The Puzzling Thyroid Panel: Answers to Tough Questions

Earlier this fall, Medscape Medical News spoke to endocrinologists Angela M. Leung, MD, and Karen Tsai, MD, on how to interpret abnormal results of thyroid hormone tests. To address some additional questions, we turned to Jenna S. Mammen, MD, PhD, an associate professor in the Division of Endocrinology, Diabetes and Metabolism at Johns Hopkins University in Baltimore. The interview has been edited for length and clarity.

Medscape: What are the most common scenarios that lead to discordant results of a thyroid hormone test?

Mammen: The most common reason is that the thyroid-stimulating hormone (TSH) level is a little bit elevated, and the thyroid hormone levels are normal or elevated. And if you’ve put somebody on thyroid hormone and their TSH level is not coming down, you start wondering: “Why can’t I make this normal?”

And that discordance where hormone levels are high is probably not a tumor in the pituitary gland making too much hormone because those are so rare. That’s probably a lab error in the TSH assay. TSH immunoassays are an indirect measure of how much hormone is in the blood, and some things can interfere with that. The most common is biotin, so patients who are using hair, skin, and nail products can get an abnormal TSH level, even with a completely normal thyroid. Just tell them to stop taking their biotin for 5 days before they get their next TSH level.

Certain antibodies that patients develop can interfere with the assays. Different labs use different assays, so if the patient is consistently going to Labcorp for their assays, you can send them to Quest.

But if you get the same result from different labs, you should measure free thyroxine 4 (T4) levels using direct dialysis. So if we’re using a direct measure of the hormone and we don’t have that interference problem, then I know that my TSH assay is unreliable in this patient and I have made them hyperthyroid by giving them too much thyroid hormone. And I’m going to stop paying attention to the TSH level and follow the free T4 level in this patient.

The other thing that can be going on that’s tricky to diagnosis is central hypothyroidism, in which TSH levels can be normal and the hormone levels — T4 and triiodothyronine (T3) — generally run low.

My husband is a neurologist, and he once called me and said: “Hey, I have a patient who looks profoundly hypothyroid. She has myxedema features and her speech is slow, but her TSH is normal.”

I said, “Well, central hypothyroidism can do that; check a free T4.” In fact, her free T4 level was very low. That can be a difficult diagnosis to make as a primary care doctor. In some situations, TSH level is not enough to make the diagnosis. It looks like a discordance, but just it’s an unusual diagnosis that you’re not used to making.

Medscape: How common is central hypothyroidism?

Mammen: It’s not that common, although we’re seeing it more often in the setting of new oncologic treatments that use autoantibodies against immune checkpoint inhibitors that can cause autoimmunity against everything. One of the organs that they cause autoimmunity against is the pituitary gland. As a result, people are now developing autoimmune disease in the pituitary that leads to central hypothyroidism.

We also see that with steroids and an acute illness in the hospital, people’s thyroid function tests can be wacky. Usually if you have something that you don’t really understand with the thyroid axis, it’s transient and we need to retest.

Medscape: Any other conditions that can cause puzzling results?

Mammen: The other situation is during the transition phase in acute thyroiditis. The first phase is hyperthyroidism: The thyroid is leaking, the TSH level is low, and thyroid hormone levels are high because the patient is hyperthyroid from the inflamed thyroid. But thyroid hormone levels can fall relatively precipitously, and people become hypothyroid. The TSH level takes longer to rise during this transition. And there is a point, if you test too early, where the thyroid hormone level can be low and the TSH level is still low because it hasn’t recovered. That patient should be retested.

The other thing that happens to primary care doctors, especially when they’re trying to adjust a patient’s thyroid hormone dose, is they check the TSH level too soon. If you check the TSH level after 3 or 4 weeks, you’re not even at a steady state yet from taking the thyroid hormone because it takes 6 weeks for TSH levels to drop.

Medscape: Suppose you see a patient who has been diagnosed with hypothyroidism, they’re taking levothyroxine, and their TSH levels have normalized. What’s your approach if they’re still reporting symptoms?

