This website is dedicated to the millions of thyroid patients who are being ignored and left to suffer unnecessarily, and to healthcare practitioners, who want to better serve those patients.

TPA Letter to the British Thyroid Association – TSH Controversy

[SIZE=5]The TSH Controversy[/SIZE]
We at Thyroid Patient Advocacy (TPA-UK) are frankly appalled that the BTA are advocating the broadening of the TSH range for use in the initial diagnosis and subsequent treatment of sub clinical and overt hypothyroidism when recent research is suggesting quite the opposite 1. This proposal is in direct contrast to the decision made by the American Academy of Clinical Endocrinologists who narrowed their reference range in 2003 2 . Their original range was, at that time, broadly similar to that currently used in the UK. Their decision has enabled 13 million more Americans to be both diagnosed with and receive treatment for thyroid dysfunction with the subsequent restoration of their health. A graph, reprinted below, taken from the results of a study of over 65000 people in 2000 3 clearly illustrates just how skewed the current reference range is, let alone the proposed broader range. This study, which backs up the case for a narrowing of the reference range, forms the basis of a very illuminating article on ‘Normal TSH’ from Thyroid Australia. The TSH graph within this article has encouraged many medical practitioners to be more flexible in their interpretation of TSH results. The full article can be found here[/LEFT]

[LEFT]If the TSH reference ranges used were truly reflective of ‘normal’ TSH then it would be expected that the bell curve would peak in the centre rather than to one side as shown. TSH is an indirect method of assessing thyroid function and should not be used as a sole diagnostic tool. This message should be promulgated throughout the medical profession without delay and given strong emphasis everywhere, where appropriate within the guidelines. We are constantly learning, and past experience or precedent does not always provide the right solution today. A full clinical picture, including signs, symptoms, family history, etc., should always be taken into account as well. If a patient appears to be clinically hypo/hyperthyroid then other tests including antibodies, T4 and T3 should be done even if TSH is deemed apparently ‘normal’. A well respected consultant endocrinologist, Dr K. Rizvi, is in complete agreement with this point of view and has given us permission to quote him and forward his e-mail of 21 st December 2005. This is reprinted at the end of this letter with other responses.

2.2.6: The Normal Healthy Adult Population

3.3.1: Diagnosis

Nature of Comment: Refutation of Recommendations regarding reference ranges

In his book “Hypothyroidism: The Unsuspected Illness”, Dr.Broda Barnes described over 47 symptoms that may be related to poor thyroid function. During his years of research and practice, Dr. Barnes condemned conventional doctors who ignored obvious clinical manifestations of thyroid deficiency. According to Dr. Barnes “The development and use of thyroid function blood tests left many patients with clinical symptoms of hypothyroidism undiagnosed and untreated.

Dr.Barnes believed that 40% of the adult population suffers from thyroid deficiency. Based on the percentage of adults now taking prescription drugs to treat depression, elevated cholesterol, high blood pressure and other conditions, Dr. Barnes’ observations about the epidemic of thyroid deficiency may now be validated.
The draft guidelines make scant reference to the clinical manifestations or clinical diagnosis of hypothyroidism. Therefore, the clinical features of hypothyroidism appear to have been relegated to the status of historical curiosities

According to an article published in the August 3, 2002 issue of The Lancet 4 , the problem with thyroid blood tests may be faulty “reference ranges” that fail to reflect what the optimal level of thyroid hormone should be in a particular individual

Standard laboratory reference ranges represent “average” populations, rather than what the optimal range should be. Back in the 1960s, for instance, the upper reference range for cholesterol extended to 300 (mg/dL). This number was based on a statistical calculation indicating that it was “normal” to have total cholesterol levels as high as 300. At that time, it was also “normal” for men to suffer fatal heart attacks at relatively young ages. As greater knowledge accumulated about the risk of heart attack and high cholesterol, the upper limit reference range gradually dropped to the point where it is now 200 (mg/dL) 5. The same situation occurred with homocysteine reference ranges. Up until recently, it was considered normal to have a homocysteine blood reading as high as 15 (mm/L) 6. Most reference ranges now provide a chart showing that homocysteine levels above 7 increase risk of heart attack and stroke 7 .

In March 2002 Danish researchers conducted an interesting study 8 , looking at the monthly thyroid levels -T4, T3 free T4 index and TSH of 16 healthy men over a period of 12 months. They found that each of the individuals had different variations of their thyroid function, around unique levels or ‘set points’. Each person had his own individual thyroid function and normal level, and people tended to fluctuate slightly within their own range.

