|Year : 2018 | Volume
| Issue : 7 | Page : 1567-1571
The clinical significance of ultrasound grayscale ratio in differentiating markedly hypoechoic and anechoic minimal thyroid nodules
Zhi-Kai Lei1, Ming-Kui Li2, Ding-Cun Luo3, Zhi-Jiang Han4
1 Department of Ultrasonic Imaging, The First Peoples Hospital of Hangzhou City, Hangzhou, China
2 Department of Ultrasound Intervention, Zhejiang Xiaoshan Hospital, Hangzhou, China
3 Department of Oncology Surgery, The First Peoples Hospital of Hangzhou City, Hangzhou, China
4 Department of Radiology, The First Peoples Hospital of Hangzhou City, Hangzhou, China
|Date of Web Publication||19-Dec-2018|
Department of Radiology, The First Peoples Hospital of Hangzhou City, Hangzhou 310000
Source of Support: None, Conflict of Interest: None
Purpose: This study explored ultrasound grayscale ratios (USGRs) for differentiating markedly hypoechoic and anechoic minimal thyroid nodules.
Materials and Methods: Longitudinal scan images of 193 markedly hypoechoic papillary thyroid microcarcinoma (PTMC) lesions from 184 patients were retrospectively reviewed using RADinfo and compared with 123 anechoic micronodular goiters (MNGs) from 110 patients. Final diagnosis was validated by pathological examination; MNGs predominantly manifested with cyst formation. Grayscale values of PTMC, MNG, and normal surrounding tissues were obtained from grayscale histograms; USGRs (grayscale ratios of pathologic tissue to surrounding normal tissue) of PTMC and MNG were calculated. Optimal USGRs for differentiating PTMC and MNG were determined with receiver operating characteristic (ROC) curves.
Results: Among 193 PTMC and 123 MNG lesions, USGRs were 0.24–0.51 (mean ± standard deviation [SD]: 0.41 ± 0.07) and 0.01–0.38 (mean ± SD: 0.12 ± 0.08), respectively. The area under the ROC curve for distinguishing markedly hypoechoic PTMC and anechoic MNG was 0.992. As USGRs decreased, sensitivity decreased and specificity increased for MNG diagnosis. At a USGR of 0.26, the Youden index was high (0.933), corresponding to 94.3% sensitivity and 99% specificity for predicting anechoic MNG. At a USGR of 0.23, sensitivity and specificity for diagnosing anechoic MNG were 92.7% and 100%, respectively. In contrast, as USGR increased, sensitivity decreased and specificity increased for predicting PTMC. At a USGR of 0.38, sensitivity and specificity for diagnosing markedly hypoechoic PTMC were 68.4% and 100%, respectively.
Conclusions: USGRs could objectively quantize grayscale values of markedly hypoechoic and anechoic lesions, enabling accurate and quantitative determination of nodular properties.
Keywords: Nodular goiters, papillary carcinoma, thyroid nodules, Ultrasound grayscale ratio (USGR)
|How to cite this article:|
Lei ZK, Li MK, Luo DC, Han ZJ. The clinical significance of ultrasound grayscale ratio in differentiating markedly hypoechoic and anechoic minimal thyroid nodules. J Can Res Ther 2018;14:1567-71
|How to cite this URL:|
Lei ZK, Li MK, Luo DC, Han ZJ. The clinical significance of ultrasound grayscale ratio in differentiating markedly hypoechoic and anechoic minimal thyroid nodules. J Can Res Ther [serial online] 2018 [cited 2020 Jan 23];14:1567-71. Available from: http://www.cancerjournal.net/text.asp?2018/14/7/1567/247710
| > Introduction|| |
Thyroid lesions are divided into five categories by comparison of their echogenicity with that of cervical strap muscle and thyroid: Anechoic, markedly hypoechoic (lower than cervical strap muscle), hypoechoic (between cervical strap muscle and thyroid parenchyma), isoechoic (similar to the thyroid parenchyma), and hyperechoic (higher than thyroid parenchyma).,,,,, The diagnostic value of “markedly hypoechoic” has been widely accepted in papillary thyroid microcarcinoma (PTMC).,,,,, In contrast, anechoic micronodules are typically indicative of benign lesions, especially for micronodular goiters (MNGs). Treatment regimens are distinct for PTMC and MNG; thus, correct diagnosis is essential.
During ultrasound examination of the thyroid gland, the examiner applies various scan techniques and scan gains to produce resulting grayscale maps; unlike computed tomography scans, there is no method to evaluate hypoechoic and anechoic nodules. Thus, the grayscale ratios of PTMC and MNG are relatively stable, despite different scan techniques and gain adjustments. In this study, we propose the concept of ultrasound grayscale ratio (USGR) for differential diagnosis of PTMC and MNG. This method is feasible, objective, and precise.
| > Materials and Methods|| |
The Institutional Review Board approved this retrospective study, and the requirement to obtain informed consent was waived. Ultrasonography scans of 2884 patients, taken between January 2012 and December 2016, were retrospectively reviewed. Exclusion criteria were included in the study: (1) Lesions >10 mm in size, (2) hypoechoic, isoechoic, and hyperechoic nodules, (3) nodules with remarkable calcification that precluded the measurement of soft tissues, and (4) thyroid nodules with complicating Hashimoto thyroiditis. Finally, this study enrolled 294 patients with 316 modules, including 60 men and 234 women, with an average age of 52 years [Figure 1].
Ultrasonic examination of thyroid lesions was performed using one of the following ultrasonic diagnostic scanners: MyLab 70 XVG (Genova, Italy), Esaote MyLab Classic C (Genova, Italy), and Esaote MyLab 90 (Genova, Italy). Broadband linear array probes (5–10 MHz) were used for this study, and the central frequency was 7.5 MHz; we adopted this central frequency for routine scanning. Patients were placed in a supine position, exposing the anterior thyroid, after a backward tilt of the head; then, transverse, longitudinal, and oblique scans were performed. Each thyroid nodule was described in terms of number, size, shape, boundary, surrounding echo halo, inner echo, calcification rate, internal and peripheral blood supply of the nodule, as well as the presence of bilateral cervical lymph nodes.
The US data, selected from the picture archiving and communication systems database, were analyzed without pathological diagnosis information by two radiologists with 15 years of experience. Nodules primarily exhibiting anechoic and markedly hypoechoic lesions (occupying >50%) were analyzed. The measured area and size of the region of interest (ROI) in nodules and surrounding normal thyroid tissues were determined. When transverse scans were performed, normal thyroid tissue was insufficient for comparison with normal tissue in superior and inferior thyroid lesions; therefore, longitudinal scans were performed. Regarding ROI measurements, the maximal ROI region was acquired for each lesion with homogenous echo [Figure 2] and [Figure 3]; for heterogeneous lesions, the main echo and transverse grayscale were measured in the same region [Figure 4].
|Figure 2: Measurement of the ultrasound grayscale ratio value for anechoic micronodular goiter micronodular goiter with homogenous echo. (a) The grayscale value of micronodular goiter was 4.99. (b) The grayscale value of normal thyroid tissue in the same transverse level was 38.37, and the ultrasound grayscale ratio was 0.13|
Click here to view
|Figure 3: Measurement of the ultrasound grayscale ratio value for papillary thyroid microcarcinoma with homogenous echo. (a) The grayscale value of papillary thyroid microcarcinoma was 38.48. (b) The grayscale value of normal thyroid tissue in the same transverse level was 86.12, and the ultrasound grayscale ratio was 0.45|
Click here to view
|Figure 4: Measurement of the ultrasound grayscale ratio value for anechoic micronodular goiter with heterogeneous echo. (a) The grayscale value of micronodular goiter was 2.49. (b) The grayscale value of normal thyroid tissue in the same transverse level was 66.33, and the ultrasound grayscale ratio was 0.04|
Click here to view
SPSS19 statistical analysis software (SPSS Inc., Chicago, IL, USA) was used to generate receiver operating characteristic (ROC) curves for differentiating PTMC from MNG, with sensitivity as the vertical axis and specificity as the horizontal axis. As the curve approached the upper left corner, a better differentiation capability was indicated. The USGR threshold was determined by comparing the Youden index.
