Introduction: This study looked at the value of ultrasound (US) features in managing thyroid nodules with initially benign cytologic results.
Methods: This retrospective study looked at the record of 6,025 consecutive patients who underwent US-guided fine-needle aspiration biopsy (FNAB) of 6118 focal thyroid nodules from October 2003 to February 2006. From this group, the authors then analyzed 1301 patients with 1,343 nodules which were 1 cm or larger and diagnosed as benign at initial cytologic evaluation. Those patients then underwent pathologic or follow-up study.
The authors compared the risk of malignancy according to the US finding. They also calculated the likelihoods of different subgroups having benign nodules.
Results: In the group of patients who were diagnosed by FNAB to have benign nodules, 26 (1.9%) were actually malignant and 1,317 (98.1%) benign, according to reference standards.
A suspicious US appearance for thyroid malignancy was defined as the presence of at least one of the following features in the thyroid nodule: marked hypoechogenicity, microlobulated or irregular margin, microcalcifications, and a greater anteroposterior than transverse dimension. If none of these features were present, the US was defined as benign.
If a thyroid nodule had benign results at the initial FNAB and benign US features, the likelihood of being benign increased from 98.1% to 99.4%. If the repeat FNAB was also benign, then the likelihood of the nodule being benign increased to 100%.
However, if a thyroid nodule had benign results at the initial FNAB but suspicious US features, the likelihood of a benign nodule decreased to 79.6% and was statistically significantly lower than the group with negative US features (99.4%, p<.001). When FNAB was repeated in 54 patients who had suspicious US and benign initial FNAB, an additional 8 patients (15%) were diagnosed with thyroid cancer and proceeded to surgery.
For a nodule with benign features at initial US, the risk of malignancy for a thyroid nodule with an increase in size at the follow-up US was 1.4%. That is slightly higher than the risk of malignancy for a thyroid nodule with no interval change or decrease in size, but that was not a significant difference (0.5%, p=.354).
Conclusion: Even if the initial cytologic results indicate that a thyroid nodule is benign, repeat FNAB should be done for thyroid nodules that have suspicious US features.
Commentary by Kresimira (Mira) Milas MD
Thyroid ultrasound (US) has become the main type of imaging used for the evaluation of thyroid nodules. It provides a number of descriptive features about a nodule that other radiologic modalities cannot. These include nodule size, shape, degree of vascularity, tissue composition (solid, cystic), smoothness of the peripheral margins, the presence and pattern of calcifications, and the presence of enlarged adjacent cervical lymph nodes. Although some sonographic features may raise suspicion of thyroid cancer, no single feature or combination of features can reliably or consistently make this diagnosis using only ultrasound. Fine-needle aspiration biopsy (FNAB), therefore, especially when performed with ultrasound guidance, is currently the most accurate diagnostic tool to identify thyroid cancer. The American Thyroid Association has identified practical guidelines that advise which nodules should undergo FNAB (Cooper DS et al. Thyroid 2009;19:1167-214).
The article by Dr. Jin Young Kwak and colleagues is highlighted in the above summary because it points to the value of repeating FNAB promptly in some patients even when the initial cytology report gives favorable, benign results. Though our best tool, FNAB still carries a low risk for inaccuracies. The false negative rate of FNAB is generally less than 5% and can be affected by several variables: skill of the clinician performing FNAB, use of palpation instead of US, the absence of clear cytologic abnormalities in some variants of thyroid cancer, and very large nodule size. The authors reported achieving a false negative rate of 1.9% overall. However, if the nodule had suspicious features – hypoechoic texture, irregular borders, hypervascularity, microcalcifications - the false negative rate would have been unacceptably high (20%) if FNAB was not repeated.
This study emphasizes the importance of considering all the clinical information available in managing patients with thyroid nodules. Some endocrinologists and surgeons perform their own US and FNAB and thus have the opportunity to see whether the cytology result is concordant with general impression given by the US images they obtained. Some may rely on other specialists for this service, and must look whether specific comments about nodule features were included in the radiology report and communicate with the sonographer if necessary to get the needed details. It is important to consider the description of the thyroid nodule on ultrasound in the context of the FNAB result. If concern for malignancy remains high, FNAB should be repeated.