June 2008, Vol 30, No. 2
Update Article

Chemical pathology case conference - interpretation of thyroid function tests

Angel O K Chan 陳安琪, C C Shek 石志忠, Michael H M Chan 陳浩明, W T Poon 潘永達, Y P Yuen 袁月冰, Chloe M Mak 麥苗, Morris H L Tai 戴學良, C W Lam 林青雲, Rossa W K Chiu 趙慧君, Albert Y W Chan 陳恩和, Sidney Tam 譚志輝, Tony W L Ma 麥永禮

HK Pract 2008;30:86-91

Summary

Thyroid function testing basically consists of the measurement of thyroid-stimulating hormone, free thyroxine and occasionally free triiodothyronine. It is one of the most commonly requested endocrine investigations in primary care. With only few exceptions, thyroid-stimulating hormone is the best first-line screening test for many thyroid disorders. Correct interpretation of these test results is essential for providing quality patient management. In this case conference, we would like to use several clinical scenarios to illustrate the common patterns of thyroid function test results and their interpretations.

摘要

甲狀腺功能測試基本上包括促甲狀腺激素,血清游離甲狀腺素,偶爾也檢測游離三碘甲狀腺原氨酸。這是基層醫療最常用的內分泌測試之一。除了某些情況外,促甲狀腺激素是眾多甲狀腺紊亂的首選篩查測試。正確地分析結果才能給予病人最適當的治療。這次病例會議中,我們會用數個臨床例子闡述如何解釋常見甲狀腺功能測試的結果。


Introduction

Secretion of thyroid hormones is under the tight regulation of the hypothalamic-pituitary-thyroid feedback loop. Thyroid-stimulating hormone (TSH) carries a reciprocal log-linear relationship with free thyroxine (FT4). A two-fold change in FT4 causes an approximately 100-fold change in the serum TSH concentration. TSH also has a narrow intra-individual variability. It has been reported that a change in TSH level by approximately 0.7 mIU/L indicates significant change in an individual's thyroid status.1 TSH is, therefore, a sensitive marker to reflect the presence of disruption in this feedback loop. Nowadays most, if not all, clinical laboratories are using third generation TSH immunoassays which can confidently measure TSH at a level of 0.02 mIU/L or less and are able to differentiate euthyroid from thyrotoxic levels (<0.01 mIU/L), further supporting the use of TSH in the initial assessment of thyroid function. Nevertheless, TSH result alone may be misleading in conditions such as central (secondary) hypothyroidism, recovery from overt hyperthyroidism, TSH-secreting pituitary tumour, and resistance to thyroid hormone (RTH).

T4 is the major thyroid hormone secreted by the thyroid gland. Peripheral deiodination of T4 accounts for over 80% of the daily triiodothyronine (T3) production. In hyperthyroidism, T3 level is usually high because of increased secretion from the thyroid and increased peripheral conversion from T4. In hypothyroidism, T3 concentration remains normal due to increased deiodination from T4 until the disease becomes severe. Many non-thyroidal factors and stress would also affect the T3 level. In fact, most of the thyroid conditions can be diagnosed by measuring the T4 level after the initial screening with TSH. T3 testing is not routinely necessary unless facing complicated or unusual thyroid problems.

Since 99.7% of T3 and 99.97% of T4 are reversibly bound to thyroxine-binding globulin (TBG), thyroxine-binding prealbumin and albumin upon secretion into the circulation, abnormalities in these binding protein levels lead to abnormal total T3 (TT3) and total T4 (TT4) results despite the absence of thyroid dysfunction, such as in patients taking estrogen which induces TBG synthesis, and in patients in hypoproteinaemic states. With the availability of free thyroid hormone immunoassays, measuring free hormone levels is preferred over the total hormone levels, although free hormone immunoassays are not problem-free.