These findings led the researchers to conclude that a thyroid test result within a laboratory’s reference limits or “normal range” is not necessarily normal for a particular individual. In fact, the researchers also concluded that the distinction between subclinical and overt thyroid disease (abnormal serum TSH and abnormal T4 and/or T3) is somewhat arbitrary, because the patient’s normal set point for T4 and T3 within the laboratory reference range is actually illustrative and needs to be taken into account.

Scientists are now examining thyroid hormone reference ranges and their findings indicate that it may be time to change the way laboratories report TSH results. The question raised by The Lancet authors, however, is whether today’s reference range for TSH reflects optimal thyroid hormone status.

The TSH reference range is so broad that most doctors will interpret a TSH reading as low as 0.2 to be as normal as a 5.5 reading. The difference between 0.2 and 5.5, however, is an astounding 27-fold . It would seem almost absurd to think that a person could be in an optimal state of thyroid health anywhere along this 27-fold parameter.

The American Academy of Clinical Endocrinologists (AACE), has confirmed that hypothyroidism, has been vastly under diagnosed. After a meeting by the AACE, the normal range for TSH was dramatically narrowed; encouraging doctors to consider treatment based on a target TSH level of 0.3 to 3.0mU/l, as noted in the AACE press release of January 2003 2 . AACE cited as evidence the guidelines issued by the National Academy of Clinical Biochemistry, part of the Academy of the American Association for Clinical Chemistry (AACC), and presented in their ‘Laboratory Medicine Practice Guidelines for the Diagnosis and Monitoring of Thyroid Disease’. Late in 2002, the group concluded that “it is likely that the current upper limit of the population reference range is skewed by the inclusion of persons with occult thyroid dysfunction.” In their guidelines, the National Academy of Clinical Biochemistry reported that: “In the future, it is likely that the upper limit of the serum TSH euthyroid reference range will be reduced to 2.5 mIU/L because 95% of rigorously screened normal euthyroid volunteers have serum TSH values between 0.4 and 2.5 mIU/L.” They also stated that “a serum TSH result between 0.5 and 2.0 mIU/L is generally considered the therapeutic target for a standard L-T4 replacement dose for primary hypothyroidism” 9 .

3.1.2 Guiding treatment with thyroxine

Nature of Comment : A Refutation of the Recommendation to rely on TSH levels as a guide to thyroxine titration :

An article in the BMJ in August 1996 10 on the management of hypothyroidism and hyperthyroidism states: ” The correct dose is that which restores the euthyroid state and relieves symptoms. In most patients these will be achieved by a dose of thyroxine resulting in a normal or slightly raised serum thyroxine concentration, a normal serum triiodothyronine concentration and a normal or below normal serum thyroid stimulating hormone concentration “. Another article in the BMJ in February 2003 -Serum thyroid stimulating hormone in assessment of severity of tissue hypothyroidism in patients with overt primary thyroid failure states: “the biological effects of thyroid hormones at the peripheral tissues and not TSH concentrations reflect the clinical severity of hypothyroidism. A judicious initiation of thyroxine treatment should be guided by clinical and metabolic presentation and thyroid hormone concentrations (free thyroxine) and not by serum TSH concentrations 11

This is backed up by a representative of the Department of Health 12 who, as recently as 23 rd November 2005, stated that: ” It is currently considered good medical practice to rely upon clinical history and examination, in addition to blood tests in the diagnosis of hypothyroidism”.

A review of published findings about TSH levels reveals that readings over 2.0 may be indicative of adverse health problems relating to insufficient thyroid hormone output. The Whickham Survey showed that individuals with TSH values over 2.0 have an increased risk of developing overt hypothyroid disease over the next 20 years . Fifteen other studies show that TSH values over 1.9 indicate abnormal pathologies of the thyroid, specifically autoimmune attacks on the thyroid gland itself that can result in significant impairment. 1

More ominous is a study showing that TSH values over 4.0 increase the prevalence of heart disease after correction of other known risk factors. 13 Another study showed that administration of thyroid hormone lowered cholesterol in patients with TSH ranges of 2.0 to 4.0, but had no effect in lowering cholesterol in patients whose TSH range was between 0.2 and 1.9. 14 This study shows that in people with elevated cholesterol, TSH values over 1.9 could indicate that a thyroid deficiency is the culprit causing excess production of cholesterol, whereas TSH levels below 2.0 would indicate no deficiency in thyroid hormone status.

In a study to evaluate psychological well-being, impairment was found in patients with thyroid abnormalities who were nonetheless within “normal” TSH reference ranges. 15

The authors of The Lancet study 4 stated that, “the emerging epidemiological data begin to suggest that TSH concentrations above 2.0 (mU/L) may be associated with adverse effects.”

