9.9 MYXEDEMA COMA

Definition and pathogenesis. Myxedema coma is a rare, life-threatening clinical condition in patients with long-standing severe untreated hypothyroidism in whom adaptive mechanisms fail to maintain homeostasis. Most patients, however, are not comatose, and the entity rather represents a form of decompensated hypothyroidism ^1,2,10. Usually a precipitating event disrupts homeostasis which is maintained in hypothyroid patients by a number of neurovascular adaptations. These adaptations include chronic peripheral vasoconstriction, diastolic hypertension and diminished blood volume; in this way a normal body core temperature is preserved. In severely hypothyroid patients homeostasis might no longer be maintained if blood volume is reduced any further (e.g. by gastrointestinal bleeding orthe use of diuretics), if respiration already compromised by a reduced ventilatory drive is further hampered by intercurrent pulmonary infection, or if CNS regulatory mechanisms are imparied by stroke, the use of sedatives or hyponatremia ².

Diagnosis. The three key features of myxedema coma are ¹:

1. Altered mental status. The patient may be entirely obtruded or may be roused by stimuli. Usually lethargy and sleepiness have been present for many months. Sleep may have occupied 20 hours or more of the day and may have interfered even with eating. There may actually have been transient episodes of coma at home before a more complete variety developed.

2. Defective thermoregulation: hypothermia, or the absence of fever despite infectious disease. Usually coma comes on during the winter months. The severely myxedematous patient becomes essentially poikilothermic. With cold weather the body temperature may drop sharply. The temperature is subnormal, often much depressed: a temperature of 74 F (23.3 C) has been recorded. A thermometer reading lower than the usual 97 F must be used, or hypothermia may be missed.

3. Precipitating event: cold exposure, infection, drugs (diuretics, tranquillizers, sedatives, analgetics), trauma, stroke, heart failure, gastrointestinal bleeding. The pulse is slow, and the absence of mild diastolic hypertension is a warning sign of impending myxedema coma ¹. Diagnosis on clinical grounds is relatively easy once the possibility is considered. Any patient with hypothermia and obtundation should be considered as having potential myxedema coma, especially if chronic renal insufficiency and hypoglycemia can be ruled out. The diagnosis can be confirmed by finding a reduced free T4 estimate and marked elevation of serum TSH. Creatine phosphokinase is often elevated. Both hypoxia and hypercapnia may be present.

Treatment. Myxedema coma is a medical emergency. Early diagnosis, rapid administration of thyroid hormones an adequate supportive measures (Table 9-14) are essential for the prognosis. The mortality with the current treatment regimen remains, nevertheless, high in the order of 20 percent ³.

Table 9-14. Recommendations for the treatment of myxedema coma.

In view of the rarity of myxedema coma, it has been difficult to perform randomized studies to resolve the issue of whether T4 or T3 is the most appropriate treatment. The mortality rate, with the current treatment regimen, is estimated to be approximately 20 percent ^3,11. There are advocates of T4 therapy alone ⁴, T3 therapy alone ^5,6, and combinations thereof ⁷. If T4 alone is used, it should be given parenterally in doses of 300 to 500 µg to replace the calculated T4 deficit ⁸. Since the average volume of distribution of T4 in a 70-kg human is approximately 7 L, 420 µg should cause an increase of 77 nM/L in the serum T4 concentration. Following this, 75 µg/day are given intravenously. Advocates of T3 therapy point to the impairment of T4 to T3 conversion characteristic of hypothyroidism that is due to a combination of the low levels of type I deiodinase as well as the generalized illness and inadequate caloric intake. Replacement should be given at a dose of 25 µg intravenously every 12 hours. The patient can be switched to oral therapy when the circulatory system is stabilized and the patient is receiving other oral medications. Other authorities recommend a combination of T4 and T3 giving 200 to 300 µg of T4 and 25 µg T3 intravenously as an initial dose ⁷. The T3 is repeated 12 hours later and 100 µg of T4 24 hours later. This is followed by 50 µg T4 daily from the third day until the patient regains consciousness. While death associated with T3 treatment in higher doses has been reported ⁷, it is difficult to know whether this reflects a direct effect of this form of treatment or individual factors in the patients' illnesses. In addition to replacement of T4, intravenous glucocorticoid should also be administred during the first days of therapy, since in severe hypothyroidism pituitary-adrenal function is impaired, and the cortisol production rate is lower. While this low production is adequate when cortisol metabolism is reduced, as it is in hypothyroidism, the rapid restoration of a normal metabolic rate from the above treatment may precipitate transient adrenal insufficiency.

