icon caret-left icon caret-right instagram pinterest linkedin facebook twitter goodreads question-circle facebook circle twitter circle linkedin circle instagram circle goodreads circle pinterest circle


Strengths and Weakness of TSH and Other Testing in Athyrotic Patients


Pardon me for starting with shameless self-promotion, but I assume since you’re contacting me you’re familiar with my book The Thyroid Paradox (2007, Basic Health Publications); if not, it addresses a lot of the issues you raise and is available from Amazon.com as a trade paperback or Kindle download; also any bookstore should be able to special order it.

I consider this book to be a good middle-of-the-road analysis of the state of modern thyroidology--criticizing what most physicians do wrong, without throwing away what they do right, which much of the hyped-up rhetoric about thyroid disease in popular books and on the internet does. In other words, I try not to throw out the baby with the bath water,

That said, I’ll address your most specific questions individually:

(1) It seems such a paradox to me to be testing the pituitary thyroid stimulating hormone in a thyroidless person

You are of course correct that TSH is a pituitary hormone—however, it is produced in response to any perceived lack of thyroid gland function by the brain, and it stimulates the thyroid gland to make thyroid hormone. Thus, TSH levels are integrally related to the proper functioning of the whole thyroid system which include the brain, pituitary and thyroid.

My disagreement with most physicians is not that they test TSH levels, but that they often only test TSH levels and ignore everything else, including FT4, FT3, and symptoms.

In other words, my position is that TSH is very useful—I just don’t assume that it tells me everything.

I know your questions relates to the athyrotic (“thyroid-less”) patient, following total thyroidectomy. Obviously in that situation the role of TSH stimulating the thyroid gland is moot--but, the pituitary still releases TSH when there is a perceived lack of thyroid hormone (in the case of the athyrotic patient, a perceived lack of Synthroid or whatever other thyroid replacement drug is being used).

So, put very simply—a patient with a high TSH is almost certainly not getting enough Synthroid—which is a very useful thing to know. Also the person not getting enough Synthroid will more often have a high TSH, and have it sooner, than an obviously low FT4 or FT3. TSH is a more sensitive test for hypothyroidism than FT4 and FT3 are. To not use it in these people would be crazy.

Now, what about monitoring for over-replacement--too much Synthroid. Almost everybody on too much thyroid medicine will have a low TSH, and they will have it before the high FT4 and FT3. In other words, TSH is also a very sensitive test for hyperthyroidism.

The problem is: TSH is NOT a very SPECIFIC test for hyperthyroidism. Other things can cause a low TSH besides too much Synthroid. That’s the mistake that gets made--almost every doctor out there--including endocrinologists--automatically assume or at least respond as if a low TSH means the person is on too much thyroid medicine and they lower the dose. That might be the right interpretation, but it might not, and the only way to know is to look at FT4, FT3, and consider the patient’s signs and symptoms.

Bottom line--TSH is a very useful test, but there are weakness which are often ignored. That is a reason to change how we think about TSH, not a reason not to use it.

(2) Shouldn't the emphasis be on the actual thyroid metabolic state reflected by serum FT4 and FT3 concentration?

It would be incorrect to assume that any given FT4 or FT3 level accurately reflects the “actual thyroid metabolic state.” And since we are talking about athyrotic patients, these levels don’t even reflect the metabolic activity of the thyroid gland to produce thyroid hormone--these levels are instead totally under the control of the Synthroid et al. prescriber’s pen or keystroke. To be clear, by “actual thyroid metabolic state” you mean the amount of thyroid hormone that is getting to the thyroid hormone receptors deep inside the nucleus of each of our cells, and the resulting intracellular actions that are triggered by those activated receptors.

Looked at that way, I think you can see that a blood level of FT4 or FT3 (which are the only measurements technically practical) does not necessarily reflect what is going on OUTSIDE the blood, deep inside cells. It’s helpful; it’s an estimate of what’s going on, but that’s all it is. That’s especially true for FT3, since much T3 is produced inside these cells and never makes it to the blood.

The fact is, just because a FT4 or FT3 blood level is “normal” doesn’t mean that cellular thyroid hormone action is normal, because there are several physical barriers and physiologic processes separating those points. Who’s to say that a low-normal FT4 of .89 is enough--maybe a high-normal level of 1.7 would be better. For that matter, who’s to say a low blood level isn’t getting the job done, or that a high level is.

“Statistically normal” which is that range printed on a lab report is not the same thing as “physiologically normal” Something else very few physicians, sadly, ever think about in my experience.

