Association of drinking-water source and use characteristics with urinary antimony concentrations

Abstract

Environmental factors, such as storage time, frequency of bottle reuse and temperature, have been shown to facilitate antimony (Sb) leaching from water- and food-packaging materials. The globally escalating consumption of water packaged in Sb-containing bottles, such as that of polyethylene terephthalate (PET), could increase human daily Sb doses. This study set out to investigate the relationship between drinking-water source, use characteristics, and urinary Sb concentrations (U-Sb) accompanied with survey responses of a healthy (n=35) Cypriot participant pool. One spot urine sample was collected during administration of questionnaire, while a second spot urine sample was collected from the same individual about 7 days later. Urinary and water Sb concentrations were measured with an inductively coupled plasma mass spectrometer. Survey responses showed that bottled water summed over various volumes and plastic types, such as polycarbonate and PET contributed to an average 61% of daily water consumption. Water sources such as tap, mobile stations (explained in a following section), and well water contributed to 24%, 14%, and 2% of an individual's daily water consumption pattern, respectively. Average daily potable water use of both bottled and tap water by individuals consisted of 65% drinking-water, while the remaining 35% was water used for preparing cold and hot beverages, such as, tea, coffee, and juices. A significant (P=0.02) association between per capita water consumption from PET bottles and urinary creatinine-unadjusted concentrations was observed, but this relationship did not remain after inclusion of covariates in a multivariate regression model. In the creatinine-adjusted regression model, only gender (female) was a significant (P<0.01) predictor of U-Sb, after adjusting for several covariates. It is proposed that consumption data collection on various water uses and sources among individuals could perhaps decrease the uncertainty associated with derivations of acceptable daily Sb intakes.