Hyperthyroidism and thyrotoxicosis

Thyrotoxicosis turns up the the CICM past papers about as often as does myxoedema coma, which is probably still more often than it turns up in clinical practice. So far, three notable SAQs have brought up this topic. 

Question 25 from the second paper of 2013 asks for some drugs which are used in the management of thyrotoxic crisis. Also, Question 15.1 from the first paper of 2017 presents a case of a patient whose TSH-secreting adenoma was massaged a little too much by the surgeon during the trans-sphenoidal resection, resulting in a post-operative thyroid storm in the recovery room. Lastly, a small 30% component of Question 13 from the second paper of 2018 asked for a list of clinical features of thyroid storm specifically. 

A reasonable reference to base one's study notes upon would be Chapter  60 of Oh's Manual,  Thyroid  emergencies by Jonathan  M  Handy  and   Alexander  M  Man  Ying  Li (pp. 652). The summary below is based largely on their work. Emphasis was given to the content of the Blue Boxes, as these tend to be mined by the examiners. They are a rich source of lists, and lists are easily made into SAQs ("name six causes of this or that, etc"). Another excellent resource for thyrotoxicosis  is Caroll & Matfin (2010); if for whatever reason the exam candidate feels the need to read beyond Oh's, they could safely stop after this article. If one is for some reason unable to safely stop reading, one could carry on to digest the expansive and definitive 2016 American Thyroid Association guidelines for diagnosis and management of hyperthyroidism and other causes of thyrotoxicosis, which carries on for ninety pages.

In brief:

Definition, diagnosis and clinical features of thyroid storm

Definition of thyroid storm

Like the electrical storm of ventricular arrhythmias, this has no widely accepted definition. Among review literature, it generally seems to be regarded as not a well-defined clinical entity of its own right, but rather part of the spectrum of hyperthyroidism. Hyperthyroidism becomes thyroid storm when its manifestations become life-threatening, which apparently happens in 1-2% of patients with clinically overt hyperthyroidism (Karger & Führer, 2008). UpToDate describes it as "an exaggeration of the usual symptoms of hyperthyroidism", but avoid giving any real defining features. Chris Nickson from LITFL defined it as a "life threatening exacerbation of hyperthyroid state with 1 or more organ dysfunction" which is as good a definition as any. Back in the 1990s,  Birch & Wartofsky (1993) had gone as far as to actually develop a scoring system to define this condition (the table from their article is reproduced somewhere below). In their system, a score in excess of 45 defined thyroid storm. However it is unclear how this definition should be used, considering that treatment for thyrotoxicosis would be offered even to patients who get a low score, and the aggression of therapy would escalate directly in proportion to the severity of the disease. 

Precipitants of thyroid storm:

We can devise a table to classify these precipitants:

Systemic disease Sepsis Surgery Trauma Child birth Eclampsia Diabetic ketoacidosis Hypoglycaemia Emotional stress Burns Pulmonary embolism Seizures Stroke

Thyroid-specific causes Cessation of antithyroid drugs "Excessive palpation" of the thyroid gland Radioactive iodine therapy Iodinated contrast Thyroxine overdose

Clinical features of thyroid storm

1. Goitre: possible airway compromise)
2. Tachypnoea due to increased CO2 production;
   Increased O₂ER (increased metabolic fuel use)
3. Tachycardia,
   Atrial fibrillation and ventricular arrhythmias
   Heart failure
   Hypertension (early), hypotension (late)
4. Tremor;
   Agitation, progressing to encephalopathy, coma and seizures.
   There is the phenomenon of "apathetic thyrotoxicosis" which presents with weakness
5. Low potassium and magnesium (particularly in "apathetic thyrotoxicosis")
   Serum cortisol should be elevated. If it is not, one might consider a relative adrenal insufficiency, and supplement some hydrocortisone.
6. Rhabdomyolysis may be present; CK may be elevated. This is "thyrotoxic myopathy"
7. Diarrhoea, nausea and vomiting
   Increased metabolic rate; increased demand for metabolic substrate.
   Nutritional requirements are increased
   Hyperglycaemia may be apparent in the non-diabetic patient
   Jaundice may develop
8. Leukocytosis; a left shift
9. Fever: in fact, may go up to 41°C. This is apparently the most characteristic feature.

