Treating Subclinical Hypothyroidism Doesn't Reduce Deaths

Becky McCall

May 05, 2014

WROCŁAW, Poland — Treatment of subclinical hypothyroidism with levothyroxine does not spare lives, according to a new study. The findings are surprising for a condition that is linked to increased cardiovascular risk, say the investigators.

"It's a patient population with increased signs of cardiovascular disease. The state of subclinical hypothyroidism increases blood lipid levels, so the implication is that treatment should help," remarked Jens Faber, MD, coinvestigator and professor of endocrinology at Copenhagen University, Denmark, in an interview with Medscape Medical News.

Research fellow and medical student Mette Nygaard Andersen from Gentofte University Hospital, Hellerup, Denmark, led the research. She presented findings from the study, the largest of its kind, at the European Congress of Endocrinology (ECE) 2014 here yesterday.

No benefit with respect to all-cause mortality was identified in patients treated with levothyroxine compared with those who didn't receive therapy (incidence rate ratio, 1.02), a nonsignificant finding.

Ongoing Controversy: To Treat or Not to Treat?

Patients with subclinical hypothyroidism have an underactive thyroid gland but a normal amount of circulating thyroid hormone. But in an effort to overcome the underactive thyroid gland, increased levels of thyroid stimulating hormone (TSH) are produced by the pituitary gland, which works overtime to try to maintain a normal circulating thyroid hormone level.

The overall prevalence of subclinical hypothyroidism is around 5% to 10% in the adult US population; the diagnosis is more common in women and increases with age. And it occurs more frequently in whites than in blacks, according to figures from 2002 .

It has long been debated whether treatment of this condition is of value, explained Dr. Faber, but recent data suggest it might be unfavorable to interfere with this subclinical situation because there is a risk of overtreatment.

"If we treat, then we also run a real risk of overtreating, which can cause heart attacks and cardiac arrhythmias in some cases. We know from questionnaires that patients tend to treat themselves, and this may lead to overtreatment because the drug [levothyroxine] aids weight loss, which patients welcome," he pointed out.

In fact, Dr. Faber added that even in symptomatic cases, it has not been unequivocally proven that treatment really helps people.

"Since subclinical hypothyroidism is related to cardiovascular disease, then some proponents of treatment say it should be treated to prolong life and reduce mortality. This is debatable according to the new results."

Effectively, to treat or not to treat remains an ongoing controversy. "In light of this we wanted to investigate whether it was worth treating these patients," said Ms. Nygaard Andersen.

Dr. Faber added that patients also have a significant voice in the debate. Many believe they should be treated, given the potential for adverse outcomes, and some studies have shown an improvement in symptoms, he noted.

But he added that subclinical hypothyroidism is a mild biochemical disturbance. "Our response is that this isn't necessarily a condition that should be treated, given the mortality figures, and that clinicians should consider being conservative in management.

"In the long run, if thyroid function continues to decline, then at this stage treatment is recommended, so the advice is to follow the development of the thyroid function rather than jump to treatment immediately."

Longer Follow-up Might Change Findings

In the new study, the researchers looked at all-cause mortality, not cardiovascular mortality, specifically due to a lack of epidemiological coding to adequately identify mortality due to cardiovascular events.

Of a total cohort of 628,953 patients 12,212 (1.9%) had subclinical hypothyroidism at baseline (defined as elevated TSH with normal free T4). The vast majority were women (79.8%), and mean age was 55.2 years. Information was obtained from individual-level linkage of nationwide Danish registries.

Those with a history of thyroid disease, related medication, or treatment with lithium, amiodarone, and glucocorticoids were excluded.

Within the first 6 months, 2483 patients (20.3%) were prescribed levothyroxine, a synthetic thyroid hormone. The remaining 9729 patients (79.7%) either initiated levothyroxine therapy later than 6 months after their initial blood test, in which case they were not considered in the analysis, or otherwise did not receive any substitute treatment at all. Patients were followed for 5 years.

The results showed that the overall mortality rate was 26/1000 person-years in the untreated patients compared with 21/1000 person-years among the levothyroxine-treated group. In total, 1566 out of 12,212 patients with subclinical hypothyroidism died.

Incidence rate ratios (IRR) of all-cause mortality were analyzed using Poisson regression models, but there was no difference between the 2 groups.

"This is a huge sample of patients from the general population," emphasized Dr. Faber. "Other studies have assessed 10 times fewer patient numbers. This is a key strength of this study."

He noted, however, that they only took a single index sample of thyroid hormone; in the future, they would like to take repeated samples to see if the levels change, to clarify activity.

And between 2% and 10% of subclinical patients convert to overt hypothyroidism annually. "It would also be interesting to observe whether these subclinical patients with repeated measurements actually progress to overt disease," he added.

Luca Persani, MD, professor of endocrinology at the Istituto Auxologico Italiano, Milan, Italy, who chaired the session, commented that if the researchers increase the length of follow-up they might find some difference in survival, because there was a trend toward increased mortality in untreated patients.

Ms. Nygaard Andersen agreed the follow-up time was short and said patients would be tracked for longer as soon as the registries were updated.

The researchers also hope to examine cardiovascular events by delving deeper into the database. "We can find out if they are still alive and have had a heart attack from a log of information on treatment for cardiovascular problems," Dr Faber explained.

They also acknowledged that further work, in the form of large randomized studies, is needed to provide more evidence to support or refute treatment in these patients.

Ms. Nygaard Andersen, Dr. Faber, and Dr. Persani have disclosed no relevant financial relationships.

European Congress of Endocrinology 2014; May 4, 2014; Wrocław, Poland. Abstract OC1.2.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.