Pregnancy

Iodine deficiency, when severe, has multiple adverse effects on growth and development in animals and human subjects, collectively termed the iodine-deficiency disorders⁽Reference Marine and Kimball^3, Reference Zimmermann⁴⁾. The adverse effects on pregnancy of mild iodine deficiency, typical of industrialised countries, remain unclear. Six controlled trials of iodine treatment in pregnant women with iodine deficiency have been conducted in Europe⁽Reference Romano, Jannini and Pepe^21–Reference Antonangeli, Maccherini and Cavaliere²⁶⁾. Overall, these studies suggest that in areas of mild-to-moderate iodine deficiency the maternal thyroid is able to adapt to meet the increased thyroid hormone requirements of pregnancy⁽Reference Zimmermann²⁷⁾. Although supplementation in these studies was generally found to be effective in minimising an increase in thyroid size during pregnancy, only two of the six studies report that maternal TSH is lower (within the normal reference range) with supplementation, and none of the studies show a marked impact of supplementation on maternal and newborn total or free thyroid hormone concentrations⁽Reference Zimmermann²⁷⁾. Thyroid hormone concentrations may be the best surrogate biochemical marker for healthy fetal development. Thus, the results of these trials are reassuring. However, because none of the trials measured long-term clinical outcomes such as maternal goitre or infant development, the potential adverse effects of mild-to-moderate iodine deficiency during pregnancy remain uncertain. In industrialised countries with iodine deficiency it has often been suggested that pregnancy is an environmental factor contributing to a higher prevalence of goitre and thyroid disorders in women, compared with men. However, the data to support this premise are scarce. In European studies data from an uncontrolled prospective study in ten women⁽Reference Glinoer, Lemone and Bourdoux²⁸⁾, a retrospective study⁽Reference Rotondi, Amato and Biondi²⁹⁾ and a cross-sectional study in smoking women⁽Reference Knudsen, Bulow and Laurberg³⁰⁾ suggest goitres formed during pregnancy may only partially regress after parturition.

Childhood

There have been many cross-sectional studies comparing cognition and/or motor function in children from chronically iodine-deficient and iodine-sufficient areas, including children from European backgrounds⁽Reference Vermiglio, Sidoti and Finocchiaro^31–Reference Vermiglio, Lo Presti and Moleti³⁵⁾. These cross-sectional studies generally report reduced intellectual function and motor skills in children from iodine-deficient areas. However, observational studies are often confounded by other factors that affect child development. Furthermore, these studies could not distinguish between the persistent effects of in utero iodine deficiency and the effects of current iodine status. Two meta-analyses have reported on this issue⁽Reference Bleichrodt, Born and Stanbury^36, Reference Qian, Wang and Watkins³⁷⁾, but they have included mainly data from regions of chronic severe iodine deficiency; in such areas populations experience a mean reduction in intelligence quotient of 12–13·5 points.

