Is placental iodine content related to dietary iodine intake?

OBJECTIVE: Delivery of iodine to the foetus depends not only on maternal dietary iodine intake but also on the presence of a functioning placental transport system. A role for the placenta as an iodine storage organ has been suggested, and this study compares the iodine content of placentas from women giving birth at term in Ireland and Iran, areas with median urinary iodine of 79 and 206 µg/l respectively. DESIGN: Placental cotyledon iodine was measured using an alkaline ashing technique with Sandell-Kolthoff kinetic colorimetry. Samples were taken from six sites from the centre and periphery of each cotyledon. Placentas (Ireland n = 58; Iran n = 45) were obtained from consecutive euthyroid women delivering at term. RESULTS: The median placental iodine (µg/g wet weight) was significantly higher in Iranian than in Irish women (187·2 µg/g vs 34·3 µg/g; P < 0·001). The distribution of individual placental iodine values showed that values >50µg/g were found in 71·0% of Iranian and in only 21·0% of Irish samples. In Irish subjects, the relationship of placental iodine to pregnant population urinary iodine (UI) (ng/g:µg/l) was 1:2 (40:79), while in Iranians this ratio is closer to 1:1 (211:206). CONCLUSIONS: These findings, by demonstrating an apparent ability of the placenta to store iodine in a concentration-dependent manner, suggest a hitherto undetected role for the placenta. Whether placental iodine has a role in protecting the foetus from inadequacies in maternal dietary iodine intake is as yet unknown.

Does iodine gas released from seaweed contribute to dietary iodine intake?

Thyroid hormone levels sufficient for brain development and normal metabolism require a minimal supply of iodine, mainly dietary. Living near the sea may confer advantages for iodine intake. Iodine (I(2)) gas released from seaweeds may, through respiration, supply a significant fraction of daily iodine requirements. Gaseous iodine released over seaweed beds was measured by a new gas chromatography-mass spectrometry (GC-MS)-based method and iodine intake assessed by measuring urinary iodine (UI) excretion. Urine samples were obtained from female schoolchildren living in coastal seaweed rich and low seaweed abundance and inland areas of Ireland. Median I(2) ranged 154-905 pg/L (daytime downwind), with higher values (~1,287 pg/L) on still nights, 1,145-3,132 pg/L (over seaweed). A rough estimate of daily gaseous iodine intake in coastal areas, based upon an arbitrary respiration of 10,000L, ranged from 1 to 20 µg/day. Despite this relatively low potential I(2) intake, UI in populations living near a seaweed hotspot were much higher than in lower abundance seaweed coastal or inland areas (158, 71 and 58 µg/L, respectively). Higher values >150 µg/L were observed in 45.6% of (seaweed rich), 3.6% (lower seaweed), 2.3% (inland)) supporting the hypothesis that iodine intake in coastal regions may be dependent on seaweed abundance rather than proximity to the sea. The findings do not exclude the possibility of a significant role for iodine inhalation in influencing iodine status. Despite lacking iodized salt, coastal communities in seaweed-rich areas can maintain an adequate iodine supply. This observation brings new meaning to the expression "Sea air is good for you!"

Short-term changes in maternal and neonatal urinary iodine excretion.

Investigation of maternal urinary iodine (UI) excretion in the immediate antenatal and early postpartum periods showed a precipitous fall in median values from 93 microg/L antenatally to 36 microg/L at delivery subsequently rising to 49 microg/L and 63 microg/L at days 3 and 10 postpartum respectively. The fate of ingested iodine not appearing in the maternal urine is unknown but measurement of UI in babies born to nursing mothers suggested transfer from the mother with median neonatal values of 117 and 159 microg/L being recorded at days 3 and 10. While maternal UI seemed to relatively unaffected by breast feeding, median UI from breast feeding babies (148 microg/L) was significantly greater than in those bottle feeding (50 microg/L). This was also reflected by the finding that no breast feeding baby had a UI values < 50 microg/L in comparison to 50% of bottle feeders. The depressed values in mothers and relatively high values in their infants could present a false picture and suggest the need to defer any investigations of iodine status at this time. The findings do however suggest a need for further investigations aimed at determining the fate of iodine ingested perinatally and its possible physiological significance in maintaining thyroid status in the mother and neonate.