Apart from direct measurement of TSH and thyroid hormones, tissue markers may also assist in the assessment of thyroid function when facing complicated thyroid cases. One of the most extensively studied tissue markers in Hong Kong is erythrocyte zinc (EZn). EZn is strongly associated with erythrocyte carbonic anhydrase. Changes in the thyroid status affect the enzyme level and therefore the EZn concentration.2,3 The most well-established clinical utility of EZn is to differentiate subacute thyroiditis or transient hyperthyroidism of hyperemesis gravidarum from established hyperthyroidism, where in the former two conditions EZn remains normal and in the latter it is decreased.4,5

Many therapeutic agents disturb the normal functioning of the pituitary-thyroid axis. Lithium causes hypothyroidism by inhibiting the synthesis of thyroid hormones. Rarely, it also causes hyperthyroidism. Propranolol and glucocorticoids inhibit the conversion of T4 to T3 and may contribute to hypothyroidism. The effect of amiodarone on the thyroid is much more complicated. Amiodarone has a high iodine content and its side effects include both hypothyroidism and hyperthyroidism by inhibiting the conversion of T4 to T3, by blocking the thyroid hormone action or by destroying the thyroid gland.6 Soybean preparation, ferrous sulphate, calcium carbonate and cholesytramine are some of the commonly used drugs which inhibit the absorption of ingested thyroxine.

In this article, we would like to use a series of clinical scenarios to illustrate how to apply and interpret thyroid function tests (TFT) in the diagnosis and management of patients with thyroid dysfunction in the primary care setting.

Case 1

A 24-year old female presented with palpitation, weight loss and oligomenorrhoea in the recent 4 months. Physical examination revealed diffuse non-tender symmetrical goiter and bilateral exophthalmos.

Serum

Unit

Reference range

TSH

<0.01

Low

mIU/L

0.27-4.20

FT4

>100.0

High

pmol/L

12.0-22.0

Question: How to interpret the TFT results?

Answer: The clinical picture together with the suppressed TSH and grossly elevated FT4 pointed towards Graves' disease as the most likely diagnosis. Testing on thyroid peroxidase antibody (anti-TPO Ab) and anti-TSH receptor antibody are not routinely required when the clinical signs and the TFT results are already diagnostic,7,8 but they are helpful in identifying the autoimmune nature of thyroid problems if there is no obvious clinical signs and if the aetiology of the disease would have an impact on patient management, e.g. hyperthyroidism during pregnancy.7,8 If thyroiditis is suspected, erythrocyte sedimentation rate should be checked and it should be elevated.


Case 2

Patient in Case 1 had been treated with carbimazole for 3 months. She was found to be clinically euthyroid on follow-up.

Serum

Unit

Reference range

TSH

<0.01

Low

mIU/L

0.27-4.20

FT4

19.8

pmol/L

12.0-22.0

Question: Should the dosage of carbimazole be increased in view of the suppressed TSH?

Answer: No. Suppression of TSH can remain for several months in patients under good control with anti-thyroid therapy, attributing to the delayed recovery of the pituitary-thyroid axis. Adjustment of anti-thyroid drug regime should be based on FT4 with monitoring once every 4-6 weeks for the first 4-6 months, and less frequently (approximately every 3 months) when the patient is on maintenance dose.7,9


Case 3

A 30-year old female complained of weight loss, tremor and heat intolerance in the recent 6 months. Physical examination revealed a right-sided 2 cm thyroid nodule.

Serum

Unit

Reference range

TSH

<0.01

Low

mIU/L

0.27-4.20

FT4

15.2

pmol/L

12.0-22.0

Question: Did the patient have hyperthyroidism?

Answer: Although FT4 was normal, the clinical findings and the suppressed TSH were suggestive of hyperthyroidism. Testing the FT3 level in this patient would be helpful in making the diagnosis:

Serum

Unit

Reference range

FT3

12.8

High

pmol/L

3.0-6.2

The elevated FT3 result confirmed the suspicion of T3 toxicosis, which is seen in a minority of patients with hyperthyroidism. FT3 instead of FT4 is the test of choice for monitoring patients with T3 toxicosis.


Case 4

A 19-year old female complained of hand tremor, insomnia and palpitation in the recent 3 months. Her body mass index was 20.0 kg/m2 three months ago, and had reduced to 18.4 kg/m2 in this visit. She had no goiter or thyroid eye sign on examination, but had profuse sweating, hand tremor and a pulse rate of 96/min.