| > Results|| |
The USGRs of 193 markedly hypoechoic PTMC lesions ranged from 0.24 to 0.51 (mean ± standard deviation [SD]: 0.41 ± 0.07), of which USGRs of 0.2–0.3, 0.31–0.4, and ≥0.41 occurred in 19, 53, and 121 lesions, respectively. The USGRs of 123 anechoic MNG lesions ranged from 0.01 to 0.38 (mean ± SD: 0.12 ± 0.08), of which USGRs of 0–0.2, 0.21–0.3, and >0.31 occurred in 107, 12, and 4 lesions, respectively [Table 1].
|Table 1: Ultrasound grayscale ratio values of markedly hypoechoic papillary thyroid microcarcinom and anechoic micronodular goiters|
Click here to view
The area under the ROC curve (AUC) for distinguishing markedly hypoechoic PTMC and anechoic MNG was 0.992 (95% confidence interval: 0.984–0.999) [Figure 5]. When the USGR decreased, sensitivity decreased and specificity increased for diagnosing MNG; at a USGR of 0.26, the Youden index was the largest (0.933), which corresponds to a sensitivity of 94.3% and specificity of 99%, for predicting anechoic MNG. At a USGR of 0.23, sensitivity and specificity for diagnosing anechoic MNG were 92.7% and 100%, respectively. In contrast to MNG, as USGR increased, sensitivity decreased and specificity increased for predicting PTMC. At a USGR of 0.38, sensitivity and specificity for diagnosing markedly hypoechoic PTMC were 68.4% and 100%, respectively.
|Figure 5: Receiver operating characteristic curve of the ultrasound grayscale ratio value for distinguishing between markedly hypoechoic papillary thyroid microcarcinoma papillary thyroid microcarcinoma and anechoic micronodular goiter|
Click here to view
| > Discussion|| |
In ultrasonography, typical hypoechoic PTMC and anechoic MNG are easily identified because of their characteristic radiographic signs: The former presents with irregular nodules, a longitudinal–transverse axis ratio of ≥1, microcalcification, and blood flow within the lesion on color Doppler flow imaging., The latter exhibits round nodules with a uniform echo and internal blood flow. Atypical hypoechoic PTMC and anechoic MNG overlap and present similar ultrasound signs such as irregular nodules, longitudinal–transverse axis ratios of ≥1, absence of blood flow within the lesions, and a lack of microcalcification. Ultrasonography is an effective method to detect blood flow signals with nodules; however, it is expensive, complicated, and difficult to popularize in clinical practice. Elastography is not accurate for small nodules, especially in cases of deep nodules. In addition, this technique requires experienced operators and high-quality instruments. Therefore, it is of great interest to improve the differential diagnosis of hypoechoic PTMC and anechoic MNG by integrating an accurate and quantitative assessment of echo levels, combined with other ultrasound signs and scanning methods.
With echo intensities of thyroid nodules ranging from weak to strong, the corresponding sonogram grayscale ranges from black to white. Currently, echo intensities of thyroid nodules are primarily determined by the subjective judgment of an examiner's naked eyes and are subsequently categorized into five levels: anechoic, markedly hypoechoic, hypoechoic, isoechoic, and hyperechoic, varying from black to white across grayscales.,,, Markedly hypoechoic lesions are an important sign of PTMC, which may be associated with a low degree of cancer cell differentiation, fewer interstitial components, and good sound transmission in the tumor. Although other scholars have reported high sensitivities of markedly hypoechoic lesions for diagnosing PTMC, these vary greatly. For example, Hong et al., Moon et al., and Kim et al. reported specificities ranging from 90.9% to 94.9% and sensitivities were distinguished from 24.5% to 41.4% this variation was associated with long distances between nodules and muscles, which resulted in differing assessments among examiners. Anechoic lesions are typically filled with liquid, and the underlying mechanisms involve acoustic waves that pass completely through the lesion. Although large papillary thyroid carcinomas also exhibit anechoic lesions that are caused by necrosis and cyst formation,, PTMC rarely displays cyst formation;, therefore, the appearance of cyst formations in microlesions is typically predicative of MNG. Although fine-needle aspiration biopsy is considered the gold standard for the diagnosis of thyroid lesions, biopsy failures still occur; one of the main causes for these failures is cyst formation., Thus, differential diagnosis of anechoic and markedly hypoechoic lesions remains difficult for clinicians and ultrasound doctors. Our results demonstrate that the USGR reflects the echo-intensity of thyroid lesions, which obviates subjective examiner variation and facilitates accurate diagnosis of markedly hypoechoic PTMC and anechoic MNG.
