Should suspicious pediatric thyroid nodules smaller than 1 cm on ultrasound undergo biopsy?
Click here to read the Cover Story, "‘Overdiagnosis’ may not fully explain rise in pediatric thyroid cancer incidence"
The American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS), published in 2017, provides a standard system for risk stratification and management of thyroid nodules detected by ultrasound. Nodules are characterized and graded on the basis of sonographic features — including composition, echogenicity, shape, margin and echogenic foci — and assigned a risk level ranging from TR1 (benign) to TR5 (highly suspicious). For TI-RADS categories 3 to 5, the recommendation to proceed to FNA is based upon nodule diameter.
Since its inception, this scoring system has been widely used to guide the decision for FNA among adults with thyroid nodules and has led to a reduction in number of biopsies without missing clinically significant malignancies —a key assumption of TI-RADS is that small cancers in adults may never require treatment. Application of ACR TI-RADS in the pediatric population has been validated only in retrospective single-institutional studies; however, most have shown a missed malignancy rate of between 20% and 25%, with papillary microcarcinoma comprising the majority of missed cancers. Although some centers feel this subgroup of papillary thyroid cancer represents a low-risk group that may be managed with lobectomy rather than total thyroidectomy, there is far-from-uniform consensus on this topic. Given that thyroid volume increases with age, many experts argue that cutoffs for nodule size should be standardized against age-adjusted average thyroid volumes. Moreover, pediatric papillary thyroid cancer may not be biologically equivalent to adult papillary thyroid cancer, with increased rates of multifocality, bilaterally, nodal metastasis and a different distribution of genomic alterations. Within our own database at Yale University, approximately one-fifth of patients diagnosed with papillary thyroid cancer have nodules smaller than 1 cm, and one-third of these have nodal involvement.
A 2021 study by Ahmad and colleagues suggested a modified algorithm for the diagnostic management of pediatric thyroid nodules in which size criteria are marginalized. Using TI-RADS 3 or greater (without size cutoff) as an indication for FNA had 100% sensitivity with no missed suspicious or malignant nodules on cytology or pathology. As children with papillary thyroid cancer will require follow-up spanning decades, early detection through FNA of small but sonographically concerning nodules may allow treatment of disease prior to wider nodal or systemic dissemination and has the potential to reduce rates of recurrence.
- Ahmad H, et al. J Pediatr Surg. 2021;doi:10.1016/j.jpedsurg.2021.02.034.
- Tessler FN, et al. J Am Coll Radiol. 2017;doi:10.1016/j.jacr.2017.01.046.
Emily Christison-Lagay, MD, is associate professor of pediatric surgery at Yale University School of Medicine. She can be reached at firstname.lastname@example.org.
“You can observe a lot by watching.” – Lawrence “Yogi” Berra
Among the central adages of pediatrics is that “children are not little adults.” This applies to both size and physiology. Both are relevant to the consideration of thyroid biopsy in children.
Two recent systematic approaches to thyroid ultrasound, the points-based ACR TI-RADS and the pattern-based ATA approach, define indications for biopsy of thyroid nodules based on sonographic appearance and size. Both advocate biopsy of nodules 1 cm or larger for the most sonographically suspicious nodules. Yet, neither considers the dynamic changes in thyroid volume in a growing child. Consider the aforementioned 1 cm nodule: In an average adult, this constitutes roughly 6.5% of the volume of a thyroid lobe, whereas in a 7-year-old child, the volume occupied by the same nodule is approximately 17%, or 2.5-fold greater. As such, the size limit of 1 cm reflects a relatively larger nodule in the context of a pediatric gland and, therefore, it stands to reason that absolute size thresholds established in adults cannot be directly extrapolated to children. Consequently, recent guidelines for evaluation of pediatric nodules exclude size as a criterion for biopsy.
Should we then biopsy all thyroid nodules in children, regardless of size? The answer is a resounding ‘no!’
First, how subcentimeter nodules come to attention bears reflection. Fewer than 7% of subcentimeter nodules are palpable. One must therefore assume that most small nodules on ultrasound are identified not in the context of characterizing a clinically apparent mass, but more commonly as incidental imaging findings.
Overuse of thyroid ultrasonography has been widely cited as an underlying contributor to the overdiagnosis of thyroid malignancy. One recent study demonstrated that 32.7% of physicians ordered ultrasound upon patient request and 28% based on abnormal thyroid function tests — an indication discouraged by international guidelines and the Choosing Wisely campaign. This overuse of ultrasound is likely contributing to the unmasking of a subclinical reservoir of small nodules that may never progress to clinical significance. Caution should also be applied in the diagnosis of small pediatric thyroid nodules, wherein intrathyroidal ectopic thymic tissue, a benign anatomic variant that does not merit biopsy, is frequently misperceived as a lesion suspicious for malignancy.
Despite higher rates of malignancy in children, among whom roughly 25% of thyroid nodules prove to be malignant, it is important to recognize the converse — that 75% of nodules are benign. Additionally, even for those lesions that prove to be malignant, the biology of thyroid carcinoma is relevant.
Although the natural history of untreated thyroid microcarcinoma in children is not known, extrapolation from adults undergoing active surveillance suggests that most follow a fairly indolent course and those that do not, can achieve successful salvage upon evidence of progression. Disease-specific survival, even for those children with distally metastatic disease, exceeds 95% at 30 years. As a result, the clinician and patient are afforded the luxury of time to monitor the behavior of a sonographic lesion that may or may not be cancer and that may or may not progress, without urgency to proceed to biopsy.
Biopsy in children can be technically more challenging than in adults, given the more compact anatomy. Moreover, among children, there is often need for sedation or general anesthesia, which introduces an additional layer of risk that must be considered in gauging the potential benefit. Finally, cytopathology may be nondiagnostic or indeterminate in up to 40% of pediatric biopsies, leading to patient and parent anxiety, as well as further biopsy or diagnostic surgery to achieve definitive diagnosis.
This is not to argue that subcentimeter nodules should never undergo biopsy. The decision to pursue biopsy of a nodule should weigh the circumstances that led to its identification, the pretest likelihood of malignancy, the experience of the clinician performing the procedure and the associated risks, against the likelihood that early diagnosis and intervention will meaningfully improve a child’s health. Thus, for the incidentally identified, clinically occult, subcentimeter thyroid nodule, even in children, we may be well-advised to follow the advice of Mr. Berra and see what we observe, just by watching.
- Brander A, et al. J Clin Ultrasound. 1992;doi:10.1002/jcu.1870200107.
- Chen DW, et al. JAMA Surg. 2020;doi:10.1001/jamasurg.2020.2507.
- Cherella CE, et al. Thyroid. 2019;doi:10.1089/thy.2018.0728.
- Francis GL, et al. Thyroid. 2015;doi:10.1089/thy.2014.0460.
- Haugen BR, et al. Thyroid. 2016;doi:10.1089/thy.2015.0020.
- Tessler FN, et al. J Am Coll Radiol. 2017;doi:10.1016/j.jacr.2017.01.046.
Jonathan Wasserman, MD, is a staff physician at The Hospital for Sick Children in Toronto. He can be reached at email@example.com.