|
|
|
|||
| |
||||
|
|
Thyroid Canada Newsletter
- Issue 3
|
|
| |
|
|
JULY 2001 SHOULD WE SCREEN FOR HYPOTHYROIDISM? The advent of diagnosing thyroid disease using primary TSH screening has led to the identification of essentially asymptomatic individuals whose TSH is elevated but whose free thyroxine levels are in the normal range. Such patients have been stated to have “subclinical hypothyroidism” (SH). In 1968, Wilson and Jungner (1) described criteria to determine when screening for a disease is justified: (1) the disease is common, (2) the disease is associated with significant morbidity and mortality, (3) a suitable screening test exists, (4) there is a simple effective treatment, (5) the potential benefits justify the costs of screening. This review examines the evidence that indicates that SH fulfills all of the criteria for screening and recommends that programs be implemented by the appropriate medical bodies.
1. The disease is common The prevalence of increased serum TSH concentrations ranges from 2.9-5.7% in males and 7.5-13.6% in females (2). The most prominent of these studies, the Whickham survey, utilized a cross-sectional design employing over 2,500 subjects living in this Northeast English village (3). In this study, 8% of women (10% of women over 55 years) and 3% of men had SH. In a recent study of 25,862 participants attending a Colorado health fair, 9% of the population, excluding those on L-thyroxine therapy, had an elevated TSH level (4). Less than 1% of these subjects had overt hypothyroidism. After reviewing 12 studies across many different cultures, Vanderpump and Tunbridge (2) concluded that hypothyroidism (TSH >6 mU/L) occurs in 5% of such populations. Virtually all studies report a higher prevalence for women, those with advancing age and in the presence of thyroid antibodies (2) with rates as high as 24% among elderly women selected from senior citizens centres (5). Clearly, the prevalence of SH is substantial and justifies a screening approach.
2. There is significant morbidity and mortality SH is clearly a misnomer (6). Some authorities suggest
the term “minimal” or “mild” hypothyroidism.
Staub et al. (7) have shown significant changes in a clinical symptom
index (the Billewicz scale) in women with SH compared to age-matched
controls. The same group again noted such differences using a uniquely-designed
symptom score which correlated with free T4 and TSH levels (8).
Patients with SH have significant alterations in memory and mood including anxiety and depression which resolved with L-T4 therapy (15). Most recently, infants of mothers hypothyroid at 15 and 18 weeks gestation were noted to have an IQ seven points lower than control infants of euthyroid mothers (16). The same group has also demonstrated an increase in fetal loss in hypothyroid mothers, the majority of whom had SH (17). Whether these outcomes are related to SH itself, autoimmunity or other factor(s) is yet to be determined. Overall, it is estimated that 20-50% of patients with SH develop overt hypothyroidism within 4-8 years. If the patient is an elderly female with TPO antibodies, the likelihood is 80%! In a 20 year follow up of the Whickham cohort, the mean annual incidence of spontaneous hypothyroidism was 3.5 cases per 1,000 persons (18). Among the surviving women, the annual risk for hypothyroidism was 4% for those who had a raised TSH and positive antibodies, 3% with a raised TSH alone and 2% with positive antibodies alone. Over 20 years, the cumulative incidence of hypothyroidism was 55%, 33% and 27%, respectively. In summary, patients with SH have altered biologic parameters that mimic those seen in overt hypothyroidism compared to euthyroid controls. This may result in hyperlipidemia, atherosclerosis and a higher rate of MI in elderly females. In pregnancy, maternal SH has been associated with fetal loss and impaired cognition in the offspring. Since the vast majority of those patients affected would not be identified otherwise, a screening program is justified.
3. A suitable screening test exists The serum TSH assay is widely available, accurate, safe and relatively inexpensive and thus, an ideal screening tool. In the absence of confounding non-thyroidal illness or drugs such as corticosteroids or amiodarone, a normal TSH has a highly negative predictive value in the outpatient setting. Pituitary or hypothalamic disease is sufficiently rare to not confound the procedure and is usually identified otherwise. In unselected populations serum TSH has a sensitivity of 89-95% and specificity of 80-90% for thyroid dysfunction compared with confirmed cases (19).
