Ultrasound of Benign Thyroid Nodules: A 120 Months Follow-Up Study

Carlo Cappelli; Ilenia Pirola; Elena Gandossi; Mario Rotondi; Claudio Casella; Davide Lombardi; Barbara Agosti; Alberto Ferlin; Maurizio Castellano


Clin Endocrinol. 2021;94(5):866-871. 

In This Article


This retrospective study showed that, among patients with benign thyroid nodules, the size of nodule remained stable for the majority (83%) during 120 months of follow-up. Moreover, FNA analysis performed in those nodules that increased in size yielded a 4% rate of thyroid malignancy.

The evaluation and management of a patient with of one or more thyroid nodules occurs frequently in the clinical setting, especially in endocrinology.[19] When a benign nodule is diagnosed by FNA, there is uncertainty over how long it should be monitored. The purpose of monitoring is to identify nodular growth, with the assumption that this variable is a surrogate for malignancy and the indication of the false diagnosis of a benign nodule at first evaluation. However, the false-negative rate of benign nodules is known to be low.[20–22]

There is no established appropriate interval for ultrasound monitoring, as there are few studies that examined this issue. The American Thyroid Association (ATA) guidelines suggest stratifying cytologically proven benign nodules according to different ultrasound features, from high-suspicious to very low-risk ultrasound patterns. ATA guidelines suggested repeating ultrasound evaluation at 6–12 months for follow-up in benign nodules with high-risk ultrasound patterns, and state that there is no need for follow-up the very low-risk group.[23] However, these suggestions are based on an average-low quality of evidence, as reported by the ATA guidelines themselves.[23] In addition, the American Association of Clinical Endocrinologists, American College of Endocrinology and Associazione Medici Endocrinologi suggested to perform ultrasound examination in benign nodules in approximately 12 months after FNA. If nodules are unchanged repeat the examination after 24 months; this assumption was again based on low quality of evidence, as reported by the guidelines themselves.[24]

Nou et al[25] reviewed the records of a large series of patients referred to the Thyroid Nodule Clinic at the Brigham and Women's Hospital in Boston, and found that most benign thyroid nodule can be safely recommended for follow-up at 2–4 years with no risk of mortality or likelihood of harm. Recently, Durante et al also found in a very elegant multicentre study that 'the majority of nodules exhibited no significant size increase during 5 years of follow-up and thyroid cancer was rare'. Based on these results, the Authors recommended the revision of current guidelines for follow-up of asymptomatic benign thyroid nodules.[20] Our data confirm and extend the previous ones on a longer period of follow-up. We showed that the majority of nodules (83%) remained stable over time, and only 11% of all lesions were evidence of nodule growth. This growth occurred in a linear fashion, starting from the first year of observation. It is important to underline the fact that, among patients who displayed growth in nodule volume at the end of the study, for 59 patients (79%) this started within the first five years of follow-up. In addition, our data emphasized the high sensitivity rate of FNA, as the false-negative score (3%) was superimposable to literature data obtained in larger series of nodules.[26,27]

In contrast to the experience of Durante and colleagues, which showed certain features predicting nodular growth (male gender, multiple nodules and nodule volumes more than 0.2 ml), we did not find any characteristics that predict an increase in nodule volume in our population (Table 2). This could be explained by different facts: firstly, the distribution of gender between the two populations is significantly different. In our series, in fact, men represent 30% of patients, while in studies by Durante et al, they represent just 18%; secondly, more patients with multinodular goitre were assessed among our patients; third, the median volume of our series is significantly larger, with 83% of nodules of volume greater than 1 ml, as compared to 31% in the other study; finally, our series includes patients (7%) undergoing levothyroxine treatment, whereas for Durante et al, none of the patients was undergoing LT4 treatment.[20] It seems worth highlighting that none of our patients under levothyroxine displayed suppressed serum TSH levels and that a similar rate of LT4-treated patients was found in relation to increased and stable nodular volumes.

Limitations of the present study could derive from the lack of data on iodine status and antiperoxidase and/or antithyroglobulin antibodies in our patients. It is well known that iodine intake is a major environmental factor in determining thyroid nodular disease.[28–30] We could say that 97% of the patients in our series were born and live in northern Italy, a region still featuring mild iodine deficiency, despite an iodine supplementation programme that has been in place for the past 15 years.[31] We cannot exclude the fact that, in a region with iodine sufficiency or with marked iodine deficiency, nodule behaviour could be different from that observed in our population.

To the best of our knowledge, there are no existing prospective or retrospective studies focusing only on the role of antiperoxidase and/or antithyroglobulin antibodies and the behaviour of benign nodular lesions. All of our patients undergoing LT4 replacement therapy (7%) were affected by Hashimoto thyroiditis, but they represent too small a number of patients to be able to draw any conclusions. However, multivariate analysis would rule out the possible role of these antibodies in nodular growth.

Another limitation could be derive from the lack of data of interobserver variability for thyroid nodule volume measurement. Unfortunately, we have no this data. However, we believe that the large number of patients, the proven large repeatability coefficient for nodule diameter and the improbability that significant interobserver volume variability could only be present in one study group, validate our results.

In conclusion, our data confirm and extend the data from previous studies over a longer follow-up. The majority of benign nodules remained stable over the period of monitoring. On the basis of our experience, we recommend ultrasound examination at a distance of two and five years following cytological evaluation, then every 4–5 years from then on.