Iodine and pregnancy: a call to action.

292 www.thelancet.com Vol 382 July 27, 2013 Iodine, a crucial micronutrient, is essential for thyroid hormone production and for normal in-utero neurodevelopment. During pregnancy, iodine intake must be increased by 50% because of physiological increases in maternal thyroid hormone production, an increase in maternal renal iodine losses, and fetal iodine needs for thyroid hormone production. Iodine defi ciency aff ects 2 billion people worldwide and is the main cause of preventable mental impairment. Every year, 38 million newborn babies in developing countries are aff ected by iodine defi ciency despite ongoing and concerted eff orts by WHO, the UN, and the International Council for the Control of Iodine Defi ciency Disorders. Although severe iodine defi ciency is not an issue in developed countries, such as the UK, USA, and Australia, moderate defi ciency has re-emerged as an important public health concern. This re-emergence is a direct consequence of insuffi cient cohesive public health policies to eradicate iodine defi ciency. Starting in the 1920s in the USA, iodine defi ciency was eliminated through voluntary salt iodisation. Later, in the USA, UK, and Australia, adequate iodine intake within populations was serendipitously assured by the use of iodophors in the dairy industry, and of iodate conditioners in bread manufacturing in the USA. Because of decreases in dairy intake in the UK, changes in the use of iodophors in dairy production in the USA and Australia, decreases in the use of iodate dough conditioners by US bread manufacturers, and probable decreases in iodised salt consumption in the USA, moderate iodine defi ciency has re-emerged. A 2011 cross-sectional study of 810 UK schoolgirls aged 14–15 years showed moderate iodine defi ciency with a median urinary iodine excretion of 80·1 μg/L (IQR 56·9–109·0). Although the association between severe iodine defi ciency and neurodevelopment is well understood, research is scarce on the eff ects of more subtle degrees of maternal iodine defi ciency on child development. In The Lancet, Sarah Bath and colleagues document their examination of associations between maternal urinary iodine concentration and child cognitive development (IQ and reading ability at age 8–9 years) in 1040 fi rst-trimester pregnant women in the UK. The pregnant women were mildly-to-moderately iodine defi cient, with a median urinary iodine concentration of 91 μg/L (IQR 53·8–143); lower than the optimum range of 150–249 μg/L recommended by WHO. The urinary iodine concentration in these study participants is similar to that in participants of other, more recent, UK studies. Such defi ciency is probably caused by poor availability of iodised salt, few UK recommendations for increased iodine intake in pregnancy, and insuffi cient use of iodine-containing prenatal supplements. In adjusted and unadjusted analyses, Bath and colleagues found that the children of women with urinary iodine to creatinine ratios of less than 150 μg/g creatinine had signifi cantly lower scores for verbal IQ (odds ratio 1·58, 95% CI 1·09–2·30), reading accuracy (1·69, 1·15–2·49), and reading comprehension (1·54, 1·06–2·23) than did children of women with ratios of 150 μg/g or more. The investigators noted a possible dose-eff ect relation, with child neurocognitive scores worsening going from maternal ratios of 150 μg/g or more, to 50–150 μg/g, to less than 50 μg/g. Strengths of the study include its prospective design, large sample size, the undertaking of several sensitivity analyses, and adjustment for many potential confounders. An important limitation is the use of urinary iodine as a marker for individual iodine status. Because of a large amount of diurnal and day-to-day variation in urinary iodine concentrations, an estimated ten spot-urine samples are needed to achieve 20% precision in estimates of individual iodine intake. In this context, that the investigators could show a clear relation based on one maternal urinary iodine concentration is surprising, even with adjustments for urine creatinine and dichotomisation by greater and less than 150 μg/g and child cognition. Other study limitations include the potential for residual confounding and for socioeconomic bias resulting from the fact that only 56% of the Avon Longitudinal Study of Parents and Children (ALSPAC) cohort had a measure of IQ. Most, but not all, previous observational studies have shown that mildly low maternal thyroid function is associated with poorer neurocognitive outcomes in children. Women in most of these studies were iodine suffi cient. Three uncontrolled, non-randomised trials in regions of mild-to-moderate iodine defi ciency have studied the eff ects of iodine supplementation in pregnancy on child developmental outcomes. Two Iodine and pregnancy: a call to action