Serum

Unit

Reference range

TSH

<0.01

Low

mIU/L

0.27-4.20

FT4

3.8

Low

pmol/L

12.0-22.0

Testing on the FT3 level would be helpful in this case:

FT3

>46.1

High

pmol/L

3.0-6.2

Question: Did the patient have T3 toxicosis?

Answer: T3 toxicosis was a possible diagnosis. However, three features in the TFT results suggested the diagnosis of thyrotoxicosis factitia where the patient was taking T3 analogue: 1) suppressed TSH, 2) suppressed endogenous production of thyroid hormones, as reflected by the low FT4, and 3) grossly elevated FT3 level. Tiratricol, which is a T3 analogue, is used for the treatment of non-tumoral pituitary resistance to thyroid hormone, and is also used for weight reduction, though its use for the latter condition is unorthodox. This drug is known to cross-react with many commercial FT3 immunoassays,10,11 giving rise to falsely high FT3 results. Testing on serum thyroglobulin (Tg) level may help to differentiate between thyrotoxicosis factitia from other causes of hyperthyroidism where in the former condition, the Tg level is suppressed or low-normal. Hyperthyroidism due to inflammation, excessive stimulation or physical destruction of the thyroid gland would have elevated Tg level.12

On further history taking, the patient admitted that she had been taking an over-the-counter weight reduction agent which consisted of tiratricol in the recent 3 months


Case 5

A 74-year old lady was found to have low mood, poor memory and constipation by her relatives and was brought to the clinic. Physical examination revealed pallor, loss of eyebrows, cold peripheries and slow-relaxing reflexes. 

Serum

Unit

Reference range

TSH

33.22

High

mIU/L

0.27-4.20

FT4

8.7

Low

pmol/L

12.0-22.0

Question: How to interpret the TFT results?

Answer: Elevated TSH and low FT4 levels were compatible with primary hypothyroidism. Testing on FT3 level is not necessary for making the diagnosis. The incidence of spontaneous primary hypothyroidism increases with age and the presence of anti-thyroid antibodies further increases the risk.13 Levothyroxine (L-T4) is the preferred drug for long-term replacement.8,12,14 Due to the slow re-equilibration of TSH, its level should only be rechecked 6-8 weeks after the commencement or adjustment of the therapy,7,8,12 with the L-T4 dosage titrated against the TSH level. Target TSH level between 0.5 to 2.0 mIU/L12 or within the reference interval7,8 have been advocated. Once the TSH level is normalized, a follow-up TSH level should be checked once every 6 to 12 months.8,12 FT4 level fluctuates according to the time of daily dosing and its level is expected to be elevated if blood is taken within the first few hours after taking L-T4 despite the simultaneous TSH level is normal in an optimized regime. Monitoring the FT4 level is not required unless compliance is a concern, in which case discordant TSH and FT4 results (both values high) might be observed if the patient only resumes the thyroxine replacement just before the blood collection.


Case 6

A 50-year old man complained of double vision, malaise and impotence.

Serum

Unit

Reference range

TSH

1.23

mIU/L

0.27-4.20

FT4

8.7

Low

pmol/L

12.0-22.0

Question: How to interpret these TFT results?

Answer: The inappropriately normal TSH in the presence of low FT4 in this clinical setting is compatible with central or secondary hypothyroidism. When there is clinical suspicion of secondary hypothyroidism, both TSH and FT4 should be requested because TSH alone would be misleading. Referral to endocrinologists for additional tests on pituitary function may be required and it is important to clarify the adrenal status before commencing L-T4 replacement therapy to avoid precipitating an Addisonian crisis. In secondary hypothyroidism, FT4 instead of TSH level should be monitored when the patient has been put on L-T4 replacement. FT4 level should be optimized to within the upper third of the reference interval, with blood collected before the daily dose of L-T4.7,12 Annual FT4 checking should be performed after stabilized on L-T4 replacement.


Case 7

A 65-year old lady who had a 5-year history of hypertension was having an annual checkup. 

Serum

Unit

Reference range

TSH

5.47

High

mIU/L

0.27-4.20

FT4

16.6

Low

pmol/L

12.0-22.0

Question: How to interpret these TFT results?