In a recent placebo-controlled double-blind 6-month intervention trial 10–12-year-old children with moderate iodine deficiency (n 310) in Albania were randomised to receive either 400 mg iodine as oral iodised oil or placebo. It was found that compared with placebo iodine treatment improves performance on tests of information processing, fine motor skills and visual problem solving⁽Reference Zimmermann, Connolly and Bozo³⁸⁾. These findings have recently been confirmed in mildly-iodine-deficient New Zealand children⁽Reference Gordon, Rose and Skeaff³⁹⁾. Thus it appears that in children born and raised in areas of moderate iodine deficiency cognitive impairment is at least partially reversible by iodine repletion⁽Reference Zimmermann, Connolly and Bozo^38, Reference Gordon, Rose and Skeaff³⁹⁾. In addition, recent controlled studies clearly demonstrate that iodine repletion in school-age children with iodine deficiency increases insulin-like growth factor-1 and insulin-like growth factor-binding protein-3 concentrations; in children who have moderate-to-severe deficiency their somatic growth is improved⁽Reference Zimmermann, Jooste and Mabapa⁴⁰⁾. Chronic iodine deficiency increases the TSH concentration and produces a thyroid hormone pattern consistent with subclinical hypothyroidism⁽Reference Zimmermann¹¹⁾, and subclinical hypothyroidism may increase the risk of CVD. Iodised oil rapidly normalises the increased TSH concentrations found in individuals who are iodine-deficient⁽Reference Zimmermann¹¹⁾ and thus corrects subclinical hypothyroidism. An uncontrolled study has reported that iodine treatment of German adolescents who are goitrous decreases plasma cholesterol concentrations⁽Reference Rönnefarth, Kauf and Deschner⁴¹⁾. A recent controlled study has reported that in children with moderate iodine deficiency and elevated TSH concentrations as a result of iodine deficiency iodine treatment of improves their lipid profile and reduces their insulin (C-peptide) levels compared with controls⁽Reference Zimmermann, Aeberli and Melse-Boonstra⁴²⁾. This previously unrecognised benefit of iodine prophylaxis may be important because iodine deficiency remains common in many industrialised countries with increasing rates of obesity and CVD.

Salt fortification with iodine

In the majority of industrialised countries affected by iodine deficiency the most effective way to control iodine deficiency is through salt iodisation⁽¹²⁾. Salt iodisation is the recommended strategy for control of iodine deficiency because: (1) salt is one of few foodstuffs consumed by virtually every individual; (2) salt intake is fairly consistent through the year; in many countries salt production and importation is limited to a few sources; (3) iodisation technology is simple and relatively inexpensive to implement; (4) the addition of iodine to salt does not affect its colour or taste; (5) the quantity of iodine in salt can be simply monitored at the production, retail and household levels. WHO/UNICEF/International Council for the Control of Iodine Deficiency Disorders recommends that iodine is added at a level of 20–40 mg iodine/kg salt, depending on local salt intake⁽¹²⁾. Salt iodisation remains the most cost-effective way of delivering iodine and of improving cognition in populations who are iodine deficient⁽Reference Engle, Black and Behrman⁴³⁾. Worldwide, the annual costs of salt iodisation are estimated to be US$ 0·02–0·05 per child covered, and the costs per child death averted are US$ 1000 and per disability-adjusted life-year gained are US$ 34–36⁽Reference Caulfield, Richard, Rivera, Jamison, Breman, Measham, Alleyne, Claeson, Evans, Jha, Mills and Musgrove⁴⁴⁾.

The importance of food industry use of iodised salt in industrialised countries

WHO recommends universal salt iodisation to control iodine deficiency⁽¹²⁾. Universal salt iodisation is a term used to describe the iodisation of all salt for human (food industry and household) and livestock consumption. Although the ideal, even in countries with successful salt iodisation programmes, universal salt iodisation is rarely achieved, as many countries do not iodise all salt for livestock and food industries are often reluctant to use iodised salt. The limited use of iodised salt by the food industry is one of the main reasons for iodine deficiency in many industrialised countries. In countries such as England and Denmark household salt, i.e. salt for cooking and table use, contributes only 8–15% to total salt intakes; most of the remainder comes from purchased processed foods⁽Reference Sanchez-Castillo, Warrender and Whitehead^45, Reference Andersen, Rasmussen and Larsen⁴⁶⁾. Thus, in industrialised countries, where on average total adult per capita salt consumption is approximately 10 g/d, only about 1 g household salt is consumed daily⁽Reference Andersen, Rasmussen and Larsen⁴⁶⁾. If the food industry uses non-iodised salt and only household salt is iodised (at WHO recommended levels of 20 mg iodine/kg salt), then total iodine intake from iodised salt will only be approximately 20 μg/d, far below the daily requirement for all age-groups. Thus, in order to successfully control iodine deficiency in industrialised countries it is critical to convince the food industry to begin to use iodised salt in their products. For example, in The Netherlands all salt used by the baking industry is iodised, and this source is the major contributor to iodine sufficiency in the country⁽Reference Andersson, de Benoist, Darnton-Hill and Delange²⁰⁾. Switzerland's long-running iodised salt programme has been successful partly because approximately 60% of salt used by the food industry is iodised on a voluntary basis. As iodine at μg/g levels in iodised salt does not cause colour or flavour changes in any foods and in most countries the price difference between iodised and non-iodised salt is negligible, there should be no major barriers to adoption of iodised salt by food manufacturers.