The authors prepared a chart based on previously published studies that provide guidance when interpreting the results from TSH blood tests. Here are three highlights from their chart that may be useful in determining what an individual’s TSH values really mean.
[TD] TSH greater than 2.0
[TD] Increased 20-year risk of hypothyroidism and increased risk of thyroid autoimmunity [15]
[TD] TSH greater than 4.0
[TD] Greater risk of heart disease [16]
[TD] TSH between 2.0 and 4.0
[TD] Cholesterol levels decline in response to thyroxine (T4) therapy [17]
Despite presenting these interesting findings, The Lancet authors stated that more studies are needed to define optimal TSH level as between 0.2 and 2.0 instead of between 0.2 and 5.5. TPA-UK feels that this type of precise information provides an opportunity to correct a medical condition that has often been unresponsive to mainstream therapies, or possibly prevent such disorders from developing in the first place.

8. Areas for further studies

Nature of Comment: Failure to delay the draft guidelines until studies have been undertaken as recommended by yourselves and Dayan C M et al. 1

It is interesting to note that since the Lancet study in 2002, these recommended further studies have not been undertaken. These should have been completed prior to drawing up these draft guidelines. Your team has highlighted no less than 6 areas under item 8 suggesting further studies be undertaken that conform to the rules of evidence based medicine in order to provide answers to some common but contentious issues in the use of thyroid function testing. Your team admits that the quality of evidence to support the recommendations in these guidelines is generally poor. T here is an urgent need for these studies, as ‘anecdotal evidence’ would suggest that the rest of the recommendations in these draft guidelines are incorrect.

A new research study just released on 15 th November 2005 from the Cardiovascular Research Institute-South Dakota Health Research Foundation and the University of South Dakota School of Medicine 15 shows a link between low thyroid function and heart problems. ” We provided strong evidence that low thyroid function alone induced in rats eventually can cause heart failure, ” said Dr. A Martin Gerdes, Director of the Cardiovascular Research Institute and co-author of the study. “We also discovered that low thyroid function severely impaired cardiac blood flow due to a dramatic loss of the heart’s small blood vessels (arterioles). Within six weeks after inducing low thyroid function in rats, half of the hearts’ small arterioles were gone”. This study in rats is giving researchers hope that more aggressive treatment of hypothyroidism and borderline hypothyroidism will result in a reduction of heart disease in human beings.While further research is needed, these results suggest that low thyroid function has the potential to cause heart failure.

Yet another study in 2005 entitled “The Evidence for a Narrower Thyrotropin Reference Range Is Compelling” 16 came to the conclusion “It has become clear that previously accepted reference ranges are no longer valid as a result of both the development of more highly sensitive TSH assays and the appreciation that reference populations previously considered normal were contaminated with individuals with various degrees of thyroid dysfunction that served to increase mean TSH levels for the group.” They further stated “Importantly, data indicating that African-Americans with very low incidence of Hashimoto thyroiditis have a mean TSH level of 1.18 mU/L strongly suggest that this value is the true normal mean for a normal population. Recognition and establishment of a more precise and true normal range for TSH have important implications for both screening and treatment of thyroid disease in general and sub clinical thyroid disease in particular.”

The recent recommendation by the AACE, that the upper limit of normal for TSH be changed from 5.0 to 3.0 mU/L should, if nothing else, raise further questions about the validity of the TSH test. If this proposed change is accepted into the mainstream, then more than four times as many people as before will be classified as hypothyroid (20.0% versus 4.64% of the population 17 ) indicating the current TSH test may be missing more than 75 percent of cases of hypothyroidism. After having been promoted for years as a nearly perfect test for thyroid function, the acknowledgment that the TSH test may be misdiagnosing so many patients is not encouraging. Certainly, lowering the upper limit of normal would increase the sensitivity of the test. On the other hand, it would likely decrease its specificity, too, resulting in some euthyroid patients being incorrectly classified as hypothyroid. Furthermore, medical practitioners who relied solely on the new TSH reference range would continue to overlook many cases of hypothyroidism. It appears that as many as half of clinically hypothyroid patients have a TSH level below 3.04 mU/L.

TPA-UK questions the reliance on diagnosing by TFT’s alone, particularly blood tests. There have been many changes in thyroid tests over the years. Each of these tests was hailed as the definitive test at the time but was found to be imperfect. Indeed there may never be a perfect test, which is universally accurate for everybody. It is unrealistic to assume therefore that any test should be the sole criteria for diagnosis. Given this history perhaps we should be looking further afield for tests that more accurately reflect the clinical status of the patient and could complement the existing blood tests. One such test 18 , which is increasingly being used in the private sector, is urine testing for thyroid status. A study into this was conducted between May 1984 and July 1997 and concluded that “The 24hr urine free T3 test seems to be a reliable test, more accurate than the serum T4, serum free T4 and serum TSH test in the diagnosis of thyroid diseases and their follow up under treatment. It correlates well with the clinical status of the patient and is not influenced by binding globulins.