Iodine intake in pregnancy in Ireland--a cause for concern?

BACKGROUND: Adequate dietary iodine intake is necessary to maintain maternal thyroid function at a level permitting normal neuropsychological development of the foetus. AIMS AND METHODS: To determine dietary iodine status by measuring urinary iodine excretion (UIE), proportional to dietary intake, in Irish mothers during the first trimester of pregnancy. RESULTS: Median UIE showed seasonal variations, being lower in summer than in winter. The median values in pregnant women were, summer 45microg/l, winter 68microg/l. Equivalent values for controls were 43 and 91microg/l respectively. UIE required to achieve WHO recommended daily iodine intakes would be 120-180microg/l. In the Irish subjects UIE values suggestive of iodine deficiency (<50microg/l) were observed in 55% of pregnant women tested in summer and 23% in winter. Dairy milk iodine, a major dietary iodine source, showed similar variation. CONCLUSIONS: While there is as yet no available evidence of widespread thyroid hypofunction in the Irish obstetric population, the findings are a cause of concern, which if confirmed by a more comprehensive investigation, may indicate the need for iodine prophylaxis.

Sequential studies on thyroid antibodies during pregnancy.

Thyroid antibodies were measured sequentially in 25 pregnant women from a Sri Lankan population. A high prevalence of antithyroid antibodies, particularly antithyroglobulin antibodies (TgAb) had previously been demonstrated in female schoolchildren drawn from this population. In the present study TgAb were detected in 36.8% of nonpregnant controls while thyroid peroxidase antibody (TPOAb) positivity was present in 26.3%. The prevalence of both antibodies in the pregnancy study group showed a progressive decline compared to nonpregnant controls throughout gestation becoming undetectable in the third trimester. The results are consistent with an immunosuppressive effect of pregnancy in a population in whom high thyroid autoantibody titers may have resulted from a recent salt iodization program.

Thyroglobulin autoantibodies in iodized subjects: relationship between epitope specificities and longitudinal antibody activity.

INTRODUCTION: We previously reported a high thyroglobulin autoantibodies (TgAb) prevalence in healthy Sri Lankans after iodine supplementation. In the present study 58 TgAb-positive schoolgirls were followed up after 5 years of continued iodination. The objectives were: (1) to observe the longitudinal profile of TgAb epitope specificities and (2) to examine the relationship between these specificities and the course of thyroid autoimmunity in this population. METHODS: Paired subjects' sera (at onset and at 5-year follow-up) were tested for TgAb, thyroid peroxidase antibody (TPOAb), and TgAb epitope-specificity. Epitope reactivity was determined by employing a panel of 10 murine monoclonal antibodies (Tg-mAbs) directed against 6 Tg antigenic clusters (I-VI) in competitive enzyme-linked immunosorbent assay (ELISA) reactions with test sera. RESULTS: The overall pattern of epitope recognition in individual subject's sera remained preserved over the time period. Nine subjects showed restricted specificities while majority of the subjects were broadly heterogeneous. At follow-up, median TgAb concentration in the restricted group was higher than in the unrestricted (1650 versus 110 kIU/L; p < 0.005). Epitope specificity was a stronger determinant of TgAb persistence than the height of the initial TgAb response or the TPOAb status of subjects. CONCLUSION: Tg epitope reactivity pattern in iodised populations may identify subjects at greater risk of developing autoimmune thyroid disease (AITD).

The prevalence of thyroid-stimulating antibodies in goitrous disease assessed by cytochemical section bioassay.