In addition, the patient should be intubated and measures taken to retain body heat. Central warming may be attempted but peripheral warming should not, since it may lead to vasodilatation and shock. The cutaneous blood flow is markedly reduced in the hypothyroid patient in order to conserve body heat. Warming blankets will defeat this mechanism. Mechanical ventilation may be needed, particularly when obesity and myxedema are combined. The thoracic and abdominal adipose tissue puts an added burden on the respiratory musculature and may lead to hypoxia, cardiac arrhythmia, and death. Hyponatremia is characteristic and free water restriction and the use of isotonic sodium chloride will usually restore normal serum sodium, as will improved cardiovascular function, which is one cause of the impaired free water clearance. Serum glucose should be monitored. Supplemental glucose may be necessary, especially if adrenal insufficiency is present. Hypotension may develop, particularly if myxedema is severe. Volume expansion is usually required to remedy this, since patients are usually maximally vasoconstricted. Dopamine should be added if fluid therapy does not restore efficient circulation.

Concomitantly, a vigorous search for precipitating factors should be instituted. Determining whether an infection is present should be a priority, since as many as 35 percent of patients with myxedema coma have infection. Since hypothyroid patients cannot mount an adequate temperature response, the usual signs of infection, including tachycardia, fever, and elevated white blood count, may be absent.

Prophylactic antibiotics are indicated until infection can be ruled out; upper respiratory infection should be eliminated. While the hypothyroid patient withstands the stress of surgery in general very well ⁹, inadvertently excessive narcotics, sedatives, and hypnotics can tip a severely hypothyroid patient into coma. Most patients begin to show increases in body temperature within the first 24 hours of treatment. The absence of an increase in body temperature within 48 hours should lead to consideration of more aggressive therapy, specifically T3 therapy if it has not already been initiated. Most patients regain consciousness within a few days.

9.10 SUBCLINICAL HYPOTHYROIDISM

Subclinical hypothyroidism is defined as an increased serum TSH in the presence of a normal serum FT4 concentration. Increased and normal refer to values above or within population-based reference ranges of these hormones. It is assumed that the increased TSH concentration is not caused by analytical interference in the TSH assay (e.g. by heterophilic TSH antibodies) or by non-thyroidal illness. If a slightly increased serum TSH is found, a second blood sample taken from the same individual may occasionally contain a normal TSH concentration; this can be due to inter-assay variation in the TSH assay (which is about 5-10%) or to ultradian and circadian TSH rhythms (but the relative risk of misjudging mean TSH serum levels by a single TSH determination between 07.00 and 17.00 hours is only 0.09% for values above 4.0 mU/l ¹).

Subclinical hypothyroidism may have endogenous causes (chronic autoimmune thyroiditis, subacute thyroiditis, postpartum thyroiditis) or exogenous causes (thyroidectomy, 131I therapy, antithyroid drugs, inadequate thyroid hormone replacement therapy). Prevalence and natural history. The prevalence of subclinical hypothyroidism is rather high. In the classical population-based study among adults in the English county of Whickham the prevalence was 75 per 1000 women and 28 per 1000 men ²; similar figures have been obtained in other studies (see §9.2). The higher prevalence of subclinical hypothyroidism in females than in males and in older than in younger subjects is in agreement with the higher prevalence of thyroglobulin and thyroid peroxidase (microsomal) antibodies in women and in elderly people. The natural history of subclinical hypothyroidism is reasonably well known. Spontaneous return of increased TSH values into the normal range occurs in 5.5% after 1 year ³.

