Relationship between prenatal methylmercury exposure and IQ: a meta-analysis.

Abstract

Developmental neurotoxicity is a well-established hazard of methylmercury (MeHg) exposure. A previous meta-analysis of three highly exposed historical cohorts (16.8-27.2 MeHg μg/L maternal blood) identified IQ decrements in MeHg-exposed children (Axelrad, 2007). It has been widely used in economic cost-benefit analyses. Their estimated decline in IQ was a decrease of 0.036 IQ points (95% CI: 0.010, -0.065) per MeHg μg/L maternal blood (converted from maternal hair). Because the cohorts used different IQ tests, scores were rescaled for this pooled analysis. Since 2007, additional epidemiology studies have examined the MeHg-IQ relationship at lower exposures (0.60-12.8 μg/L maternal blood). Yet, their findings have not been systematically synthesized. To address this gap, we conducted a meta-analysis, drawing on a comprehensive literature search of studies published between 1998 and 2025. Each study was evaluated for potential risk of bias and sensitivity. Quantitative findings from studies rated high or medium confidence were extracted and converted into a standardized exposure metric. A total of seven populations from five new cohorts examining the relationship between IQ (measured at ages 3.5-12 years) and prenatal MeHg exposure (0.60-12.8 μg/L maternal blood) were identified. A meta-analysis was performed using R package Metafor on the new studies that reported full-scale IQ. Based on the recent cohorts using cord or maternal blood biomarkers a decrement of 0.407 (-0.962, 0.147) IQ point per MeHg μg/L maternal blood was observed. Future cost-benefit analyses could consider the results of this updated analysis.