Thyroid-stimulating antibodies (TSAb) were detectable using a highly sensitive cytochemical section bioassay in plasma from all 56 hyperthyroid patients studied, including those who had either diffuse hyperplasia or nodular goiters. The maximum dilution at which TSAb was detectable ranged from 10(-2) - 10(-6). Six of these patients had scintigraphic evidence of single functioning nodules, and, surprisingly, TSAb was present in all. Of 27 patients who had nontoxic goiter, 14 (52%) had positive titers for TSAb ranging from 10(-2) - 10(-4). The mean serum T3 value in nontoxic goitrous patients who had TSAb was significantly higher than that in subjects in whom TSAb was absent; in contrast, mean serum T4 values were not significantly different in those two groups. It is concluded that idiopathic nontoxic goiter, toxic nodular goiter with functioning nodules (Plummer's disease), and toxic diffuse goiter (Graves' disease) share, in part, a common pathogenesis.

Association of thyroid-stimulating immunoglobulins and thyrotropin-releasing hormone responsiveness in women with euthyroid goiter.

Thyroid-stimulating immunoglobulins (TSI) were measured, using a highly sensitive cytochemical bioassay, in plasma from 26 euthyroid women with idiopathic diffuse or multinodular goiter selected on the basis of their serum TSH responses to TRH stimulation. Thirteen were chosen because they were previously identified to have impairment in TRH responsiveness and were compared with 13 consecutive patients who had normal responses to TRH. TSI were present in a significantly greater number of those who had subnormal TRH responses (11:13) compared to those who had normal responses (3:13) P less than 0.005. Although serum T4, T3, and basal TSH values were all within the normal range, mean serum T4 and T3 values were significantly higher and basal TSH significantly lower in the 14 patients who had TSI than in the 12 in whom TSI was absent. The coexistence of impaired TRH responsiveness and TSI was associated with a family history of thyroid disease. The data suggest that TSI in patients with euthyroid goiter cause a modest increase in thyroid secretion sufficient to blunt the TSH response to TRH but not to cause clinical hyperthyroidism.

Maternal iodine status and thyroid volume during pregnancy: correlation with neonatal iodine intake.

Differences in pregnancy-associated alterations in thyroid volume and urinary iodine (UI) excretion have been attributed to geographical variations in dietary iodine intake. In this study, ultrasound-measured thyroid volume and UI excretion were assessed during the 3 trimesters of pregnancy, at delivery, and at 6 weeks postpartum. Urine specimens also were obtained from mothers and both breast- and formula-feeding infants at 3 days after delivery. Thyroid volume showed a significant increase (maximum 47.0%), compared with nonpregnant control values over the 3 trimesters of pregnancy, which occurred as early as the first trimester and was paralleled by increased UI excretion, followed in turn by a precipitous fall at delivery. UI excretion in breast-feeding neonates (100 +/- 6.8 micrograms/L) was significantly higher than in their mothers (76 +/- 5.6 micrograms/L; p < 0.01) but was significantly lower (43 +/- 3.5 micrograms/L) in formula-fed infants. The results suggest that in an area of moderate dietary iodine intake, UI loss during pregnancy may result in maternal thyroid enlargement. The ability of the breast to transport iodine compensates for this loss in breast-fed infants, but this protection may be lost in formula feeding.

Adrenal abnormalities in polycystic ovary syndrome.

This study was undertaken to examine the role of adrenal androgen excess in the pathogenesis of polycystic ovary syndrome (PCOS) and, if such was present, to assess its reversibility using dexamethasone given in physiological dosage at night. Mean plasma testosterone (T), T/sex-hormone binding globulin (T/SHBG) ratio, androstenedione, and 17-OH-progesterone levels were elevated in the 19 patients studied. Plasma estrone values were elevated, whereas estradiol levels were normal. Plasma FSH was decreased and LH responsiveness to LHRH was exaggerated. Metyrapone, an 11-hydroxylase inhibitor, was administered at 2400 h to induce hypocortisolemia and compensatory ACTH secretion so that adrenal androgen and glucocorticoid responsiveness to endogenous stimulation could be examined. Plasma T, androstenedione, and 11-deoxycortisol responses to metyrapone were excessive in PCOS patients, thus indicating a specific adrenal abnormality. After 3 months treatment with dexamethasone, 0.5 mg at night, mean plasma T/SHBG and androstenedione declined to normal, and mean plasma dehydroepiandrosterone and dehydroepiandrosterone sulfate declined to below normal. The mean estrone value was slightly lower during dexamethasone. Plasma LH responsiveness to LHRH was no longer significantly different from normal, but FSH was suppressed. During treatment androgen responsiveness to metyrapone stimulation was normal, whereas 11-deoxycortisol responsiveness was suppressed. Fifteen patients completed 3 months of treatment with dexamethasone. Of these, 10 resumed regular menstruation. The latter group had suppression of plasma T, T/SHBG, androstenedione, dehydroepiandrosterone, and dehydroepiandrosterone sulfate. Only plasma androstenedione fell significantly in the remainder. These observations support the hypothesis that, in at least some patients, PCOS develops in response to abnormal gonadotropin secretion induced by hyperestronemia occurring as a consequence of excessive adrenal androgen secretion.

