Thyroid disorders

Thyrotoxicosis

Recognition of thyrotoxicosis is generally straightforward if there is weight loss, a tachycardia and thyroid eye signs. Pitfalls occur in pregnancy, patients with known valvular heart disease, where the deterioration is unthinkingly ascribed to the valve, and in the elderly, who may simply dement or become apathetic without much in the way of peripheral signs.

Most patients have Graves’ disease (the so-called diffuse toxic goitre) but other causes are toxic multinodular goitre, toxic solitary adenoma, and various forms of thyroiditis such as that associated with the post-partum period and the subacute variety known as de Quervain’s (thyroid pain with fever and raised ESR).

Initial treatment of Graves’ disease is with carbimazole 20 mg per day, which may also have mild immunosuppressant effects (on the thyroid). Subsequently, titrate the dose against thyroid function (usually 5 to 15 mg daily.) Patients with a poor response should be considered for early I¹³¹ rather than escalating the dose, but the explanation is often poor adherence to treatment. Combining thyroxine and carbimazole does not yield superior outcomes,¹ and the side-effect and withdrawal profile may be worse. Side effects of carbimazole include agranulocytosis in 0.5% of patients and a maculopapular skin rash in 5%. Patients should be encouraged to present for a FBC check if they develop a sore throat.

Efficacy – after 9 months about 40% will be in remission. If not in remission by then, or if there is relapse, refer for radioactive iodine treatment (stop carbimazole about five days before I¹³¹ and re-start 5 days after.)

Thyrotoxic eye disease – treat with carbimazole and prednisone (e.g. 20mg daily) to settle before considering I¹³¹.

Patients with toxic multinodular goitre should have their initial symptoms controlled with carbimazole and then be referred for radioactive iodine – induction of remission with carbimazole is unlikely.

Patients with subacute thyroiditis, where symptoms may respond nicely to non-steroidal anti-inflammatory treatment alone. Very occasionally one may need to give a short course of prednisone to control the pain and sometimes beta-blockade is helpful.

Perspective – thyroid function testing

Modern TSH assays have high sensitivity and specificity, yielding LR+ of >99 and LR- of <0.01 for the diagnosis of thyrotoxicosis or hypothyroidism. Confusion may arise with the sick euthyroid syndrome, where patients with other medical conditions who are clinically euthyroid have deranged TFTs (TSH up or down, T4 up or down). If in doubt, repeat after a few weeks – it should have normalised.

Perspective – beta-blockers in hyperthyroidism.

Most texts endorse the use of beta-blockers for the relief of symptoms of thyrotoxicosis. There are a few caveats.

• Patients started on beta-blockers but not on definitive therapy may feel better and carry on with this, then present severely ill after stopping or running out of beta-blockers. Patients should clearly understand that beta-blockers are only intended for short-term symptom control while waiting for more definitive treatment to work.
• Clinicians sometimes think that there is an automatic need to give beta-blockers simply because the T4 is very high. In patients without very distressing symptoms, carbimazole usually starts to settle things within a couple of weeks.
• Patients with any evidence of cardiac dysfunction should be evaluated very cautiously and treatment started carefully with a short acting agent in a low dose,² which can then be titrated to response, e.g. propranolol 20 mg 8 hourly. Metoprolol is possibly preferable³ but more difficult to access. In patients without cardiac disease, choice of beta-blocker is less important, and atenolol is a reasonable option.⁴ If in doubt, ask for help and get echocardiographic evaluation of LV function.

Perspective – dose of carbimazole.

The starting dose of carbimazole is often fairly haphazard, and there isn’t firm evidence on which to base recommendations. In an unblinded randomised trial,⁵ 20 mg/d was as effective as 40 mg/d in terms of clinical response at 6 and 12 weeks. In those given the higher dose, biochemical resolution at 4 weeks was greater, but drug-related hypothyroidism was more common. If the initial T4 was very high (>260 nmol/l) 40 mg/d was more effective.

Perspective – predicting risk of relapse after stopping carbimazole

A review⁶ looked at the predictive value for relapse of various features present at the time of diagnosis in Grave’s. Overall less than 50% of risk was explained by the features, with TSH receptor antibodies accounting for less than 10% and thyroid stimulating antibodies for 1%. It is arguable whether doing these two tests is worthwhile if intended for this purpose only :

Thyroid storm

Patients with poorly controlled Graves’ get an infection, undergo surgery or some other stressor, and then develop a constellation of:

• fever
• tachycardia
• cardiac failure
• altered sensorium
• GIT problems (vomiting, diarrhoea, abdominal pain, even jaundice).

A grading system has been proposed,⁷ based on giving points for the severity of each of the items mentioned above, but adding up the score probably adds little to recognising the possibility based on some or all of the signs mentioned above.

Management:

• Carbimazole 20 mg 8 hourly orally or via NGT.
• Potassium iodide solution (USP) or Lugol’s iodine are very slightly different solutions of potassium iodide which most pharmacies will delight in making up. The dose in thyroid storm is 10-20 mg of free iodine 8 hourly for up to 10 days, which works out to about 4 drops of Lugol’s iodine 8 hourly, or about 3 drops of potassium iodide USP 8 hourly. Give well diluted with water, as it is a gastric irritant. It should be started only AFTER the carbimazole has been on board for a few hours at least, as the iodine load on its own may stimulate more hormone synthesis initially before settling in to inhibit hormone release.
• Hydrocortisone 300 mg IV, followed by 100 mg 8 hourly is often recommended because of ‘relative’ adrenal insufficiency and a purported reduction in peripheral T4 to T3 conversion.
• Propranolol 40-80 mg 8 hourly is commonly recommended ‘after ascertaining that LV function is normal’ If unsure, diltiazem may have an adequate effect on pulse rate without such concerns about LV function.

(If it is difficult to get hold of potassium iodide, an alternative is lithium 250mg orally 8 hourly, but toxicity is a concern, so either monitor levels, or stop it after a day or so. One other option is iodinated contrast media, but dosing is poorly established and obviously varies from agent to agent.)

Hypothyroidism

This is a relatively straightforward management issue once the diagnosis is considered. Common causes are post-autoimmune thyroiditis (Hashimoto), post I¹³¹ for thyrotoxicosis, and long after thyroid surgery. Possible drug associations are with both amiodarone and stavudine. Transient hypothyroidism may be a component of deQuervain’s thyroiditis (fever, malaise, tender thyroid) and may also be due to peripartum thyroiditis, which usually only lasts about three months, although very occasionally can be permanent. The positive predictive values of individual symptoms commonly associated with hypothyroidism is about 10%.⁸ False positive TSH results may occur in renal failure, adrenal failure, and in hypothermia.

Treatment is with thyroxine, but this drug has quite a narrow therapeutic index, with some studies showing that about 40% of patients are either under or over treated.⁹

Unless frail, with ischaemic heart disease, or in heart failure, start with 100 mcg per day of thyroxine, and increase by monthly increments of 25-50 micrograms to a target maintenance dose of 1.8 mcg/kg/day. Aim for a TSH and T4 within the normal range.

In the elderly, the expected target dose is lower at 0.5 mcg/kg/d, but again titrate according to TSH.

In IHD, start at 25 mcg/day and increase by monthly increments of 25 mcg/day.

Patients with hypothyroidism of pituitary origin obviously can’t be monitored using a TSH (it will be low) – in this group keep the T₄ in the upper half of the normal range. Note: if you do see a hypothyroid patient with a low TSH, consider panhypopituitarism, which means he or she may well be cortisol deficient. If this is a real possibility, then send off a serum cortisol and start hydrocortisone replacement BEFORE starting thyroxine, as the latter can precipitate an adrenal crisis in this situation.

Patients on warfarin may need a warfarin dose increase once the hypothyroidism is brought under control.

Once fully stabilised, patients only need their TSH checking yearly (but earlier if there is a change in state of health)

Perspective – diagnosis of hypothyroidism.

A coarse or cool and dry skin, and hypothyroid speech each have a LR+ of about 5; of note is that a pulse of <70 bpm only has a LR- of 0.8.¹⁰ In other words, absence of a bradycardia is not a good way of excluding hypothyroidism. The LR+ for eyebrow loss was 1.9 and for periorbital puffiness 2.8. The combination of coarse skin, bradycardia and delayed ankle jerks had an unimpressive LR+ of 3.75 and LR- of 0.48.¹¹

Perspective: how low should the TSH go on treatment?