Answer: Mildly elevated TSH together with normal FT4 were compatible with subclinical hypothyroidism. The findings should be confirmed by repeating the TFT within 6 months.7,12 Testing on thyroid autoantibodies is warranted because patients with positive antibodies are prone to develop overt hypothyroidism. L-T4 replacement will benefit those with TSH level greater than 10 mIU/L but routine treatment for those with TSH level between 5 to 10 mIU/L remains debatable.7,8,15


Case 8

A 40-year old man had a pre-employment checkup.

Serum

Unit

Reference range

TSH

1.75

mIU/L

0.27-4.20

FT4

24.2

High

pmol/L

12.0-22.0

FT3

6.7

High

pmol/L

3.0-6.2

He did not complain of any thyrotoxic symptoms and he was clinically euthyroid. TFT was repeated a few months later, showing similar results.

Question: Was the patient thyrotoxic?

Answer: Despite the elevated thyroid hormone levels, the patient remained euthyroid and the TSH level was not suppressed. This condition is known as euthyroid hyperthyroxinaemia. The differential diagnoses of euthyroid hyperthyroxinaemia include RTH, familial dysalbuminaemic hyperthyroxinaemia (FDH) and TBG excess. RTH is an autosomal dominant condition. About 90% of the cases were caused by mutation in the THRB gene,16,17 producing defective thyroid hormone receptor beta and disrupting the negative feedback loop in the pituitary-thyroid axis. Patients with RTH may be hypothyroid, euthyroid or hyperthyroid, depending on the basis of their genetic defect. FDH is another autosomal dominant condition and the defect lies in the ALB gene. The mutant albumin has abnormally high affinity to T4, leading to falsely high FT4 results in homogenous or one-step FT4 immunoassays using T4 analogue.18 T3 measurement is not affected. TBG excess could be an inherited or iatrogenic condition. Its effect is best appreciated in TT4 measurement. The disturbed binding equilibrium with displacement of thyroid hormones from their binding proteins on initiation of some therapies such as salicylate, phenytoin, and frusemide, and interference in the TSH or thyroid hormone immunoassays might also give rise to TFT pattern seen in this case.19 If the patient exhibits symptoms and signs of hyperthyroidism, TSH-secreting pituitary tumour should be considered. When faced with these TFT results, anti-thyroid treatment should be withheld and referral to an endocrinologist for further management is recommended.

In this case, further investigations on the patient and his family members showed that the patient and some of his family members had RTH. Several cases on local patients with RTH have been reported.11,20,21


Case 9

A 23-year old female had a routine checkup. She did not have any history of thyroid problem. She had no symptoms or signs of hypothyroidism.

Serum

Unit

Reference range

TSH

35.82

High

mIU/L

0.27-4.20

FT4

20.4

pmol/L

12.0-22.0

FT3

5.2

pmol/L

3.0-6.2

Question: Did the patient have subclinical hypothyroidism?

Answer: The extent of TSH elevation and the high-normal thyroid hormone levels make the diagnosis of subclinical hypothyroidism highly unlikely. When there are discordances in the clinical findings and the biochemical results, the laboratory should be contacted to discuss about the possibility of errors in any stage of the pre-analytical, analytical or post-analytical results. In this case, there was interference in the TSH assay, causing the spuriously high TSH result stage. The patient's TSH was normal when measured by another TSH assay platform and she did not have any thyroid dysfunction. Interference of the thyroid hormone immunoassays by autoantibodies, heterophile antibodies and rheumatoid factors are well-recognised.22,23 Grossly different TFT results when performed in different laboratories using different immunoassay platforms in an otherwise euthyroid patient should prompt the clinicians to think about immunoassay interference.


Case 10

A 78-year old man with history of hypertension on anti-hypertensive was having a routine body check. There was no personal or family history of thyroid problem. Physical examination revealed a bilaterally enlarged multinodular goiter. ECG showed atrial fibrillation.

Serum

Unit

Reference range

TSH

<0.01

Low

mIU/L

0.27-4.20

FT4

20.2

pmol/L

12.0-22.0

FT3

5.8

pmol/L

3.0-6.2

Question: Was the patient thyrotoxic?

