The prevalence of hypothyroidism increases with age, and this condition is more common in women than in men. Framingham study data show the prevalence to be 5.9 percent in women and 2.3 percent in men older than 60 years. Results from another study conducted in Massachusetts also showed the overall prevalence of hypothyroidism in the 60-and-older age group to be 5.9 percent. The diagnosis of hypothyroidism in both studies was confirmed by a thyroid-stimulating hormone (TSH) level above 10 μU per mL (10 mU per L). Previously published research data indicated that many patients from community-dwelling settings were taking thyroid hormone replacement therapy inappropriately. Some elderly patients may have begun therapy before new assays for TSH were available or as treatment for fatigue, hair loss or hyperlipidemia. Coll and Abourizk recently performed a study to determine if thyroid hormone replacement therapy could be withdrawn from adult patients without producing any adverse effects.

Participants in the study were residents of four nursing homes who were taking thyroid hormone therapy. Initial exclusion criteria included prior laboratory documentation in the patient's medical record of a TSH level above 10 μU per mL. Patients were also excluded if they were taking lithium or amiodarone, or if they had a history of thyroid nodule or goiter.

A TSH assay was obtained for eligible patients, and those with a level above 7 μU per mL were excluded from the study. Patients with a TSH level below 7 μU per mL had their current dosage of thyroid hormone reduced by as close to one half as practical (e.g., from 125 to 75 μg per day or from 75 to 50 μg per day). A repeat TSH assay was obtained four weeks later and, if the level exceeded 7 μU per mL, the patient was excluded from the study and resumed taking the original dosage. In the remaining patients, thyroid replacement therapy was discontinued, and another TSH assay was obtained approximately four weeks later. If the follow-up TSH was 7 μU per mL or less, a free thyroxine (T₄) level was obtained from the same sample of blood. If the free T₄ level was above 0.8 ng per dL (103 pmol per L), the patient remained off thyroid therapy, and the TSH was measured again two months later. Therapy was permanently discontinued if the level was still below 7 μU per mL.

The study selected 915 patients from the four nursing homes, of whom 115 were taking thyroid hormones. Thirty-one patients were excluded because of lack of informed consent or unwillingness on the part of their physicians to participate. Other patients were excluded because of death, elevated baseline TSH level, or use of lithium or amiodarone. Twenty-two patients were ultimately included in the study. Their mean age was 78 years, and 20 participants were women. On completion of the study, 11 of the 22 participants were successfully withdrawn from thyroid replacement therapy. Patients who were taking more than 50 μg per day of thyroid hormone were more likely to exhibit a significant increase in TSH level following dosage reduction.

The authors conclude from the results of this study that thyroid hormone therapy can be withdrawn in a significant number of nursing home patients. Their findings are similar to those from studies of patients in community-based settings. The authors note * that it was previously reported that over-treating with thyroid hormone may adversely affect patients' myocardial contractility and decrease bone density.

editor's note: This study should encourage physicians who care for elderly patients—especially those in nursing homes—to confirm the diagnosis of hypothyroidism by appropriate use of a sensitive TSH assay. One of the geriatricians in our practice has taken this a step further and at times has boldly suggested stopping all medications in select elderly patients, many of whom are concurrently taking 10 or more prescription and over-the-counter medications, and reassessing which medications they truly need.—j.t.k.