High TSH Levels Associated With Frailty in Older Women

Carole VanSickle Ellis

May 18, 2010

Dr. George Wang

May 18, 2010 (Orlando, Florida) — Women with thyroid-stimulating hormone (TSH) levels greater than 5 mU/L have a higher risk of incident frailty, George C. Wang, MD, assistant professor of medicine at Johns Hopkins School of Medicine in Baltimore, Maryland, reported here at the American Geriatrics Society (AGS) 2010 Annual Scientific Meeting.

"When we looked at participants according to their TSH levels, we could clearly see that the individuals with TSH between 3 and 5 [mU/L] and individuals with TSH greater than 5 [mU/L] had higher incidence rates than [those with] TSH between 1 and 3 [mU/L] [85.7% and 92.9% vs 55.2%]," he told Medscape Internal Medicine. Ultimately, those numbers translated into a hazard ratio of 1.76 in the group with THS greater than 5 [mU/L]. After the hazard ratio was adjusted for confounders, it actually rose to 2.28 for THS greater than 5 [mU/L], said Dr. Wang.

The study sample consisted of 608 women between 70 and 79 years of age from the Women's Health and Aging Studies I and II. Serum TSH and total thyroxine (T4) were measured by immunochemiluminometric assay and immunoassay, respectively. Frailty status was measured using the "validated 5-point criteria of shrinking, exhaustion, low energy expenditure, slowness, and weakness," said Dr. Wang. Participants were classified as frail if they met 3 or more of the criteria, prefrail if they met 1 or 2 of the criteria, and nonfrail if they met none of the criteria.

"Incident frailty" was defined as the development of new-onset frailty during 3 years of follow-up in patients not frail at baseline, Dr. Wang reported. "We had 530 participants at risk at baseline who were not frail; over the 3-year follow-up, 86 developed frailty," leading to an incidence rate of 56.8 cases per 1000 person-years, Dr. Wang said.

The team also adjusted for age, race, education, body mass index, smoking history, cardiovascular disease, and thyroid medication. "These associations are statistically significant," he told Medscape Internal Medicine, adding that, "to me, the adjusted analysis is the most important because we have taken into consideration the additional confounding effects of other participant factors and comorbidities."

Current guidelines do not recommend treating patients with TSH levels between 4.5 and 10 mU/L who have normal T4 levels and no symptoms of hypothyroidism. "Thyroid hormone replacement therapy might be beneficial for older adults with TSH greater than 5 [mU/L]," said Dr. Wang. "Of course, further research is needed to determine the benefits of such a treatment, and the risks, if any."

The more than 2-fold increased risk for frailty in older women with high TSH levels "suggests that the loss of thyroid function is associated with frailty," Sharon A. Brangman, MD, FACP, ASGF, president of the AGS, professor of medicine, and division chief of geriatric medicine in the Department of Medicine at Upstate Medical University in Syracuse, New York, told Medscape Internal Medicine.

Currently, "there may be some hesitation in treating hypothyroidism in older women because the treatment can sometimes cause cardiac arrhythmias, increases in blood pressure, and other problems," said Dr. Brangman. "These results indicate that we should probably treat — and treat very carefully — so that we can help reduce some of the side effects that come with aggressive thyroid replacement."

This research was supported by Atlantic Philanthropies, the American Geriatrics Society, the John A. Hartford Foundation, the Association of Subspecialty Professors, the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute on Aging, Johns Hopkins Hospital, and Johns Hopkins Bayview Medical Center General Clinical Research Center. The researchers have disclosed no relevant financial relationships.

American Geriatrics Society (AGS) 2010 Annual Scientific Meeting: Abstract P32. Presented May 15, 2010.