I would like to applaud a 2019 article by Mitsuru Ito and colleagues from the Center for Excellence in Thyroid Care at Kuma Hospital, Kobe, Japan.
Dear researchers, your article, “Thyroid function related symptoms during monotherapy in athyreotic patients” showcased research on Free T3 levels and symptoms that is potentially meaningful to the life and health of more thyroid patients than just those who have had a thyroidectomy.
Thank you for caring about examining thyrotoxic and hypothyroid symptoms that may arise while following ATA recommendations regarding TSH during therapy after thyroid cancer removal.
The evidence from your research was lacking when the American Thyroid Association (ATA) wrote their most recent guidelines for the treatment of hypothyroidism (Jonklaas et al, 2014).
Six years ago, this ATA guideline had stated that there was still
- “insufficient evidence of benefit to recommend that treatment with LT4 be targeted to achieve low-normal TSH values or high-normal T3 values in patients with hypothyroidism who are athyreotic.”
The ATA committee drew this conclusion largely based on the earlier research led by its lead author, Jonklaas, in 2008, which had serious flaws they did not admit, such as the failure to assess symptoms and health outcome biomarkers, and the use of Total T3 rather than Free T3.
The field of thyroid endocrinology should hang its head in shame. They chose to idolize TSH monitoring and LT4 therapy in the 1970s and 1980s, but as of 2014 they admitted they had still not yet discovered how to use their favored pharmaceutical to optimize patients’ thyroid hormones FT3 and FT4 to eliminate chronic hypothyroid symptoms while preventing diseases like thyroid cancer.
What has been holding back the field from doing this research?
Are scientists finally beginning to fill the embarrassing knowledge gap?
Thanks to your research, Ito and team, combined with the more in-depth studies by Larisch et al (2018) which you have respectfully cited, and newer research by Hoermann et al (2019), we now have much more evidence of the following:
- Our thyroid symptoms correlate with lower or higher FT3 levels, even within the reference range,
- Lower FT3 levels are often associated with hypothyroid symptoms and a normal TSH,
- Low TSH levels may be euthyroid and do not not necessarily cause thyrotoxic symptoms unless both FT4 and FT3 are higher, and
- Thyroid therapy must be carefully adjusted to the individual based on multiple factors, without making the TSH lower boundary into a prison.
However, even the best research could achieve higher levels of excellence with a patient’s respectful yet challenging input.
After praising your article and describing your methods and results, I’d like to point out some flaws in research methods and conclusions, such as:
1) The TSH-centric cohorts that framed the study and perpetuate the idea that TSH ranges must still be a therapy target.
2) The symptom questionnaire that was too simplistic and incomplete.
3) The reluctance to suggest that findings have implications for patients who lack thyroid function for reasons other than a thyroidectomy.
Thank you for your conclusions
First of all, thanks are in order.
Dear researchers, you are truly on your way to achieving “excellence in thyroid care,” as you have nobly named your hospital’s center. Who else in the world has set up a center for excellence in thyroid care? Bravo!
Thank you, Ito and team, for not presuming that the secretion of one gland, the TSH, can unilaterally define euthyroidism in the context of thyroid therapy in people without healthy thyroids.
You considered the response of the other tissues and organs to FT4 and FT3 levels as manifested in symptoms.
Your article has stated very clearly that a low TSH may achieve euthyroid status:
- “mild TSH suppression [by LT4] can be recommended to achieve preoperative native FT3 levels and euthyroid status.”
To determine euthyroid status, you considered symptoms.
Your courage addresses a common failure in both medical logic and compassion. Too many thyroid scientists have considered all our hypothyroid symptoms “nonspecific” and have given up studying them in the context of TSH-guided thyroid therapy. Our symptoms are discounted not only because they are also symptoms of other disorders but because they often occur when an idolized tissue response biomarker, the TSH, is interpreted as being in disagreement with our symptoms.
You seem to agree with many patients that it is unfair to expect a disease that influences ALL organs and tissues to manifest in ways that always agree with TSH levels co-produced by only TWO organs (hypothalamus & pituitary). Medicine has chosen to treat us with drugs that are often more potent at reducing our TSH than providing our cellular receptors throughout our bodies with enough T3. It is very strange that thyroid therapy guidelines are not focused on adjusting our T3 and T4 concentrations and ratios wherever they need to be to remove as many signs and symptoms of thyroid hormone imbalance as possible.
Thank you also for not considering merely “normal” FT3 levels as if low-normal euthyroid FT3 levels post-thyroidectomy are always physiologically adequate (as Jonklaas et al presumed in 2008).
You have learned that it matters where FT3 falls in relation to pre-thyroidectomy levels.
- If FT3 falls too far below that standard, hypothyroid symptoms are likely to arise, even if FT4 is higher than it was before. The extra FT4 can’t compensate for loss of FT3.
- If FT3 rises too high while FT4 is also very high in LT4 monotherapy, thyrotoxic symptoms are more likely to arise.
Thank you for considering that the achievement of truly euthyroid levels of FT3 may require the lowering of TSH.
You found that a low TSH did not always cause symptoms thyrotoxicosis.
- “A mildly TSH-suppressive treatment (approximately 0.1 mU/L [TSH]) may not necessarily cause thyrotoxicosis. Rather, it may result in euthyroid status in patients who have undergone a total thyroidectomy.”
You wisely considered that the absence of thyrotoxic symptoms alone was not enough to establish euthyroidism.
Therefore, you also pointed to the evidence from your earlier study (Ito et al, 2017) where you discovered that metabolic markers such as cholesterol, bone metabolism markers, and others confirmed that a TSH that was mildly low was generally not thyrotoxic, but euthyroid for tissues beyond the pituitary gland.
Thank you for stressing the need for multiple parameters in clinical decision making.
You asserted that more data than TSH was needed for monitoring and adjustment:
“In each case, it is considered necessary to adjust the LT4 dose comprehensively using
- not only serum TSH levels
- but also serum thyroid hormone levels,
- physical symptoms,
- and metabolic markers.”
You seem to understand that the desire for one thyroid biomarker to rule them all is not only futile but has been damaging to individuals who cannot achieve true euthyroid status within a normal TSH due to the loss of their vital gland.
Thank you for acknowledging our diversity of TSH- thyroid hormone relationships as thyroid patients, even within the narrow category of post-thyroidectomy cancer patients.
You discovered that individual adjustment for each patient was necessary for their health.
Your team are among the few who realize any medical system that permits doctors to leave an individual symptomatically hypothyroid just because their TSH and FT3 are “normal” fails at achieving “excellence” in therapy.
As we now know, even in thyroid health, each individual’s healthy FT3 can fit within 38% of the population’s reference range. Where’s your healthy range within the range? According to this research, you’re not likely to stay in the lower part of the FT3 reference range, but you never know if it’s temporarily where a TSH-driven metabolism would have placed your FT3. Will your doctor rescue you when it’s not enough FT3 and you become symptomatic? Not if they worship the normalized TSH.
Leave no thyroid patient behind.
Background and methods
Here’s how the scientists came to these conclusions.
The article set out to investigate the wide variability among patients post-thyroidectomy on Levothyroxine (LT4) with regard to their Free T3 levels and their symptoms.
It had been noticed by other scientists that in the thyroidless population on LT4 monotherapy,
- some people have “mildly low” Free T3 and “normal” TSH
- some people have “normal” Free T3 and “mildly suppressed” TSH
- some people have “elevated” Free T3 and “strongly suppressed” TSH.
The researchers asked the important question:
“Which of these three patient groups are closer to their preoperative euthyroid condition based on reported subjective symptoms”?
They studied 148 patients aged 18-78 years old before and 12 months after their thyroidectomy.
They excluded anyone who had health conditions that could cause symptoms that mimic hypo- and hyperthyroid symptoms and anyone with health conditions or medications that could cause or unusual thyroid hormone levels.
All the patients had had only papilliary thyroid carcinoma and were neither hypothyroid nor hyperthyroid prior to thyroidectomy.
In the first year post-thyroidectomy, they adjusted the patients’ LT4 therapy to TSH levels according to ATA guidelines based on cancer risk stratification (Cooper et al, 2009).
They asked patients’ to complete symptom questionnaires 2 days before surgery and 12 months after thyroidectomy, after their dose had been adjusted according to guidelines.
They found that not only did all patients have significantly more Free T4 than preoperative levels, but that
- 33 patients with “normal” TSH had Free T3 levels “significantly lower” than preoperative levels, but within reference range. They had significantly more hypothyroid symptoms.
- 50 patients with “mildly suppressed” TSH had Free T3 levels that averaged equal to their preoperative levels, and the average symptom scores were unchanged from preoperative levels.
- 65 patients with “strongly suppressed” TSH had Free T3 levels that were higher than preoperative values, but only 3 patients had levels above the reference range. They had more thyrotoxic symptoms than the other two groups, but individuals varied.
In addition, they found that certain symptoms were associated with higher and lower Free T3 levels and their inverse TSH levels.
The most significant thyrotoxic symptoms were
- mild heat intolerance (+1), or feeling hot in a temperate room (+2)
- softer bowel movements (+1), or movements several times a day (+2)
- tremor of hands on examination (+1), or tremor at rest (+2)
The most significant hypothyroid symptoms were
- mild sensitivity to cold (-1), or feeling cold in a warm room (-2)
- decreased productivity (-1), or slow movements (-2)
However, symptoms were individualized. In each patient group, there were some patients who experienced a significant increase in thyrotoxic symptoms and some who experienced an increase in hypothyroid symptoms (+2 or -2 change).
