The trusted source for
healthcare information and
Abstract & Commentaries
Synopsis: A recent study concluded that undiagnosed hypothyroidism in pregnant women may adversely affect neuropsychological development in the fetus, and that screening for hypothyroidism early in pregnancy may be worthwhile, even in asymptomatic women.
Source: Haddow JE, et al. N Engl J Med 1999;341: 549-555.
Normal maternal thyroid function is known to play an important role in fetal neurologic development. To determine if maternal hypothyroidism effects neuropsychological development in the fetus, Haddow and associates measured serum thyrotropin (TSH) in blood samples from 25,216 women collected in the second trimester as part of a screening program for open neural tube defects and Down’s syndrome. Forty-seven women were identified with TSH levels at or above the 99.7th percentile and 15 with values between the 98th and 99.6th percentiles in combination with low thyroxine levels. These women were matched with 124 women whose values were normal. Serum TSH levels averaged 13.2 mU/liter in women with hypothyroidism as compared with 1.4 mU/liter in controls. Serum thyroxine and free thyroxine concentrations were also significantly lower in hypothyroid women. Of note, antithyroid antibodies were found in 77% of women with hypothyroidism vs. 14% of controls. None of the children of the hypothyroid women were hypothyroid at birth. At 7-9 years of age, all children were tested to evaluate intelligence, attention, language and reading skills, school performance, and visual-motor performance. The study subjects were matched for their parent’s educational level and occupation, maternal age of delivery, and sex of the infant. Overall, the children of hypothyroid mothers did significantly less well in word discrimination and measurements of sustained vigilance and attention. Of the 62 women with hypothyroidism, 15 had a diagnosis made prior to pregnancy and 14 were being treated. When the evaluations of the 48 children of untreated women with hypothyroidism were compared to those of control women, a reduction of 7 IQ points was observed while there was no significant difference found for the children of treated women with hypothyroidism. Nineteen percent of the children of women whose hypothyroidism was not treated had an IQ score of 85 or less as compared with 5% of the control children.
Haddow et al conclude that undiagnosed hypothyroidism in pregnant women may adversely affect neuropsychological development in the fetus. They propose that screening for hypothyroidism early in pregnancy may be worthwhile, even in asymptomatic women. They suggest that testing be performed at the first prenatal visit, preferably in the first trimester, and that women with a positive screen receive timely follow-up and treatment as needed.
Comment by Steven G. Gabbe, MD
If your practice is like mine, your patients have heard about this study and have begun requesting testing for hypothyroidism. Should this become part of our routine prenatal battery? I don’t think so. Only one in 400 women were found to have elevated TSH levels and hypothyroidism. And, of the 62 women identified with hypothyroidism in this study, 15 were known to have this disorder prior to pregnancy. The children of mothers known to be hypothyroid and treated prior to pregnancy were found to have no neuropsychological deficits on testing at 7-9 years of age, even though their mothers had elevated TSH levels and lower serum thyroxine concentrations. In contrast, the children of the 48 women with untreated hypothyroidism had lower IQ scores. It should be noted that all of the children born to hypothyroid women in this study had normal neonatal thyroid screening tests.
Should all pregnant women be screened for hypothyroidism? Before this question can be answered, we need more information. Haddow et al have not provided data on the costs of screening and the benefits expected, or on the occurrence of pregnancy-related complications associated with hypothyroidism such as preeclampsia and preterm delivery.
Comment by Sarah L. Berga, MD
This is an interesting study with important results. Haddow et al conclude that asymptomatic maternal hypothyroidism can adversely affect the offspring’s later neuropsychological development. They cite a similar, but smaller, study showing that lower maternal-free thyroxine in the first trimester was associated with impaired psychomotor development at 10 months of age.1 Thyroid hormone is also important for the later stages of fetal brain development when neuronal migration and organization occur. Thus, a mechanism for the deleterious consequences of maternal hypothyroidism in the second and third trimester is provided. In an accompanying editorial, Utiger and associates point out that the mother is the sole source of fetal thyroxine in the first trimester, but also the predominant source in the second and third trimesters.2 Thus, the fetus remains critically dependent upon maternal thyroidal function throughout gestation. Haddow et al suggest that maternal screening should be considered, but Utiger et al point out the prevalence of maternal hypothyroidism in the United States, as estimated from another survey, was only 2.5%. Still, screening for hypothyroidism involves only a serum TSH test, the cost of which is generally around $30. This determination could easily be made from the blood that is routinely obtained at the first prenatal visit. Utiger et al also caution us to remember that iodine deficiency is the only preventable cause of maternal hypothyroidism. Since 15% of women of childbearing age are iodine-deficient, Utiger et al suggest that iodine deficiency in pregnancy is best prevented by adding extra iodine to prenatal vitamins, or increasing the amount in table salt and other foods. The latter strategy would benefit others as well.
There is one other less common cause of hypothyroidism that needs to be considered in this discussion. Women with functional hypothalamic amennorrhea (FHA), particularly those with anorexia nervosa, have what can best be thought of as "hypothalamic hypothyroidism." This is because stress-related increases in cortisol secretion lead to reduced TSH and TRH secretion. This is true whether the stress is performance pressure, excercise, or inadequate nutrition. In our study population of women with FHA of normal body weight who were without overt eating disorders, thyroxine levels were roughly 25% lower than those of the eumenorrheic comparison group.3 The reduction of thyroxine in FHA is comparable to the reductions seen in the present study, which were in the range of 30%. Ovulation induction in women with FHA overcomes the decrease in gonaddotropin-releasing hormone and gonadotropins that are the proximate cause of anovulation, but ovulation induction does not correct the underlying concomitant metabolic deficits, including the relative hypothyroidism due to decreased hypothalamic drive. Are the offspring of women who undergo ovulation induction for FHA and conceive at increased risk for poor neuropsychological development? I know of no study that has addressed this issue, but the present data certainly heightened my sense of alarm. At the least, women with stress-related anovulation need to be counseled about this possibility. Since FHA is theoretically reversible with lifestyle management, clearly that type of intervention is warranted before ovulation induction is undertaken. Should we give thyroxine to women with FHA undergoing ovulation induction? I suppose that is another possibility, but that strategy also has limitations. First, there are metabolic alterations other than thyroid that would persist and might have fetal consequences. Second, the hypothalumus has made an "executive decision" in shutting down hypothalmic-pituitary-ovarian and hypothalmic-pituitary axes and it seems cavalier to think we are smart enough to intelligently overide this mechanism.
1. Pop VJ, et al. Clin Endocrinol 1999;50:149-155.
2. Utiger RD, et al. N Engl J Med 1999;341:601-602.
3. Berga SL, et al. J Clin Endocrinol Metab 1989;68: 301-308.
a. total thyroxine
b. free thyroxine
c. anti-thyroid antibodies
e. thyroxine-binding globulin