Improving the long-term management of benign thyroid nodules.

Nodules have been reported to be present in 50% of thyroids in autopsy series.1 Aminority of these thyroidnodules arepalpable, leaving a broad opportunity for discovery of incidental thyroid nodules through imaging tests performed for other indications, including carotid ultrasound and chest computed tomography. Indeed, detection of asymptomatic thyroid nodules as incidental findings on radiological evaluations that include the neck has increased over the past 3 decades.2 Once detected, these thyroid nodules require additional evaluation. Themajor concern with thyroid nodules is identification of the 7% to 9% that are cancerous3 and require treatment. Appropriate initial evaluation of a thyroid nodule includes thyrotropin measurement and a thyroid ultrasound. Recommendations for decidingwhich thyroidnodules require ultrasound-guided fine-needle aspiration to rule out thyroid cancer have evolved from reliance exclusively on size criteria. Refinements inultrasoundtechnologyhaveallowedfiner resolutionof thyroidnodulesanddetectionof thyroidnodulecharacteristics, several of which are associatedwith higher risk of a malignant nodule. The presence of these higher-risk sonographic characteristics, which include hypoechogenicity, irregular margins, taller-than-wide shape, intranodular vascularity, andmicrocalcifications,4,5 are also indications for initial fine-needle aspiration. There are fewdata to guide the long-termmanagement of the majority of patients with thyroid nodules, which includes those with benign initial fine-needle aspiration results or in whom aspiration was not indicated based on size andultrasound characteristics. The clinical concern is detecting the cancerous nodules that were not identified during initial evaluation, both aspiration false-negatives and ultrasound false-negatives, small thyroid nodules lacking highrisk ultrasound features. Current information about thyroid nodule biology suggests that benign thyroid nodules do not transform into malignant nodules; the nodule either arose as a benign nodule or as a malignant nodule. Therefore, ultrasound surveillance has been advocated to detect falsenegatives from the initial evaluation.4 Differentiated thyroid cancer is a low-mortality cancer, with overall 5-year survival rates surpassing97%.6Even thoughdifferentiated thyroidcancer is generally a slow-growing tumor, an untested assumption in followinguppatientswith thyroidnodules is that nodules with significant growth, detected through serial ultrasounds, are more likely to harbor cancer. A reduction in the frequency of thyroid ultrasound assessment would lead to cost savings andprovide a schedule that is concordantwith the low risk of detecting a clinically relevant thyroid cancer. The study by Durante et al7 in this issue of JAMA documents the frequency and factors associatedwith thyroidnodule growth and demonstrates just how low the risk of detecting a thyroid cancer is during follow-up surveillance. The authorsperformedaprospective, observational study in8 Italian centers. For study inclusion, participantshadat least 1 thyroid nodule and no thyroid dysfunction, and they either had a benign fine-needle aspiration test result or did notmeet criteria for aspiration due to size and sonographic characteristics. The 992 enrolled patients returned annually for 5 years for thyroid ultrasound evaluation, unless significant nodule growth occurred (defined as greater than 20% and at least 2 mm in 2 dimensions) or concerning sonographic characteristics appeared, defined as hypoechogenicity, irregular margins, taller-than-wide shape, intranodular vascularity, andmicrocalcifications, triggering fine-needle aspirationevaluation. In addition, patients who had enrolled after a benign test result at baseline were offered a second aspiration at year 5. A total of 63% of patients with a benign nodule at baseline had repeat aspiration performed during follow-up. Among the 1567 nodules in the 992 enrolled patients, significant nodule growth and detectable nodule shrinkage occurred at similar frequency, with nodule growth in 15.4% of patients and shrinkage in 18.5%. Thyroid nodule growth was independently associated with multinodular glands and younger age of diagnosis. Thyroid cancer was detected during follow-up in only 0.3% (95% CI, 0.0%-0.6%) of the originallydetectednodules.Notsurprisingly,withadditional followup, there was additional nodule discovery, in 9.3% (95% CI, 7.5%-11.1%) of patients, 1 ofwhichwas subsequently found to be a thyroid cancer. This studyhas 4 important implications for the follow-up of thyroid nodules. First, these prospective data provide reassurance about the validity of a benign cytology result obtained by ultrasound-guided fine-needle aspiration and confirm a very low false-negative rate, at 1.1%.8 Second, the practice of routine sonographic surveillance with repeat fine-needle aspiration for growth, as recommended by published guidelines,4 is not the most efficient strategy to detect the very small number of missed cancers amongpreviously sampled cytologically benignnodules. The one-size-fits-all approach simplydoesnotwork. Instead, surveillance strategies should be individualized based on a nodule’s sonographic appearance.All 4missed cancers in thepreRelated article page 926 Opinion