A series of main symptoms can be used as an efficient tool in the discovery of thyroid disease.” Blood tests are of course a very accurate indication of the levels of thyroid hormone in the blood but they do not reflect what is going on at a cellular level where the thyroid hormone is actually needed. This gives rise to the oft-repeated scenario of a patient being told that he/she is euthyroid when the clinical picture would indicate not. This is backed up by some of the responses to TPA-UK from doctors, reprinted below, and by many thousands of patients worldwide. TPA-UK asks you to urgently consider trials into this type of testing. Tests may prove useful in confirming or questioning a clinical diagnosis but any tests of themselves should not be the ultimate arbiter of that diagnosis. Expertise and judgment are critical; a full clinical diagnosis supported by test results would enable a more accurate outcome for patients. The key issue is that tests should be used to inform part of a clinical appraisal and not solely form that diagnosis.

TPA-UK is asking BTA to take note of the points that we have raised. The proposed new guidelines, if implemented, would have a devastating effect upon the lives of the many thousands within the UK population who are suffering from hypothyroidism, and who would remain undiagnosed and/or under-treated as a consequence.

Yours sincerely
Sheila Turner
Thyroid Patient Advocate.


Some responses received recently from Doctors by TPA-UK, regarding the BTA Proposed Draft Guidelines for TFT’s

“I have had the opportunity to read the draft, you are right, they should not be advocating the increase in the TSH value for initiating treatment for sub clinical hypothyroidism as even the current so called normal is disputed because of the shape of the distribution curve. The patient’s symptoms, family history, presence of microsomal antibdies / goitre must be taken into account. I entirely agree with your concerns. I feel they have not allowed enough time for consultation either as most of the endocrinologists I know have not had the time to read / comment on these guidelines”. Dr K Rizvi Consultant Physician GIM/ Endocrinology and Diabetes

“I have no problem with the current guidelines and follow the 1996 advice. You have my support in challenging an increased TSH level”. Dr J L

“I’m a GP so not an expert, but I can say that for years I’ve had the impression that some people seem hypothyroid but have a ‘normal’ TSH also, that several patients on replacement Thyroxine need more of it than the lab. Tests suggest they need ” Dr. C B

“Thank you for your letter forwarded to me as a GP. I trained at the Royal London where I was taught to give thyroxine replacement to suppress TSH, and to go on listening to the patient and their symptoms and work together to find the optimum dose for them. The info you enclosed will help me to continue in this wa”y. Dr. A P

“Thank you for your letter. The debate regarding the TSH normal range is not new. The Patient Advocacy Group has been pushing for lowering of the upper limit of TSH normal range amongst other things, including the use of thyroid extract. I understand that the current thyroid function testing guidelines are in the drafting stage and, to me, in their current form, they will not change what we do in practice at present. I don’t see the reference ranges for TSH changing based on the current evidence of very small animal studies. Even if the upper limit of TSH changes, say, to 3.0 iu/l, this would result in enlarging the sub-group of patients with subclinical hypothyroidism. It will not result in an increase in the diagnosis of overt hypothyroidism, which would require, in addition to a higher TSH, a subnormal free T4. One thing I would like to see is an agreement or standardisation of TSH ranges among the laboratories as we currently see with HbA1c. I await the results of the final thyroid function testing guideline”. Dr. F I A. Consultant Physician

“Thanks for this info. Personally I will continue to use my usual cut off level of TSH of 4.4 or above. I still feel a lot of my (usually female) patients know when they are Hypothyroid, before the blood test shows it. I would be very wary of using a cut off of 10 for TSH” . Dr. L O’M

” The argument is quite complex particularly as there is no single standard reference range laboratories determine their own and this is necessary because they use different tests and equipment. I get results from 2 labs and the upper limit for one is 3.5 and the other is 5.0. This does not surprise but it needs to be taken into account. I have just been co-author on a short review on this area it has not yet been accepted for publication (and so I cannot send it to you) but it does reflect a Europe-wide view which was that there is no need at present to lower the reference range to the lower half of what it currently is (which is typically 0.5-2.0) for the diagnosis of new thyroid disease but we also recognise that in individual patients there may be a benefit from using doses of thyroxine to bring the TSH down to this lower level. I also hope our BTA guidelines which are in preparation will cover this point” . Prof. A W

