UC Berkeley

Background

Arsenic is an immunotoxicant. Clinical reports observe the reactivation of varicella zoster virus (VZV) in people who have recovered from arsenic poisoning and in patients with acute promyelocytic leukemia that have been treated with arsenic trioxide.

Objective

We evaluated the association between arsenic and the seroprevalence of VZV IgG antibody in a representative sample of the U.S.

Methods

We analyzed data from 3,348 participants of the National Health and Nutrition Examination Survey (NHANES) 2003-2004 and 2009-2010 pooled survey cycles. Participants were eligible if they were 6-49 years of age with information on both VZV IgG and urinary arsenic concentrations. We used two measures of total urinary arsenic (TUA): TUA1 was defined as the sum of arsenite, arsenate, monomethylarsonic acid, and dimethylarsinic acid, and TUA2 was defined as total urinary arsenic minus arsenobetaine and arsenocholine.

Results

The overall weighted seronegative prevalence of VZV was 2.2% for the pooled NHANES sample. The geometric means of TUA1 and TUA2 were 6.57 μg/L and 5.64 μg/L, respectively. After adjusting for age, sex, race, income, creatinine, and survey cycle, odds ratios for a negative VZV IgG result in association with 1-unit increases in natural log-transformed (ln)-TUA1 and ln-TUA2 were 1.87 (95% CI: 1.03, 3.44) and 1.40 (95% CI: 1.0, 1.97), respectively.

Conclusions

In this cross-sectional analysis, urinary arsenic was inversely associated with VZV IgG seroprevalence in the U.S.

Population

This finding is in accordance with clinical observations of zoster virus reactivation from high doses of arsenic. Additional studies are needed to confirm the association and evaluate causal mechanisms.