Issue: August 2012
August 08, 2012
11 min read

New data shed light on subclinical thyroid disease, CVD link

Issue: August 2012
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

The association between thyroid disease and cardiovascular events has been established in the literature. However, recent studies published in the Archives of Internal Medicine have highlighted the link between subclinical thyroid disease and cardiovascular events as well as the controversy over diagnostic testing and potential treatments. Endocrine Today spoke with leading experts in the fields of endocrinology and cardiology about the clinical implications and future research initiatives surrounding this observation.

“Subclinical or mild hypothyroidism affects roughly 5% to 10% of the general population and tends to be more prevalent in women and as one gets older. It has been controversial for a few decades whether this needs treatment, and frequently it is not treated by a number of people, as the evidence for any benefits is not available,” Salman Razvi, MD, FRCP, of Gateshead Health National Health Service Foundation Trust and Newcastle University in the United Kingdom, and a lead author on the study, told Endocrine Today.

Salman Razvi, MD, FRCP, and colleagues published a paper on the treatment of subclinical hypothyroidism this year that has created a buzz among endocrinologists and cardiologists alike.

Photo by  Mr. Gary Turner

The population prevalence of subclinical hyperthyroidism, however, is dependent on age, sex and iodine intake, and rates have been shown to vary among clinical studies, according to a seminal paper by David S. Cooper, MD, of the division of endocrinology and metabolism, and professor of medicine at The Johns Hopkins University School of Medicine, and Bernadette Biondi, MD, of the department of clinical and molecular endocrinology and oncology at the University of Naples Federico II in Naples, Italy.

“For most patients who have more mildly low or high serum [thyroid-stimulating hormone] concentrations in between the extremes (ie, 0.1-0.5 mU/L and 5-10 mU/L, respectively), no firm recommendations can be made, and the decision to treat or not to treat a patient will be based on various clinical factors,” Cooper and Biondi wrote.

Thyroid dysfunction, CVD link

Data from a prospective cohort study published in the Archives of Internal Medicine in May demonstrated that the association between the risk for coronary heart disease, incident atrial fibrillation (AF) and mortality, and subclinical hyperthyroidism is especially evident when thyrotropin levels are lower than 0.10 mIU/L.

David S. Cooper

The study of 52,674 patients (median age, 59 years; 58.5% women) selected from 10 cohorts was conducted by Tinh-Hai Collet, MD, of the department of ambulatory care and community medicine at the University of Lausanne in Switzerland, and colleagues. There was a median follow-up of 8.8 years.

Of those patients, 50,486 were euthyroid and 2,188 (4.2%) had endogenous subclinical hyperthyroidism.

CHD events were analyzed in 22,437 patients from six cohorts (3.2% with subclinical hyperthyroidism), researchers wrote, and incident AF was analyzed in 8,711 patients from five cohorts (9.3% with subclinical hyperthyroidism). The researchers reported that 8,527 patients died during follow-up, including 1,896 patients who died of CHD.

During age and sex-adjusted analyses, subclinical hyperthyroidism was related to increased total mortality (95% CI, 1.06-1.46), CHD mortality (95% CI, 1.02-1.62), CHD events (95% CI, 0.99-1.46) and AF (95% CI, 1.16-2.43; see chart).

In a separate study presented at the Joint 15th International Congress of Endocrinology and 14th European Congress of Endocrinology meeting, Christian Selmer, MD, a research fellow at Gentofte University Hospital in Copenhagen, Denmark, and colleagues found a physiological relationship between all levels of thyroid dysfunction and the risk for AF. This was substantiated with a higher risk in patients with overactive thyroid glands, even when the condition was very mild or at the high end of “normal,” and a lower risk in patients with underactive thyroid glands.

Kenneth D. Burman

Selmer and colleagues identified individual-level linkage among 525,100 patients (mean age, 51.7 years; 39.5% men) who consulted their general practitioner from 2000 to 2009 using nationwide registries. Of these, 504,113 (96%) patients were euthyroid, 1,474 (0.3%) had clinical hypothyroidism, 10,679 (2%) had SCH, 3,421 (0.7%) had clinical hyperthyroidism and 5,414 (1%) had subclinical hyperthyroidism.