Progression to overt hypothyroidism ranges from 7.8% to 17.8% in various studies ^3,4,5. Another report indicates that approximately 30% of patients with subclinical hypothyroidism had developed overt hypothyroidism after 10 years; the higher the initial TSH, the greater the risk ⁶. The 10-year probability to develop overt hypothyroidism at an initial TSH value of 12 mU/l was ~27% in antibody-negative patients but ~57% in antibody-positive patients. The importance of thyroid antibodies is also evident from a Dutch study ⁷: 39.6% of 55-year old women with thyroid microsomal antibodies had raised TSH levels 10 years later, in contrast to 3.2% of women without antibodies ⁷. derived A Swiss follow-up study among subclinically hypothyroid women reports a 10-year risk for progression to overt hypothyroidism of 0% in women with baseline serum TSH >4-6 mU/l, of 43% (3 percent per year) at baseline TSH levels of >6-12 mU/l, and of 77% (11 percent per year) at baseline TSH >12 mU/l; the cumulative incidence of overt hypothyroidism was 23% among women without TPO-antibodies and 58% among women with TPO-antibodies⁵⁶. The most extensive data are from a 20-year follow-up in the participants of the Whickham survey ⁸: the incidence of overt hypothyroidism was 4.1 per 1000 women per year and 0.6 per 1000 men per year. Risk factors were the initial presence of either a raised TSH or thyroid antibodies (see also § 9.2). The annual incidence of hypothyroidism in women was 5% ^8,9. Systemic manifestations. A number of abnormalities listed in Table 9-15, has been reported in some but not all subjects with hypothyroidism. The described abnormalities are in general minor, and more frequent in subjects with the highest TSH values ¹⁰. Hypothyroid symptoms (especially dry skin, cold intolerance and easy fatigability) occur more often than in controls ^10,11, as also evident from a clinical symptom score for hypothyroidism ¹² (see § 9.7.1). An association with depression is reported. A study among healthy females with a family history of thyroid disease, recruited by advertisement, indicated a higher lifetime frequency of depression in subjects with subclinical hypothyroidism (56%) than in euthyroid subjects (20%) ¹⁴. Patients with major depression also have a poorer response to antidepressive drugs if they are subclinically hypothyroid ¹⁵.

Table 9-15. Abnormalities reported in some but not all subjects with subclinical hypothyroidism.

