Elevated drinking water manganese and fetal and child health and development

Abstract

Manganese (Mn) is an essential element that functions as a cofactor in the metabolism of carbohydrates and proteins, and it is also incorporated in antioxidants such as superoxide dismutase. In general, intake of Mn mainly occurs via food. Additional or excess exposure might occur via drinking water, as elevated concentrations of Mn in ground water are prevalent worldwide. In recent years, several studies have shown associations between Mn concentrations in drinking water and adverse health effects in children. A few studies have concerned early-life exposure, but these have mostly been cross sectional in design or with a very short follow-up of the children. Prospective studies which follow the children from early intrauterine life to childhood and explore associated adverse effects are lacking.

The overall aim of the present thesis was to assess the potential effects of Mn exposure via drinking water during pregnancy and early-life on fetal outcomes (spontaneous abortion and perinatal mortality), size at birth, and child development (cognitive function and behavior). These prospective cohort studies were nested in a population-based trial involving food and micronutrient supplementation (Maternal and Infant Nutrition Interventions in Matlab; MINIMat), conducted from November 2001 to October 2003, in 4,436 pregnant women in Matlab, Bangladesh. A subsample of the children born within the MINIMat trial was selected for longitudinal follow-up of growth and development until 10 years of age. In Matlab, 70% of the drinking water wells contained high concentration of arsenic (>10 µg/L, mostly in shallow wells), a potent toxicant and carcinogen. To reduce the exposure to arsenic the inhabitants were recommended to install deeper wells. Later we revealed that over 40% of the wells (mainly deeper well >50 m), contained >400 µg Mn/L (previous WHO guideline value). The Mn exposure in the present studies was assessed by the concentrations in drinking water used by the mothers during pregnancy and by the children at 5 and 10 years of age, and measured by inductively coupled plasma mass spectrometry.

The median water Mn concentration has increased from about 200 µg/L during pregnancy to 339 µg/L when the children were 10 years old, whereas the arsenic concentrations decreased (median from 33 to 2.3 µg/L) during the same period. This is probably due to the ongoing installation of deeper wells. In early pregnancy (n=1,875), women in the highest tertile of water Mn concentrations (median=1,292 µg/L) had an approximately 35% reduced risk (Odds Ratio = 0.65, 95% CI 0.43, 0.99) of spontaneous abortion, compared with women in the lowest tertile (median=56 µg/L). This is possibly related to the role of Mn in the placental antioxidant defense. Elevated water Mn concentrations were not related to any increased risk of perinatal mortality.

The newborns (n=1,177) to mothers in the highest tertile of water Mn (median=1,495 µg/L) were on average 0.5 cm shorter (0.20 SD) compared to those in the lowest tertile (median 56 µg/L). The association was strongest in the girls, but apparent also in the boys of mothers with low hemoglobin (Hb) values (<114 g/L), likely due to higher absorption of Mn at low iron status. The findings indicate that elevated levels of Mn in drinking water during pregnancy may impair fetal growth. The association between erythrocyte Mn concentrations and size at birth was less apparent.

The prenatal Mn exposure, but not the postnatal, was associated with aggravation of the difficult behavior at 10 years of age (n=1,295). For each increase of one mg Mn/L, the difficult behavior scores increased by 0.5 points (0.13 SD) in boys and 0.7 points (0.18 SD) in girls of mothers with anemia (Hb<110 g/L). Associations of water Mn (especially at 5 years) with boys´ cognitive function were generally inverse (effect size ~0.15 SD per mg/L of water Mn). In girls, there was a positive association of maternal water Mn below 3 mg/L with cognitive function and pro-social behavior. However, a tendency of inverse associations with cognitive function were observed at higher Mn concentrations (>3 mg/L) and in girls of mothers with anemia. Early-life appeared to be a particularly susceptible period to inverse effects on development by elevated water Mn concentrations, although the effect differed by outcome, time-point of exposure and child gender.

In conclusion, the Mn concentrations in well water varied widely. Elevated concentrations during pregnancy appeared to be protective for early fetal loss. On the other hand, the exposure during pregnancy decreased fetal growth and impaired the children´s cognitive function and worsened their difficult behavior, especially in children of anemic women. Thus, Mn concentration in drinking water is an important public health concern worldwide. However, in arsenic contaminated areas, the benefit of deep wells with less arsenic is most likely higher than the modest adverse effects of Mn. In order to reduce the exposure to arsenic and Mn through well water, screening for both elements by sediment color tools could be done during installation of new wells, as well as digging wells in aquifers that are low in both arsenic and Mn (deeper than 100 m). As Mn absorption tends to be higher in anemic women, existing routine programs, including assessment of anemia during pregnancy and supplementation with iron and folic acid, should be strengthened. Also, screening and treating children with anemia can be proposed as an important public health intervention.