4. There is a simple, effective treatment Many of the biological parameters found to be abnormal in SH improve or normalize with L-thyroxine therapy, including symptom score, systolic time intervals and lipid profile. Tanis et al. (20) in a retrospective analysis of intervention studies, demonstrated significant reductions in total and LDL cholesterol levels after L-thyroxine therapy if the initial TSH is greater than 10 mU/L. On average, the total cholesterol decreased by 6% (0.4 µmol/L) with decreases of 1.2 mmol/L and 3.4 mmol/L if the initial cholesterol level was 8 or greater than 8 mmol/L, respectively. In one study, Lp(a) returned to normal in patients with SH following treatment with L-thyroxine (9). An accelerated progression of coronary angiographic lesions has been reported in SH patients compared with patients whose TSH was maintained within the normal range (21). DiBello et al demonstrated that early alterations in myocardial function and structure using conventional echocardiography and videodensitometric analysis exist with SH (22). Recently, the same group studied the effects of L-thyroxine on cardiac function and structure in SH in a double-blind, placebo-controlled study of 20 subjects (23). SH patients had significantly higher isovolumic relaxation times, PEP/ET ratios, and lower cyclic variation indexes compared to euthyroid controls which improved or resolved with L-thyroxine treatment. In general, L-thyroxine has proven to reverse many of the biological abnormalities associated with SH.
5. The potential benefits justify the costs A detailed cost analysis is beyond the scope of this article. Besides, no high-quality randomized controlled trials of a screening program for SH exist. However, using a computer decision model, it was estimated that testing women aged 35 years of age or older every 5 years for 50 years would be beneficial (19,24). Over 50% of the benefit in quality-adjusted life years was accounted for by preventing progression to overt hypothyroidism, 30% by improving associated mild symptoms and 2% by preventing cardiovascular disease. In this model, the cost effectiveness of screening for SH was comparable to screening for hypertension, hypercholesterolemia or breast cancer.
CONCLUSION: A screening program for SH: The time has come. A screening program for SH fulfills the criteria outlined for diseases worth screening. Although most North American TSH assays have an upper normal limit of 4-5 mU/L, some European authorities believe it should be lowered to 2 mU/L (6). Clearly, the “ideal” TSH for euthyroidism is yet to be determined. The American Thyroid Association (ATA) recommends that all adults be screened for thyroid dysfunction beginning at age 35 years and every 5 years thereafter (25). Tunbridge and Vanderpump recommend that screening be limited to women over 50 years (19). I endorse the ATA view and would extend it to pregnant women. Although the effects on the offspring of hypothyroid mothers may be due to autoimmunity and no intervention studies have been performed, it seems inappropriate to delay intervention until such data is available considering the emotional and familial consequences of the demonstrated outcomes. In such a screening program for SH, the risk of false positives (thyroiditis), non-compliance, and potential over-treatment (with the consequence of atrial fibrillation and osteoporosis) have yet to be determined but will likely be minimal. Over 10 years ago, the ATA supported TSH screening in women over 50 years who had suggestive symptoms in the primary care setting (26). The last decade has provided further evidence that SH represents true hypothyroidism and a potentially serious health burden. It is time to ensure that provincial health authorities create the appropriate guidelines for a national, comprehensive screening policy for hypothyroidism in adulthood as they did for newborns in a previous generation. Jody Ginsberg, MD, FRCPC
REFERENCES:
**********
EDITORIAL COMMENT: From ROBERT VOLPÉ,
M.D., FRCP (C), MACP It is evident from the currently available studies that these criteria have been met and screening for adults for thyroid dysfunction should begin at age 35 and every five years after. There may be some slight controversy as to whether the screening should be limited to women over 50 years of age, but, in any event, the principles for screening have now been established. In Canada, one of the hindrances delaying such screening
has been the attitude of some provincial governments who remain
hesitant about this matter. Endocrinologists should encourage their
provincial departments of health so that screening for sub-clinical
hypothyroidism becomes a reality. ********* Upcoming Events:
If you would like to see an upcoming thyroid event published in the next issue of Thyroid Canada, please fax the information to Theramed Corporation at (905) 564-4776 as soon as possible. Published with a grant from Theramed Corporation.
|
|
Send mail to Webmaster
with questions or comments about this web site. |
