| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
TRACK to:
|
|
Electronic letters published:
![[Read Comment]](/icons/misc/sectionDOWN.gif){kind=icon}
REPLY
Hypothyroidism and Dyslipidemia
So What Does A Mildly Elevated TSH Mean?
|
|
|||
|
Wiliam J. Hueston, Charleston, SC Chair, Department of Family Medicine, Medical University of South Carolina
Send response to journal:
|
Dr. Garrison raises some important points in his letter that I feel need to be clarified. First, while Dr. Garrison derides our study as an "outlier," there is a key difference between our study and previous ones examining this issue. The important difference is that our study reflects a population-based sampling frame. The only previous study that comes close to being population-based is the Colorado Thyroid Study, but this study was not population-based since study participants were self-selected attendees at health fairs across the state; one must assume that people who attend health fairs are representative of the general populations (which I do not believe) and that those who migrate to a "thyroid testing" booth are a random sample of that group. Other studies that showed improvement in well-being with thyroid supplementation in patients with elevated TSH were frequently populated with patients who had previous thyroid abnormalities (most often hyperthyroid that had been treated with radioactive iodine) and patients referred to endocrinologists, presumably because of symptoms (otherwise why would they be at an endocrinologist's office?). Based on the skewed populations in previous studies looking at cholesterol and thyroid disease, I believe that the supposition that these people were asymptomatic is likely false. What these previous studies likely showed is that if a patient exhibits symptoms similar to hypothyroidism and has a high TSH, thyroid supplementation may be appropriate. But the ability to generalize these results to the truly asymptomatic individual (the vast majority of people who have subclinical hypothyroidism) is still open to question. Second, the contention that the connection between subclinical hypothyroidism and hyperlipidemia or heart disease is well established is not supported by the evidence. A recent scientific review and guidelines for diagnosis and management of subclinical thyroid disease sponsored by the Endocrine Society (1) stated that the strength of evidence between subclinical hypothyroidism and heart disease was insufficient and that there was no evidence of benefits of treatment. For cholesterol and subclinical hypothyroidism, most evidence also was judged to be insufficient. Consequently, the association between subclinical hypothyroidism and heart disease or risk factors for heart disease, such as hyperlipidemia, is far from settled. As noted in a previous letter from Dr. Scherger, the lack of evidence that subclinical hypothyroidism is directly harmful has not stopped industry- supported experts from taking what little data there is and drawing conclusions. However, I would urge us to be more cautious in our interpretation of studies performed on narrowly-drawn patient populations. William Hueston, MD References: 1. Surks MI, Ortiz E, Daniels GH, et al. Subclinical thyroid disease. JAMA 2004:291:228-238. Competing interests: None declared |
|||
|
|
|||
|
Richard L. Garrison, Baytown, TX San Jacinto Methodist Hosp FP Residency
Send response to journal:
|
Hueston and Pearson perform a substantial work in this article. Nevertheless, I ask whether the science has already progressed beyond the simple association of hypothyroidism and dyslipidemia. Elevated TSH with normal T4 (regretably in this case, total rather than free) can no longer be properly referred to as subclinical hypothyroidism, except to establish a link with a historical literature. This particular brand of hypothyroidism is associated with dyslipidemias in a number of other articles, making this article an outlier in that regard. Far more importantly, articles reporting treatment of these patients with T4 demonstrate improvement of lipid profiles in addition to the improved wellbeing cited by the authors. Even more importantly, independent of whether hypothyroidism, subclinical or otherwise, has an effect on lipids, it has an effect on cardiovascular risk, increases arterial intimal thickening that is reversed by T4 treatment, and adversely affects endothelial function in a manner reversed by T4 treatment. In short, treatment makes a difference, so this is not subclinical disease. We now have so many examples of treatments that reduce cardiovascular risk or have demonstrated direct effects on the actual lesion of atherosclerosis (T4, folate, Omega-3 FA, allicin to name several) that function independently of changes in lipids due to treatment, that it is no longer wise to evaluate potential benefits of treatment by looking only at the earliest surrogate marker, lipids. Our experience with clofibrate should be enough to teach us that improvement in that parameter alone is not enough, and the effects of statins above and beyond that accountable by lipid changes take us further down that road. I hope that further work in the area will look more at effects on the lesion of atherosclerosis and its reversal, and better still, on clinically significant outcomes, than on surrogate markers. Competing interests: None declared |
|||
|
|
|||
|
Joseph E. Scherger, San Diego, CA USA Clinical Professor, University of California, San Diego
Send response to journal:
|
Hueston and Pearson have made a valuable contribution to the controversy over mildly elevated TSH levels in euthyroid patients. Is the so-called "subclinical hypothyroidism" really subclinical? Advocates for treating mildly elevated TSH levels (myself included) have been convinced that this condition has true clinical consequences and is better termed mild hypothyroidism. I recently attended a half day continuing education course (funded by the makers of Synthroid) in which leading academic thyroidologists advocated the lowering of the upper limit of normal TSH to 3.5. They cited numerous clinical consequences of elevated TSH levels, including hyperlipidemia. Hueston and Pearson, using a community population data set, show that the differences in lipids among people with mildly elevated TSH levels are small compared to people with normal TSH levels. From their study, it appears that the supposed hyperlipidemia of elevated TSH levels has been exaggerated. This study should not be used to dispute any clinical consequences of elevated TSH levels. Many questions remain. Are the mildly increased total cholesterol and triglyceride levels significant to the health of the population? Even small differences matter when the condition is common and leads to the most common cause of death. Do patients with elevated TSH levels respond as readily to lipid lowering therapy? Would a small dose of levothyroxine eliminate the need for a lipid lowering drug, or reduce the dose of drug if needed? What about weight loss? With the epidemic of overweight and obesity, do patients with elevated TSH levels have more difficulty losing weight if untreated? What about fertility? What about mood disorders? It is tempting to think that normal T4 levels are what matters, but elevated TSH levels imply that autoimmune hypothyroidism is developing and does early treatment matter? Who should get anti-thyroid antibody levels? There are many questions to answer in this area in which the problem and potential comorbidities are common and have major health consequences. More reseach like this needs to be done to put the issue of elevated TSH levels into clinical perspective. We need less belief and emotion and more good science. It would help if the NIH or other research agencies independent of commercialism supported this work. Hueston and Pearson lead by example. Competing interests: None declared |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
TRACK: