Little is known about the optimum level of iodine intake for iodine supplementation programmes, or about the effects of the high levels of iodine intake that are found in some countries. We compared the incidence of different types of hyperthyroidism in East-Jutland Denmark with a low average iodine intake but no endemic goitre, and the incidence in Iceland with a relatively high iodine intake. Hyperthyroidism was more common in East-Jutland than in Iceland, due to a much higher incidence of multinodular toxic goitre and also of single toxic adenoma. Most of the patients with these diseases were over 50 years of age. By contrast, the incidence of Graves' disease was significantly higher in Iceland than in East-Jutland. This difference was most marked in the younger age groups, in which hyperthyroidism was more than twice as common in Iceland as in East-Jutland. These results demonstrate that even mild iodine deficiency has a significant effect on population health, since it leads to a high incidence of autonomous thyroid nodules with hyperthyroidism in the elderly population. However, population iodine intake probably should not exceed a level much higher than that necessary to avoid iodine deficiency, otherwise Graves' disease may be induced in the young population.

High individuality causes laboratory reference ranges to be insensitive to changes in test results that are significant for the individual. We undertook a longitudinal study of variation in thyroid function tests in 16 healthy men with monthly sampling for 12 months using standard procedures. We measured serum T(4), T(3), free T(4) index, and TSH. All individuals had different variations of thyroid function tests (P < 0.001 for all variables) around individual mean values (set points) (P < 0.001 for all variables). The width of the individual 95% confidence intervals were approximately half that of the group for all variables. Accordingly, the index of individuality was low: T(4) = 0.58; T(3) = 0.54; free T(4) index = 0.59; TSH = 0.49. One test result described the individual set point with a precision of +/- 25% for T(4), T(3), free T(4) index, and +/- 50% for TSH. The differences required to be 95% confident of significant changes in repeated testing were (average, range): T(4) = 28, 11-62 nmol/liter; T(3) = 0.55, 0.3--0.9 nmol/liter; free T4 index = 33, 15-61 nmol/liter; TSH = 0.75, 0.2-1.6 mU/liter. Our data indicate that each individual had a unique thyroid function. The individual reference ranges for test results were narrow, compared with group reference ranges used to develop laboratory reference ranges. Accordingly, a test result within laboratory reference limits is not necessarily normal for an individual. Because serum TSH responds with logarithmically amplified variation to minor changes in serum T(4) and T(3), abnormal serum TSH may indicate that serum T(4) and T(3) are not normal for an individual. A condition with abnormal serum TSH but with serum T(4) and T(3) within laboratory reference ranges is labeled subclinical thyroid disease. Our data indicate that the distinction between subclinical and overt thyroid disease (abnormal serum TSH and abnormal T(4) and/or T(3)) is somewhat arbitrary. For the same degree of thyroid function abnormality, the diagnosis depends to a considerable extent on the position of the patient's normal set point for T(4) and T(3) within the laboratory reference range.

The iodine intake level in a population is determined in cross-sectional studies. Urinary iodine varies considerably and the reliability of studies of iodine nutrition and the number of samples needed is unsettled. We performed a longitudinal study of sixteen healthy men living in an area of mild to moderate iodine deficiency. Iodine and creatinine concentrations were measured in spot urine samples collected monthly for 13 months. From these data we calculated the number of urine samples needed to determine the iodine excretion level for crude urinary iodine and for 24 h iodine excretion estimated from age-and gender-specific creatinine excretions. We found that mean urinary iodine excretion varied from 30 to 87 mg/l (31 to 91 mg/24 h). Sample iodine varied from 10 to 260 mg/l (20 to 161 mg/24 h). Crude urinary iodine varied more than estimated 24 h iodine excretion (population standard deviation 32 v. 26; individual standard deviation 29 v. 21; Bartlett's test, P,0·01 for both). The number of spot urine samples needed to estimate the iodine level in a population with 95 % confidence within a precision range of^10 % was about 125 (100 when using estimated 24 h iodine excretions), and within a precision range of^5 % was about 500 (400). A precision range of 20 % in an individual required twelve urine samples or more (seven when using estimated 24 h iodine excretions). In conclusion, estimating population iodine excretion requires 100 -500 spot urine samples for each group or subgroup. Less than ten urine samples in an individual may be misleading.

Knowledge of the effect of differences in iodine intake levels on public health in areas with no endemic goiter is limited. Groups at risk when iodine intake is relatively low are pregnant and lactating women and their newborns. A prospective randomized study was performed to evaluate the effect of iodine supplementation in an area where the median daily iodine excretion in urine is around 50 micrograms. Fifty-four normal pregnant women were randomized to be controls or to receive 200 micrograms iodine/day from weeks 17-18 of pregnancy until 12 months after delivery. In the control group, serum TSH, serum thyroglobulin (Tg), and thyroid size showed significant increases during pregnancy. These variations were ameliorated by iodine supplementation. Iodine did not induce significant variations in serum T4, T3, or free T4. Cord blood Tg was much lower when the mother had received iodine, whereas TSH, T4, T3, and free T4 levels were unaltered. The results suggest that a relatively low iodine intake during pregnancy leads to thyroidal stress, with increases in Tg release and thyroid size. However, the thyroid gland is able to adapt and keep thyroid hormones in the mother and the child normal, at least under normal circumstances, as evaluated in the present study. It is not known whether this stress is sufficient to be of importance for late development of autonomous thyroid growth and function.