Diagnostic criteria for thyroid storm

An attempt has been made to produce some diagnostic criteria (Burch and Wartofsky, 1993); the authors proposed a scoring system whereby one attributes points to different clinical features.

Thermoregulatory dysfunction Temperature (°F | °C) 99 to 99.9 | 37.2 to 37.7 5 100 to 100.9 | 37.8 to 38.2 10 101 to 101.9 | 38.3 to 38.8 15 102 to 102.9 | 38.9 to 39.4 20 103 to 103.9 | 39.4 to 39.9 25 ≥104.0 | >40.0 30 Central nervous system effects Mild 10 Agitation Moderate 20 Delirium Psychosis Extreme lethargy Severe 30 Seizure Coma Gastrointestinal-hepatic dysfunction Moderate 10 Diarrhea Nausea/vomiting Abdominal pain Severe 20 Unexplained jaundice

Cardiovascular dysfunction Tachycardia 99 to 109 5 110 to 119 10 120 to 129 15 130 to 139 20 ≥140 25 Atrial fibrillation 10 Heart failure Mild 5 Pedal edema Moderate 10 Bibasilar rales Severe 15 Pulmonary edema Precipitant history Negative 0 Positive 10

If one scores over 45, thyroid storm is very likely, and a score of 25-44 suggests that it is imminent.

Diagnostic investigations and their characteristic findings

• Bloods:
  • Raised T3 and T4
  • Low TSH
  • Deranged LFTs
• Ultrasound of the thyroid:
  • Hypervascularity (eg. "thyroid inferno" of Graves disease)
  • Solid masses (eg. nodular goitre)

Management of thyroid storm

Prevention of thyroid hormone synthesis

Thiouracils: propylthiouracil blocks synthesis of T3 and T4 as well as peripheral T4-T3 conversion. In the thyroid gland, propythiouracil blocks the organification of iodide by inhibiting thyroperoxidase, the enzyme responsible for oxidising the iodide anion. Thus, iodide cannot be added to tyrosine residues of thyroglobulin, and thyroid hormone synthesis dies off. It is therefore important to give the patient some propylthiouracil before giving them iodide.

Imidazoles: methimazole or carbimazole (which is metabolised into methimazole) also block synthesis of T3 and T4 by interfering with the action of thyroperoxidase. it has a longer half-life than propylthiouracil.

Imidazoles should probably be the first-line choice in patients who are not critically ill.  Nakamura et al (2013) compared the efficacy of methimazol and propylthiouracil in a prospective randomised study of non-storming patients with Graves disease. The authors concluded that even lowish-dose methimazole (15mg/d) is very effective, and that propylthiouracil cannot be recommended as an initial treatment because of poor efficacy and severe side effects.

On the other hand, propylthiouracil is also able to block peripheral T4-T3 conversion. UpToDate (American) authors recommend propylthiouracil over methimazole in thyroid storm because of this factor. There is also some evidence of more rapid action - T3 levels decrease more quickly with propylthiouracil (Cooper et al, 1982). Therefore, propylthiouracil should be the first choice in the ICU setting, particularly in the context of thyroid storm (Lechner & Angell, 2020)

Given that the two drugs compete for the same enzyme, there may be no synergistic effect in giving both at the same time.

Prevention of thyroid hormone release

Inorganic iodine therapy, eg. potassium iodide (given after synthesis is blocked with propylthiouracil or methimazole) is effective because iodine blocks the release of T4 and T3 from the gland within hours. Also, large doses of iodine can inhibit the organification of iodine, which is called the Wolff-Chaikoff effect (Abrahams, 2005). However, the iodide transport system soon adapts to higher concentrations of iodine, and there may be an "escape" effect where thyroid hormone synthesis starts again. This is why iodide therapy is best combined with a syntheis blocking agent like methimazole.