Answer: Suppressed TSH and normal FT4 and FT3 were compatible with subclinical hyperthyroidism. The causes of subclinical hyperthyroidism are similar to those of overt hyperthyroidism, but patients remain asymptomatic or only mildly symptomatic, and could be found in less than 2% of the adult or elderly population.8,15 Repeating the assessment of the thyroid status is necessary to confirm the biochemical abnormality and to exclude the progression to overt hyperthyroidism. The risks of atrial fibrillation and osteoporosis are increased in patients with subclinical hyperthyroidism and periodic monitoring of the thyroid function is warranted. Therapy should be individualized and referral to endocrinologist should be considered.

Conclusion

Bearing in mind that the TSH and the thyroid hormone levels normally have reciprocal relationship due to the negative feedback regulation between the pituitary and the thyroid, most of the TFT results are straightforward and easy to interpret. Still, there are times when the TSH and the thyroid hormone levels point towards the same direction. In these circumstances, euthyroid hyperthyroxinaemia, TSH-secreting pituitary tumour, central hypothyroidism and immunoassay interferences are the possible differential diagnoses. As mismanagement of thyroid conditions could have serious consequences, endocrinologist referral or discussion with the laboratory should be considered.

Key messages

  1. Thyroid-stimulating hormone (TSH) carries a reciprocal log-linear relationship with free thyroxine.
  2. TSH is the best first-line screening test for various thyroid disorders except for a few conditions where TSH result alone may be misleading, such as central (secondary) hypothyroidism, recovery from overt hyperthyroidism, TSH-secreting pituitary tumour, and resistance to thyroid hormone.
  3. Most of the thyroid disorders can be diagnosed by measuring the thyroxine level after an initial abnormal TSH result. Triiodothyronine (T3) testing is not routinely necessary.
  4. Free hormone is preferred to total hormone measurement since the former is less affected by the binding proteins.
  5. T3 analogue is not uncommonly used as an unorthodox slimming agent. The diagnosis of thyrotoxicosis factitia due to such consumption should be considered when the T3 result is grossly elevated while TSH and free thyroxine levels are suppressed.
  6. TSH is the test of choice for monitoring patients on L-thyroxine replacement for primary hypothyroidism. Free thyroxine level is needed only when compliance is a concern.
  7. Analytical interference is not uncommon in thyroid hormone assays. When there is discrepancy in the clinical and laboratory findings, consultation with chemical pathologists is recommended to verify the results.

Angel O K Chan, MBChB(CUHK), FRCPA, FHKAM(Pathology)
Associate Consultant,

C C Shek, MBBS(HK), FRCPath, FRCPA, FHKAM(Pathology)
Consultant,
Department of Pathology, Queen Elizabeth Hospital.

Michael H M Chan, MBChB(CUHK), FRCPA, FHKAM(Pathology)
Associate Consultant,

C W Lam, MBChB(CUHK), PhD(CUHK), FRCPA, FHKAM(Pathology)
Associate Professor,

Rossa W K Chiu, MBBS(Qld), PhD(CUHK), FRCPA, FHKAM(Pathology)
Professor,
Department of Chemical Pathology, The Chinese University of Hong Kong, Prince of Wales Hospital.

Morris H L Tai, MBChB(CUHK), FRCPA, FHKAM(Pathology)
Associate Consultant,

Sidney Tam, FRCP(Edin), FRCPA, FHKAM(Medicine), FHKAM(Pathology)
Head and Consultant,
Division of Clinical Biochemistry, Department of Pathology, Queen Mary Hospital.

W T Poon, MBChB(CUHK), FHKAM(Pathology)Associate
Consultant,

Tony W L Mak, MBChB(CUHK), MBA, FRCPA, FHKAM(Pathology)
Consultant,
Hospital Authority Toxicology Reference Laboratory

Y P Yuen, MBChB(CUHK), FRCPA, FHKAM(Pathology)
Associate Consultant,

Chloe M Mak, MBBS(HK), FRCPA, FHKAM(Pathology)
Associate Consultant,

Albert Y W Chan, MBChB(Glasg), MD(CUHK), FHKCP, FHKAM(Pathology)
Consultant,
Department of Pathology, Princess Margaret Hospital.