A slight increase in serum TSH is associated with a fall in resting energy expenditure ¹⁶ and lengthening of the Achilles tendon reflex time ^10,17. Muscle energy metabolism is impaired: during exercise, blood lactate is significantly higher in subjects with subclinical hypothyroidism than in controls (18). Systolic time intervals like the Qkd time (the interval from the Q wave of the electrocardiogram to the pulse wave arrival time at the brachial artery) and the PEP/LVET ratio (pre-ejection period divided by the left ventricular ejection time) were not different from controls in one study (19), but the PEP/LVET ratio decreased upon thyroxine replacement in two placebo-controlled clinical trials (although only in subjects with the highest pre-treatment PEP/LVET ratio's in one of these studies) ^11,20. Left ventricular ejection function at rest or with moderate exercise does not change upon thyroxine therapy in subclinical hypothyroidism; it is improved by T4 treatment only at maximal exercise ^21,22. The biologic significance of these subtle changes in myocardial contractility appears to be very small. Studies in the early seventies suggested preclinical hypothyroidism as a risk factor for coronary heart disease, presumably via increased cholesterol levels ^23,24. Much interest was thus aroused to evaluate whether or not subclinical hypothyroidism is associated with hypercholesterolemia. The bulk of the evidence indicates that serum total cholesterol, LDL-cholesterol and HDL-cholesterol in subclinical hypothyroidism are not different from those in age- and sex-matched controls ^10,25-29; only a few studies report higher LDL- and lower HDL-cholesterol values ^30,31. The 20-year follow-up study in participants of the Whickham Survey, however, lend no support to the view that thyroid antibodies or subclinical hypothyroidism are risk factors for the development of ischemic heart disease ³². In contrast, a population-based cross-sectional study of elderly women (mean age 69±7.5 years) living in The Netherlands, reports that subclinical hypothyroidism is associated with a greater age-adjusted prevalence of aortic atherosclerosis (odds ratio 1.7, 95% CI 1.1 to 2.6) and myocardial infarction (odds ratio 2.3, 95% CI 1.3 to 4.0)⁴¹. Additional adjustment for body mass index, serum cholesterol, blood pressure, smoking and the use of $-blockers did not affect these estimates. The odds ratio?s were slightly higher in women with subclinical hypothyroidism and antibodies to thyroid peroxidase, but thyroid autoimmunity itself was not related to cardiovascular disease. The population attributable risk for subclinical hypothyroidism associated with myocardial infarction was 14%, within the range of that for known major risk factors for cardiovascular disease (hypercholesterolemia 18%, smoking 15%, hypertension 14%, diabetes 14%). In this respect it is noteworthy that the increased cardiovascular risk associated with subclinical hypothyroidism seems to extend itself into the normal range of thyroid function ⁴². For, in euthyroid healthy volunteers the association of TSH with LDL cholesterol is modified by insulin sensitivity, being absent in insulin-sensitive and strongly positive in insulin-resistant subjects ⁴².

Treatment. Several double-blind placebo-controlled studies have been performed on the usefulness of thyroxine replacement in subclinical hypothyroidism; some report a greater benefit of thyroxine than of placebo, but others do not find any improvement of symptoms by T4 treatment. Hypothyroid symptoms improved in 47% of T4-treated subjects and in 19% of placebo-treated subjects ¹¹. During T4 treatment 24% of the subjects improved as judged by psychometric test and their own rating ²⁰. A small improvement in a composite memory score was observed in actively treated versus placebo-treated subjects ³⁶. In another study of patients with subclinical hypothyroidism free from neuropsychological complaints, thyroxine treatment improved memory skils and decreased somatic complaints and obsessionality rating ¹³. Two clinical scores assessing symptoms and signs of hypothyroidism (Billewicz and Zulewski scores) improved significantly in T4-treated but not in placebo-treated patients⁵¹. In contrast, still another study did not find significant differences in the changes from baseline to 6 months between women in the thyroxine group and the placebo group for body mass index, resting energy expenditure, LDL-cholesterol, or general health questionnaires⁵².With respect to serum lipids, a meta-analysis concludes that normalization of serum TSH in subclinical hypothyroidism decreases serum cholesterol on average by 0.4 mmol/l (95% CI 0.2-0.6 mmol/l) ³⁷. A more recent meta-analysis also concludes that normalization of serum TSH decreases serum LDL-cholesterol by 0.26 mmol/l (95% CI 0.12-0.41 mmol/l)⁴³. The reduction in serum total and LDL cholesterol may be larger in individuals with higher pretreatment cholesterol levels ^43,51. The observed decrease in LDL-cholesterol is estimated to decrease the risk of cardiovascular mortality by 9-31%⁵¹. Lipoprotein (a) levels do not change upon normalization of TSH⁵³. Levothyroxine treatment of subclinical hypothyroidism results in 6% reduction in supine mean arterial pressure, 14% increase in upright cardiac output, and 13%-20% decrease in systemic vascular resistance⁵⁴.In a study among women with subclinical to mild hypothyroidism, progression of arterial disease after 1 year had occurred in 20% of the T4-treated patients and in 60% of the untreated patients ³⁸. The impaired left ventricular function at rest and systolic dysfunction on effort are reversed by restoration of euthyroidism⁴⁴, as also evident from a double-blind placebo-controlled study⁵⁵.