How to reconcile the policy of reducing salt consumption to prevent chronic diseases and the policy of universal salt iodisation to eliminate iodine deficiency

Current WHO recommendations to prevent chronic diseases state that the population average per capita consumption of salt should be <5 g/d (<2 g Na/d)⁽⁴⁷⁾. However, the WHO recommendation that salt be fortified with iodine at a level of 20–40 μg/g is based on the assumption of an average salt intake of approximately 10 g/d at the population level⁽¹²⁾. Thus, the level of iodine fortification may need to be adjusted by national salt iodisation programmes if health authorities are successful in reducing salt consumption. Since iodisation methods can easily fortify salt with adequate concentrations of iodine to provide recommended iodine intakes even if per capita salt intake falls to <5 g/d, these two important health policies are not in conflict⁽⁴⁷⁾. Interaction with food manufacturers in industrialised countries will be fundamental to the success of both salt-reduction strategies and to universal salt iodisation.

Other fortification vehicles

Bread can be an effective vehicle for iodine by including baker's salt enriched with iodine⁽Reference Seal, Doyle and Burgess⁴⁸⁾. Although iodising drinking water or irrigation water can also be effective⁽Reference Squatrito, Vigneri and Runello⁴⁹⁾, the higher cost and the complexity of monitoring are disadvantages. In Finland iodine-fortified animal fodder has increased the iodine content of foods derived from animal sources. In countries with iodine deficiency iodine should be routinely added to complementary foods or add-on formulas for infants during weaning to provide approximately 90 μg iodine/d⁽Reference Dunn⁵⁰⁾.

Iodine supplementation

In some regions iodisation of salt may not be practical for control of iodine deficiency, at least in the short term, and iodine supplements can be used (Table 4)⁽¹²⁾. Iodised oil supplements are prepared by esterification of the unsaturated fatty acids in seed or vegetable oils and addition of iodine to the double bonds. It can be given orally or by intramuscular injection⁽¹²⁾. The intramuscular route has a longer duration of action, but oral administration is more common because it is simpler. Usual doses are 200–400 mg iodine/year⁽¹²⁾ and it is often targeted to women of child-bearing age, pregnant women and children. Its disadvantages are an uneven level of iodine in the body over time and the need for direct contact with individuals with the accompanying increased costs. Iodine can also be given as potassium iodide or potassium iodate as drops or tablets. Single oral doses of potassium iodide monthly (30 mg) or biweekly (8 mg) can provide adequate iodine for school-age children⁽Reference Todd and Dunn⁵¹⁾.

Table 4.

Recommendations for iodine supplementation in pregnancy and infancy in areas in which <90% of households are using iodised salt and the median urinary iodine is <100 μg/l in schoolchildren⁽¹²⁾

RNI, reference nutrient intake.

Strategies to prevent or correct deficiency during pregnancy and lactation

For nearly all industrialised countries the primary strategy for sustainable elimination of iodine deficiency in pregnancy remains salt iodisation⁽¹²⁾. In countries or regions in which a salt iodisation programme covers ≥90% of households and has been sustained for ≥2 years, and the median UI indicates iodine sufficiency (Table 4), pregnant and lactating women do not need iodine supplementation⁽¹²⁾. Several countries with long-standing successful iodised salt programmes have reported an optimal median UI in pregnant women⁽Reference Zimmermann²⁷⁾. Moreover, in countries affected by mild or moderate iodine deficiency (Republic of Ireland, Germany, Belgium, Italy, Denmark) thyroid volume increases 15–31% during pregnancy, while in iodine-sufficient countries (Finland, The Netherlands) there is little or no increase in thyroid volume during pregnancy⁽Reference Berghout and Wiersinga⁵²⁾. These data suggest that effective salt iodisation can provide adequate iodine intake during pregnancy, but iodine-containing supplements taken during the prenatal period may have contributed to iodine intakes in these studies.