The study concluded that patients with TSH levels of 0.1 mU/L and between 0.1 mU/L and 0.2 mU/L had a 1.8 (80%) and 1.5 (50%) increased relative risk for AF, respectively. Patients who exhibited a higher range of normal TSH levels of 0.2 mU/L to 0.4 mU/L had a 1.3 (30%) increased risk.

The researchers found clinical and SCH related to a lower risk for AF, whereas subclinical hyperthyroidism and “high-normal” thyroid function levels displayed a significant risk factor for AF.

Another study led by Fen-Yu Tseng, MD, PhD, of the department of internal medicine at National Taiwan University College of Medicine in Taipei, Taiwan, evaluated the relationship between SCH and all-cause and CVD mortality.

The researchers used a baseline cohort of 115,746 Taiwan patients aged at least 20 years without a history of thyroid disease.

“Our data revealed that SCH was associated with increased risk for all-cause and CVD mortality, especially in older subjects,” the researchers wrote.

During the 10-year follow-up period, there were 3,669 deaths (680 due to CVD), according to data. Additional data confirmed that compared with patients with euthyroidism, after adjustments, the RR for death from all-cause mortality was 1.3 (95% CI, 1.02-1.66), and CVD mortality was 1.68 (95% CI, 1.02- 2.76).

“We have found that old age and female sex increase the prevalence of SCH. Patients with SCH had higher BMIs and increased frequency of hyperlipidemia, diabetes, and hypertension compared with euthyroid subjects. Furthermore, SCH is independently associated with an increased risk for all-cause and CVD mortality after adjusting for the aforementioned confounders,” the researchers concluded.

Some data now suggest that levothyroxine could be the solution. Retrospective information gathered by researchers in the United Kingdom provides a glimpse into what may be the next treatment option for those with subclinical thyroid disease.

The role of levothyroxine

In the absence of any randomized control trial data suggesting or showing any benefit of treating subclinical thyroid disease with levothyroxine, Razvi said he and colleagues thought it would be best to tackle the issue from a “real-life” approach.

To do so, Razvi and colleagues obtained data for 3,093 patients aged 40 to 70 years and 1,642 patients aged older than 70 years using the United Kingdom General Practitioner Research Database, a large primary care database of more than 10 million patients. Patient data were recorded in 2001, and outcomes were examined until March 2009. All patients had new SCH (serum thyrotropin levels of 5.01-10 mIU/L).

Separate analyses were performed for younger and older patients. HRs for fatal and nonfatal events were calculated following adjustments for conventional ischemic heart disease risk factors, baseline serum thyrotropin levels and initiation of levothyroxine treatment as a time-dependent covariate.

Levothyroxine was used to treat 52.8% of younger patients and 49.9% of older patients during a median follow-up period of 7.6 years.

“My hypothesis, even before I did the analysis, was that, based on previous meta-analyses, age may be a significant factor that affects [CVD] and this condition. Therefore, we analyzed these two groups separately,” Razvi said.

Sixty-eight incident ischemic heart disease events occurred among 1,634 younger patients treated with levothyroxine vs. 97 events in 1,459 untreated patients (HR=0.61; 95% CI, 0.39-0.95). In older patients, however, 104 events occurred among 819 treated patients vs. 88 events in 823 untreated patients (HR=0.99; 95% CI, 0.59-1.33).

Richard Stein

“At baseline, there wasn’t any difference between who was treated vs. who wasn’t, with reference to the other CV risk factors,” Razvi said. “But when we looked for a number of surrogate risk markers, things like contact with a health professional and number of different CV medications, they were similar in both groups.”

According to Richard Stein, MD, spokesman for the American Heart Association and professor of medicine and director of the urban community cardiology program at New York University School of Medicine, this is one of the more interesting papers he has seen this year.

“It’s an impressive-looking trend in a relatively small number of cardiac patients, but a large group of patients in a study, and it raises an interesting question: Should we treat people with [SCH] to reduce their risk for a cardiac event?”

Clinical implications of treatment

Although it is a topic of debate among endocrinologists and cardiologists alike, Stein said the study does not form the basis for changing treatment.

“In real life, patients who are treated for this condition seem to do well, and it is entirely safe to be treating. But, it’s probably only worthwhile in the younger age groups, and it may not have an impact in older individuals,” Razvi said.

However, due to the retrospective nature of the study, Razvi said it is slightly more difficult to conclude that it could be translated into clinical practice.

