Inorganic Arsenic in Drinking Water: An Evolving Public Health Concern

Abstract

In recent years, ingestion of inorganic arsenic from drinking water has emerged as an important scientific issue and public health concern. The International Agency for Research on Cancer has identified sufficient evidence in humans that inorganic arsenic causes lung, bladder, and nonmelanoma skin cancers ( 1 ) (summary at http://monographs.iarc.fr/ ). Reports by expert committees have suggested associations with several other cancer and noncancer outcomes ( 2 – 4 ). As a result, the US Environmental Protection Agency lowered the maximum contaminant level for arsenic in drinking water supplied by community water systems from 50 µ g/L, a level set in 1942, to 10 µ g/L, effective from January 2006. The reduction will necessitate changes to water systems serving approximately 13 million people ( http://www.epa.gov/safewater/arsenic/compliance. html ). The World Health Organization has similarly recommended 10 µ g/L for arsenic in water as a “guideline value” for setting national standards ( http://www.who.int/mediacentre/factsheets/fs210/en/ index.html ). Substantial numbers of people worldwide are exposed to elevated concentrations of arsenic. An estimated 25 million people in Bangladesh (19% of the population) and 6 million people in West Bengal, India (8% of the population), consume water with arsenic exceeding 50 µ g/L ( 5 ), with perhaps twice these numbers in these countries consuming water with arsenic levels exceeding 10 µ g/L. High levels of inorganic arsenic levels are found in drinking water in many other parts of the world, including areas of Argentina (county means ranged to 178 µ g/L) ( 6 ), the Inner Mongolia and Xinjiang autonomous regions in China (levels in excess of 600 µ g/L) ( 7 ), Finland (levels were generally low but ranged to 64 µ g/L) ( 8 ), northern Mexico (ranges in exposed areas to 160 – 740 µ g/L) ( 9 ), and the United States (with some areas in California, Nevada, Alaska, Michigan, New England, New Mexico, and Utah exceeding 50 µ g/L) (see US Geological Survey map of 31 350 ground water measurements at http://co.water.usgs.gov/trace/ pubs/arsenic_fi g1.html ). Drinking water arsenic levels in excess of 150 µ g/L also occur in Nepal ( 10 ), Thailand ( 11 ), Vietnam ( 12 ), Hungary ( 13 ), Ghana ( 14 ), and elsewhere ( 15 ). Although arsenic contamination of water supplies arises mainly from natural sources, human activities such as mining and ore processing may also contribute to elevated arsenic levels in some areas. Although ingestion of high levels of inorganic arsenic is most strongly linked to lung, bladder, and nonmelanoma skin cancers, there is evidence for associations with cancers of the kidney, liver, and possibly prostate ( 16 ). In addition, high levels of drinking water arsenic have been linked to cutaneous effects (hyperpigmentation, hypopigmentation, palmar – plantar hyperkeratoses, and leucomelanosis), gastrointestinal effects (gastrointestinal distress and liver cirrhosis), vascular effects (peripheral vascular diseases, such as blackfoot disease and Raynaud’s syndrome; arterial occlusions; cardiovascular disease; cerebrovascular disease; and hypertension), diabetes mellitus, and peripheral neuropathy ( 2 ), as well as to chronic cough, shortness of breath, and other respiratory effects ( 17 ). Arsenic can cross the placental barrier, and ingestion of high levels by pregnant women may cause adverse reproductive and developmental effects and increase rates of stillbirths, spontaneous abortions, low – birth weight deliveries, and neonatal and postneonatal mortality ( 18 , 19 ). Many of these associations were identifi ed from ecologic studies conducted in a limited number of populations; there have been relatively few analytic studies. Thus, many of the associations between arsenic ingestion and cancers other than lung, bladder, and nonmelanoma skin cancers and noncancer outcomes have not been conclusively demonstrated, and additional studies are needed. In this issue of the Journal, Marshall et al. ( 20 ) analyzed 50 years of mortality data for lung and bladder cancers, substantially extending a previous analysis ( 21 ). Data are ecologic and come from the climatologically and geologically unique setting of Region II in northern Chile. The climate is extremely arid, and the population lives primarily in cities and towns and relies almost exclusively on municipally supplied drinking water. Based on arsenic measurements made since the 1950s, arsenic levels within Region II have varied spatially and temporally, from highs of