Implementation of salt iodisation is not always feasible, and non-implementation may result in inadequate iodine supply for women of child-bearing age and pregnant women. In this situation supplementation of these groups should be considered. WHO recommends that countries assess their salt iodisation programmes and then decide whether supplementation is indicated⁽¹²⁾. In order to ensure adequate iodine supply during pregnancy women should ideally be provided with adequate iodine intake (≥150 μg/d) for a long period before conception to ensure plentiful intrathyroidal iodine stores⁽Reference Moleti, Lo Presti and Campolo⁵³⁾. In Italy thyroid function in pregnant women from a mildly-iodine-deficient area who had regularly used iodised salt for ≥2 year before becoming pregnant was compared with that in women who began using iodised salt on becoming pregnant⁽Reference Moleti, Lo Presti and Campolo⁵³⁾. The findings suggest that prolonged use of iodised salt is associated with better maternal thyroid function, possibly as a result of having greater intrathyroidal iodine stores to draw on during pregnancy. In iodine-deficient countries or regions that have weak iodised salt distribution, i.e. in countries or areas in which <90% of households are using iodised salt and the median UI is <100 μg/l in schoolchildren, supplements should be given to pregnant women, lactating women and infants, according to the strategy shown in Table 4⁽¹²⁾.

France

Iodised household salt was introduced in France on a voluntary basis in 1952 (at 10–15 mg sodium iodide/kg salt)⁽Reference Valeix, Zarebska and Preziosi⁵⁴⁾. In the mid-1990s iodine status was investigated in a nationally-representative sample of French adults (4860 men aged 45–60 years and 7154 women aged 35–60 years)⁽Reference Valeix, Zarebska and Preziosi⁵⁴⁾. The median UI in males and females was reported to be 85 and 82 μg/l respectively, indicating mild iodine deficiency, with an overall goitre rate of 11% for males and 14% for females. The prevalence of individuals with an adequate UI was found to be lowest among 55–60 year olds, with the only 35% of men and 34% of women in this age-group having a UI ≥100 μg/l. The age-related decrease in iodine intake was thought to be a result of a reduction in energy intake with aging and/or increased awareness of the health benefits of reducing Na intake, which may have reduced iodine intake from salt. Median UI were found to display wide marked regional variations; median UI were higher among residents living in the north-west and west than among those living in the centre and east, independent of age and gender. The central region (Auvergne, Limousin) and the north-eastern areas (Alsace, Lorraine, Champagne-Ardennes, Franche-Comté) were found to have the highest percentages (33 and 24 for women and 22 and 23 for men respectively) with UI values <50 μg/l. It was suggested that these regional variations in median UI were a result of variations in the iodine contents of local foodstuffs, as they are dependent on native soil and water iodine content as well as incidental addition of iodine to foods by agricultural practices, including iodine-containing fertilisers, herbicides, disinfectants and/or iodine-supplemented livestock feed⁽Reference Valeix, Zarebska and Preziosi⁵⁴⁾.

In contrast to the results in adults, a study of approximately 1500 French children aged 6–14 years conducted in the mid-1990s has reported iodine sufficiency in four areas of France: Lorraine; Rhône-Alpes; Languedoc-Roussillon; Midi-Pyrénées⁽Reference Caron, Jaffiol and Leclère⁵⁵⁾. The mean UI was found to be 130 (sd 74) μg/l (only 10% of UI values <50 μg/l), with the prevalence of goitre, determined by ultrasound, being 4·1% in boys and 3·1% in girls.