Peter A. Singer, MD, professor of clinical medicine and chief of the clinical endocrinology department at the Keck School of Medicine of the University of Southern California, told Endocrine Today that doctors should rely on clinical judgment, as well as these findings.

Peter A. Singer

“There are certainly no contraindications to treating, and people treat all the time; but I don’t think this single article alone is enough to say you should go ahead and treat, but it is one new piece of the puzzle,” Singer said.

According to Kenneth D. Burman, MD, chief of endocrinology at Washington Hospital Center and professor in the department of medicine at Georgetown University, Washington, D.C., further long-term, prospective studies are needed to determine the relationship between SCH, hyperthyroidism (both treated and untreated) and CV events.

“At present, levothyroxine therapy should be used mainly with consideration of thyroid function tests, TSH and the clinical context. I would tend to agree with the guidelines that suggest levothyroxine treatment should be considered in patients with TSH values above the upper normal range and less than 10 mU/L, and should be administered to patients with TSH values greater than 10 mU/L,” Burman said.

The plan for treating SCH (serum TSH concentrations between 5 mU/L and 9 mU/L) was published in Cooper and Biondi’s seminar. They wrote that SCH may be associated with greater CV risk in young and middle-aged patients compared with those aged older than 65 years, and that levothyroxine may be justified as a form of treatment.

“If levothyroxine replacement has a beneficial effect, treatment should be continued and serum TSH concentrations should be assessed every 6 to 12 months to ensure that they remain within the normal range. Patients can progress to overt hypothyroidism; therefore, increases in levothyroxine might be needed during follow-up,” they wrote.

However, in the absence of any “clear-cut benefits,” Cooper and Biondi wrote that levothyroxine therapy should be stopped and serum TSH concentrations should be screened at annual visits. Additionally, the researchers wrote that the treatment of SCH is not recommended in elderly patients (aged older than 75 years) due to a lack of quality-of-life and symptomatic evidence.

Age-associated TSH levels

Anne R. Cappola, MD, ScM, associate professor of medicine at Penn Medicine and physician at Perelman Center for Advanced Medicine in Philadelphia told Endocrine Today, older patients may not benefit from this treatment because of an “age-associated shift in the distribution of TSH levels toward higher levels with increasing age.”

Anne R. Cappola

Studies of older patients suggest no increase in CV risk from leaving those with mild elevations untreated, she said.

“Not surprisingly, the older population (>70 years) had twice as many nonfatal and fatal CV events as the younger population (aged 40 to 70 years). In the younger population, the absolute difference in incidence of these events was 2.2% over a 7.6-year period. This is less of an effect on CV risk than statins, but still noteworthy,” Cappola said of the study by Razvi and colleagues.

“One of the reasons that treating an older person with [SCH] might not help them is because there is nothing wrong with them in the first place,” Cooper said. “The high TSH that you’re diagnosing as being [SCH] is just the normal aging process and just the normal level of TSH elevating, and is not necessarily indicative of an underlying thyroid problem.”

Cooper said the Razvi and colleagues paper adds some evidence to the idea that even if patients feel healthy, treating them is reasonable and might even benefit them in terms of CV outcomes.

Future directions

“It is probably time we did a proper randomized controlled trial of the right patient age group (younger individuals) to look at CV disease and outcomes,” Razvi said, adding that the problem has mainly been funding such a trial, with little to no commercial interest in levothyroxine. However, such a trial is about to develop in Europe, he said.

Currently, Nicolas Rodondi, MD, MAS, of the department of general internal medicine, Inselspital University of Bern in Switzerland, is collaborating with University of Glasgow (linking with Greater Glasgow Health Board) in Scotland; Leiden University Medical Centre in the Netherlands; University College Cork in Ireland, and thyroid experts from the University of California for the Thyroid Hormone Replacement for Subclinical Hypothyroidism Trial (TRUST).

According to the TRUST website, the objective is to determine whether there are benefits and/or drawbacks to administering levothyroxine to older patients with SCH.

During a 5-year period, the TRUST researchers plan to follow 3,000 patients aged older than 65 years to understand how to treat SCH. Half of the patients will be treated with levothyroxine, and the other half will be given placebo. Both of the groups will be monitored and evaluated on their response to the treatments. The European Union is funding the research. – by Samantha Costa

  • Dr. Burman is deputy editor of the Journal of Clinical Endocrinology and Metabolism, editorial board member for Thyroid, consultant for Medscape and UpToDate, with clinical research protocols and support to his institution from Pfizer, Amgen, Genzyme and Eisai. Dr. Cooper is an editor for UpToDate. All other researchers report no relevant financial disclosures.