Mammen: This situation presents a challenge. In most of these cases, symptoms are persisting because they probably weren’t due to hypothyroidism in the first place. The Colorado Thyroid Disease Prevalence Study interviewed nearly 25,000 people attending a statewide health fair and looked at thyroid function in all of them. They asked respondents a huge list of symptoms of hypothyroidism, such as constipation, fatigue, brain fog, and feeling cold, and then they determined who had an elevated TSH level or a normal level of the hormone. And the ability of four or more symptoms to predict who had a slightly high TSH level is terrible. The positive predictive value is less than 15%.

Looked at the other way, bringing the TSH level into the reference range is not going to change symptoms because in half of cases, the symptoms and the elevated TSH level are unlinked. This situation requires a lot of education.

I talk about what TSH means and how it reflects how the pituitary and the thyroid talk to each other. You can think about the pituitary as the driver of the car, TSH is the pressure on the gas pedal, and thyroid hormone is the speedometer.

If the pituitary is driving a new car on a flat road, which is analogous to a normal thyroid, it puts a gentle pressure on the gas pedal, and it moves along fine. But for someone with Hashimoto disease, it’s more like driving a 20-year-old Toyota up a little bit of a hill, and the pituitary makes more TSH to push harder on the gas pedal to keep the car moving at the speed that it wants.

If I give a patient a pill with thyroid hormone, I’m leveling the road. The pituitary says, “Oh, great, it’s much easier to drive now.” The TSH level normalizes because it stops having to push so hard on the gas pedal. Less of the gas is coming from the thyroid gland because now it’s coming from the pill, and the car can maintain the same speed.

A TSH level of 7-10 uIU/mL is the threshold for starting thyroid hormone in most guidelines, where the data begin to demonstrate a benefit in terms of health outcomes. But for lower elevations of TSH — so-called subclinical hypothyroidism — the benefits are less clear because the speed of the car is the same throughout, with or without treatment, and most people with subclinical hypothyroidism don’t notice much difference with treatment.

Patients might feel a little bit better initially because when I give them thyroid hormone, I am increasing the speed of the car transiently because their TSH level hasn’t come down yet to adjust the amount coming from the thyroid. Some people come back at follow-up and say, “Oh, I felt great for the first 3 weeks, and now, I don’t feel so great anymore.” What happens is people were a little bit hyperthyroid.

It’s important to recognize that being hyperthyroid might feel good, but levothyroxine is not approved as a stimulant. There are harms associated with excess thyroid hormone, like atrial fibrillation and lower bone density, and we just published a paper in JAMA Internal Medicine last year, showing an increased risk over time for dementia.

Medscape: What other conditions could explain a patient’s persistent hypothyroid symptoms even after their TSH level has normalized?

Mammen: The differential is broad but includes things like menopause, sleep apnea, and insomnia. Sleep issues are a huge. There’s also a growing recognition of the role of post-infectious fatigue syndromes. COVID-19 was so widespread, and lots of people are now really struggling with that.

Medscape: Some clinicians believe a combination of levothyroxine and T3 has greater efficacy than levothyroxine alone. What’s your stance?

Mammen: My personal practice is that physiologic replacements are appropriate, as long as your targets are physiologic. T3 is harder to use because it’s short acting. It can cause TSH suppression, and it wears off in the afternoon. I use it to achieve a normal TSH level, just like I would use levothyroxine, but not with a goal of using it as a pharmacologic stimulant.

Medscape: Which patients are candidates for combination therapy?

Mammen: One category is the rare patients, that only endocrinologists would see, who have malabsorption syndromes, such as short bowel. But some patients just want to try it. They’ve been on T4 and say, “Hey, I’ve still got a lot of symptoms, and I wonder if I would feel better on combination therapy.” Then we have the conversation about how their symptoms probably are not related to their thyroid, which can be helpful.

But for those who still want to try combination therapy, you need to proceed with caution and the understanding that you’re not going to do it to suppress the TSH level; you’re doing it to maintain TSH level within the reference range. If done correctly, most patients find this approach is not much different in the end, just more complicated. I do plenty of consultations going the other way.

Mammen had no financial disclosures.

A former pediatrician and disease detective, Ann Thomas is a freelance science writer living in Portland, Oregon.