ABSTRACT
Objective:
The differentiation between benign and non-benign thyroid nodules is a complex problem to solve in clinical practice. We aimed to observe and describe the role of shear-wave elastography (SWE) to distinct benign and non-benign thyroid nodules before a fine-needle aspiration biopsy (FNAB).
Methods:
Ninety-seven patients were prospectively included in the study from March 2019 to January 2020. Patients with a history of thyroid infections, surgery or trauma of thyroid tissue, autoimmune diseases associated with thyroid gland, non-diagnostic histopathology (Bethesda I) were excluded from the study. Thyroid imaging reporting and data system (TI-RADS) based on the 2017 American College of Radiology (ACR) was used for the radiological classification of nodules. Patients’ age, number of thyroid nodules, SWE value of nodules, and TI-RADS categories were compared to pathological classification.
Results:
The mean age of patients was 49.80±11.42 years. Benign thyroid nodules were classified as “Group 1” (G1) (n=79) and non-benign thyroid nodules as “Group 2” (G2) (n=12) according to pathological diagnosis. The median SWE values of patients in G1 and G2 were 9.47 (7.48) and 47.38 (51.46) kPa, respectively. The median SWE values of G2 were statistically significantly higher than G1 (p=0.001). While nearly 50% of the patients in the G1 were in TI-RADS category 3, the ratio of TI-RADS 5 was over 40% in the G2 and the difference was statistically significant in terms of the TI-RADS category (p=0.001)
Conclusion:
In addition to TI-RADS classification based on 2017 ACR, SWE measurements of thyroid nodules may differentiate benign and non-benign thyroid nodules before an FNAB. For this reason, both methods can be combined to increase the diagnosis’s specificity, sensitivity, and accuracy.
Introduction
Thyroid nodules are frequently seen in radiology and endocrinology practice and the incidence of thyroid nodules is increasing, especially in the population who is in the first two decades of life (1). While about 5% of diagnosed thyroid nodules have malign pathology, the majority has benign pathology and it is clinically significant to diagnose malignant nodules.
Although some gray scale ultrasound imaging findings are related to the malignancy of thyroid nodules, the prediction of histopathological diagnosis of thyroid nodules is complicated by using ultrasonography (US) (2).
Ultrasound elastography is a recent and advanced modality that raises the diagnostic performance of US by distinguishing varied tissues elasticity (3). Shear-wave elastography (SWE) is a less operator-dependent imaging technique that directly determines absolute stiffness without a maneuver (4,5). In recent years, elastosonography is widely used for the diagnosis of malignities, fibrosis, or inflammatory (edematous appearance) process in breast, thyroid or prostate tissue and musculoskeletal system. SWE uses shearwaves to quantitatively measure tissue stiffness that allows the differentiation of benign and non-benign thyroid nodules. Further, the best method for this differentiation is fine needle aspiration biopsy (FNAB). Thyroid Imaging Report and Data System (TI-RADS) is a guide for the decision of FNAB. TI-RADS classificaiton system was suggested firstly based on the Breast Imaging Report and Data System model, that was reported by the American College of Radiology (ACR) in 2003 (6). ACR published the revised version of TI-RADS in 2017 based off wide-ranging evidence and clinical confirmation. TI-RADS based on 2017 ACR is a risk grading system to guide decisions concerning FNAB and follow-up of thyroid nodules acording to their gray scale ultrasound features. It identifies the most clinically significant malignancies while decreasing biopsies applied on benign nodules (7). However, FNAB has disadvantages as it is a time-consuming and invasive procedure (8-10).
In this first prospective study, we aim to analyze and investigate the diagnostic efficacy of the TI-RADS based on 2017 ACR classification using the SWE values in detecting benign andnon-benign thyroid nodules before FNAB.
Methods
The local ethics committee of our hospital approved this prospective study (GEAH-KEK-2019/0057). The procedure was explained in detail to each participant prior to measurement and it was performed from March 2019 to January 2020 in our clinic. Informed consent was obtained from each patient before the measurement and FNAB, and the study was conforming to the principles of the Declaration of Helsinki.
Participants with a history of thyroid infections, surgery or trauma of thyroid tissue, autoimmune diseases associated with thyroid gland, non-diagnostic histopathology (Bethesda I) were excluded from the study. Ninety-one thyroid nodules diagnosed in 91 patients were prospectively comprised to the study from March 2019 to January 2020.
We used a Ultrasound 2D SWE system to examine the thyroid nodules, initially the gray scale US, then quantified with SWE software without compression (Esaote QElaXto 2D ultrasound, Genova, Italy). Evaluation of thyroid nodule stiffness was made by using SWE velocity and quantitative measurements were received as kilopascal (kPa). A linear transducer of 6-9 MHz was preferred for the process. The patients laid down with supine position and were told to hold their breath while during the SWE examination and measuring. Quantitative elastographic assessment was performed using a 2x2 mm region of interest (ROI). We placed the ROI within the biopsy considered nodule. Course calcified and cystic areas were excluded from the ROI area. The preset of grayscale ultrasound was set to a depth of 0-2.5 cm. Five ROI areas were randomly selected for each area for quantitative measurement. We obtained the mean value of the 5 ROI areas to use in the statistical analysis as shown in Figure 1.