The other symptoms on the questionnaire were not significantly correlated with Free T3 or TSH levels (sweating, appetite, wet/dry skin, temperature of hands and feet).
Room for improvement
Despite the strengths of the study, it had weaknesses that can limit the value of the study for thyroid patients’ health.
Biased division of cohorts
Ito and colleagues, unfortunately, the way you set up the three cohorts of patients in your study was biased and limiting. Cohorts were pre-defined by TSH-centric guidelines and only considered two variables, TSH and FT3.
The study did not attempt to discover the cohorts of highly symptomatic vs. relatively symptom-free patients that formed organically from the population’s data set.
With a design that permitted patterns and cohorts to arise from data, you might have gained further insight into both the symptoms and the thyroid hormone levels.
If you wanted to analyze the data also by the TSH levels, you could have done that as well, but this method only offered one guideline-driven grid through which one could analyze the data.
As a result, your research design unfortunately led to conclusions that tended to reinforce or shift TSH-centric targets. The final paragraph beginning “In conclusion” stated
“patients with mildly suppressed TSH levels were closest to euthyroid status, whereas those with normal TSH levels had mild hypothyroidism and those with strongly suppressed TSH levels had mild hyperthyroidism.”
Only one sentence recommended using “serum thyroid hormone levels, physical symptoms, and metabolic markers” to adjust the LT4 dose.
Another sentence suggested “that mild TSH suppression [by] LT4 can be recommended to achieve preoperative native FT3 levels and
euthyroid status in athyreotic patients.”
But let’s re-read that quotation just above. Suppressing the TSH is not the biological method “to achieve” a native FT3 and euthyroid status.
Instead, you know well as thyroid scientists that suppressing the TSH is a side effect of raising FT4, and FT3 rises only insofar as the individual patient is capable of converting T4 to T3 in cells at a healthy rate.
When people lack a thyroid, they lack a gland that not only secretes, but also converts T4 to T3.
No matter how many studies show that TSH enhances T4-T3 conversion in peripheral tissues, raising the TSH by reducing LT4 dose has not resulted in the appearance of more FT3 in the blood of an athyreotic person.
In addition, lowering the TSH by raising LT4 cannot always achieve preoperative FT3 levels or remove symptoms enough within an individual, even if it achieves it in your cohort on average. Remember that your cohort was not complicated by any other disease factors. Real thyroid therapy is messy.
Conclusions may continue to reinforce to some people the backwards idea that a TSH level alone, one that is approximately 0.3 or even 0.1 mU/L, can be a safe or effective “target” of thyroid therapy for a diverse cohort of athyreotic patients.
I hope you would agree that making the TSH a sole target of thyroid therapy is as illogical as trying to make a dog happy by forcefully wagging his broken tail. The wagging tail is the effect, not the cause, of his happiness. It could also be the effect of apprehensiveness.
I dare anyone to disprove this claim: No evidence exists to support the idea that eliciting a TSH within a certain target range will cause or induce euthyroid Free T3 and Free T4 levels and ratios in all thyroidless people.
Without a thyroid to mediate its influence, TSH is impotent in inducing health or disease by itself.
If the TSH concentration is not capable of normalizing tissue thyroid hormone levels throughout the body in a person without a thyroid gland, its use as a sole surrogate endpoint of therapy, and as a target of therapy, is flawed.
No target TSH can predict euthyroidism for a large population of patients. The diverse population will inevitably contain an unknown amount of people with mild undiagnosed central hypothyroidism (TSH hyposecretion). Some people use medications or have TSH-Receptor stimulating antibodies that depress TSH. There will be people with genetic flaws in thryoid hormone transport, metabolism and receptor sensitivity. Dismissing these people as “rare” is unfair when the medical system is set up to conceal them by monitoring by TSH alone.
Instead of new TSH targets, what do thyroid patients need from excellent thyroid researchers like you?
We need advice on how to compassionately remove symptoms and achieve true euthyroidism at the same time as reducing the health impact and medical cost of concurrent health conditions. Thyroid patients do not just have thyroid cancers, but conditions well beyond cancer and heart disease and osteoporosis. We need help interpreting relationships between thyroid hormone levels and biomarkers of tissue health like brain, kidney and liver function. And what is wrong with measuring ankle reflex if we truly care about tendon health?
Please continue to tell the world that once the more important biomarkers and health outcomes are in place on a particular therapy such as LT4, here or there is where the TSH usually falls in your purified cohort of patients. But why imply it can be another TSH target for all?
In your future work, please help patients by bringing FT3 adjustment to symptoms into the spotlight as a more rational target for therapy in relationship to FT4 and many tissue biomarkers.
Let TSH fall where it may, for the sake of health.
