Blood Lead Is Associated with Latent Subgroups of Cognitive Performance in a Cross-Sectional Study of Uruguayan Children

摘要

Introduction. Lead exposure, even at low levels, harms child cognition. The effect of lead across individual IQ domains, however, has been inconsistent in the literature. Latent cognitive subpopulations were identified to elucidate effects of lead across several cognitive domains simultaneously. Methods. Among of 357 children 5-8 years old from Montevideo, blood lead levels (BLL) were measured with atomic absorption spectroscopy. Cognitive performance was assessed using 3 tests from the Cambridge Neuropsychological Test Automated Battery (CANTAB): Intra-Extra Dimensional Shift (IED), Spatial Span (SSP) and Stockings of Cambridge (SOC), as well as the Woodcock-Munoz cognitive battery (WM): Visual-Motor Integration, Verbal Comprehension, Visual-Auditory Comprehension, Concept Formation, Visual Spatial Thinking, Number Inversion and Spatial Relationship test. Bender Gestalt Test was also included. A total of 10 cognitive outcomes, reflecting differing domains of IQ were entered into a latent class analysis (LCA). After extracting cognitive classes, BLL was entered into an ordinal regression predicting cognitive class membership, controlling for socioeconomic and demographic confounders. Results. Mean BLL was 4.15 mg/dL and 31% children had BLL >5 mg/dL. Four latent classes were identified: A 'High Performing' class (14%), an 'Average' class (54%), a class performing well on SSP, but poorly on WM 'Poor WM / Intact Memory' (17%), and a class performing poorly on WM, SOC and IED 'Global Cognitive Impairment' (15%). Upper blood lead tertiles were associated with lower odds of membership in the higher performing classes: 2nd fertile OR = 0.50 (0.18, 1.40) and 3rd quintile OR = 0.54 (0.23, 1.25), although this did not reach statistical significance. Conclusions. Even at low concentrations, there is some evidence of a blood lead association with impairment across multiple cognitive domains.