Original Contribution
Diagnostic Performance Evaluation of Practice Guidelines, Elastography and Their Combined Results for Thyroid Nodules: A Multicenter Study

https://doi.org/10.1016/j.ultrasmedbio.2020.03.031Get rights and content

Abstract

The purpose of this multicenter study was to compare the differential diagnostic value of the 2015 American Thyroid Association (ATA) and 2017 American College of Radiology (ACR) practice guidelines and elastography in thyroid nodules. This study also investigated whether the diagnostic value of practice guidelines can be improved, and the unnecessary biopsy rate decreased in combination with elastography. A total of 498 thyroid nodules were evaluated using the ATA and the ACR guidelines. Strain elastography, acoustic radiation force impulse imaging and point-shear wave elastography were used to assess the nodules. The suspicious levels were downgraded or upgraded after combination and unnecessary biopsy rates were calculated, respectively. The diagnostic performance of the practice guidelines was better than that of elastography. The ACR guidelines had a lower unnecessary biopsy rate and similar diagnostic performance compared with the ATA guidelines. The unnecessary biopsy rates significantly decreased when the ACR guidelines were combined with elastography, but the rates did not decrease when the ATA guidelines were combined with elastography.

Introduction

Thyroid ultrasound (US) has been widely used in clinical applications to differentiate benign thyroid nodules from malignant nodules and is highly sensitive for these cases. The US characteristics of thyroid nodules have been widely studied (Moon et al. 2008). Because of the complex structure of thyroid nodules and the common features of benign and malignant cases, the diagnostic sensitivity of the individual US features for malignancy is only 27%–63% (Remonti et al. 2015). In 2015, the American Thyroid Association (ATA) proposed 5-tier risk stratification guidelines to estimate the risk of malignancy. The ATA guidelines demonstrated relatively high diagnostic performance and have been widely used in clinical research (Haugen et al. 2016). However, some atypical thyroid nodules, such as those with a US pattern of an isoechoic or hyperechoic solid or mixed cystic nodules with suspicious US characteristics, cannot be categorized according to the ATA guidelines (Ha et al. 2019). In 2017, the American College of Radiology (ACR) proposed a 5-tier quantitative scoring system based on various US features and risk levels (Tessler et al. 2017). Both guidelines have been widely used for the clinical diagnosis of thyroid nodules. The standard treatment for thyroid nodules has a relatively high accuracy for the evaluation of benign and malignant nodules according to their ultrasonic characteristics. Suspicious nodules are primarily treated using fine needle aspiration (FNA), which is the most accurate method of diagnosing thyroid carcinoma. The accurate evaluation of benign or malignant thyroid nodules via non-invasive diagnostic methods will guide the next treatment stages (follow-up, FNA or surgery), which is a concern in clinical research. Although FNA is a relatively efficient examination method for evaluating benign and malignant thyroid nodules, this diagnostic technique at times yields indeterminate results or a non-diagnosis, and repeated punctures or surgeries are needed to confirm the diagnosis, which may cause extra burden to patients (Polyzos and Anastasilakis 2009). Among the nodules recommended by the ATA/ACR guidelines to undergo FNA, some benign nodules also receive unnecessary biopsies, which is a problem that must be addressed (Koseoglu Atilla et al. 2018). Elastography, as a supplementary technique, has also been used for the differentiation of benign and malignant thyroid nodules (Xu et al. 2014; Dobruch-Sobczak et al. 2016; Bardet et al. 2017; Cosgrove et al. 2017; Gregory et al. 2018).