Lithium has also been used to block the release of thyroid hormone (Bogazzi et al, 2002). Shek et al (2006) also demonstrated satisfactory control of thyrotoxicosis in non-storming patients who had contraindications to propylthiouracil and carbimazole. The authors both complained that the response was slow. The savvy candidate should be aware of this agent for the exams, but its use for thyroid storm is hampered by its  pointlessness and by the existence of less toxic agents. One can potentially envision a situation where a patient who is for some reason intolerant of potassium iodide might benefit from lithium instead.

Blockade of peripheral T4⇒T3  conversion

Many drugs block peripheral T4-T3 conversion.

• Propylthiouracil
• Propanolol
• Corticosteroids
• Iodinated contrast

Of these, iodinated contrast agents are probably the most potent. Roti et al (1988) had demonstrated a significant and rapid decrease in circulating T3 levels with sodium ipodate, which normally goes by the name Oragrafin.  Apparently, yuou can also use ipanoic acid (Telepaque). Standard Gastrografin (sodium amidotrizoate) will not cut it.

Of course, this arm of the therapy is not a very effective pathway, as it only works on the small pool of potentially harmful T4. Treatments which prevent the synthesis of T4 and block its release are probably more effective in the long run, and in the short term more good will be done by focusing on blocking the peripheral effects of thyroid hormone.

Blockade of the peripheral effects of thyroid hormone

β-blockade with propanolol (which also decreases T4-T3 conversion) is the standard of acute care. Most of the immediately life-threatening consequences of thyoroid storm are cardiovascular. Propanolol is effective in controlling heart rate; with a slower rate the cardiac failure may actually improve and the blood pressure may paradoxically increase. The alternative agent is the easily titratable esmolol, but you miss out on the peripheral T4-T3 conversion blockade. 

Management of coexisting adrenal insufficiency

Corticosteroids are routinely used in thyroid storm to address the coexisting hypoadrenal state. Their peripheral effects on T4-T3 conversion are trivial and probably have little influence on their effectiveness in thyrotoxicosis. Thyroid disease (particularly long-standing hyperthyroidism) is associated with a diminished adrenal reserve (Tsatsoulis et al, 2000) and the addition of a small amount of hydrocortisone seems to correct this. An additional benefit is the treatment of whatever autoimmune disease is responsible (Graves, Hashimoto, etc)

Blockade of enterohepatic recirculation of thyroxine

Cholestyramine is a bile acid sequestrant which prevents the reabsorption of thyroid hormone which is excreted with the bile. The normal metabolism of T3 involves conjugation with glucuronide and sulfate, and the excretion of this product  in the bile. Free thyroid hormones are released in the intestine and are then reabsorbed. This recirculation makes the gut a potentially massive reservoir of thyroid hormones. The addition of cholestyramine to other medical management strategies tends to decrease the levels of circulating T3 and T4 (Solomon et al, 1993). The key is not to administer it together with other medications (it may bind them and reduce their efficacy).

Extracorporeal clearance of circulating thyroid hormone

Plasma exchange is an effective treatment for this problem, and is one of the effective last resort measures in severe thyroid storm with multiorgan system failure. Experience with it is based on case series only, albeit encouraging ones.  Binimelis et al (1987) reported that the rate of T4 clearance was about thirty times higher than with standard medical management (that was a case series of six patients with massive thyroxine overdose).

Haemoperfusion is even better than plasma exchange. Binimelis et al (1987) reported that with charcoal haemoperfusion the clearance rate for T3 was even higher than with plasmapheresis. This technique has been known about since the ancient times (Herrmann et al, 1977), but more recently there has ben a resurgence in interest with the development of safer filters. Kreisner et al (2010) report a more recent case. Three 2-hour sessions were required.

Thyroidectomy

This seems drastic, but it may be the only choice in patients who have failed medical therapy. Weber et al (1999) presents such a case series, where all medically refractory patients underwent subtotal thyroidectomy and did well therafter.