Methods for accounting for urine dilution in arsenic analyses: comparisons to blood and water arsenic in FACT and FOX

Abstract

BACKGROUND AND AIM: Millions of people worldwide are exposed to arsenic (As). Urinary As (uAs) is a biomarker of As exposure. Urinary creatinine (uCr), a breakdown product of creatine, is used to correct uAs for urine dilution. However, uCr levels are influenced by meat intake, muscle mass, and factors related to As methylation/elimination. An alternative method utilizes specific gravity (SG), which has limitations in individuals with kidney damage and/or diabetes. We compared uAs corrected for uCr or SG vs. blood As (bAs) and water As (wAs), which do not require dilution adjustment. METHODS: We used data from 539 participants from the Folic Acid and Creatine Trial (FACT) and 343 participants from the Folate and Oxidative Stress (FOX) study. bAs and wAs were log2-transformed. Models were adjusted for log2-transformed uAs, age, sex, and BMI, and further adjusted for log2-transformed uCr or SG, separately. RESULTS:Median uAs/bAs/wAs concentrations were 113/8.4/102 and 140/12.3/114 µg/L in FACT and FOX, respectively. uCr and SG were highly correlated with each other (Spearman correlations0.86). A two-fold increase in total uAs was related to 35% and 24% increases in bAs and 18% and 13% increases in wAs concentrations with respective uCr and SG adjustment, in FACT; 40% and 23% increases in bAs and 14% and 8% wAs in FOX. CONCLUSIONS:The association of uAs with bAs and wAs was stronger after adjustment for uCr vs. SG in a Bangladeshi population exposed to high water As levels, supporting that adjusting for uCr is an adequate method to account for urine dilution. However, individual studies should consider whether their outcomes are impacted by factors that also influence uCr. Additionally, evaluation of populations with differing sociodemographic characteristics and levels of As exposure are needed to assess the consistency of these findings and optimization of urine dilution adjustment methods in As-related research. KEYWORDS: Biomarkers of exposure, exposure assessment