According to various tissue hardness evaluation methods, elastography is categorized as strain imaging and shear wave imaging. Strain imaging is a qualitative evaluation method, and shear wave imaging is a quantitative evaluation method. According to mechanical excitation, strain imaging is categorized as strain elastography (SE) (manual compression) and acoustic radiation force impulse (ARFI) imaging. The change in length during compression is measured using strain imaging, which qualitatively indicates the tissue stiffness. Harder tissues undergo smaller deformation during compression. Shear wave imaging includes the following: (i) point shear wave speed (SWS) measurements (which determine the average stiffness from a localized tissue region indicated by the size and position of a region of interest [ROI] box), (ii) SWS imaging (which generates a 2-D image of the stiffness over a larger tissue region, with generally improved resolution but decreased precision at each image pixel) and (iii) transient elastography. In this method, focused acoustic radiation force pushing pulses with a short duration generate shear waves within an organ of interest, and the speed of the shear waves propagating away from the pushing location is measured (Shiina et al. 2015). The induced SWS is related to the tissue stiffness, and stiffer tissues conduct faster shear waves. Quantitative and qualitative assessment methods have advantages and disadvantages. SE and ARFI imaging indicate the stiffness of the entire lesion and surrounding tissue but cannot provide specific stiffness values. Although point-wave sheer elastography (p-SWE) quantifies the stiffness of a specific lesion area, it cannot reflect the whole lesion's stiffness because the elasticity of lesions may vary in different locations. When the ROI is selected, the representative region should be chosen, and the liquefaction and calcification areas should be avoided. As recommended by the World Federation of Ultrasound in Medicine and Biology guidelines, elastography—as a supplementary method for determining benign and malignant thyroid nodules—is useful for selecting suspicious thyroid nodules for surgery. Relevant studies have also reported that elastography can aid the diagnosis of FNA (Cantisani et al. 2016). Earlier studies of elastography for differentiating thyroid nodule malignancy and benignity demonstrated that stiffer nodules have a higher likelihood of malignancy than softer nodules (Bojunga et al. 2012). Studies reported that combining elastography with conventional US findings and the elastic index demonstrated increased sensitivity and the area under the receiver operating characteristic (AUROC) curve for predicting the malignancy of thyroid nodules. However, debate remains regarding elastography and the combination method's diagnostic efficiency for detecting malignant thyroid nodules (Zhang et al., 2014a, Zhang et al., 2014b; Park et al. 2015; Samir et al. 2015; Dobruch-Sobczak et al. 2016; Gregory et al. 2018).

The first purpose of this study was to compare the differential diagnostic value of the ATA and ACR practice guidelines and elastography in thyroid nodules. The second was to investigate whether the diagnostic value of the practice guidelines can be improved, and the unnecessary biopsy rate decreased after combination with elastography.

Section snippets

Patients

This retrospective study was approved by the local ethics committee, and informed consent was obtained from all patients. From May 2011 to May 2015, a total of 712 patients were included in this study. The 4 centers were involved were as follows: the Department of Medical Ultrasound, Shanghai Tenth People's Hospital (center 1, n = 548); the Department of Ultrasound, Affiliated Hospital of Guangdong Medical University (center 2, n = 92); the Department of Medical Ultrasound, Central Theater

Patient and pathologic data

A total of 498 nodules were included in this study, there were 99 males and 399 females (demographic data presented in Table 1). The final pathologic and cytologic results were as follows: 312 benign cases, among which 232 were confirmed by surgical pathology; 2 cases were confirmed as subacute thyroiditis; 37 cases were confirmed as adenoma; 18 cases were confirmed as Hashimoto's disease; 174 cases were confirmed as nodular goiter; and 1 case was confirmed as a parathyroid adenoma. A total of

Discussion

Research has reported that the diagnostic efficiency of elastography alone is lower than that of gray-scale US (Unluturk et al. 2012). Studies on the identification of benign and malignant thyroid nodules, using this method, have indicated that its diagnostic accuracy is limited with many restrictions based on certain conditions, including minimizing pre-compression, assessing the ROI size and positions, avoiding regions with artifacts, gross calcifications, or cystic areas and instructing

Conclusion

Elastography had diagnostic efficiency for detecting thyroid cancer, but the combination did not significantly improve the diagnostic performance for detecting thyroid carcinoma. The combination of the ACR guidelines and elastography decreased the unnecessary biopsy rate, but it did not decrease in combination with the ATA guidelines.

Conflict of Interest disclosure

The authors have no conflicts of interest to disclose.

References (29)

  • V. Cantisani et al.

    Strain ratio ultrasound elastography increases the accuracy of colour-Doppler ultrasound in the evaluation of Thy-3 nodules. A bi-centre university experience

    Eur Radiol

    (2016)
  • E.S. Cibas et al.

    NCI Thyroid FNA State of the Science Conference. The Bethesda system for reporting thyroid cytopathology

    Am J Clin Pathol

    (2009)
  • D.S. Cooper et al.

    Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer

    Thyroid

    (2009)
  • S.M. Ha et al.

    Diagnostic performance of practice guidelines for thyroid nodules: Thyroid nodule size versus biopsy rates

    Radiology

    (2019)
  • View full text