Abstract
Beyond its nutritional benefits, seafood is a source of toxicant exposure including polychlorinated biphenyls (PCB). The association of PCB exposure from seafood intake during pregnancy and/or lactation (PL) and child growth outcomes is uncertain. This systematic review investigated the evidence and quantified the association between PCB exposure during PL from seafood intake and child growth outcomes. Embase, PubMed, and CENTRAL databases were searched from their inception for peer-reviewed English articles. Records were screened independently by 2 researchers at title and abstract, and then full-text levels. Studies were included if they: 1) were conducted in a country with a high Human Development Index, 2) measured maternal PCB exposure directly, 3) assessed the relationship between PCB and seafood exposures or PCB or seafood associations with a child growth outcome, and 4) were randomized or nonrandomized interventions, cohort, or nested case-control studies. Pooled partial correlations (r(p)) were calculated using random-effects models for studies with sufficient data and narratively for the remaining studies. Cochrane ROBINS-E and GRADE tools were used to assess risk of bias and certainty of evidence, respectively. Child growth outcomes included birthweight, birth length, head and chest circumference at birth, and small for gestational age (SGA). Seven studies were included. PCB exposure during PL was weakly but significantly associated with lower birthweight [r(p) = -0.07; 95% confidence interval (CI): -0.12, -0.02; n = 5], but showed no association with birth length (r(p) = -0.04; 95% CI: -0.09, 0.02; n = 4) and head circumference (r(p) = -0.03, 95% CI: -0.09, 0.03; n = 3). Studies on SGA and chest circumference yielded inconclusive results. The certainty of the evidence was low or very low because of the risk of bias from confounding, missing data, and exposure misclassification. The evidence suggests minimal to no link between PCB exposure from seafood during PL on child growth outcomes but with low to very low certainty. This trial was registered at PROSPERO as CRD42023493302.