Medical Treatment of Benign Thyroid Nodules: Have We Defined a Benefit?

Thyroid nodules, detected by physical examination, are present in 4% to 7% of the U.S. population [1, 2]. This figure represents only the tip of the iceberg because ultrasonography done in patients without clinically apparent thyroid disease and autopsy studies show that thyroid nodules are present in 30% to 50% of adult thyroid glands [3]. The chance that a nodule detected by physical examination is malignant is less than 10%; most studies show a prevalence of approximately 4% [4, 5]. Fine-needle aspiration biopsy of thyroid nodules has emerged as the most important diagnostic study to perform to distinguish malignant from benign thyroid nodules. Because at least 90% of thyroid nodules are benign, physicians and patients are presented with the interesting problem of how to manage benign nodules. The magnitude of the problem is illustrated by considering that as many as 10 to 20 million persons in the United States have a benign and clinically apparent thyroid nodule. Surgical excision of all such nodules is clearly not acceptable to patients, physicians, or the health care industry: This practice could consume in excess of $100 billion annually. A more popular option has been medical therapy, consisting of oral thyroxine therapy to suppress pituitary secretion of thyroid-stimulating hormone, thereby removing an important growth factor for thyroid follicle cells. Such therapy is relatively inexpensive: One year of thyroxine therapy costs $50 to $100. However, even this conservative form of therapy would cost $1 to $2 billion annually. A third option is to do nothing except examine the patient annually and determine whether the clinical characteristics of the nodule have changed.

The critical issue is to define not only the question but also the desired outcomes from any therapeutic choice and then prospectively test the efficacy and benefit of the intervention. Herein lies the problem. Desired outcomes for benign hypofunctioning thyroid nodules have not been rigorously defined. Physicians and patients may have varying and sometimes different goals for desired outcomes, ranging from prevention of malignant transformation to reduction in nodule size, prevention of additional growth, prevention of new nodules else-where in the thyroid, prevention or relief of symptoms caused by the thyroid nodule, and cosmetic relief.

Each of these outcomes may have merit, but none should be accepted without challenge. Clearly, prevention of malignant transformation would be desirable, but the actual rate of such transformation is very small. Thus, the net efficacy and benefit would be difficult, if not impossible, to prove. Reduction in nodule size may have benefit if the nodule was causing a problem, such as dysphasia, dysphonia, pain, or an unacceptable cosmetic appearance. However, most thyroid nodules are small and have no symptoms associated with them. Thus, it is not clear what benefit can be ascribed to only a reduction in size. The idea that a therapeutic intervention may prevent additional growth of the thyroid nodule may also be beneficial, assuming that the additional growth would cause a problem. Certainly, nodules must grow to become clinically detected, and one might assume that all such nodules will continue to grow. However, such an intuitive truth has not been verified in one large study on the natural history of thyroid nodules done in Japan [6]. In this study, 34% of solid nodules remained the same size over a 15-year period. These data are instructive, but whether they apply to other populations is unclear. For example, in four prospective trials of the efficacy of treatment with L-thyroxine, controls showed neither an increase nor a decrease in nodule size 46% to 86% of the time, as assessed by ultrasonography over the relatively short interval of 6 to 18 months [7-10].

In this issue, Gharib and Mazzaferri [11] give a balanced review and analysis of the efficacy of thyroid hormone suppression therapy for benign thyroid nodules on many clinical outcomes. Only one of these potential outcomes, reduction in benign nodule size, has been tested in several prospective trials. The results are mixed, some showing significant reduction and others not. Among the seven referenced reports [7-10, 12-14] the size of the nodule was reduced more than 50% by thyroid hormone suppression therapy 14% to 39% of the time, whereas no therapy or placebo accomplished the same goal in 0% to 35% of instances. Thus, clinicians still do not have a definitive answer. Of note, only one of these studies [12] was from the United States; the others were from other countries in which iodine intakes differ. Furthermore, the value or benefit derived from reducing thyroid nodules by more than 50% is inescapably vague. In fact, the practice of treating a benign nodule of an endocrine gland is restricted to the thyroid gland. Do physicians routinely use hormone suppression to treat benign, single, nonfunctional nodules of the pituitary, adrenal, gonads, prostate, or breast?

Perhaps a more reasonable strategy for the medical treatment of thyroid nodules would be to prevent their growth. As Gharib and Mazzaferri point out, such an outcome has not been studied. One could certainly speculate on the benefit of this outcome; for example, preventing growth may prevent such future symptoms as dysphasia, dysphonia, or pain and may prevent the need for surgery (most nodules that grow with or without medical therapy eventually receive surgical therapy).

Do we have any indication of how often thyroid nodules grow and whether growth is preventable by thyroxine suppressive therapy? In the report from Japan by Kuma and colleagues [6] on the natural history of nodules in 140 untreated patients followed for an average of 15 years, only 14% of nodules grew; of these, 26% were malignant. Interestingly, four of the referenced prospective studies analyzed size increase of more than 50% [7-10]. Among patients who received thyroid hormone suppression therapy, 0% to 14% of patients had nodules that grew more than 50%, whereas 14% to 22% of patients who did not receive therapy or placebo had nodules that grew more than 50%. The lack of overlap between patients who did and did not receive thyroid hormone suppression therapy suggests that such therapy may prevent growth of existing benign thyroid nodules. Clearly, prospective and randomized studies should be done to directly answer this question. Such data may also allow careful cost–benefit analyses to be performed.

In the meantime, is there a middle ground that can provide a rational approach to treatment of a benign solitary thyroid nodule? Although algorithms are never perfect, I offer a potential strategy for analysis and study (Figure 1). Assume that a patient has a single, solid, benign thyroid nodule (determined by fine-needle aspiration biopsy) and normal thyroid function test results. The patient is put in a no-therapy observation period for 1 year. At the end of 1 year, the patient is evaluated for symptoms, thyroid function, and nodule size by ultrasonography. If the nodule has not changed or has decreased in size, observation alone is continued. If the nodule has increased in size, fine-needle aspiration biopsy is repeated. If the biopsy result shows malignancy or is suspicious, the patient is sent to surgery. All other patients whose nodules have grown are given L-thyroxine therapy in doses sufficient to suppress the serum thyroid-stimulating hormone level to 0.1 to 0.5 mU/L. Observation and L-thyroxine therapy are continued for another year. If the nodule grows during L-thyroxine therapy, the patient is referred for surgery; patients whose nodules do not grow or shrink during L-thyroxine therapy continue to receive medical therapy.

View larger version:

• In this window
• In a new window

Figure 1. Decision making for treatment of patients with benign thyroid nodules.

Although this medical approach is not perfect, it eliminates treatment of patients with thyroid hormone suppression who have spontaneous resolution of or no change in their nodules. It will restrict the use of thyroid hormone suppression therapy to those who might benefit from it: specifically, patients whose nodules have shown the potential for growth. The goal of this approach is not to reduce nodule size but to prevent growth and, thus, the potential for symptoms (and, more rarely, malignancy). The benefit becomes prevention of clinical disease rather than reduction of nonclinical disease. Making a hypothesis-thyroid hormone suppression therapy will prevent growth of benign thyroid nodules-is easy; proving efficacy by controlled prospective studies is the unmet challenge.

• Copyright ©2004 by the American College of Physicians