Two studies in pregnant women in southern France have confirmed iodine deficiency in this group. Mean UI in pregnant women in south-west France (n 347) was found to be 69 (sd 4) μg/l at initial presentation (before 12 weeks of gestation) and 86 (sd 6) μg/l during the ninth month of pregnancy⁽Reference Caron, Hoff and Bazzi⁵⁶⁾. During pregnancy a marked decrease was found in free thyroxine and triiodothyronine concentrations, with an increase in TSH and thyroglobulin concentrations. A thyroid ultrasound was performed 1–5 d after delivery in 246 mothers. Thyroid hypertrophy (defined as thyroid volume >18 ml) was found to affect 15·4% of women whose first trimester UI was <50 μg/l, but was present in only 3·5% of women whose first trimester UI was >100 μg/l; 11% of women were found to be goitrous (thyroid volume >22 ml). Similar findings have recently been reported in pregnant women in the third trimester of pregnancy in the Nice area (n 330); median UI was found to be 64 μg/l, with 86% of women having a UI <150 μg/l⁽Reference Hiéronimus, Bec-Roche and Ferrari⁵⁷⁾. Median free thyroxine was found to be 12·3 pmol/l, with 41·2% of values being <12 pmol/l. Median TSH was reported to be 1·93 mU/l, with 26·3% having a TSH >2·5 mU/l. Thus, iodine deficiency is common in pregnant women in southern France and may be associated with hypothyroxinaemia and hyperthyrotopinaemia.

The USA

Recent data from the USA indicate the general population is iodine sufficient, and iodine nutrition has improved since the 1960s–1970s when much of the population was consuming excess dietary iodine. Based on the success of the classic iodine supplementation studies in Ohio, general prophylaxis with iodised salt was first introduced in the state of Michigan in 1924⁽Reference Zimmermann⁴⁾. In 1948 the USA Endemic Goitre Committee tried to introduce iodised salt to all states by federal law, but the bill failed⁽Reference Zimmermann⁴⁾. However, the voluntary use of iodised salt increased across the country and adventitious iodine entered processed foods as iodine-containing compounds were used as bread conditioners and in the dairy industry. These trends were associated with a decline in the prevalence of goitre⁽Reference Hollowell and Haddow⁵⁸⁾.

By the 1970s iodine excess had become a greater concern than iodine deficiency, as iodine intakes were estimated to be 150–700 μg/d⁽Reference Robbins⁵⁹⁾. In a ten-state study of approximately 36 000 individuals in 1975 the median UI was found to be 250 μg creatinine/g, with the goitre prevalence in all age-groups at 3·1%; a higher prevalence of goitre was found among individuals with high UI values than those with low values⁽Reference Trowbridge, Hand and Nichaman⁶⁰⁾. In a subsequent study in approximately 7800 USA children aged 9–16 years the prevalence of goitre was reported to be 6·8%; again with goitre being associated with higher, not lower, UI⁽Reference Trowbridge, Matovinovic and McLaren⁶¹⁾. However, the higher iodine intakes were not found to be associated with discernable thyroid dysfunction.

In the early 1970s the NHANES began to track iodine nutrition in the USA population. NHANES I (1971–1974) reported a median UI of 320 μg/l⁽Reference Hollowell and Haddow⁵⁸⁾. Subsequently, results from the USA Total Diet Study suggested a decreased total iodine content of the food supply between 1984 and 1991⁽Reference Pennington and Schoen⁶²⁾, at least partially because of a reduction in the amount of iodine in milk and the replacement of iodine compounds with bromine salts by the baking industry. Although there are no national data on iodised salt use, it is estimated that only about 60% of salt consumed in the USA is iodised (added as potassium iodide at a concentration of 77 mg iodine/kg salt)⁽Reference Hollowell and Haddow⁵⁸⁾.

