Original Contribution
Real-Time Elastography in the Evaluation of Diffuse Thyroid Disease: A Study Based on Elastography Histogram Parameters

https://doi.org/10.1016/j.ultrasmedbio.2014.03.033Get rights and content

Abstract

The purpose of this study was to evaluate the diagnostic performance of quantitative histogram parameters using real-time tissue elastography (RTE) in the diagnosis of patients with diffuse thyroid disease. One hundred and sixteen patients (mean age, 43.7 ± 10.97 y) who had undergone pre-operative staging ultrasonography and RTE were included. For each patient, 11 parameters were obtained from RTE images, from which the “elastic index” was calculated. Diagnostic performance of the elastic index and that of the 11 parameters on RTE were calculated and compared. Of the 116 patients, 31 had diffuse thyroid disease and 85 had normal thyroid parenchyma. Area under the receiver operating characteristic curve (Az) of MEAN (average relative value) elasticity was high (0.737), without significant differences from other elasticity values. Diagnostic performance of the elastic index was higher than the MEAN, Az = 0.753, without significance (p = 0.802). In conclusion, RTE using the elastic index was found to have good diagnostic performance and may be useful in the diagnosis and management of patients with diffuse thyroid disease.

Introduction

Ultrasonography (US) is a tolerable and easily applicable diagnostic method that has been used worldwide in the differential diagnosis of thyroid nodules. US mostly represents the morphologic features of the targeted thyroid mass, and studies have reported high levels of performance with combinations of gray-scale US features in the diagnosis of thyroid nodules (Kim et al., 2002, Moon et al., 2008, Papini et al., 2002). In addition to gray-scale US, there has recently emerged elastography, which provides an estimate of tissue stiffness by measuring the degree of displacement secondary to external compression applied (Asteria et al., 2008, Itoh et al., 2006, Rago et al., 2007). Based on the concept that malignant masses are relatively stiff compared with the adjacent normal parenchyma (Itoh et al., 2006, Yoon et al., 2011), elastography has been used mostly in the differentiation between benign and malignant nodules within the thyroid (Asteria et al., 2008, Dighe et al., 2008, Ding et al., 2012, Hong et al., 2009, Moon et al., 2012, Rago et al., 2007). In addition to evaluation of masses, elastography has been reported to have a role in evaluating the extent of inflammatory infiltrations and fibrosis within the diseased parenchyma (Friedrich-Rust et al., 2007, Sandrin et al., 2003, Wang et al., 2012).

Real-time elastography (RTE) visualizes the elasticity of the target area by capturing echo signals derived secondary to repetitive compressions applied with a freehand technique (Kanamoto et al., 2009, Wang et al., 2012, Yoon et al., 2011). A quantitative analysis method based on RTE has been developed that can be used on diffuse pathologic lesions, such as liver fibrosis, as reported in several studies (Friedrich-Rust et al., 2007, Kanamoto et al., 2009, Wang et al., 2012). Results of these studies have proved that the quantitative data obtained from RTE constitute a useful, non-invasive imaging method with promising results in the assessment of liver fibrosis in patients with chronic hepatitis. Similarly, thyroid stiffness assessment has been used in the diagnosis of diffuse thyroid gland pathology with acoustic radiation force impulse elastography (Sporea et al. 2011b) or shear wave elastography (Kim et al. 2014) in several recent studies, with promising results. Quantitative parameters obtained from RTE may also be helpful in the diagnosis and management of patients with diffuse thyroid disease (DTD). The 11 elastography parameters used in RTE, which are calculated by maneuvering the ultrasound unit, each display characteristics related to the elastography image, factors related to the elastography histogram and factors related to the gray-scale co-occurrence matrix. Also, the “elastic index,” calculated on the basis of the 11 parameters derived from RTE by principal component analysis used in the diagnosis of liver fibrosis (Wang et al. 2012), may be applied to the thyroid gland. Although RTE parameters, including elastic index, have been used to evaluate fibrosis in liver parenchyma, to our knowledge, there are no other studies reporting the application of quantitative RTE parameters and the elastic index to the diagnosis of DTD. On this basis, we evaluated the diagnostic performance of quantitative histogram parameters using high-technology RTE in the diagnosis of patients with DTD.

Section snippets

Methods

Informed written consent was obtained from all patients. The study was performed in accordance with the ethics guidelines of the Helsinki Declaration and was approved by the institutional review board of Severance Hospital, Seoul, Korea.

Results

According to the pathology results, 31 patients were positive for DTD, and the remaining 85 patients were negative for DTD. Mean ages of the patients with (46.8 ± 11.0 y) and patients without (42.3 ± 10.3 y) DTD did not significantly differ (p = 0.054). The ratio of men to women among patients with DTD (2/29) did not significantly differ from that among patients without DTD (19/66) (p = 0.681). The mean interval between pre-operative staging US and thyroid surgery was 0.9 mo (range, 0.1–2.3 mo).

Discussion

Gray-scale US features that are known to be useful in the evaluation of DTD are decreased or increased parenchymal echogenicity, coarse-appearing echotexture, changes in vascularity, decreased or increased anteroposterior diameter of each gland, marginal nodularity and the presence of scattered microcalcifications (Kim et al. 2010). Although the aforementioned US features may suggest the presence of DTD, clinical or laboratory features are more accurate and generally used in the evaluation and

Conclusions

We evaluated the usefulness of RTE, including the elastic index, in the differential diagnosis of thyroid nodules. The results indicated that the MEAN and elastic index performed well and comparably. On the basis of our results, we conclude that RTE using the elastic index performs well diagnostically and may be useful in the diagnosis and management of patients with diffuse thyroid disease.