References :[/LEFT]

  1. Vanderpump MP, Tunbridge WM, French JM, Appleton D, Bates D, Clark F, Grimley Evans J, Hasan DM, Rodgers H, Tunbridge F, et al. The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham Survey. Clin Endocrinol (Oxf) 1995 Jul;43(1):55-68.
  2. AACE 2003 Campaign Encourages Awareness of Mild Thyroid Failure, Importance of Routine Testing.
  3. T Bjro et al, ‘Prevalence of thyroid disease, thyroid dysfunction and thyroid peroxidase antibodies in a large, unselected population. The Health Study of Nord-Trndelag (HUNT).’ European Journal of Endocrinology 2000 143 639-647.
  4. Colin M Dayan, Ponnusamy Saravanan, Graham Bayly Whose normal thyroid function is better-yours or mine? Commentary The Lancet 2002 Aug 03; 360 (9330): 353.
  5. ADVANCEDATA, from Vital & Health Statistics of The National Center for Health Statistics, US Department of Health and Education Welfare, No. 5, February 22, 1977, Public Health Service Health Resources Administration. “A Comparison of Levels of Serum Cholesterol of Adults 18-74 Years of Age in the Untied States in 1960-62 and 1971-74.”
  6. Mahanonda N, Leowattana W, Kangkagate C, Lolekha P, Pokum S.Homocysteine and restenosis after percutaneous coronary intervention. J Med Assoc Thai 2001 Dec;84 Suppl 3:S636-44.
  7. Robinson K, Mayer EL, Miller DP, Green R, van Lente F, Gupta A, Kottke-Marchant K, Savon SR, Selhub J, Nissen SE, et al. Hyperhomocysteinemia and low pyridoxal phosphate. Common and independent reversible risk factors for coronary artery disease. Circulation 1995 Nov 15;92(10):2825-30.
  8. Narrow Individual Variations in Serum T4 and T3 in Normal Subjects: A Clue to the Understanding of Subclinical Thyroid Disease,” The Journal of Clinical Endocrinology & Metabolism, Vol. 87, No. 3 1068-1072, 2002.
  9. Pollock MA, Sturrock A, Marshall K, Davidson KM, Kelly CJ, McMahon AD, McLaren EH. Thyroxine treatment in patients with symptoms of hypothyroidism but thyroid function tests within the reference range: randomised double blind placebo controlled crossover trial. BMJ 2001 Oct 20;323(7318):891
  10. Consensus statement for good practice and audit measures in the management of hypothyroidism and hyperthyroidism (BMJ 1996;313:539-544, 31 August)
  11. Christian Meier, Peter Trittibach, Merih Guglielmetti, Jean-Jacques Staub, and Beat Mller
    BMJ 2003 326: 311-312. [Full Text]
  12. Matt Bolton, Customer Service Centre, Department of Health. Ref. Ref:DE00049269
  13. Hak AE, Pols HA, Visser TJ, Drexhage HA, Hofman A, Witteman JC. Subclinical hypothyroidism is an independent risk factor for atherosclerosis and myocardial infarction in elderly women: the Rotterdam Study. Ann Intern Med 2000 Feb 15;132(4):270-8.
  14. Michalopoulou G, Alevizaki M, Piperingos G, et al. High serum cholesterol levels in persons with “high-normal” TSH levels: should one extend the definition of subclinical hypothyroidism? Eur J Endocrinol 1998; 138: 141-45
  15. Yi-Da Tang et al, ‘Low Thyroid Function Leads to Cardiac Atrophy with Chamber Dilation, Impaired Myocardial Blood Flow, Loss of Arterioles, and Severe Systolic Dysfunction.’ This Cardiovascular Research Institute (CRI) study is published in the Nov. 15 issue of Circulation, the journal of the American Heart Association. The study can be viewed online at: . The news article is at
  16. Leonard Wartofsky and Richard A Dickey, The Evidence for a Narrower Thyrotropin Reference Range Is Compelling’ The Journal of Clinical Endocrinology & Metabolism Vol. 90, No. 9 5483-5488
  17. Fatourechi V, Klee GG, Grebe SK, et al. Effects of reducing the upper limit of normal TSH values. JAMA 2003;290:3195-3196.
  18. W. V. BAISIER MD,1. Thyroid Insufficiency. Is TSH Measurement the Only Diagnostic Tool?
    Journal of Nutritional & Environmental Medicine (2000) 10, 105-113, Published:10-Jan-2000


You must be logged in to post a comment.

Previous comments