NHANES III (1988–1994) reported a median UI of 145 μg/l, less than half the median in NHANES I⁽Reference Hollowell and Haddow⁵⁸⁾. An increase in individuals with UI concentrations <50 μg/l was found (11·6% in 1988–94 v. 2·4% in 1971–4). However, the frequency of UI concentrations >500 μg/l was reported to be reduced from 27·8% to 5·3%, with the frequency of UI >1000 μg/l falling from 5·3% to 1·3%. Thus, using WHO criteria⁽¹²⁾, the data from NHANES III suggested an improvement in iodine status in the USA population, indicated by a sharp decrease in the prevalence of iodine excess, with the median UI in the middle of the range (100–200 μg/l) signifying adequate iodine status. However, concern was expressed that iodine intakes would continue to fall, putting the population at risk of deficiency. This situation has not arisen; in NHANES III the median UI was shown to be stable over the 6 years of sampling and NHANES 2000 and 2001–2002 have reported median UI of 161 μg/l and 168 μg/l respectively⁽Reference Hollowell and Haddow⁵⁸⁾. Thus, it appears the iodine intakes in the USA have stabilised. However, in the 2001–2002 NHANES the UI of pregnant women was reported to be 173 (95% CI 75, 229) μg/l, within the range recommended by WHO (150–249 μg/l), but the lower 95% CI was <150 μg/l, partly because of small sample size⁽Reference Becker and Braverman⁶³⁾. Thus, until additional data become available, in order to ensure adequate iodine intake to maintain euthyroidism for the mother and fetus during pregnancy, the American Thyroid Association has recommended that women receive 150 μg iodine supplements daily during pregnancy and lactation and that all prenatal vitamin–mineral preparations contain 150 μg iodine⁽Reference Becker and Braverman⁶³⁾.

Australia

Australia is an industrialised country in which mild iodine deficiency has reappeared, after having been controlled for decades mainly by adventitious iodine in milk and milk products. In 1992 the median UI in the Australian population was reported to be >200 μg/l, indicating iodine sufficiency⁽Reference Li, Eastman and Waite⁶⁴⁾. A national survey in 2003–4 of schoolchildren in mainland Australia aged 8–10 years (n 1709) has reported a median UI of 104 μg/l⁽Reference Li, Eastman and Waite⁶⁴⁾. In New South Wales and Victoria children were found to be mildly iodine deficient, with median UI of 89 μg/l and 73·5 μg/l respectively. Based on WHO reference values by age, 5·6 and 9·3% of boys and girls respectively were found to have increased thyroid volume, but no significant association was demonstrated between UI and thyroid volume. These results indicate marginal iodine intake in the Australian population at the national level, with several of the states being mildly iodine deficient. Tasmania was not included in this survey, but in a study in 2001, before bread fortification with iodine began, the median UI in Tasmanian schoolchildren was reported to be 84 μg/l, with 20% of values <50 μg/l⁽Reference Guttikonda, Burgess and Hynes⁶⁵⁾.

These data are supported by a recent cross-sectional study of 815 pregnant women (≥28 weeks of gestation) and 824 newborns in New South Wales⁽Reference Travers, Guttikonda and Norton⁶⁶⁾. Spot urine samples were measured for iodine concentration in the women and TSH was measured in newborns at 48–72 h after birth using a sensitive fluoroimmunoassay. The median UI in pregnant women was reported to be 85 μg/l, indicating mild iodine deficiency, with eighteen newborns (2·2%) having TSH values >5 mIU/l. No correlation was found between neonatal TSH and maternal UI, but mothers with a UI <50 μg/l were shown to be 2·6 times (relative risk 2·65 (95% CI 1·49, 4·73)) more likely to have a baby with a TSH level >5 mIU/l.

