Correspondence
Subclinical Hypothyroidism
N Engl J Med 2001; 345:1855-1856December 20, 2001
- Article
To the Editor:
In his review of subclinical hypothyroidism, Cooper (July 26 issue)1 did not fully address the complex issue of prenatal screening. Since Haddow et al.2 reported that the offspring of women with elevated thyrotropin concentrations had slightly but significantly lower IQ scores (by 4 points), there has been considerable controversy over the potential impact of thyroid screening during pregnancy on subsequent developmental delays in the children of these women.3
The efficacy at prenatal screening is unproved. Fetal neurologic development is most dependent on maternal thyroid hormone early in pregnancy, since fetal thyroxine is not detectable until 10 weeks' gestation.4 Thus, it seems unlikely that screening and treatment after pregnancy is established would reverse damage related to maternal hypothyroidism.
As a result, the American College of Obstetricians and Gynecologists5 concluded that “it is possible that screening for hypothyroidism as early as the initial prenatal visit in the first trimester may be too late to obviate the potential deficiency in fetal brain development.” As with folic acid supplementation, which it is recommended be initiated before conception in order to prevent neural-tube defects, screening for hypothyroidism may need to be done before pregnancy in order to affect fetal development. But before mandating prenatal screening for 4 million women annually in the United States, we must carefully study the efficacy of such a strategy.
5 ReferencesJodi S. Dashe, M.D.
F. Gary Cunningham, M.D.
University of Texas Southwestern Medical Center, Dallas, TX 75235-9032
jodi.dashe@utsouthwestern. edu 1
Cooper DS . Subclinical hypothyroidism. N Engl J Med 2001;345:260-265
Full Text | Web of Science | Medline2
Haddow JE ,Palomaki GE ,Allan WC . Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N Engl J Med 1999;341:549-555
Full Text | Web of Science | Medline3
Davies TF . The ATA, the Endocrine Society, and AACE confuse endocrinologists on thyroid disease in pregnancy. Thyroid 2000;10:107-107
CrossRef | Web of Science | Medline4
Burrow GN ,Fisher DA ,Larsen PR . Maternal and fetal thyroid function. N Engl J Med 1994;331:1072-1078
Full Text | Web of Science | Medline5
Committee on Obstetric Practice. Screening for hypothyroidism. Washington, D.C.: American College of Obstetricians and Gynecologists, September 2000. (ACOG Committee Opinion 241.)
To the Editor:
Dr. Cooper suggests treating some patients who have mildly elevated thyrotropin levels (range, 5 to 10 mU per liter) even if tests for antibodies against thyroperoxidase are not positive. I have repeatedly found discrepancies in thyrotropin measurements between reputable laboratories; for example, values reported by one laboratory as in the range of 5 to 10 mU per liter and thus abnormal are reported by another laboratory as in the range of 3 to 5 mU per liter and thus normal. I am concerned that the proposed strategy could lead to unwarranted treatment in some patients.
Alexander Mazziotti, M.D., Ph.D.
Valley Hospital, Ridgewood, NJ 07506
doc@mdmazz.com To the Editor:
In patients with subclinical hypothyroidism, free thyroxine levels are, by definition, within the normal range, but it should be emphasized that the normal range refers to the whole population. In contrast, in a patient with subclinical hypothyroidism, the free thyroxine level is below the normal range for that particular person (this range is much narrower). Otherwise, the thyrotropin level would not be considered to be increased.
Martin Press, F.R.C.P.
Royal Free Hospital, London NW3 2QG, United Kingdom
martin.press@rfh. nthames. nhs. uk Author/Editor Response
Dr. Cooper replies:
To the Editor: Drs. Dashe and Cunningham point out one of the many complex issues surrounding a screening program for maternal hypothyroidism — namely, the possibility that preconception screening may be more effective than screening during the initial prenatal visit. Although this makes intuitive sense, it should be recalled that Haddow et al. found that the IQ scores were not significantly different between control children and children whose mothers had been treated for hypothyroidism that had been recognized during pregnancy.1 Similar data were reported by Man and Serunian decades ago.2 It is unlikely that there will ever be a prospective, randomized trial comparing the outcomes of a preconception screening strategy with those of screening during early pregnancy. And, in my view, it would be unethical to assign women with subclinical hypothyroidism to a placebo group in a randomized trial. Therefore, the disparate opinions expressed by various organizations on this issue must be reconciled, perhaps by means of a consensus development conference.
Dr. Mazziotti wonders about variations between laboratories in serum thyrotropin measurements. One possible explanation for the observed differences is the pulsatile nature of thyrotropin secretion. With respect to the issue of variability between laboratories, my clinical sense is that thyrotropin levels on the high end are generally reproducible. It may be true that one laboratory will report a value of 6 mU per liter and another will report a value of 4 mU per liter in the same patient. I believe, however, that the upper limit of the normal range is actually closer to 2.5 or 3 mU per liter, rather than the traditional value of 5 mU per liter used by most laboratories. Although the magnitude of the risk of eventual hypothyroidism is less clear for thyrotropin levels in the high-normal range than for frankly elevated levels, I would view a thyrotropin level of 4 to 6 mU per liter as suggestive of subclinical hypothyroidism. Whether to treat or follow patients with such levels depends on a variety of factors discussed in my review.
As Dr. Press suggests, the clinically incontrovertible point is that one does not know what free thyroxine value is normal for an individual person unless the serum thyrotropin level is measured.
2 ReferencesDavid S. Cooper, M.D.
Sinai Hospital of Baltimore, Baltimore, MD 21215
dcooper@lifebridgehealth.org 1
Haddow JE ,Palomaki GE ,Allan WC , et al. Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N Engl J Med 1999;341:549-555
Full Text | Web of Science | Medline2
Man EB ,Serunian SA . Thyroid function in human pregnancy. IX. Development or retardation of 7-year-old progeny of hypothyroxinemic women. Am J Obstet Gynecol 1976;125:949-957
Web of Science | Medline