The symptom questionnaire
Unfortunately, the symptom questionnaire was deeply flawed.
Fortunately, it still yielded some results of significance for a few items.
And fortunately, the symptom questionnaire was only one of two major variables (preoperative FT3 levels was the other).
The researchers’ symptom questionnaire had only 8 items drawn from three well-known symptom scores:
- Billewicz et al. (1969) and Zulewski et al (1997) for hypothyroid symptoms
- Klein et al (1988) for hyperthyroid symptoms.
Ito et al’s team appeared to cherry-pick items rather than explain any systematic reasons that guided their choice of symptoms to include or exclude. It is unclear why Wayne’s index was not considered for hyperthyroid symptoms (see Kalra et al, 2011).
The questionnaire could have relied more on the originals which had been clinically tested. These systems, such as Billewicz’s system, are more highly refined. They are enriched by the experience of clinicians who had evaluated hypothyroid and thyrotoxic symptoms over a long career without being biased by looking at one biomarker above all.
Ito and team’s questionnaire also clearly suffers from the common mistaken presumption that hypothyroid symptoms are always the direct opposite of hyperthyroid symptoms, and the concept that there is a neutral midpoint of symptom-free status that is equidistant from both extremes. This is false. The human body can overcompensate for hypothyroidism through various mechanisms. Often the resulting symptoms in hypothyroid patients can mimic transient or mild signs of hyperthyroidism. The dualistic belief is a common source of misdiagnosis and delayed diagnosis.
Ito’s scoring system was oversimplified to scales from -2 to +2 for most items and only hand tremor was from 0 to +2. In contrast, Billewicz’s research team weighted symptoms. They gave a symptom’s presence a +6 but its absence a -2, as in the case of “diminished sweating.”
In addition, visual and verbal survey design can skew response, and it is not clear what design was presented to patients. A patient who believes they are hypothyroid when they do not have enough experience with thyrotoxic symptoms can be tricked by a visual scale to choose the items on the hypothyroid side of the page.
Ito’s instrument must have been delivered in Japanese, but word choice like “decreased productivity” as a symptom implies a poor work ethic, which carries moral or cultural stigma. Choosing such a symptom in a list feels like an act of blaming and shaming oneself rather than objectively reporting symptoms. Items must be randomized and synonyms should be offered.
Ultimately, the scoring system could have (and should have) been co-designed by thyroid patient communities in their own nation and language.
If the study had been done in Canada, Canadian thyroid patients might well have said that their scoring system lacked some of the most prominent thyroid symptoms that our patients self-report on social media support forums. It was missing key symptoms such as depression, anxiety, weight gain or loss, slow or fast heart rate, cognitive difficulty, and hair loss.
Much work remains to be done on thyroid symptoms, since we have not advanced beyond Zulewski in 1997 and ignorance about thyroid symptoms abounds. It is shocking and dismaying that no large study has yet been done to create a better scoring system by crowdsourcing patient-reported thyroid symptoms and then clinically testing them against multiple biomarkers in addition to TSH and thyroid hormones.
Until more such studies are done, thyroid diagnosis and therapy adjustment will remain a highly contentious and imprecise field overruled by medical prejudices and biases, such as the belief that laboratory tests tell the only truth worthy of the name of “evidence.”
Conclusions may apply to others who lack living thyroid tissue
Dear researchers, your research on “athyreotic” patients ought to be compared with and potentially applied to others who lack functional thyroid tissue, not just patients whose thyroid glands have been surgically removed after thyroid cancer.
The continual focus on thyroidectomized patients sets up an unfair hierarchy within thyroid therapy in which only thyroid cancer patients get red carpet treatment.
If it is true that lower TSH does not always result in thyrotoxicosis, then we cannot restrict to a small subclass of patients an “elite” form of customized therapy that removes symptoms by means of elevating FT3 and permission to reduce TSH.
Why would your findings not apply equally to:
- humans with severe autoimmune Atrophic Thyroiditis whose formerly healthy thyroids can be shrivelled at any stage of life to a volume as small as 0.5 mL (one tenth of a teaspoon) consisting of mostly fibrosed tissue. (This is the volume of my thyroid.)
- people who were born without functional thyroids or any thyroid tissue.
- individuals whose thyroid glands have been thoroughly neutralized by radioiodine therapy after Graves’ disease. Indeed, another study by your own team has examined this patient population.
- citizens whose glands have been fully fibrosed by the lymphocytic infiltration that kills thyroid cells in Hashimoto’s thyroiditis. Why do so many people seem to believe that all patients with Hashimoto’s have residual thyroid function?
Thyroid patients look forward to more research that permits anyone who needs more FT3 to attain true euthyroidism at a lower TSH, if that’s where the TSH happens to fall.
Tania S. Smith
References – See Page 2
Pages: 1 2