Serum thyroid peroxidase autoantibodies, thyroid volume, and outcome in breast carcinoma.

The prevalence of thyroid peroxidase autoantibodies (TPO.Ab) was assessed in patients with either breast carcinoma or benign breast disease, and its association with disease outcome in breast carcinoma was studied. TPO.Ab were detected by direct RIA in serum from 121/356 (34.0%) of patients with breast carcinoma, compared with 36/194 (18.5%) of controls (P < 0.001); and in 31/108 (28.7%) with benign breast disease, compared with 12/88 (13.6%) of controls (P < 0.05). Survival analysis in a group of 142 women with breast carcinoma demonstrated that TPO.Ab titres > or = 0.3 U/mL were associated with a significantly better disease-free [relative risk (RR) = 1.84, P < 0.05] and overall survival (RR = 3.46, P < 0.02), compared with those who were TPO.Ab-negative. Better outcome associated with higher TPO.Ab titres was confined to those who had thyroid volumes within the intermediate range (10.1-18.8 mL) and did not further enhance the good outcome recorded when volumes were < or = 10.0 mL or > 18.8 mL. Multivariate survival analysis showed that both TPO.Ab and thyroid volume were independently associated with prognosis in breast carcinoma and that RRs for disease-free survival were of a similar order of magnitude to well-established prognostic indices such as axillary nodal status or tumor size. These findings supply evidence that manifestations of thyroid autoimmunity are associated with a beneficial effect on disease outcome in breast carcinoma and provide the strongest evidence to date of a biological link between breast carcinoma and thyroid disease.

A direct relationship between thyroid enlargement and breast cancer.

Despite extensive study, evidence to support a direct relationship between diseases of the thyroid and breast has not been established. In this study thyroid volume was assessed by ultrasound in 200 patients with breast cancer and 354 with benign breast disease. Results were compared to appropriate female control groups. Both mean thyroid volume (21.1 +/- 1.4 mL) and the percentage of individual patients with enlarged (> 18.0 mL) thyroid glands (41.5%) were significantly greater in the breast cancer group than equivalent values (13.2 +/- 0.5 mL and 10.5%, respectively) in age-matched controls (P < 0.01 in both cases). The mean thyroid volume of 14.5 +/- 0.34 mL in patients with benign breast disease was also significantly greater than that of 12.5 +/- 0.38 mL in younger controls (P < 0.01). The results support a direct association between breast cancer and increased thyroid volume as mean thyroid volumes and the percentage of individual patients with enlarged thyroid glands were similar in those studied both before (20.8 +/- 1.3 mL and 43.0%) and after (21.4 +/- 1.6 mL and 40.0%) therapies for breast cancer. Although there is no evidence that thyroid enlargement represents a risk factor for breast cancer, the results emphasize the importance of raising the consciousness of the coincidence of both disorders.

Immunological features of nonimmunogenic hyperthyroidism.

