Chapter 1 (Section 3) - Gerard N Burrow, MD and Lauren H Golden, MD
October 28, 2002
Estimates of the incidence of gestational hypothyroidism in iodine-sufficient areas are in the range of 0.1% to 0.3%101,102 The most common cause is related to thyroid autoimmunity (Table 3). The requirement for thyroid hormone replacement therapy is increased during pregnancy, suggesting an increased demand on the thyroid gland.103,104,105,106
Myxedematous patients have been reported to carry their pregnancies to term successfully.107,108,109,110 Prior to the second trimester, thyroid hormone requirements necessary for normal fetal growth are provided by the maternal thyroid, with associated transplacental passage of thyroid hormone. In severely hypothyroid mothers, this hormone would be lacking. There is no evidence for an increased risk of birth defects associated with a history of maternal hypothyroidism,111 although an interesting association has been made between the presence of high titers of thyroid autoantibodies in women, and Down syndrome in their children.112 A hypothetical explanation for this is that maternal thyroid autoimmunity may predispose to aneuploidy in the child and may play a major role in the birth of children with Down syndrome in younger mothers. In addition, as mentioned above, there is increasing evidence that maternal thyroid hormone plays a role in the development of the fetal brain.113,114 Inadequate thyroid hormone replacement during pregnancy may result in psychoneurological and developmental deficits in the progeny.115,116
Animal data suggest that mild or moderate hypothyroidism has minimal effect on fertility, although hypothyroid animals do have difficulty maintaining pregnancy. A study of 244 pregnant hypothyroid women found the rate of stillbirth was double that in control subjects.117 Another study evaluated the outcome of a group of children whose mothers had proved or suspected thyroid disorders during pregnancy. This smaller study revealed that outcomes were poor in six of seven pregnancies in women with clinically suspected hypothyroidism and low butanol-extractable iodine (BEI).
The work of Man et al suggested that the progeny of inadequately treated "hypothyroxinemic" women suffered psychoneurological deficits. Developmental deficits were demonstrable as early as 8 months of age118. A 7-year follow-up study revealed that these children had persistent deficits, manifested in lower psychological test scores although confounding factors, such as socioeconomic status, may have played a role in the poor outcome. More recently, a 7-9 year follow-up study of 62 euthyroid progeny of hypothyroid mothers confirmed these findings. Children of hypothyroid mothers who remained untreated during pregnancy achieved lower IQ scores than controls. In addition, a larger percentage (15%) of children of untreated mothers had IQ scores of 85 or less, compared with 5% among controls.119 Together these findings suggest that thyroid hormone influences the developing fetal brain, impacting neuronal differentiation and multiplication among other events.
It has been observed that hypothyroid women experience increased fetal loss. As a result of this observation, serum thyroid hormone determinations were evaluated in women who aborted spontaneously and were found to be low.164 Normal values were found in the majority of patients who had elective abortions, although a minority of these women did have low thyroid hormone levels as well. Follow-up studies in those patients with low thyroid hormone levels who had undergone elective abortions determined that they were euthyroid and presumably had not been hypothyroid during pregnancy. The data suggested that lower estrogen levels post-termination were associated with concomitant decreases in TBG and serum thyroid hormone concentration. Therefore, the low serum thyroid values were felt to be secondary to the abortion, rather than the cause. Because the low serum T4 merely reflects the decreased estrogen production and TBG, there is no reason to suppose that thyroid hormone would be helpful in these situations.120
The balance of evidence suggests that the great majority of women with early spontaneous abortions have normal thyroid function. However, an association has been made between the presence of thyroid autoantibodies (anti-thyroid peroxidase and anti-thyroglobulin antibodies) and an increased risk of spontaneous abortion. In a study of 552 first trimester, antibody-positive pregnant women the rate of spontaneous abortion was 17%, compared with 8.4% in antibody-negative women.121 The miscarriage rate was found to be unrelated to the titer of autoantibody or the level of serum TSH. It is important to note that secondary causes of abortion were not excluded in this study, and may have had a confounding effect on the results. Nonetheless, this initial observation prompted further investigation into the potential association between thyroid autoantibodies and pregnancy loss. A recent review of the available evidence122 cites successive corroborative studies in Belgium123 and Japan124, involving hundreds of women, that found 2-fold to 4-fold increased rates of spontaneous abortion in autoantibody-positive women compared to controls. Subsequent studies in smaller study populations found an increased incidence of thyroid antibodies only in patients with recurrent, but not spontaneous, abortion.125,126,127. The etiology of pregnancy loss in women who are thyroid antibody-positive remains unknown. It has been hypothesized that the presence of thyroid autoantibodies reflects a generalized activation of the immune system, and that the antibodies are not the causative agents of spontaneous abortion. It is also possible that subtle, subclinical deficiencies in thyroid hormone levels occur in these women, predisposing them to pregnancy loss. The question of intervention has been raised. Small studies128,129,130,131 have addressed the issue of immunomodulation using IVIG. Due to sub-optimal study designs and size constraints, the impact is unclear and the results are inconclusive.