Taken together, the findings (especially the high rate of progression towards overt hypothyroidism) argue against a "wait-and-see" policy and favor early thyroxine treatment ³⁹. Improvement of hypothyroid symptoms, mood and cognitive functions may be expected in 25-30% of patients. When in doubt because of nonspecific complaints, a trial of thyroxine treatment for 3-6 months can be considered. However, whether or not subclinical hypothyroidism should be treated is still hotly delated; there are strong defenders as well as strong opponents to levothyroxine treatment^45-48. A recent scientific review by a panel of experts concluded that data supporting associations of subclinical thyroid disease with symptoms or adverse clinical outcomes or benefits of treatment are few, and that the consequences of subclinical thyroid disease are minimal⁴⁹; consequently, the panel recommended against routine treatment of subclinical hypothyroidism, albeit recognizing the possible need for treatment in selected individual cases. Given the current state of affairs with a lack of controlled trials reporting on long-term outcome, an algorithm as proposed by Cooper⁵⁰ might be useful. A slightly modified algorithm is outlined in figure 9, providing guidelines on the appropriate management of the individual patient. The topic of subclinical hypothyroidism has been extensively reviewed ⁴⁰. A summary of key data is listed in Table 9-16.

Table 9-16. Summary of data on subclinical hypothyroidism.

9.11 SCREENING FOR HYPOTHYROIDISM

In view of the rather high prevalence of thyroid function disorders and the availability of a suitable screening test in the form of the sensitive TSH assay, the question arises if screening programs are warranted in the general adult population ¹. Case-finding strategies have been employed successfully: previously unknown hypothyroidism was found in 0.64% of middle-age women in connection with screening for cervical carcinoma ², and in 0.3% of women attending a primary care unit ³; the prevalence of subclinical hypothyroidism in the latter study was 1.2%.Case-finding in women over 40 years of age can be useful. Patients admitted to geriatric units also benefit from routine testing as 2% to 5% have treatable thyroid disease, but patients hospitalized with acute illness do not benefit from routine thyroid function tests due to frequent interference of test results by the sick euthyroid syndrome ⁴. The cost-effectiveness of periodic screening for mild thyroid failure has been investigated using a state-transition computer decision model that account for case-finding, medical consequences of mild thyroid failure, and costs of care during 40 years of simulated follow-up ⁵. The cost-effectiveness of screening 35-year old patients with a serum TSH assay every 5 years was $ 9223 per QALY (quality-adjusted life year) for women and $22595 for men. The cost-effectiveness compares favorably with other generally accepted prevention programs. The authors recommend screening in the general community for mild hypothyroidism with serum TSH (combined with serum cholesterol) every five years at the age of 35 years ^5,6. A recent update on screening for thyroid disease in the general adult population, however, argues that the evidence of the efficacy of treatment for subclinical thyroid dysfunction is inconclusive and that large randomized trials are needed to determine the likelihood that treatment will improve the quality of life in otherwise healthy subjects who have mildly elevated TSH levels ⁷. On the other hand, the update favors office-based screening to detect overt thyroid dysfunction in women older than 50 years of age: in this group, 1 in 71 women screened would benefit from relief of symptoms. Taken together, the presently available data do not justify yet screening of the healthy adult population for hypothyroidism. Case-finding, i.e. testing on patients visiting their physician for unrelated reasons, seems currently the best approach to detect previously unsuspected hypothyroidism; it is especially worthwhile in women over 40 years of age. Table 9-17 provides a useful list of indications for screening' for hypothyroidism ⁸.

Table 9-17. Indications for screening for hypothyroidism. (Reproduced with permission) ⁸

Return to Top