Is Arsenic a Contributor to CKD?

 

 

 

Arsenic has a fascinating and checkered history. It has been a popular poison for at least 2,000 years because it is colorless and tasteless, yet it also commonly was used therapeutically in the 19th and early 20th century in Fowler's solution (1% potassium arsenite), Donovan's solution (arsenic iodide), and deValagin's solution (arsenic trichloride) to treat a range of maladies,1 and arsenic trioxide currently is prescribed as part of many leukemia treatments. Stories about arsenic appear periodically in the public sphere as a result of accidental poisonings from ingestion of old arsenical herbicides or use of arsenic in treated wood, although this use of chromated copper arsenate ceased in 2003. However, for the vast majority of the world's population, the dominant route of exposure is ingestion of naturally occurring arsenic in drinking water or in food cooked or irrigated with contaminated water. There are estimated to be more than 100,000,000 people worldwide exposed to arsenic levels >10 μg/L in drinking water (the recommended maximum level by the United States and the World Health Organization), with the largest population of concern in Bangladesh.2 Inorganic arsenic exposure through drinking water has been a public health concern since we first began testing for it in the 1940s; however, it is only in the past few decades that epidemiologic studies have been able to quantify the risk due to exposure.

 

Research in high-exposure populations (occupational settings or regions with naturally contaminated groundwater) has identified increased risk for a range of adverse health outcomes. Long-term exposure to inorganic arsenic in drinking water is a well-established risk factor for cardiovascular disease,3 diabetes,4 cancer,5 and skin disease,6 with some suggestion of an association with chronic kidney disease (CKD),7, 8, 9 among other adverse health outcomes. Given the known risks at higher levels of exposure, investigators now are turning to understanding associations from lower level exposures (<50 μg/L in drinking water) for which risk is equivocal.

 

In this issue of AJKD, Zheng et al10 look to build on findings from areas with high arsenic exposure by being the first to use a cohort to investigate the association between arsenic exposure and CKD in a low-to-moderate arsenic-exposure area. Zheng et al10 conducted a cross-sectional study in a well-powered cohort derived from the Strong Heart Study, a prospective study of risk factors for cardiovascular disease in American Indian men and women. The outcome measure, albuminuria, was defined as urine albumin level corrected for creatinine ≥30 mg/g. Exposure to arsenic was based on urine arsenic species concentration relevant to inorganic arsenic exposure (As3+, As5+, monomethylarsonic acid, and dimethylarsenic acid). After adjusting extensively for potential risk factors for kidney disease, regression models showed increasing prevalence with increasing creatinine-corrected urine arsenic levels (prevalence ratios of 1.16 [95% CI, 1.00-1.34], 1.24 [95% CI, 1.07-1.43], and 1.55 [95% CI, 1.35-1.78] for concentrations in the ranges of 5.8-9.7, 9.7-15.6, and ≥15.6 μg/g, respectively). These results were robust in analyses stratified by several covariates,10 indicating a consistent association across many subsets of their study population. Importantly, these findings are significant at relatively low urine arsenic levels similar to those commonly seen in the upper quartile of the US population per the National Health and Nutrition Examination Survey (NHANES),11 which suggests that inorganic arsenic exposure at low to moderate levels may have adverse kidney effects. If one were to assume a causal relationship, the attributable fraction of albuminuria due to arsenic exposure would be ∼25% in this study population and ∼6% in the total US population, for which ∼25% of the population has urine arsenic concentrations (As3+ + As5+ + monomethylarsonic acid + dimethylarsenic acid) higher than ∼8 μg per gram of creatinine.

 

The study by Zheng et al10 had several limitations. Most importantly, it is cross-sectional in design and also did not associate urinary arsenic concentrations to measured arsenic concentrations in drinking water directly. Scientific literature confirms a strong correlation between urine arsenic species and drinking water arsenic, especially in populations that do not smoke or do not have occupational exposures.12, 13, 14, 15 However, inorganic arsenic levels in groundwater in the 4 states of the Strong Heart Study are highly variable, with ranges from nondetectable to >50 μg/L, and urine arsenic concentrations cannot be linked directly with drinking water levels to inform public health policy.

 

Results from this study quantify the dose-response relationships of inorganic arsenic at low exposure, which supports research in areas with high exposure, including a recent case-control study in Taiwan8 that showed that patients with CKD had significantly greater total urinary arsenic levels compared with controls. This stratagem in arsenic research of determining adverse outcomes in high-exposure areas and then testing the association in moderate to low areas also has been seen in cardiovascular disease, cancer, and diabetes. As more physiologic systems are identified as being adversely affected by inorganic arsenic exposure through drinking water, the premise that arsenic is a systemic toxin and that there is no safe level of long-term exposure becomes more authenticated. However, substantially more research in low to moderate areas needs to be conducted in the future.

 

Future research of inorganic arsenic exposure, especially related to nephrotoxicity, can follow several directions. Experimental research to discern the complex mechanistic pathway is necessary to understand the nephrotoxicity of inorganic arsenic and answer questions such as whether inorganic arsenic is acting on the kidneys directly or perhaps also through diabetogenic effects. Epidemiologic studies need to continue to replicate and quantify associations at low levels of exposure, seeking to eliminate possible confounding factors. Exposure science research needs to develop methodology that can discern the additive or synergistic effects of other metals with known nephrotoxic effects, such as cadmium, thallium, and lead. Although Zheng et al10 did not identify an interaction between urine arsenic and cadmium levels and urinary albumin levels, another study in an area with high exposure found possible interactions with biomarkers for kidney disease.16 Last, research needs to determine the biologically relevant exposure period. Kidney disease is a complex syndrome with many mechanisms for initiation; determining when exposure to inorganic arsenic initiates disease action is important to assess exposure, and therefore risk, accurately.

 

In closing, Zheng et al10 made commendable efforts to ensure the strongest analytic plan to address the main hypothesis given the cross-sectional study design. This research is the first to identify an association between urinary arsenic concentrations and a biomarker for kidney disease, albuminuria, in a population with moderate to low exposure, thereby identifying another physiologic system adversely affected by long-term exposure to inorganic arsenic.

 

 

 

http://www.ajkd.org/article/S0272-6386(12)01508-9/fulltext