Classification of thyroid nodules and biopsy indications were made according to ACR 2017 TI-RADS. Then, we performed FNAB to thyroid nodules under ultrasound guidance after SWE measurements by two radiologists who had three years of experience in FNAB. 1% lidocaine was used for skin local anaesthesia then 22- gauge PTC needle was used for biopsy under ultrasound guidance. Three aspirations and smears were made for each nodule. A pathologist classified the FNA thyroid cytopathology according to the Bethesda System as follows; Bethesda I (non-diagnostic or unsatisfactory), Bethesda II (benign), Bethesda III (AUS/FLUS, atypia of undetermined significance/follicular lesion of undetermined significance), Bethesda IV (FN/SFN, follicular neoplasm/suspicious for follicular neoplasm), Bethesda V (suspicious for malignancy), and Bethesda VI (malignant)(11).
Benign (Bethesda II) thyroid nodules were defined as “Group 1” (G1) (n=79) and non-benign (Bethesda II-III-IV-V-VI) thyroid nodules as “Group 2” (G2) (n=12) according to pathological diagnosis. Both groups were evaluated in terms of age, the number of thyroid nodules, SWE values, and TI-RADS categories.
Statistical Analysis
We used SPSS version 22.0 software in the data analysis of the study (IBM SPSS, USA). The normality of distributions was assessed by using the Shapiro-Wilk test. We used the descriptive statistics analysis such as mean, Standard deviation, frequency or median and interquartile range. Student-t test was performed to compare normally distributed parameters and Mann-Whitney U test was performed as a non parametric test when the parameters did not have normal distribution. Values were assessed within 95% confidence range at p<0.05 significance level.
Results
The general data of the participants are summarized in Table 1. The mean age of all patients with SD was 49.80±11.42 years. Compared to the patients in G1, the mean age of patients in G2 was statistically lower (p=0.02). However, no statistically significant difference was found in the number of thyroid nodules between the groups (p>0.05). The median SWE values of patients in G1 and G2 were 9.47 (7.48) kPa and 47.38 (51.46) kPa, respectively. The median SWE values of G2 were statistically significantly higher than G1 (p=0.001). While nearly 50% of the patients in the G1 were in TI-RADS category 3, the rate of TI-RADS 5 was over 40% in the G2 and the difference was statistically significant in terms of the TI-RADS category (p=0.001)
Discussion
The differentiation of benign and non-benign thyroid nodules is a complex problem to solve in medical practice. Recently, elastography has been investigated in the distinction of benign and non-benign thyroid nodules, and different studies have reported that SWE is helpful for the distinction of benign and non-benign thyroid nodules (12-24). However, some studies did not support this positive contribution (25-29). Thyroid nodules were not classified according to the ACRTI-RADS based on 2017 in all studies mentioned above. According to the results of a study in the literature evaluating thyroid nodules based on the last reported TI-RADS grading system in ACR 2017 and investigating the effect of SWE on thyroid nodules before biopsy; it is recommended that SWE must be used especially unclear fine-needle aspiration cytology (Bethesda III, IV, and V) (30). Zhang et al. (30) suggested that using of the SWE and TI-RADS based on 2017 ACR together might contribute to diagnostic accuracy and sensitivity and help determine benign and malignant thyroid nodules with uncertain fine-needle aspiration cytology. In another study, Xu et al. (31) suggested a similar idea that the combination of the SWE and 2017 ACR TI-RADS could increase the precision of the diagnosis. Bora Makal and Aslan (32) concluded that SWE was more successful than the 2017 version of ACR TI-RADS categorization in the distinction of benign and non-benign thyroid nodules. However, one of the limitations of the studies is retrospective planning. In a prospective study, “modified TI-RADS” [the combination of ACR TI-RADS, SWE and CEUS (contrast-enhanced ultrasound)] was recommended to provide a decrease in the number of biopsiesapplied on benign nodules (33). In our study, it was found that the non-benign nodules had higher SWE values, thus SWE had a diagnostic effect to differentiate between benign and non-benign thyroid nodules prior to FNAB. We suggest ACR TI-RADS + SWE or “modified TI-RADS” in the distinction of benign and non-benign thyroid nodules. Although elastosonography is exciting, it has some limitations, such as lack of standardization (34). Therefore, the presence of interventional radiologists with sufficient experience on the subject is significant.
Study Limitations
Our study has some limitations. First of all, measurement on SWE is an operator-dependent imaging tecnique. Also, we have a small sample size; but the design of our prospective study enhances the reliability of our outcomes.
Conclusion
Measurements with SWE on thyroid nodules that we suggested as a non-invasive and an easily approachable tecnique with the use of the ACR TI-RADS (2017 version) might be advantageous to differentiate benign and non-benign thyroid nodules prior to FNAB.
Ethics
Ethics Committee Approval: The local ethics committee of our hospital approved this prospective study (GEAH-KEK-2019/0057).
Informed Consent: Informed consent was obtained from each patient before the measurement and FNAB, and the study was conforming to the principles of the Declaration of Helsinki.
Peer-review: Externally peer reviewed.
Authorship Contributions
Surgical and Medical Practices: S.Ş.E., M.T., Concept: S.Ş.E., A.K., Design: S.Ş.E., M.T., A.K., Data Collection or Processing: S.Ş.E., M.T., Analysis or Interpretation: S.Ş.E., M.T., Literature Search: S.Ş.E., A.K., Writing: S.Ş.E., A.K.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.