A smallish (56 patients) double blind randomised cross-over trial¹² comparing clinical benefit after eight weeks (quality of life, cognitive function, and personal assessment of well-being) failed to demonstrate any difference between patients with average TSHs of 2.8, 1.0 or 0.3 mU/l. Previous observational studies had suggested that pushing the TSH lower might make patients feel better. A very much larger study supports keeping the TSH in the normal range, obviously with considerably worse outcomes for mortality, cardiovascular events and fractures when the TSH was very high (>10)¹³

Myxoedema coma

Patients in coma or near coma who are hypothermic, have a bradycardia, are hyponatraemic or simply ‘look’ hypothyroid should be investigated for this – TSH high and T4 very low. There is often some delay in getting results back, and so a decision about starting treatment often needs to be made in their absence. There is a paucity of information on whether the relatively high mortality rate in this condition is related to delayed therapy or to intercurrent infection and its complications – for obvious reasons there are no RCTs looking at withholding treatment in myxoedema coma.

• Give hydrocortisone 100 mg IV 12 hourly until there is clear evidence of clinical improvement, then wean down to 50 mg 12 hourly, and stop once you have a random cortisol result back with a value excluding the diagnosis of concurrent hypoadrenalism (>550 nmol/l).
• Unless there is a strong history of angina, start on 200 mcg per day of thyroxine. It is rarely possible to access intravenous preparations in less than 12 hours, by which time the oral will be well absorbed.
• Monitor glucose regularly
• Treat infections with intravenous antibiotics – in the absence of an indicator as to the site of sepsis, consider treating as a gram-negative septicaemia.
• Manage hypothermia

---

1. Abraham P, Avenell A, Watson WA, et al. Antithyroid drug regimen for treating Graves’ hyperthyroidism (Cochrane Review). In: The Cochrane Library, Issue 3, 2004. Chichester, UK: John Wiley & Sons, Ltd ↩

2. Fraser T, Green D. Weathering the storm: beta-blockade and the potential for disaster in severe hyperthyroidism. Emerg Med 2001;13:376-80 ↩

3. Vickers P, Garg KM, Arya R, et al. The role of selective beta 1-blocker in the preoperative preparation of thyrotoxicosis:  a comparative study with propranolol. Int Surg 1990;75:179-83 ↩

4. McDevitt DG, Nelson JK. Comparative trial of atenolol and propranolol in hyperthyroidism. Br J Clin Pharmacol. 1978 Sep;6(3):233-7. doi: 10.1111/j.1365-2125.1978.tb04590.x. PMID: 356867; PMCID: PMC1429452. ↩

5. Page SR, Sheard CE, Herbert M, et al.  A comparison of 20mg or 40mg per day of carbimazole in the initial treatment of hyperthyroidism. Clin Edocrinol. 1996;45:511-5 ↩

6. Struja T, Fehlberg H, Kutz A, Guebelin L, Degen C, Mueller B, Schuetz P. Can we predict relapse in Graves’ disease? Results from a systematic review and meta-analysis. Eur J Endocrinol. 2017 Jan;176(1):87-97. doi: 10.1530/EJE-16-0725. Epub 2016 Oct 25. PMID: 27780830 ↩

7. Burch HB, Wartofsky L. Life-threatening thyrotoxicosis. Thyroid storm. Endocrinol Metab Clin North Am. 1993;22:263-77 ↩

8. Canaris GJ, Steiner JF, Ridgway EC. Do traditional symptoms of hypothyroidism correlate with biochemical disease? J Gen Intern Med. 1997;12:544-50 ↩

9. Roberts CG,  Ladenson PW. Hypothyroidism. Lancet. 2004;363:893-803 ↩

10. McGee S. Evidence based physical diagnosis. WS Saunders 2001 p284 ↩

11. Indra R, Patil SS, Joshi R, et al. Accuracy of physical examination in the diagnosis of hypothyroidism: a cross-sectional double-blind study. J Postgrad Med. 2004;50:7-11 ↩

12. Walsh JP, Ward LC, Burke V, et al. Small changes in thyroxine dosage do not produce measurable changes in hypothyroid symptoms, well-being, or quality of life: results of a double-blind randomised clinical trial. J Clin Endocrinol Metab. 2006;91:2624-30 ↩

13. Thayakaran R, Adderley NJ, Sainsbury C, et al. Thyroid replacement therapy, thyroid stimulating hormone concentrations, and long term health outcomes in patients with hypothyroidism: longitudinal study. BMJ. 2019 Sep 3;366:l4892. doi: 10.1136/bmj.l4892. PMID: 31481394; PMCID: PMC6719286. ↩

Was this article helpful?

0 out Of 5 Stars

5 Stars		0%
4 Stars		0%
3 Stars		0%
2 Stars		0%
1 Stars		0%

5 Submit

Send Feedback

How can we improve this article?

Please submit the reason for your vote so that we can improve the article.

Submit