The Sixteenth Conference of the International Society for Environmental Epidemiology (ISEE): Abstracts
Weaver, Virginia*; Jaar, Bernard*; Schwartz, Brian*; Ahn, Kyu-Dong†; Todd, Andrew‡; Wen, Jiayu*; Parsons, Patrick*; Lee, Byung-Kook†
*Johns Hopkins University Bloomberg School of Public Health; †SoonChunHyang University; ‡Mount Sinai Medical Center
We previously reported that tibia and blood lead were predictors of adverse renal function in lead workers. Recent evidence suggests that uric acid (UA) is nephrotoxic. Since lead is also reported to increase serum UA (urate), we evaluated whether these lead biomarkers were predictors of UA and whether relations between these biomarkers and renal function outcomes were altered after adjustment by UA.
UA was obtained in 803 lead workers. Renal function was assessed with blood urea nitrogen (BUN), serum creatinine, measured creatinine clearance, calculated creatinine clearance, and urinary N-acetyl-b-D-glucosaminidase (NAG) and retinol-binding protein (RBP). Lead measures included tibia and blood lead.
Mean (SD) UA, tibia and blood lead were 4.8 (1.2) mg/dl, 37.2 (40.4) mg/g bone mineral and 32.0 (15.0) mg/dl, respectively. After adjustment for age, gender, body mass index, systolic blood pressure and alcohol use, neither lead biomarker was associated with UA. However, when effect modification by age on these relations was examined, blood lead was a significant predictor of UA among workers in the oldest age tertile (age < 46.0 years; b=0.0105; 95th CI=0.0025-0.019). Analysis was next directed towards hypothesis generation regarding causal relations among lead, UA, and renal function. First, when serum creatinine was added to this model, the association was reduced (b=0.0071; 95th CI=-0.001-0.015). Next, effect modification by serum creatinine on relations between UA and the lead biomarkers was evaluated among the 266 oldest lead workers; blood lead was a predictor of UA only in workers with serum creatinine above the median (b=0.0150; 95th CI=0.004-0.026). Finally, the six renal function measures were modeled as outcomes. Previously published analyses in this population showed tibia and blood lead to be predictors of worse renal function primarily among workers in the oldest age tertile. When UA was added to these age interaction models, it was a significant predictor of all renal function measures except NAG. Betas for the lead biomarkers in NAG and RBP models were unchanged. However, betas for the lead measures in models with the four other renal outcomes decreased by as much as 20%.
These data suggest that older workers, particularly those with worse renal function, comprise a susceptible population for increased UA due to lead. Our data may also indicate that UA is one mechanism for lead-related nephrotoxicity.
© 2004 Lippincott Williams & Wilkins, Inc.