Variability of Metal Levels in Spot, First Morning, and 24-Hour Urine Samples over a 3-Month Period in Healthy Adult Chinese Men

BACKGROUND: Metals in single spot urine samples are often used to estimate individual exposure in human studies. However, measurements in urine could vary greatly over time due to variable exposure, potentially leading to exposure misclassification. OBJECTIVE: We examined the variability of metal levels in the urine of 11 men who provided 529 samples on 8 days during a 3-month period, which corresponds to the duration of spermatogenesis. METHOD: The urinary levels of arsenic (As), cadmium (Cd), cobalt (Co), copper (Cu), lead (Pb), molybdenum (Mo), and nickel (Ni) were measured using inductively coupled plasma-mass spectrometry. We calculated the intraclass correlation coefficients (ICCs) to assess the reproducibility of metal measures and computed the sensitivity and specificity to evaluate how well spot urine samples determined the individuals' 3-month average exposure. RESULTS: Fair to good reproducibility was observed for the serial measurements of Cd (ICC = 0.53) in spot samples collected during the 3-month period, whereas the serial measurements of As, Co, Cu, Pb, Mo, and Ni showed poor reproducibility (ICCs = 0.01-0.29). Use of single spot urine samples to classify the high (top 33%) 3-month average metal levels had uniformly high specificities (0.70-0.84) but relatively low sensitivities (0.40-0.57), except for Cd (0.77). The minimum number of specimens (kappa) required to estimate the participant-specific mean for the seven metals within 20% of the "true" values ranged from 3 for Cd to 27 for Ni. CONCLUSIONS: The high variability observed in the urinary levels of As, Co, Cu, Pb, Mo, and Ni suggests that a single measurement provides only a brief snapshot in time of the exposure levels of an individual, which can result in a moderate degree of exposure misclassification.