Adaptation to iodine deficiency (ID) requires changes in thyroid and pituitary function that have been well characterized in animals. The present studies were undertaken to analyze the relationships between serum thyroid hormones and TSH concentrations in iodine-deficient pregnantwomen as well as their newborns. The broad range of iodine intake of the population studied, from very low to adequate, allowed us to describe quantitatively the relationships among iodine intake,thyroid hormones, and TSH. The interpretation of the data was supported by the effects that iodinesupplementation had on these various hormones and is consistent with animal observations. About 250 pregnant women from an iodine-deficient area were studied. Fifty percent had a urinary iodine excretion of 50 μg creatinine (cr) or less; 25% had 50∓100 ug I/g-cr, and about 20% had 100∓200 ug I/g cr. Baseline serum hormone concentrations on first examination (mean ± SD) were: T4, 8.0± 0.7 jug⁄dl; T3> 179 ° 45ng⁄dl; rT3, 25 ° 9 ng⁄dl, and TSH, 2.9 μ 1.7 μ U/ml. All values were independent of the age of gestation between 10−40 weeks. Serum T4 and urinary I correlated linearly in a double reciprocal i lot (r = 0.90; P < 0.001). Serum TSH correlated with the reciprocal of serum T4 (r = 0.76; P < 0.005). In 36 women, oral iodine supplementation (OIS; 300 μg I/day for 4∓16 weeks)increased total serum T4 from 8.7 ± 2.4to 12.5 ± 1.8 μg/dl (P <0.001), decreased TSH from 2.8 ° 1.4 to 1.5 °08juU/ml (P <0.001), and increased rT3 from 21 ± 7 to 25 ± 9 ng/dl (P0.02; all by paired t tests); serum T3 remained unchanged (187 ±0 25 and 180 ±46 ng/dl), as did T4 and T3 free fractions. The free T4 concentration, which was lower than that in nonpregnant adults before treatment (1.63 ± 0.4 vs. 2.30 ± 0.4 ng/dl; P < 0.001), increased to 2.3 ± 0.4 with OIS (P < 0.001). No change inany of these variables was observed in 10 untreated controls followed for similar lengths of time.Cord serum T4 and rT3 were higher in the offspring of OIS mothers (11.3 ± 1.5 μg/dl and 226 ± 55 ng/dl) than in those of untreated mothers (9.3 ± 1.7 μg/dl and 183 ±0 37 ng/dl; P<0.001 and p< 0.005, respectively). Newborn serum T4 correlated with the mothers' value (P 0.001) but was found to be less affected by iodine intake than the mothers' results.
We conclude that: 1) the decrease in TSH observed with OIS was due to the increase in serum T4,since serum T3 was unchanged, this stressing the critical role of the reduced serum T4 in the adaptation to ID; 2) the small difference in serum TSH was physiologically significant, since this wasassociated with a compensatory increase in thyroidal T3 secretion during ID; 3) the larger increase in serum T4 induced by OIS in maternal as opposed to corresponding cord sera suggests that the fetus is relatively protected against ID; and 4) these findings are in agreement with previous animal studies, suggesting applicability of the latter to human thyroid physiology





