Mercury Exposure and Poor Nutritional Status Reduce Response to Six Expanded Program on Immunization Vaccines in Children: An Observational Cohort Study of Communities Affected by Gold Mining in the Peruvian Amazon

Abstract

Background: Poor nutritional status combined with mercury exposure can generate adverse child health outcomes. Diet is a mediator of mercury exposure and evidence suggests that nutritional status modifies aspects of mercury toxicity. However, health impacts beyond the nervous system are poorly understood. This study evaluates antibody responses to six vaccines from the expanded program on immunization (EPI), including hepatitis B, Haemophilus influenzae type B, measles, pertussis, tetanus, and diphtheria in children with variable hair mercury and malnutrition indicators. Methods: An observational cohort study (n = 98) was conducted in native and non-native communities in Madre de Dios, Peru, a region with elevated mercury exposure from artisanal and small-scale gold mining. Adaptive immune responses in young (3–48 months) and older children (4–8 year olds) were evaluated by vaccine type (live attenuated, protein subunits, toxoids) to account for differences in response by antigen, and measured by total IgG concentration and antibody (IgG) concentrations of each EPI vaccine. Mercury was measured from hair samples and malnutrition determined using anthropometry and hemoglobin levels in blood. Generalized linear mixed models were used to evaluate associations with each antibody type. Results: Changes in child antibodies and protection levels were associated with malnutrition indicators, mercury exposure, and their interaction. Malnutrition was associated with decreased measles and diphtheria-specific IgG. A one-unit decrease in hemoglobin was associated with a 0.17 IU/mL (95% CI: 0.04–0.30) decline in measles-specific IgG in younger children and 2.56 (95% CI: 1.01–6.25) higher odds of being unprotected against diphtheria in older children. Associations between mercury exposure and immune responses were also dependent on child age. In younger children, one-unit increase in log10 child hair mercury content was associated with 0.68 IU/mL (95% CI: 0.18–1.17) higher pertussis and 0.79 IU/mL (95% CI: 0.18–1.70) higher diphtheria-specific IgG levels. In older children, child hair mercury content exceeding 1.2 µg/g was associated with 73.7 higher odds (95% CI: 2.7–1984.3) of being a non-responder against measles and hair mercury content exceeding 2.0 µg/g with 0.32 IU/mL (95% CI: 0.10–0.69) lower measles-specific antibodies. Log10 hair mercury significantly interacted with weight-for-height z-score, indicating a multiplicative effect of higher mercury and lower nutrition on measles response. Specifically, among older children with poor nutrition (WHZ = −1), log10 measles antibody is reduced from 1.40 to 0.43 for low (<1.2 µg/g) vs. high mercury exposure, whereas for children with good nutritional status (WHZ = 1), log10 measles antibody is minimally changed for low vs. high mercury exposure (0.72 vs. 0.81, respectively). Conclusions: Child immune response to EPI vaccines may be attenuated in regions with elevated mercury exposure risk and exacerbated by concurrent malnutrition.