In 2013, Erol et al. demonstrated grayscale value histograms of normal and pathologic fat lobules in mammary glands; the ratio of former to the latter was termed the lesion echogenicity ratio (LER), which was higher in malignant lesions than in benign lesions. The determination of mammary LER has several limitations that are influenced by specific physiological factors, including menstrual cycle and age. In addition, the disperse distribution of fat around lesions leads to variable results in different regions. This disadvantage would not occur in thyroid tissue, and normal or relatively normal tissue could function as a control to reduce variation among examiners and allow more objective results. Our results demonstrated that USGRs in markedly hypoechoic PTMC were higher than in anechoic MNG (0.24–0.51 and 0.01–0.38, respectively). The AUC of USGR for differentiating markedly hypoechoic PTMC and anechoic MNG was 0.992. At a USGR of 0.26, the Youden index was highest (0.993), yielding 94.3% sensitivity and 99% specificity for differentiating anechoic MNG. At a USGR of 0.23, the sensitivity and specificity of differentiating anechoic MNG were 92.7% and 100%, respectively. Therefore, with decreasing USGRs, sensitivity decreased and specificity increased for diagnosing MNG; this is important to reduce unnecessary fine-needle aspiration biopsies and surgical traumas. Theoretically, water-like liquid is completely anechoic and the USGR should be 0; however, in this cohort, no USGRs reached 0, and there were 12 and 4 lesions with USGRs of 0.2–0.3 and >0.3, respectively. We suspect that this occurred because the cyst fluid mixed with protein, blood, glial factors, fibrosis debris, and other solid components that increased USGR and confound the differential diagnosis of markedly hypoechoic PTMC. Therefore, for accurate diagnoses, other image features should be considered. In addition, when USGR values exceeded 0.38, the specificity for the diagnosis of markedly hypoechoic PTMC was 100%; this confirmatory diagnostic value is of great significance for efficient clinical treatment.
There are several limitations to this study. First, the grayscale histogram was adopted to quantize PTMC and MNG; this procedure was complex and required dedicated software. We are actively researching and developing a simplified and feasible method for grayscale measurement, to facilitate its use with ultrasound apparatuses. Second, for heterogeneous lesions, variations in ROI selection may exist; cooperation between two or more experienced radiologists could reduce such variation. Third, during ultrasound scanning, differences in focusing and depth gain also result in variation; however, when we measured the ROI of lesions and surrounding normal tissues at the same level, the images displayed sufficient quality to minimize variability. The five-level classification system of echo intensity was also based on a variety of gains. Fourth, the results of this study were mainly derived from the use of three Esaote ultrasound systems; thus, whether these results are applicable to different ultrasound systems (e.g., those from other companies) requires further controlled studies. Finally, selection bias inevitably existed in this retrospective study.
In summary, USGR could qualify the grayscale value of markedly hypoechoic and anechoic lesions objectively and accurately and could quantitatively determine nodular properties. This is important to reduce unnecessary needle aspiration biopsies and surgical traumas.
This research is supported by 2013 Hangzhou Major Science and Technology Innovation Project (20131813A08), Zhejiang Province Medical and Health Plan Project (2015KYB293), Zhejiang Provincial Education Department Project (Y201636958), and Zhejiang Public Service Technology Application Project (2017C33180).
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Sharma R, Verma N, Kaushal V, Sharma DR, Sharma D. Diagnostic accuracy of fine-needle aspiration cytology of thyroid gland lesions: A study of 200 cases in Himalayan belt. J Cancer Res Ther 2017;13:451-5.
Shah AA, Jain PP, Dubey AS, Panjwani GN, Shah HA. A study of clinicopathological characteristics of thyroid carcinoma at a tertiary care center. J Cancer Res Ther 2018;14:357-60.
Selemetjev S, Savin S, Paunovic I, Tatic S, Cvejic D. Concomitant high expression of survivin and vascular endothelial growth factor-C is strongly associated with metastatic status of lymph nodes in papillary thyroid carcinoma. J Cancer Res Ther 2018;14:S114-9.
Cosgrove D, Barr R, Bojunga J, Cantisani V, Chammas MC, Dighe M, et al.
