 times (95% CI: 0.16-0.93), and 0.35 times (95% CI: 0.13-0.99), respectively, that in the groups with T1 concentrations, and the multi-metal model showed that the risk of hyperuricemia in the zinc T3 concentration group was 4.96 times (95% CI: 2.09-11.81) that in the zinc T1 concentration group. The RCS function showed a negative dose-response relationship between urinary rubidium/strontium and hyperuricemia after adjustment for confounding factors (P < 0.05, P_nonlinear>0.05).

Conclusion Urinary levels of zinc, rubidium, and strontium may be associated with hyperuricemia in elderly men in this place of Guangxi. Urinary zinc is positively correlated with the risk of hyperuricemia, while urinary rubidium and strontium are negatively correlated with the risk of hyperuricemia.