Hypothyroidism is most commonly iatrogenic, the result of either thyroid surgery or the administration of radioactive iodine therapy. Idiopathic hypothyroidism is most often attributable to underlying Hashimoto's disease, and its onset may be more insidious. Hashimoto's disease is more common in patients with diabetes mellitus; in one study of 100 diabetic women, 20% of patients with type I diabetes also had Hashimoto's disease.132
The symptoms referable to underlying hypothyroidism are common to all of the underlying etiologies. Patients complain of constipation, cold intolerance, cool, dry skin, coarse hair, irritability and inability to concentrate. Of note, however, there is significant overlap with complaints common to euthyroid pregnant women, making the clinical diagnosis difficult. The presence of paresthesias may be helpful, as it is an early symptom in approximately 75% of patients with hypothyroidism. The presence of delayed deep tendon reflexes is also suggestive of hypothyroidism. In addition, signs of gross myexedma, including a low body temperature, large tongue, hoarse voice, and periorbital edema are not found in normal pregnancy, and their presence should prompt an immediate evaluation for hypothyroidism. Patients may complain of excessive fatigue. Gestational hypertension is common.133 Postpartum amenorrhea and galactorrhea associated with hyperprolactinemia may be indicative of hypothyroidism.134,135
An elevated TSH in association with a low serum T4 concentration is the most sensitive indicator of primary hypothyroidism. Because of the elevated TBG of pregnancy, the serum T4 determination may not be as low as would be expected, and may appear inappropriate in the setting of an elevated TSH. Thyroid autoantibody positivity provides supporting evidence for the presence of hypothyroidism, particularly in the absence of a past history of thyroidectomy or radioactive iodine therapy. Elevated serum cholesterol concentrations are common to hypothyroidism but are not helpful in the diagnosis, as serum cholesterol concentrations increase during gestation, up to 60% above pre-pregnancy values.
Full replacement doses of T4 should be given immediately upon the diagnosis of hypothyroidism in the pregnant woman, regardless of the degree of thyroid function. This will minimize further fetal exposure to the hypothyroid milieu. Therapy can be titrated rapidly136 in young pregnant women with no other co-morbid conditions. One reasonable schedule is to start with 0.100 mg of T4 daily for 3 to 5 weeks, with subsequent dosage adjustments depending on thyroid function test results. T4 need be given only once a day because of the long half-life. With adequate treatment, the serum TSH concentration should decrease to values below 6 mU/ml, usually within four weeks, and the serum T4 concentration should increase to normal values for pregnancy. The optimal range for TSH during pregnancy is <3.0 mU/ml. Note that normal serum T4 concentrations in pregnancy are above the upper limit of the normal range for non-pregnant women. This is due to an increase in thyroxine binding in the setting of pregnancy-induced increases in serum TBG concentrations. The free T4 concentrations should be brought into the upper range of normal. If the values do not return to normal, the dose of T4 should be increased by 0.05-mg increments. The serum TSH concentration may take longer to return to normal values.
The majority of pregnant women receiving thyroid hormone during gestation have been diagnosed with hypothyroidism prior to conception. In many cases, the reason for the initial diagnosis of hypothyroidism may be obscure. The number of women in whom hypothyroidism is diagnosed during pregnancy is small. To ensure that the pregnant woman is receiving adequate thyroid hormone, full replacement doses must be given. As mentioned previously, the dose of thyroid hormone commonly has to be increased by up to 50% during pregnancy.137 The recommended replacement dose of T4 is approximately 0.10 mg/day, based on the amount of T4 necessary to suppress the elevated serum TSH concentration.138 Women on thyroid hormone therapy require close follow-up during pregnancy to maintain optimal thyroid hormone concentrations.139,140,141,142
In women with no prior history of hypothyroidism in whom the diagnosis is made during pregnancy, thyroid hormone therapy may be discontinued during the postpartum period, with reassessment of thyroid function 5 to 6 weeks later. Recovery may be delayed, as normal thyroid function may remain suppressed for a number of weeks after prolonged thyroid hormone therapy. Most will remain hypothyroid and will ultimately require chronic replacement therapy.