The major sources of dietary iodine in the Australian diet are thought to be milk and milk products, seafood and iodised salt⁽Reference Li, Waite and Ma⁶⁷⁾. However, few individuals purchase iodised salt and, except in Tasmania, few if any food manufacturers use iodised salt in the preparation and manufacture of foods. It has been suggested that for the last 40–50 years the largest source of iodine in the Australian diet has been milk containing iodine residues from sanitising iodophors used in dairying. In a 1975 survey mean iodine concentrations in cow's milk in southern Australia were reported to be 583–593 μg/l⁽Reference Li, Waite and Ma⁶⁷⁾. As a result of concerns that high iodine levels in milk could lead to iodine toxicity Food Standards Australia specified a milk iodine limit of 500 μg/l in the Food Standards Code of 1982⁽⁶⁸⁾, which led to the replacement of iodophors by chlorine-containing sanitisers. Iodine content of milk samples in Sydney from 2001 to 2004 (thirteen samples per year) were found to be highly variable but generally much lower than in the 1970s, with median concentrations of 140 (range 60–220) μg/l in 2001 and 195 (range 66–412) μg/l in 2004⁽Reference Li, Waite and Ma⁶⁷⁾. It was suggested that the reduced amount of iodine in milk is a major reason why iodine deficiency has reappeared in regions of Australia. In 2009 a decision was made by Food Standards Australia New Zealand to address iodine deficiency in the population by requiring the salt in bread to be replaced with iodised salt⁽⁶⁹⁾.

The UK and the Republic of Ireland

The re-emergence of iodine deficiency in Australia should raise concern in the UK, as the pattern of a rise and fall in iodine intakes as a result of changes in milk iodine levels may recur in Britain. Studies in the UK in the 1800s reported a high prevalence of goitre and occasional cretinism⁽Reference Phillips⁷⁰⁾. By 1900 the English goitre belt was well-defined; it extended from the West Country through Somerset and northwards into Derbyshire and the Peak District, with offshoots into North and South Wales. The high regional goitre rates in adults and children in the 1940s led to recommendations by the Medical Research Council in 1944 and 1948 for a programme of iodised salt throughout the UK, but no action was taken⁽⁷¹⁾. In Oxfordshire the goitre prevalence between 1948 and 1958 was found to be essentially unchanged in boys but increased in girls, from 27% to 40%⁽Reference Hughes, Rodgers and Wilson⁷²⁾. Goitre remained a health problem into the early 1960s in Sheffield and Wales⁽Reference Trotter, Cochrane and Benjamin^73, Reference Kilpatrick, Milne and Rushbrooke⁷⁴⁾, as well as in Glasgow, where patients presenting with non-toxic goitre were reported to have low dietary iodine intake and low plasma inorganic iodine concentrations⁽Reference Phillips⁷⁰⁾. Although monitoring data are scarce, it appears that since the 1960s the prevalence of iodine deficiency and goitre has fallen in the UK. A 1990 survey has reported the absence of thyroid enlargement in children in a traditionally iodine-deficient area of South Wales⁽Reference Lazarus, Phillips, Parkes, Delange, Dunn and Glinoer⁷⁵⁾.

The improvement in iodine intakes dating from the 1960s in the UK may have been a result, at least in part, of increasing iodine intake from milk and milk products. Iodine supplementation of livestock improves their reproductive performance and many UK dairy herds, beginning as early as the 1920s and 1930s, have been given iodine via iodised salt licks and/or iodine-enriched cattle feed⁽Reference Phillips⁷⁰⁾. This practice increases the iodine content of milk, particularly during the winter months, when cattle are more dependent on iodine-rich artificial feed. In addition, the use of iodophor disinfectants for cleaning teats and milk tankers and giving iodinated casein to cows to promote lactation have also contributed to higher iodine levels in milk⁽Reference Phillips⁷⁰⁾. The simultaneous increase in per capita milk consumption has resulted in the amount of iodine provided by milk in the UK sharply increasing; it was estimated that average daily iodine intake from milk increased between 1930 and 1980 from approximately 5 μg to 25 μg (in summer) and >100 μg (in winter)⁽Reference Phillips⁷⁰⁾. The increasing contribution of milk iodine is likely to be a major reason why estimated dietary iodine intakes in the UK increased between 1952 and 1982 from 80 μg to 255 μg/d⁽Reference Wenlock, Buss and Moxon⁷⁶⁾.

