Arsenic Exposure From Drinking Water and the Incidence of CKD in Low to Moderate Exposed Areas of Taiwan: A 14-Year Prospective Study

Background Arsenic exposure is associated with decreased kidney function. The association between low to moderate arsenic exposure and kidney disease has not been fully clarified. Study Design The association between arsenic exposure from drinking water and chronic kidney disease (CKD) was examined in a long-term prospective observational study. Setting & Participants 6,093 participants 40 years and older were recruited from arseniasis-endemic areas in northeastern Taiwan. Arsenic levels were 28.0, 92.8, and 295.7 μg/L at the 50th, 75th, and 90th percentiles, respectively. Predictor Well-water arsenic and urinary total arsenic (inorganic plus methylated arsenic species) concentrations, adjusted for urinary creatinine concentration. Outcomes Kidney diseases (ICD-9 codes: 250.4, 274.1, 283.11, 403.*1, 404.*2, 404.*3, 440.1, 442.1, 447.3, or 580-589) and CKD (ICD-9 code: 585) ascertained using Taiwan’s National Health Insurance database 1998 to 2011. Measurements HRs contrasting CKD risk across arsenic exposure levels were estimated using Cox regression. Prevalence ORs for proteinuria (protein excretion ≥ 200 mg/g) comparing quartiles of total urinary arsenic concentrations were estimated using logistic regression. Results We identified 1,104 incident kidney disease cases, including 447 CKD cases (incidence rates, 166.5 and 67.4 per 104 person-years, respectively). A dose-dependent association between well-water arsenic concentrations and kidney diseases was observed after adjusting for age, sex, education, body mass index, cigarette smoking, alcohol consumption, and analgesic use. Using arsenic concentration ≤ 10.0 μg/L as reference, multivariable-adjusted HRs for incident CKD were 1.12 (95% CI, 0.88-1.42), 1.33 (95% CI, 1.03-1.72), and 1.33 (95% CI, 1.00-1.77) for arsenic concentrations of 10.1 to 49.9, 50.0 to 149.9, and ≥150.0 μg/L, respectively (P for trend = 0.02). The association between arsenic concentration and kidney diseases was stronger for women (P for interaction = 0.06). Arsenic values in the range of 50th to 75th and 75th to 100th percentiles of total urinary arsenic concentrations were associated with 50% and 67% higher prevalences, respectively, of proteinuria. Limitations Kidney diseases and CKD outcomes were based on diagnostic codes. Glomerular filtration rates were not available. Other heavy metals were not measured. Conclusions This study describes the temporal relationship between arsenic concentrations ≥ 10 μg/L in drinking water and CKD. A dose-dependent association between well-water arsenic concentration and kidney diseases was observed. Higher creatinine-adjusted urinary total arsenic concentrations were associated with a higher prevalence of proteinuria. Arsenic exposure is associated with decreased kidney function. The association between low to moderate arsenic exposure and kidney disease has not been fully clarified. The association between arsenic exposure from drinking water and chronic kidney disease (CKD) was examined in a long-term prospective observational study. 6,093 participants 40 years and older were recruited from arseniasis-endemic areas in northeastern Taiwan. Arsenic levels were 28.0, 92.8, and 295.7 μg/L at the 50th, 75th, and 90th percentiles, respectively. Well-water arsenic and urinary total arsenic (inorganic plus methylated arsenic species) concentrations, adjusted for urinary creatinine concentration. Kidney diseases (ICD-9 codes: 250.4, 274.1, 283.11, 403.*1, 404.*2, 404.*3, 440.1, 442.1, 447.3, or 580-589) and CKD (ICD-9 code: 585) ascertained using Taiwan’s National Health Insurance database 1998 to 2011. HRs contrasting CKD risk across arsenic exposure levels were estimated using Cox regression. Prevalence ORs for proteinuria (protein excretion ≥ 200 mg/g) comparing quartiles of total urinary arsenic concentrations were estimated using logistic regression. We identified 1,104 incident kidney disease cases, including 447 CKD cases (incidence rates, 166.5 and 67.4 per 104 person-years, respectively). A dose-dependent association between well-water arsenic concentrations and kidney diseases was observed after adjusting for age, sex, education, body mass index, cigarette smoking, alcohol consumption, and analgesic use. Using arsenic concentration ≤ 10.0 μg/L as reference, multivariable-adjusted HRs for incident CKD were 1.12 (95% CI, 0.88-1.42), 1.33 (95% CI, 1.03-1.72), and 1.33 (95% CI, 1.00-1.77) for arsenic concentrations of 10.1 to 49.9, 50.0 to 149.9, and ≥150.0 μg/L, respectively (P for trend = 0.02). The association between arsenic concentration and kidney diseases was stronger for women (P for interaction = 0.06). Arsenic values in the range of 50th to 75th and 75th to 100th percentiles of total urinary arsenic concentrations were associated with 50% and 67% higher prevalences, respectively, of proteinuria. Kidney diseases and CKD outcomes were based on diagnostic codes. Glomerular filtration rates were not available. Other heavy metals were not measured. This study describes the temporal relationship between arsenic concentrations ≥ 10 μg/L in drinking water and CKD. A dose-dependent association between well-water arsenic concentration and kidney diseases was observed. Higher creatinine-adjusted urinary total arsenic concentrations were associated with a higher prevalence of proteinuria.