References (28)

  • Y. Hong et al.

    Real-time ultrasound elastography in the differential diagnosis of benign and malignant thyroid nodules

    J Ultrasound Med

    (2009)
  • A. Itoh et al.

    Breast disease: Clinical application of US elastography for diagnosis

    Radiology

    (2006)
  • M. Kanamoto et al.

    Real time elastography for non-invasive diagnosis of liver fibrosis

    J Hepatobiliary Pancreat Surg

    (2009)
  • D.W. Kim et al.

    Sonographic differentiation of asymptomatic diffuse thyroid disease from normal thyroid: A prospective study

    AJNR Am J Neuroradiol

    (2010)
  • Cited by (23)

    • Texture Analysis to Differentiate Malignant Renal Tumors in Children Using Gray-Scale Ultrasonography Images

      2019, Ultrasound in Medicine and Biology
      Citation Excerpt :

      Radiomics data are now widely and rapidly adopted in radiology, including characterization of the lesions, building prognostic significance and validation (Aerts et al. 2014). This approach has been studied in various organs in adults, including thyroid, lymph nodes, brain, liver, lung and kidneys (Choi et al. 2015; Daginawala et al. 2016; Kim et al. 2015; Lubner et al. 2016; Nam et al. 2016; Peikert et al. 2018; Yoon et al. 2014). However, there are only a few studies that have evaluated the usefulness of radiomics in pediatric diseases (Alizadeh et al. 2018; Vajapeyam et al. 2018; You et al. 2015), and no studies have assessed pediatric renal tumors.

    • A Comparison Study of Real-Time Ultrasound Elastography and Electron Microscopy for the Assessment of Liver Damage Induced by Brain Death

      2019, Ultrasound in Medicine and Biology
      Citation Excerpt :

      There were five continuous and stable troughs, one of which was selected for observation of the elastogram. Eleven quantitative parameters were obtained, including the mean value of the relative strain within the ROI (MEAN); the standard deviation of the relative strain ratio within the ROI (SD); the area of the low-strain region (%AREA); the complexity of each low-strain area (boundary length2/area) (COMP); kurtosis (KURT), which evaluates if the histogram distribution can be concentrated into an average value; a scale of asymmetry that indicates the degree in which the histogram is skewed (SKEW); the feature value of the textual variations (the greater the resolution, the greater the contrast) (CONT); the feature value of the textual randomness, entropy (ENT); the feature value of the textual homogeneity, the inverse difference moment (IDM); the feature value of textual uniformity, the angular second moment (ASM); the directivity of the texture, reflecting the degree of similarity of the texture in the row and column directions, correlation (CORR) (Shi et al. 2014; Yoon et al. 2014). The average of five repeated measurements was taken as the final result.

    • WFUMB Guidelines and Recommendations on the Clinical Use of Ultrasound Elastography: Part 4. Thyroid

      2017, Ultrasound in Medicine and Biology
      Citation Excerpt :

      They had a classification accuracy of 93.6% in differentiating between benign and malignant nodules. Histogram analysis of the color elastograms (elasticity index and mean) were used for diagnosis and management of patients with diffuse thyroid disease (Cantisani et al. 2015a, 2015b; Yoon et al. 2014), and another study indicated excellent inter-observer agreement (Lim et al. 2012). Recommendation 5.

    • Histogram and gray level co-occurrence matrix on gray-scale ultrasound images for diagnosing lymphocytic thyroiditis

      2016, Computers in Biology and Medicine
      Citation Excerpt :

      To overcome these limitations, there have been many quantitative studies that have focused on diagnosing DTD with USE (US elastography) as well as gray-scale US [1,8,9,11,20–22] (Table 5) and both methods have shown good diagnostic performances. The Az value of the mean was 0.737 in a study using USE [9] and 0.8 [21] in a study using conventional US combined with USE utilizing shear waves. The study which investigated and compared the diagnostic performance of conventional US and USE using shear waves to diagnose DTD revealed that USE using shear waves had a higher Az value (0.619) than conventional gray-scale US (Az=0.585) [8].

    • Variability in Interpretation of Ultrasound Elastography and Gray-Scale Ultrasound in Assessing Thyroid Nodules

      2016, Ultrasound in Medicine and Biology
      Citation Excerpt :

      Gray-scale ultrasound (US) is the most sensitive test currently available for detecting thyroid lesions; however, differentiation of benign and malignant nodules is not highly accurate with gray-scale ultrasound (Takashima et al. 1995), and thus, its diagnostic value varies considerably from study to study (Fish et al. 2008; Frates et al. 2006; Kim et al. 2002; Kovacevic and Skurla 2007; Lim et al. 2012). Ultrasound elastography (USE) enables the assessment of tissue consistency by differentiating stiff nodules from soft nodules, and it supplements the diagnostic limitations of gray-scale US (Asteria et al. 2008; Azizi et al. 2013; Bamber et al. 2013; Cantisani et al. 2014; Hong et al. 2009; Kagoya et al. 2010; Kim et al. 2014; Mehrotra et al. 2013; Moon et al. 2012; Rago et al. 2007; Rubaltelli et al. 2009; Shuzhen 2012; Shweel and Mansour 2013; Trimboli et al. 2012; Unluturk et al. 2012; Yoon et al. 2014). Previous studies suggested that with respect to diagnostic performance, USE is better or comparable to gray-scale US when differentiating benign from malignant thyroid nodules (Asteria et al. 2008; Azizi et al. 2013; Cantisani et al. 2014; Hong et al. 2009; Rago et al. 2007; Shuzhen 2012; Trimboli et al. 2012).

    View all citing articles on Scopus
    View full text