Blood lymphocyte subpopulations (Leu 4+ cells = pan-T cells, Leu 3a+ cells = helper/inducer cells, and Leu 2a+ cells = suppressor/cytotoxic cells), thyroid-stimulating immunoglobulins, microsomal antibodies and antibodies against thyroglobulin were determined in 10 patients with hyperthyroidism due to single autonomously functioning thyroid nodules (ATN), 11 patients with hyperthyroidism due to Graves' disease (GD) and in 20 normal subjects. Thyroidectomy was performed in 8 of the patients with ATN and in 6 of those with GD after 3 weeks of antithyroid drug treatment with methimazole. Lymphocytic infiltration of thyroid tissue, the amount of the various lymphocyte subsets (Leu 4+, Leu 3a+, and Leu 2a+ T cells as well as B+ B cells) in the thyroid gland, as well as the expression of the histocompatibility antigen HLA-DR on thyrocytes and intrathyroidal lymphocytes were examined. Blood Leu 4+ cells were reduced due to a lack of Leu 2a+ cells in patients with ATN and GD when compared to normal subjects. Thyroid-stimulating immunoglobulins were detected in all patients with ATN and GD, but in none of the normal subjects. Lymphocytic infiltration of thyroid tissue was present in patients with ATN and GD. The various lymphocyte subsets in the thyroid gland did not differ between the two patient groups. DR expression on thyrocytes was seen in 6 of the patients operated for ATN and in 5 of those who underwent surgery for GD. Infiltration with DR+-T lymphocytes was found in all thyroid glands investigated. Thus immunological findings usually classified as proof for the autoimmune origin of GD exist also in patients with ATN. An overlap in the pathogenetic background of both diseases seems possible.

Tissue iodine content and serum-mediated 125I uptake-blocking activity in breast cancer.

In the thyroid, active transport of iodide is under control of the TSH-dependent Na+/I- symporter (NIS), whereas in the breast such control is less well understood. In this study, NIS expression was demonstrated by RT-PCR in 2 of 2 fibroadenomata and 6 of 7 breast carcinoma messenger ribonucleic acid isolates. In addition, mean total tissue iodine levels of 80.9 +/- 9.5 ng I/mg protein in 23 benign tumors (fibroadenomata) were significantly higher than those in 19 breast cancers taken from either the tumor (18.2 +/- 4.6 ng I/mg) or morphologically normal tissue taken from within the tumor-bearing breast (31.8 +/- 4.9 ng I/mg; P < 0.05 in each case). Inhibition of 125I uptake into NIS-transfected CHO cells was observed in serum from 20 of 105 (19.0%) breast carcinoma, 8 of 49 (16.3%) benign breast disease, and 27 of 86 (31.4%) Graves' patients, but in only 1 of 33 (3.0%) age-matched female controls. IgG purified from serum of patients showing positive 125I uptake inhibition also inhibited iodide uptake, suggesting that such inhibition was antibody mediated. 125I uptake inhibition was significantly associated with thyroid peroxidase antibody positivity (P < 0.05) in sera from breast cancer patients, but not in those with benign breast disease, once again suggesting an association between thyroid autoimmunity and breast carcinoma.

Quantitative cytochemistry of glucose-6-phosphate dehydrogenase in benign and malignant breast tumours.

Glucose-6-phosphate dehydrogenase (G6PD) activity was quantified cytochemically in mammary epithelial cells within frozen tissue sections from 38 patients with breast cancer and 44 with benign breast disease. G6PD activities were measured under atmospheres of both N2 and O2. The mean (S.E.) G6PD value 2.5 (0.23) IE U/min measured in N2 in mammary epithelial cells from the group of malignancies was significantly greater than that of 1.6 (0.37) IE U/min in the benign group (P less than 0.001), but there was considerable overlap between individual values. G6PD measured in O2 was detectable in 84% of malignancies compared to only 14% of benign biopsies and the group mean of 1.3 (0.18) IE U/min in the former was significantly greater than that of 0.35 (0.20) IE U/min in the latter (P less than 0.001). Significant correlations between G6PD activities measured in N2 and O2 were observed in both groups. The techniques present a sensitive method of identifying increases in G6PD activity in mammary epithelial cells and provide an assay that in a majority of cases permits the separation of malignant from benign tissues.

Increased prevalence of thyroglobulin antibodies in Sri Lankan schoolgirls--is iodine the cause?

