Associations Between Urinary Metal Levels and Incident Heart Failure: A Multi-Cohort Analysis

Environmental metals are recognized cardiovascular disease risk factors, yet the role of metal exposure in heart failure (HF) risk remains understudied.

This study aims to evaluate the prospective association of urinary metals with incident HF across 3 geographically and ethnically/racially diverse cohorts: MESA (Multi-Ethnic Study of Atherosclerosis) and SHS (Strong Heart Study) in the United States, and the Hortega Study in Spain.

Adults 18-85 years of age in MESA (n = 6,601), SHS (n = 2,917), and Hortega (n = 1,300) were followed up to 20 years. Urinary levels of a multi-metal panel were measured at baseline and corrected for urine dilution. Cox proportional hazards and Cox elastic-net models were used to estimate the multi-adjusted (sociodemographic/clinical/lifestyle covariates) HR of incident HF by individual metals and the mixture of 5 metals available in all cohorts, respectively. The pooled HR (95% CI) of HF by 1-unit increase in log2-transformed levels of individual metals (ie, doubling of the dose) across cohorts was estimated using a fixed effects meta-analysis. Analyses by left ventricular ejection fraction were conducted in a subset.

A total of 1,001 participants developed HF. In adjusted models, significant associations (pooled HRs [95% CI] per doubling of urinary metal) were identified for cadmium (HR: 1.15 [95% CI: 1.07-1.24]) molybdenum (HR: 1.13 [95% CI: 1.05-1.22]), and zinc (HR: 1.22 [95% CI: 1.14-1.32]). The HRs (95% CIs) for the association of 1 IQR increase in the multi-metal mixture levels and incident HF were 1.38 (95% CI: 1.00-1.86) in MESA, HR: 1.55 (95% CI: 1.28-1.97) in SHS, and HR: 1.08 (95% CI: 0.85-1.63) in Hortega in fully adjusted models. Stratified models by left ventricular ejection fraction were consistent with the pooled results.

Urinary metals are risk factors of HF across 3 diverse populations, supporting the role of reducing metal exposures to lower HF risk.