Subclinical Hypothyroidism in Elderly: OK to Watch and Wait?

Anne R. Cappola, MD, ScM


May 01, 2017

Editorial Collaboration

Medscape &

Hello. I'm Dr Anne Cappola from the University of Pennsylvania, where I am a professor in the Division of Endocrinology, Diabetes, and Metabolism. I am a clinical investigator, and my focus is on thyroid disease in the elderly. Today I will be talking about the management of hypothyroidism in older people.[1]

Overt vs Subclinical Hypothyroidism

Hypothyroidism is more common in older people than in younger people, particularly the milder form that we call "subclinical hypothyroidism." There seems to be an age-adjusted shift to higher thyroid-stimulating hormone (TSH) levels as people get older. Sometimes, it is difficult to recognize hypothyroidism because the symptoms are so nonspecific. Other times, people present with a variety of different complaints. It is critical that we obtain a thyroid test to distinguish which things are thyroid-related and which things are not.

Anybody you find with overt hypothyroidism should be treated. Right now, we treat anybody who has a TSH level ≥ 10 mIU/L, although we usually like to "start low and go slow," because we do not want to overdo it.

The controversy is for those whose TSH level is a little bit lower than 10 mIU/L, and even more so for patients whose TSH levels are in the 4-7 mIU/L range—just outside the reference range. With these patients, it is a dilemma whether or not to treat. Already, these are people taking many medications. It is very tempting to treat when you see a result outside the reference range, particularly because there are so many different symptoms.

You think you are going to help your patient by treating them, but it is not clear whether that is the case or not.

Does Treating Subclinical Hypothyroidism Help or Hurt?

We have done many studies[2,3,4] looking at people with mildly elevated TSH levels or subclinical hypothyroidism. For those who have the TSHs in the 4-7 mIU/L range, we have been unable to find any associations with cardiovascular disease, fractures, strokes, dementia, or other outcomes.

Do not go ahead, pull the trigger, and treat them.

Recently, the New England Journal of Medicine[5] published a placebo-controlled study that followed several hundred older people (mean age, 74 years) treated with levothyroxine (average dose, 25 µg). Primary outcomes were two symptom scores on a validated thyroid questionnaire—one regarding hypothyroidism, the other regarding fatigue.

The mean level of TSH for all patients at time of enrollment was 6.4 mIU/L. At 1 year, mean TSH drifted down to 5.48 mIU/L in the placebo group and to 3.63 mIU/L in the treated group—within the normal range, but not the lower part of normal. There was no difference in the treated group or the untreated group in terms of symptoms.

These are really important data. This tells us that it is okay to just watch older people who have TSH levels in that 4-7 mIU/L range. Do not go ahead, pull the trigger, and treat them. You can follow them with annual thyroid tests and see if they are progressing. In many cases, their TSH levels will get better.

We are still in a little bit of a gray zone for TSH levels of 7-10 mIU/L. There are some data[6] to suggest from observational studies that people in this range may have slightly higher risk for cardiovascular mortality. Certainly, we treat those with levels 10 mIU/L or above.


Definitely treat overt hypothyroidism, but be careful and ginger with treatment in terms of dosing, pace of dose escalation, and monitoring. For those with subclinical hypothyroidism, it appears from the data that it is okay to watch them when the TSH is in the 4-7 mIU/L range. There are some individual opinions on the 7-10 mIU/L range, but some data suggest that those patients should be treated as well. Target levels are still an area under consideration, but it certainly seems reasonable to let them drift up within the normal range.

Thank you very much. It has been a pleasure talking to you about hypothyroidism in the elderly.


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.