OBJECTIVE: Iodine deficiency was the likely cause of a high prevalence of goitre previously in Sri Lankan schoolchildren. Salt iodination was made compulsory in 1993 but there has been no recent study, using modern techniques, of its benefits or harmful effects. METHODS: Three hundred and sixty-seven schoolgirls between the ages of 11 and 16 years had ultrasound thyroid volume, free thyroxine (T4), free tri-iodothyronine (T3), thyrotrophin (TSH), anti-thyroglobulin (TgAb) and thyroid peroxidase (TPOAb) antibodies, and urine iodine concentrations measured. RESULTS: Median ultrasound thyroid volume ranged from 4.8 ml (11-year-old girls) to 8.6 ml (16-year-old girls) with an age-related increase. Median urine iodine concentrations ranged from 105 to 152 microg/l. Free T4 and free T3 were normal in all, but TSH was elevated in four subjects (5. 53-41.29 mU/l). However, the prevalence of TgAb was markedly raised, ranging between 14.3% (11-year-old girls) and 69.7% (16-year-old girls) (P<0.03). In contrast, the prevalence of TPOAb was 10% or less in all age groups. CONCLUSIONS: Normal median thyroid volumes, iodine concentrations and thyroid function would indicate that iodine deficiency is not a major problem in this group. The high prevalence of TgAb, hitherto unreported, most likely reflects excessive iodination of Tg resulting in increased immunogenicity. There is an urgent need to continuously monitor the adequacy and risks of iodination in this population.

Evolution of thyroid autoimmunity during iodine prophylaxis--the Sri Lankan experience.

OBJECTIVE: To study the evolution of thyroid autoimmunity, in relation to the change in goitre prevalence, during 3 Years of iodine prophylaxis in Sri Lanka. METHODS: Two groups of Sri Lankan schoolgirls between the ages of 10.8 and 17.5 Years were studied in 1998 (401 girls) and 2001 (282 girls). A prospective study was performed in 42 schoolgirls who were thyroid autoantibody (Ab)-positive (+ve) in 1998. Anthropometric measures, urinary iodine excretion (UIE), thyroid Volume, free thyroxine, free tri-iodothyronine, TSH, and thyroglobulin (Tg) and thyroid peroxidase (TPO) Ab were evaluated in all 683 girls. RESULTS: Goitre prevalence was significantly lower in 2001 compared with 1998 related to age (2.9% compared with 20.2%) and body surface area (11.6% compared with 40.8%), although UIE was unchanged. Prevalence of thyroid Ab in 2001 was also lower (23.4% compared with 49.9%); among those with the Ab, 34.8% had TgAb alone and 46.9% had a combination of TgAb+TPOAb, compared with 82.0% TgAb alone in 1998. In 2001, subclinical hypothyroidism was more frequent in Ab+ve (6.3%) than Ab-negative girls (1.0%). A cohort of 42 Ab+ve schoolgirls in 1998 (34 with TgAb alone, eight with TgAb+TPOAb) were evaluated again in 2001. Only 10 of them (23.8%) remained Ab+ve (mostly TPOAb+/-TgAb) in 2001. CONCLUSIONS: This study demonstrates that: (1) in 2001, goitre prevalence and thyroid autoimmunity rates were significantly lower than in 1998; (2) the pattern of thyroid Ab was different in the two surveys; (3) in 2001 alone, the occurrence of hypothyroidism was correlated with the presence of thyroid autoimmunity. These results indicate an evolution of thyroid autoimmune markers during the course of iodine prophylaxis, which has not been described before.

Altered androstenedione and estrone dynamics associated with abnormal hormonal profiles in amenorrheic subjects with weight loss or obesity.