However, several UK authors have suggested that the increasing trend to replace iodine compounds used in the dairy and baking industries with non-iodine alternatives may be decreasing iodine intakes, similar to the situation that has occurred in Australia⁽Reference Lazarus and Smyth^77–Reference Smyth, Hetherton and Smith⁸²⁾. If adventitious dietary iodine falls, the population may be vulnerable to iodine deficiency because iodised salt makes up <5% of all salt consumed in the UK and the Republic of Ireland⁽Reference Lazarus and Smyth⁷⁷⁾. Recent measurement of the iodine content in retail salt (thirty-six samples) in Cardiff (Wales) has shown iodine concentrations to be undetectable in thirty-two samples, with two samples, labelled as iodised, containing approximately 20 mg iodine/kg⁽Reference Lazarus and Smyth⁷⁷⁾.

Limited data from Scotland⁽Reference Barnett, Visser and Williams⁷⁸⁾ and north-east England⁽Reference Kibirige, Hutchison and Owen⁷⁹⁾ suggest many pregnant women are iodine deficient. In a study in Middlesbrough (north-east England) in 2000–2001 pregnant women attending the antenatal clinic at 15 weeks of gestation (n 227) and non-pregnant age-matched controls (n 227) were enrolled⁽Reference Kibirige, Hutchison and Owen⁷⁹⁾. Sixteen of the pregnant women (7%) and twenty of the non-pregnant controls (8·8%) were found to have UI <50 μg/l, with ninety (40%) of the pregnant women having UI between 50 μg/l and 100 μg/l. The 227 pregnancies resulted in 218 live births and no newborn was reported to have a clearly elevated blood TSH. Although it was suggested that the data indicate borderline iodine deficiency in pregnancy, using the current WHO cut-off value for inadequate iodine intake for this group (150 μg/l)⁽¹²⁾ most of the women were clearly iodine deficient.

In the Republic of Ireland the distribution of newborn TSH as a measure of iodine nutrition has been examined in the population⁽Reference Burns, Mayne and O'Herlihy⁸⁰⁾. Samples drawn from the National Neonatal Screening Programme were assessed during the years 1995–2006. In 1995–6 the frequency of TSH values >5·0 mIU/l was found to be >3%, suggesting iodine deficiency in the population. However, a marked decline in the proportion of elevated values was observed in the subsequent years. At the same time, a small but significant increase (P<0·001) in mean blood TSH within the 0–5·0 mIU/l interval was observed between 1999 and 2006. It was suggested that the trend towards increasing TSH is of concern and should be monitored carefully. UI was measured in Irish women (n 54) during the first trimester of pregnancy (median 8 weeks of gestation) attending the National Maternity Hospital in Dublin⁽Reference Nawoor, Burns and Smith⁸¹⁾. The samples were taken at different times during the year with thirty-six samples taken in the summer months. The UI were found to show seasonal variation; the median UI in the women being 45 μg/l in summer and 68 μg/l in winter. These values are lower than those in a study of pregnant women in Dublin in the mid-1990s from the same research group⁽Reference Smyth, Hetherton and Smith⁸²⁾, in which the median UI was reported to be 83 μg/l in summer and 135 μg/l in winter. Although these studies were small, they suggest dietary iodine intake in Irish pregnant women continues to fall well short of WHO recommendations. This outcome is not surprising, as only about 3–4% of total salt sales in Ireland are iodised⁽Reference Nawoor, Burns and Smith⁸¹⁾. Thus, dietary iodine intake in the Irish population depends on consumption of sea fish and marine products, which varies widely among regions and individuals, and milk and dairy products, the iodine content of which varies seasonally and may be decreasing as a result of reduced use of iodophor disinfectants by dairies. As in the UK, relying on unpredictable amounts of iodine in the food supply, driven mainly by commercial rather than health interests, is precarious. The few available data suggest that the Republic of Ireland and the UK are mildly iodine deficient. Representative data on current iodine status in children and pregnant women are urgently needed to inform health policy.