VICTORIA, B.C. – , according to findings from a study reported at the annual meeting of the American Thyroid Association.
“In the U.S., individuals are frequently treated either for just a number – just because their thyroid-stimulating hormone (TSH) is elevated – or for nonspecific hypothyroid-type symptoms, such as weight gain, cold intolerance, and such. It’s extremely common,” lead investigator Douglas Bauer, MD, professor and internist at the University of California, San Francisco, commented in an interview.
Susan London/Frontline Medical News
Dr. Douglas Bauer
“On average, this study suggests that you shouldn’t be using hypothyroid-type symptomatology to treat subclinical hypothyroidism,” he said, while also acknowledging that not writing that prescription can be challenging. “It’s hard to do nothing, I know.”
Dr. Bauer and his coinvestigators performed a subgroup analysis of the randomized controlled TRUST trial (Thyroid Hormone Replacement for Subclinical Hypothyroidism), conducted in Switzerland, Ireland, the Netherlands, and the United Kingdom. In the trial, 737 adults aged 65 years or older with persistent subclinical hypothyroidism (TSH level, 4.60-19.99 mIU/L, with normal free thyroxine level) were given either levothyroxine or placebo on a double-blind basis.
Results for the entire trial population, previously reported, showed that at 1 year, patient-reported symptoms on a thyroid-specific quality-of-life questionnaire had improved by a similar extent in both groups, with no significant differences between them (N Engl J Med. 2017;376:2534-44).
The new analysis focused on two subgroups that might be especially expected to benefit: 132 patients with a hypothyroid symptoms score greater than 30 on a 100-point scale and 209 patients with a tiredness score greater than 30 on a 100-point scale.
Results reported in a poster session showed that at 1 year, scores had improved by about 10 points with levothyroxine and placebo alike, with no significant difference, in both the group with higher hypothyroid symptoms scores (P = .90) and the group with higher tiredness scores (P = .80).
“This provides additional evidence that it’s unlikely that the treatment of subclinical hypothyroidism, at least in this population, is going to lead to symptomatic improvement,” Dr. Bauer commented.
“I would speculate that we’ve overestimated the benefit [of thyroid hormone therapy] on symptoms based on the fact that previous studies haven’t been blinded,” he said, noting that the TRUST trial used a rigorous blinding protocol, even going so far as to change the appearance of placebo pills to convince patients in the placebo group that their dose was being titrated.
A potential criticism is that the treatment was not aggressive enough, with patients in the levothyroxine group achieving a mean TSH level of 3.6 mIU/L, according to Dr. Bauer. “I think many people treating with thyroxine would like to see this TSH fall into the 2 mIU/L to 3 mIU/L range,” he acknowledged.
Taken together, the trial’s overall and subgroup findings do not rule out potential benefit for certain patients, he cautioned. For example, patients with very high symptom burden and patients with baseline TSH levels greater than 10 mIU/L were too few for separate analysis, and younger adults were not included at all.
Additionally, treatment impact on other important clinical outcomes – cardiovascular events and fractures – could not be assessed in TRUST because of insufficient enrollment.
Dr. Bauer disclosed that he had no relevant conflicts of interest. The trial was funded by the European Union, Swiss National Science Foundation, Swiss Heart Foundation, and Velux Stiftung. Merck supplied study drug.