The present study was designed for exploration of hormonal disturbances underlying common forms of amenorrhea. Polycystic ovary syndrome (PCO) patients and obese amenorrheic subjects had significantly elevated estrone (E1) levels, elevated luteinizing hormone/follicle-stimulating hormone ratios, and an exaggerated luteinizing hormone response to luteinizing hormone-releasing hormone. However, androstenedione (delta 4A), the precursor of E1, was elevated only in PCO. Thus, the E1/delta 4A ratio, which provides an indirect index of aromatase activity in extraglandular sites, was raised in obese subjects as a group but not in PCO subjects. These findings suggest that elevated E1 levels, which give rise to abnormal gonadotropin secretion, arise from increased available androgens in PCO but from an increased effect of aromatase (present in adipose tissue) in obese subjects. Measurement of androgens and the E1/delta 4A ratio provides insights into the relative contributions of hyperandrogenemia and enhanced aromatase activity to the genesis of amenorrhea in these groups. In patients with suppressed estradiol levels associated with hyperprolactinemia or weight loss, follicle-stimulating hormone levels were suppressed, while luteinizing hormone was not elevated. Prolactin excess explains these findings in hyperprolactinemia. Plasma E1 levels and the E1/delta 4A ratio were suppressed in patients with weight loss, possibly as a consequence of reduced adiposity. This finding suggests that hypothesis that a minimum level of E1, dependent upon adequate adiposity, is critical for the normal mature function of the hypothalamic-pituitary-ovarian axis. Abnormal E1/delta 4A ratios, high in obesity-associated amenorrhea and suppressed in weight loss-associated amenorrhea, may provide specific markers for these groups of patients.

Thyroid disorders and breast cancer.

We have investigated the controversial association between diseases of the thyroid gland and breast carcinoma using methodology which allows positive exclusion of cases of breast disease from control groups and the detection of subclinical alterations in thyroid volume using high resolution ultrasonography, thus addressing the deficiencies of earlier studies. Whereas the prevalence of hyperthyroidism and hypothyroidism in patients with breast carcinoma and in healthy controls without clinical evidence of breast disease was similar, non-toxic goitre was more than twice as common in the breast carcinoma patients. Thyroid volumes were also significantly higher in breast carcinoma patients than in controls; using World Health Organisation criteria, 45.5% of breast carcinoma patients had thyroid enlargement compared with only 10.5% of controls. Finally, antithyroid peroxidase autoantibodies were twice as common in breast cancer patients than in controls. These findings provide clear evidence of a relationship between thyroid disease and breast carcinoma, although the mechanisms underlying this relationship require further study, future studies of breast cancer risk factors should therefore include assessment of thyroid function, antibody status and volume.

Variation in iodine handling during normal pregnancy.

Reports in the literature are divided on changes in thyroid volume and urinary iodine excretion (UI) during normal pregnancy. Reports from Ireland, an area of modest dietary iodine intake (median UI 70 microg/L) showed an increase in UI that rose to a median value of 135 microg/L in the first trimester (T1) and continued at 124 microg/L in the second (T2) and 122 microg/L in the third trimester (T3). In parallel with the increase in UI, mean ultrasound measured thyroid volume increased by a maximum of 47% over nonpregnant values in the third trimester (T3). Although these findings were consistent with studies in Cardiff, UK (median UI 73 microg/L), which also showed a pregnancy-associated rise in UI excretion (maximum 176 microg/L) accompanied by a 30% increase in median thyroid volume, they differed from findings in Sri Lanka (median UI 146 microg/L), a country in which a successful program of salt iodination has recently been implemented, which showed no significant changes in UI excretion (T3 maximum 154 microg/L) but did show a modest (20%) maximum increase in median thyroid volume at T3. Prospective studies on Irish subjects showed that median UI fell precipitously to nonpregnant control values (76 microg/L) at delivery. In addition, UI in neonates sampled at 3-days postdelivery showed that excretion was greater in breast-fed than in bottle fed infants. Differences in reported UI excretion patterns during pregnancy may may reflect the existence of a threshold above which increased renal clearance results in increased iodine loss but that is masked at higher iodine intakes. Assuming constant dietary iodine intake during pregnancy, any increased urine loss will inevitably lead to negative iodine balance and thyroid depletion. In these circumstances, increased thyroid volume may in part be a compensatory mechanism to allow for greater iodine storage.