Getting back to TSH--if the pituitary is working properly (I don’t think it always is in athyrotic patients) but IF IT IS then TSH gets produced by the pituitary as a result of the brain’s perception of the adequacy of intracellular thyroid-hormone action. So really, in fact, it could be argued that the blood TSH level does, in many cases, more accurately reflect “actual thyroid metabolic state” than blood FT4 and FT3 levels. You might say that TSH reflects the “ACTUAL thyroid metabolic state,” while FT4 and FT3 reflect “POTENTIAL thyroid metabolic state.”

Bottom line--none of these tests are perfect--all have strengths and weaknesses which need to be better understood by all of us. The right answer is to evaluate them all to come up with what is never going to be any better than an estimate of what is really going on.

(3) Could you please cite any studies/information showing that a low but not suppressed TSH is acceptable as long as the free T4 and free T3 are within range and there are no hyperthyroid symptoms

As you probably know, there is a virtual obsession amongst most physicians about avoiding low, and especially undetectable TSHs. The concern is that the low TSH equals hyperthyroidism, which is of course wrong—it is a marker for it, but the problem is hyperthyroidism (too much T4 and/or T3 for the person), not low TSH. Hyperthyroidism does need to be avoided; it does do harm. There are other reasons, though, why the TSH might be low, and those other reasons are almost never considered. This obsession with avoiding low TSHs often in my experience leads to an underdosing of patients, and a lot of frustration and poor quality of life for patients—especially athyrotic ones.

I’m frankly dumbfounded why endocrinologists smarter than I am don’t see and teach the “disconnect” between “low TSH” and actual hyperthyroidism. It has a lot to do, though, with the Hippocratic principle of “first do no harm.” If we never get the TSH low then we will never do harm with hyperthyroidism. If we let TSHs get low then some of the time we might err and do harm with hyperthyroidism, and that is a possibility that mainstream medicine isn’t comfortable with.

Hopefully, most physicians out there at least see that a mildly low TSH of say 0.31 is probably okay relative to one below, say, 0.1--but alas, many do not split these hairs in actual practice.

Your specific question was about “low but not suppressed TSH” being “acceptable as long as…”

I would modify that by saying that in some but not all cases even a suppressed (undetectable, that is) TSH might be acceptable.

Unfortunately there really isn’t much literature to support this position. Common sense supports it, but the modern obsession with evidence-based medicine doesn’t give as much credence to common sense as one would like--hopefully we are seeing a slow backswing.

I will list and comment upon a few articles I referenced in my book that might be of help:

--Shimon, I., et al. “Thyrotropin Suppression by Thyroid Hormone Replacement is Correlated with Thyroxine Level Normalization in Central Hypothyroidism.” Thyroid 12 (2002): 823–827.

(This is the strongest argument for our position. Central hypothyroidism is hypothyroidism caused by a defect in TSH secretion by the pituitary--it is thought to be rare--I think it’s common but under-recognized. When I say that it’s okay for a patient to have a low or undetectable TSH, what I’m really saying is that the patient is either all or partly centrally hypothyroid, rather than pure primary hypothyroidism.)

--Zulewski, et al. “Estimation of Tissue Hypothyroidism by a New Clinical Score: Evaluation of Patients with Various Grades of Hypothyroidism and Controls.” Journal of Clinical Endocrinology and Metabolism 82 (1997): 771–777.

(Good analysis of symptoms vs. labs in the diagnosis of hypothyroidism.)

--Andersen, S., et al. “Narrow Individual Variations in Serum T4 and T3 in Normal Subjects: A Clue to the Understanding of Subclinical Thyroid Disease.” Journal of Clinical Endocrinology and Metabolism 87 (2002): 1068–1072.

--Dickey, R., L. Wartofsky, and S. Feld. “Optimal Thyrotropin Level: Normal Ranges and Reference Intervals are Not Equivalent.” Thyroid 15 (2005): 1035–1039.

(Two papers that telling us that what is “statistically” normal for the population isn’t the same as “physiologically” normal for the individual.)

--Demers, L., and C. Spencer. “Laboratory Medicine Practice Guidelines: Laboratory Support for the Diagnosis and Monitoring of Thyroid Disease.” Thyroid 13 (2003): 3–126.

(Any paper by Carol Spencer from UCLA, I believe, will be very open-minded about the weaknesses of and precautions to take when using modern thyroid function tests.)

(4) Shouldn't dosage be based more on free T4 and free T3 levels and clinical symptoms rather than so much emphasis on TSH?

Yes, it should be based on all 4 parameters: FT4, FT3, TSH, and clinical evaluation. Yes, there is too much emphasis on TSH—just don’t make the mistake of thinking that TSH is worthless.

Good questions—I hope this helps.