Large variations exist in thyrotropin (TSH) and thyroid hormones in serum. The components of variation include preanalytical, analytical, and biologic variation. This is divided into between- and within-individual variation. The latter consists of circadian and seasonal differences although there are indicators of a genetically determined starting point. The ratio of within- to between-individual variation describes the reliability of population-based reference ranges. This ratio is low for serum TSH, thyroxine (T(4)) and triiodothyronine (T(3)) indicating that laboratory reference ranges are relatively insensitive to aberrations from normality in the individual. Solutions are considered but reducing the analytical variation below the calculated analytical goals of 7%, 5% and 12% for serum T(3), T(4), and TSH does not improve diagnostic performance. Neither does determination of the individual set-point and reference range. In practice this means that population-based reference ranges are necessary but that it is important to recognize their limitations for use in individuals. Serum TSH responds with amplification to minor alterations in T(4) and T(3). A consistently abnormal TSH probably indicates that T(4) and T(3) are not normal for the individual even when inside the laboratory reference range. This underlines the importance of TSH in diagnosis and monitoring of thyroid dysfunctions. Also, it implies that subclinical thyroid disease may be defined in purely biochemical terms. Under critical circumstances such as pregnancy where normal thyroid function is of importance for fetal brain development, subclinical thyroid disease should be treated. Even TSH within the reference range may be associated with slightly abnormal thyroid function of the individual. The clinical importance of such small abnormalities in thyroid function in small children and pregnant women for brain development remains to be elucidated.

Lack of iodine for thyroid hormone formation during the fetal stage and/or the first years of life may lead to developmental brain damage. During the period of breastfeeding, thyroid function of the infant depends on iodine in maternal milk. We studied healthy, pregnant women admitted for delivery and their newborn infants. Cotinine in urine and serum was used to classify mothers as smokers (n = 50) or nonsmokers (n = 90). Smoking and nonsmoking mothers had identical urinary iodine on d 5 after delivery, but smoking was associated with reduced iodine content in breast milk (smokers 26.0 micro g/liter vs. nonsmokers 53.8 micro g/liter; geometric mean, P < 0.001) and in the infants' urine (smokers 33.3 micro g/liter, vs. nonsmokers 50.4 micro g/liter, P = 0.005). Results were consistent in multivariate linear models and by logistic regression analysis. The odds ratio for smoking vs. nonsmoking mothers to have lower breast milk than urinary iodine content was 8.4 (95% confidence interval, 3.5-20.1). In smokers, iodine transfer into breast milk correlated negatively to urinary cotinine concentration. Smoking mothers had significantly higher serum levels of thiocyanate, which may competitively inhibit the sodium-iodide symporter responsible for iodide transport in the lactating mammary gland. Smoking during the period of breastfeeding increases the risk of iodine deficiency-induced brain damage in the child. Women who breastfeed should not smoke, but if they do, an extra iodine supplement should be considered.

The iodine intake level of the population is of major importance for the occurrence of thyroid disorders in an area. The aim of the present study was to evaluate the importance of drinking water iodine content for the known regional differences in iodine intake in Denmark and for the iodine content of infant formulas.Iodine in tap water obtained from 55 different locations in Denmark varied from <1.0 to 139 mg/l. In general the iodine content was low in Jutland (median 4.1 mg/l) with higher values on Sealand (23 mg/l) and other islands. Preparation of coffee or tea did not reduce the iodine content of tap water with a high initial iodine concentration. A statistically significant correlation was found between tap water iodine content today and the urinary iodine excretion measured in 41 towns in 1967 (r = 0.68, P < 0.001). The correlation corresponded to a basic urinary iodine excretion in Denmark of 43 mg/24 h excluding iodine in water and a daily water intake of 1.7 l. The iodine content of infant formulas prepared by addition of demineralized water varied from 37 to 138 mg/l (median 57 mg/l, n = 18). Hence the final iodine content would depend heavily on the source of water used for preparation.We found that iodine in tap water was a major determinant of regional differences in iodine intake in Denmark. Changes in water supply and possibly water purification methods may influence the population iodine intake level and the occurrence of thyroid disorders.

Thyroid abnormalities are common in all populations, but it is difficult to compare results of epidemiological studies, because different methods have been used for evaluation. We studied the importance of the population iodine intake level for the prevalence rate of various thyroid abnormalities in elderly subjects. Random samples of elderly subjects (68 yr) were selected from the central person registers in Jutland, Denmark, with low (n = 423) and, in Iceland, with longstanding relatively high (n = 100) iodine intake. Females from Jutland had a high prevalence of goiter or previous goiter surgery (12.2%), compared with males from Jutland (3.2%) and females (1.9%) and males (2.2%) from Iceland. Abnormal thyroid function was very common in both areas, with serum TSH outside the reference range in 13.5% of subjects from Jutland and 19% of those from Iceland. In Jutland, it was mainly thyroid hyperfunction (9.7% had low, 3.8% had high serum TSH), whereas in Iceland, it was impaired thyroid function (1% had low, 18% had high serum TSH). All subjects with serum TSH more than 10 mU/L had autoantibodies in serum, but antibodies were, in general, more common in Jutland than in Iceland. Thus, thyroid abnormalities in populations with low iodine intake and those with high iodine intake develop in opposite directions: goiter and thyroid hyperfunction when iodine intake is relatively low, and impaired thyroid function when iodine intake is relatively high. Probably, mild iodine deficiency partly protects against autoimmune thyroid disease. Thyroid autoantibodies may be markers of an autoimmune process in the thyroid or secondary to the development of goiter.