In 1990, a study was carried out looking at the causes of post-partum thyroiditis (1). This uncovered, by chance, an association between the presence of thyroid autoantibodies (thyroid peroxidase antibody) and the rate of miscarriage and preterm birth. At the time, this association was largely disregarded as a plausible causative hypothesis did not exist. Twenty one years later, a wealth of studies carried out support the initial observation. This meta-analysis on 31 studies (case control and cohort) demonstrates an increased risk of both miscarriage and preterm birth with the presence of thyroid peroxidase antibodies.

In an attempt to determine whether treating the positive antibody status with thyroxine leads to any benefit, the authors cite two studies. Both these studies, carried out by the same group of investigators, have some serious flaws. First of all, the antibody positive group are older, secondly, one of the studies is not placebo controlled and thirdly, the studies are carried out in Italy, in an iodine deficient population. However, both studies demonstrated a significant reduction in miscarriage and preterm birth with thyroxine treatment.

Many GP’s and endocrinologists are currently checking the thyroid antibody status on infertile or subfertile women. The question about whether or not to treat these women with thyroxine is hotly debated. Some feel that thyroxine will do no harm and may be beneficial. Others feel that the appropriate evidence for treatment is not yet available. An important caveat in all of this would be to remember that rendering a pregnant woman hyperthyroid could also have serious metabolic consequences to mother and unborn child.