Is thyroxine administration the answer?

TOPIC: Autoimmune thyroid disease during pregnancy

Title: Levothyroxine treatment in euthyroid pregnant women with autoimmune thyroid disease: effects on obstetrical complications.

Authors: Negro R , Formoso G , Mangieri T , Pezzarossa A , Dazzi D , & Hassan H.

Reference: Journal of Clinical Endocrinology & Metabolism 91:2587-2591, 2007



Euthyroid women with autoimmune thyroid disease show impairment of thyroid function during gestation and seem to suffer from a higher rate of obstetrical complications.

Objective: The authors sought to determine whether these women suffer from a higher rate of obstetrical complications and whether levothyroxine (L-T4) treatment exerts beneficial effects. This was a prospective study conducted in a Department of Obstetrics and Gynecology. The literature on the impact of thyroid abnormalities on pregnancy and the postpartum has expanded rapidly over the last two decades.


A total of 984 pregnant women were studied from November 2002 to October 2004; 11.7% were thyroid peroxidase antibody positive (TPO-Ab(+)). TPOAb(+) women were divided into two groups: Group A (n = 57) was treated with L-T4, and Group B (n = 58) was not treated. The 869 TPOAb(-) women (Group C) served as a normal population control group. Rates of obstetrical complications in treated and untreated groups were compared.

Main Results: At baseline, TPOAb(+) had higher serum TSH compared with TPOAb(-); TSH remained higher in Group B compared with Groups A and C throughout gestation. Free T4 values were lower in Group B than Groups A and C after 30 weeks and after parturition. Groups A and C showed a similar rate of miscarriage (3.5% and 2.4%, respectively), which was lower than in Group B (13.8%) [P < 0.05; relative risk (RR) = 1.72; and P < 0.01; RR = 4.95]. Group B displayed a 22.4% rate of premature deliveries, which was significantly higher than Group A (7%) (P < 0.05; RR = 1.66) and group C (8.2%) (P < 0.01; RR = 12.18).


Euthyroid pregnant women who are positive for TPOAb may develop impaired thyroid function during pregnancy. The presence of thyroid autoantibodies is associated with an increased risk of miscarriage and premature deliveries. In this study substitutive treatment with L-T4 was able to lower the chance of miscarriage and premature delivery among women who were positive for these antibodies in the first trimester of pregnancy.


This unique study has generated much interest, and also has presented physicians with a therapeutic quandary. Numerous retrospective studies have found a 3-6 fold increment in the incidence of miscarriage among women with positive anti-TPO antibody assays. The reason for this association is not entirely known. The association does not seem to be related to a generalized autoimmune process. One hypothesis is that thyroid autoimmunity may be associated with mild or subclinical hypothyroidism, which is known to develop in up to 40% of previously euthyroid antibody-positive women during the course of pregnancy. To date no reported prospective randomized study has examined this issue, although two studies are now underway that will (hopefully) be informative.

The study by Negro et al. was not specifically designed to examine the role of mild hypothyroidism in relation to miscarriage, but certainly offers suggestive data. The authors assayed serum TSH levels in the women to be given L-T4, and assigned doses of either 1.0, 0.75, or 0.5 -g/Kg b.w., depending on the initial TSH level. The women were followed throughout pregnancy. Serum TSH levels in the treated group paralleled those of the antibody-negative group during the last two trimesters, as did their free T4 levels, while serum TSH levels in the untreated but antibody-positive group climbed to the -top normal- upper limit by the end of pregnancy. While these results are very interesting, it must be remembered that none of the treated patients had proven hypothyroidism. Perhaps more importantly, this is one observation, and has not yet been confirmed by other groups.

What message should treating physicians take from the study? Should all antibody-positive women be treated with thyroid hormone? Obviously, this might risk complications including hyperthyroidism. From the opposite view, the possibility of nullifying the significant risk of miscarriage and prematurity by administering a safe dose of thyroid hormone seems quite attractive. This writer can not recommend such therapy as a general rule. But it is also clear that many antibody-positive pregnant women will be given thyroid hormone on an individual basis, while we await the results of other trials.

Summary and commentary prepared by Leslie DeGroot (related to Chapter 14 of TDM)