WFUMB guidelines and recommendations on the clinical use of ultrasound elastography: Part 4. Thyroid. Ultrasound Med Biol 2017;43:4-26.
Cantisani V, Grazhdani H, Drakonaki E, D'Andrea V, Di Segni M, Kaleshi E, et al.
Strain US elastography for the characterization of thyroid nodules: Advantages and limitation. Int J Endocrinol 2015;2015:908575.
Park JY, Lee HJ, Jang HW, Kim HK, Yi JH, Lee W, et al.
A proposal for a thyroid imaging reporting and data system for ultrasound features of thyroid carcinoma. Thyroid 2009;19:1257-64.
Cheng SP, Lee JJ, Lin JL, Chuang SM, Chien MN, Liu CL, et al.
Characterization of thyroid nodules using the proposed thyroid imaging reporting and data system (TI-RADS). Head Neck 2013;35:541-7.
Kim GR, Kim MH, Moon HJ, Chung WY, Kwak JY, Kim EK, et al.
Sonographic characteristics suggesting papillary thyroid carcinoma according to nodule size. Ann Surg Oncol 2013;20:906-13.
Jeh SK, Jung SL, Kim BS, Lee YS. Evaluating the degree of conformity of papillary carcinoma and follicular carcinoma to the reported ultrasonographic findings of malignant thyroid tumor. Korean J Radiol 2007;8:192-7.
Kwak JY, Han KH, Yoon JH, Moon HJ, Son EJ, Park SH, et al.
Thyroid imaging reporting and data system for US features of nodules: A step in establishing better stratification of cancer risk. Radiology 2011;260:892-9.
Horvath E, Majlis S, Rossi R, Franco C, Niedmann JP, Castro A, et al.
An ultrasonogram reporting system for thyroid nodules stratifying cancer risk for clinical management. J Clin Endocrinol Metab 2009;94:1748-51.
Cappelli C, Castellano M, Pirola I, Gandossi E, De Martino E, Cumetti D, et al.
Thyroid nodule shape suggests malignancy. Eur J Endocrinol 2006;155:27-31.
Choi YJ, Kim SM, Choi SI. Diagnostic accuracy of ultrasound features in thyroid microcarcinomas. Endocr J 2008;55:931-8.
Zhang XL, Qian LX. Ultrasonic features of papillary thyroid microcarcinoma and non-microcarcinoma. Exp Ther Med 2014;8:1335-9.
Hong YJ, Son EJ, Kim EK, Kwak JY, Hong SW, Chang HS, et al.
Positive predictive values of sonographic features of solid thyroid nodule. Clin Imaging 2010;34:127-33.
Moon WJ, Jung SL, Lee JH, Na DG, Baek JH, Lee YH, et al.
Benign and malignant thyroid nodules: US differentiation – Multicenter retrospective study. Radiology 2008;247:762-70.
Sharma A, Gabriel H, Nemcek AA, Nayar R, Du H, Nikolaidis P, et al.
Subcentimeter thyroid nodules: Utility of sonographic characterization and ultrasound-guided needle biopsy. AJR Am J Roentgenol 2011;197:W1123-8.
Wang Y, Li L, Wang YX, Feng XL, Zhao F, Zou SM, et al.
Ultrasound findings of papillary thyroid microcarcinoma: A review of 113 consecutive cases with histopathologic correlation. Ultrasound Med Biol 2012;38:1681-8.
Choi YS, Hong SW, Kwak JY, Moon HJ, Kim EK. Clinical and ultrasonographic findings affecting nondiagnostic results upon the second fine needle aspiration for thyroid nodules. Ann Surg Oncol 2012;19:2304-9.
Choi SH, Baek JH, Lee JH, Choi YJ, Hong MJ, Song DE, et al.
Thyroid nodules with initially non-diagnostic, fine-needle aspiration results: Comparison of core-needle biopsy and repeated fine-needle aspiration. Eur Radiol 2014;24:2819-26.
Erol B, Kara T, Gürses C, Karakoyun R, Köroğlu M, Süren D, et al.
Gray scale histogram analysis of solid breast lesions with ultrasonography: Can lesion echogenicity ratio be used to differentiate the malignancy? Clin Imaging 2013;37:871-5.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]