Does the study by Razvi et al provide enough evidence to encourage routine thyroid hormone treatment in middle-aged patients with subclinical hypothyroidism (SCH)?


The reason this study by Razvi and colleagues is so clinically important is, at the present time within endocrinology, there is a major, honest difference of opinion about whether patients with SCH should be treated.

Our basic science and clinical research work approaches this from the cardiovascular (CV) point of view. From the cardiac perspective, subclinically hypothyroid patients have a variety of modifiable CV risk factors. They have elevated cholesterol, hypertension, impaired cardiac function and impaired endothelium-derived relaxing factor, the factor that allows you to dilate your blood vessels when you exercise.

Despite the fact that those things were known in the cardiac field, there has never been an outcome study to establish that if you treat hypothyroid patients, the reversal of these risk factors leads to an improved outcome. This is the first study to demonstrate that CV mortality could be lowered in subclinically hypothyroid patients who were treated.

Irwin L. Klein

You can criticize the study in a sense that it wasn’t a prospective, randomized controlled trial, but it is what we call a ‘real-world study.’ These are practitioners choosing to treat or not treat based upon existing treatment paradigms. It turns out they chose to treat or not treat on a 1:1 ratio and there was no obvious treatment bias.

The results clearly demonstrated, in patients aged younger than 70 years, that there was a survival advantage. So that is a very relevant finding. If I’m in the office with a 50-year-old patient with SCH, I have further rationale to treat. Not only do I have the potential to improve the various thyroid hormone mediated aspects of metabolism and the standard but often subtle thyroid symptoms, but I can also potentially improve CV morbitity and mortality.

With that in mind, it is my opinion that patients with SCH should be given levothyroxine to reduce their CV risk, with the understanding that there is good reason to suspect you could improve their overall CV mortality.

Irwin L. Klein, MD, is a Thyroidologist and Professor of Medicine, North Shore University Hospital, Manhasset, NY. Disclosure: Dr. Klein reports no relevant financial disclosures.


No, it does not. The authors point out that although this is a large observational study, it is retrospective in nature and subject to potential bias. In addition, the authors conclude by suggesting that a randomized, prospective trial is indicated (by their calculations, approximately 1,450 participants would be needed).

Stuart R. Chipkin

So, even the authors are not rushing to suggest widespread prescribing of levothyroxine under these circumstances. As clinicians weigh the choice of treating patients with similar biochemical parameters, it is worth remembering who was excluded from this study:

  • Those with a history of ischemic heart disease.
  • Those with cerebrovascular disease.
  • Those receiving lithium, amiodarone or steroids within the past year.

Manuscript notes the high compliance of patients treated with levothyroxine. This may be related to improvement in some subtle symptoms experienced by patients. In that case, it is not the CV benefit that may be the reason to treat patients, but the presence of symptoms despite only mild elevations in TSH and normal thyroxine concentrations. In addition, the loss in benefit-to-risk, which occurs after age 70 raises the question of, ‘What factors are at play?’ Are there other factors (or illnesses) that might mitigate the benefit of levothyroxine that appears to be present at an earlier age? Since there are several questions and insufficient answers, I would not advocate the routine use of thyroid hormone in middle-aged patients with SCH. Patients who do not have any of the exclusionary criteria and who have symptoms suggestive of hypothyroidism and who are willing to accept the addition of another medication can have the potential advantages (and disadvantages) explained to them. Given our imperfect knowledge, a mutual decision based on available information would seem like a reasonable course of action.

I would not rush to treat elderly patients with biochemical evidence of SCH. These data do not suggest any benefit. Elderly patients are on more medications (confirmed in this study) and the unclear benefit in clinical outcomes would not speak toward recommending the addition of another medication to these individuals.

Stuart R. Chipkin, MD, is a research professor in the School of Public Health and Health Sciences at the University of Massachusetts Amherst and an endocrinologist practicing with the Valley Medical Group in Amherst, Mass. Disclosure: Dr. Chipkin reports no relevant financial disclosures.