Correspondence to: Dr Angel O K Chan, Department of Pathology, Queen Elizebeth Hospital, Kowloon, Hong Kong SAR


References
  1. Browning MCK, Ford RP, Callaghan SJ, et al. Intra- and interindividual biological variation of five analytes used in assessing thyroid function: implications for necessary standards of performance and the interpretation of results. Clin Chem 1986;32:962-966.
  2. Swaminathan R, Segall NH, Chapman C, et al. Red-blood-cell composition in thyroid disease. Lancet 1976;25:1382-1385.
  3. Yoshida K, Kiso Y, Watanabe T, et al. Clinical utility of red blood cell carbonic anhydrase I and zinc concentrations in patients with thyroid diseases. Metabolism 1991;40:1048-1051.
  4. Chow CC, Shek CC, Mak YT, et al. Erythrocyte zinc concentrations in subacute (de Quervain's) thyroiditis. BMJ 1989;299:1505-1506.
  5. Chan AYW, Mak YT, Shek CC, et al. Changes in erythrocyte zinc in a case of transient thyrotoxicosis. Ann Clin Biochem 1991;8:524-525.
  6. Basaria S, Cooper DS. Amiodarone and the thyroid. Am J Med 2005;118:706-714.
  7. The Association for Clinical Biochemistry, British Thyroid Association and British Thyroid Foundation. UK guidelines for the use of thyroid function tests. July 2006.
  8. American Association of Clinical Endocrinologists. Medical guidelines for clinical practice for the evaluation and treatment of hyperthyroidism and hypothyroidism. Endocr Pract 2002;8:457-469.
  9. Cooper DS. Hyperthyroidism. Lancet 2003;362:459-468.
  10. Bauer BA, Elkin PL, Erickson D, et al. Symptomatic hyperthyroidism in a patient taking the dietary supplement tiratricol. Mayo Clin Proc 2002;77:587-590.
  11. Kong APS, Lam CW, Chan AOK, et al. Resistance to thyroid hormone in a Chinese family with R429Q mutation in the thyroid hormone receptor beta gene. Hong Kong Med J 2005;11:125-129.
  12. Demers LM, Spencer CA. Laboratory support for the diagnosis and monitoring of thyroid disease. National Academy of Clinical Biochemistry. 2002.
  13. Vanderpump MP, Tunbridge WM, French JM, et al. The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham Survey. Clin Endocrinol 1995;43:55-68.
  14. Singer PA, Cooper DS, Levy EG, et al. Treatment guidelines for patients with hyperthyroidism and hypothyroidism. Standards of Care Committee, American Thyroid Association. JAMA 1995;273:808-812.
  15. Fatourechi V. Subclinical thyroid disease. Mayo Clin Proc 2001;76:413-417.
  16. Weiss RE, Hayashi Y, Nagaya T, et al. Dominant inheritance of resistance to thyroid hormone not linked to defects in the thyroid hormone receptor alpha or beta genes may be due to a defective cofactor. J Clin Endocrinol Metab 1996;81:4196-4203.
  17. Pohlenz J, Weiss RE, Macchia PE, et al. Five new families with resistance to thyroid hormone not caused by mutations in the thyroid hormone receptor beta gene. J Clin Endocrinol Metab 1999;84:3919-3928.
  18. Ruiz M, Rajatanavin R, Young RA, et al. Familial dysalbuminemic hyperthyroxinemia: a syndrome that can be confused with thyrotoxicosis. N Engl J Med 1982;306:635-639.
  19. Burtis CA, Ashwood ER. Tietz Textbook of Clinical Chemistry 3rd edition. WB Saunders. 1999; p1504.
  20. Lam CW, Chan AOK, Tong SF, et al. A novel THRB mutation associated with mild resistance to thyroid hormone. Clin Chim Acta 2005;358;55-59.
  21. Pohlenz J, Weiss RE, Macchia PE, et al. Five new families with resistance to thyroid hormone not caused by mutations in the thyroid hormone receptor b gene. J Clin Endocrinol Metab 1999;84:3919-3928.
  22. Despres N, Grant AM. Antibody interference in thyroid assays: a potential for clinical misinformation. Clin Chem 1998;44:440-454.
  23. Ward G, McKinnon L, Badrick T, et al. Heterophilic antibodies remain a problem for the immunoassay laboratory. Am J Clin Pathol 1997:108;417-421.