Fish consumption in pregnancy, cord blood mercury level and cognitive and psychomotor development of infants followed over the first three years of life Krakow epidemiologic study

Background: Although the maternal fish consumption is supposed to have beneficial effects on development of infants, it may be harmful for child cognitive development since fish is a common source of methylmercury. Purpose of the study: Purpose of the study was to describe the usual pattern of fish consumption during pregnancy in Poland and explain the variability of prenatal mercury exposure due to fish intake by mothers. The other endpoint of the study was the assessment of the cognitive and psychomotor development of infants related to prenatal mercury exposure over the 3-year follow-up. Material and methods: The study sample consisted of 374 infants born at 33-42 weeks of gestation between January 2001 and March 2003 to mothers attending ambulatory prenatal clinics in the first and second trimesters of pregnancy. Total mercury level in whole cord blood was measured and the Bayley Scales of Infant Development (BSID-II) was used to assess the mental (MDI) and psychomotor developmental index (PDI) in children at 12, 24 and 36 months of age. Results: Self-reported weekly amount of fish consumption during the first two trimesters of pregnancy correlated positively with umbilical cord mercury concentrations (r(s)=0.22, p<0.0001). The corresponding correlation coefficient for the fish consumption in the third trimester of pregnancy was 0.21, p<0.0001. There was an inverse association between mercury exposure and both MDI (beta regression coeff.=-2.8, p=0.01) and PDI scores (beta coeff.=-2.3, p=0.04) at 12 months of age. Subsequent BSID-II testing at 24 and 36 months did not confirm significant association between exposure and cognitive or psychomotor function. The estimates of association between mercury prenatal exposure and the development of infants, which were based on the longitudinal analysis of all BSID-II measurements done in the follow-up (generalized estimating equations statistical model) showed that the performance deficit observed at 12 months of age was of border significance. (C) 2007 Elsevier Ltd. All rights reserved.