Usefulness of semiquantitative elastography in predicting ...

Radiología. 2016;58(5):366---372

www.elsevier.es/rx

ORIGINAL REPORT

Usefulness of semiquantitative elastographyin predicting malignancy in thyroid nodules!

C. Franco Uliaquea,∗, F.J. Pardo Berdúna, R. Laborda Herrerob, C. Pérez Lórenza

a Servicio de Radiodiagnóstico, Centro Médico de Especialidades Ramón y Cajal, Hospital Universitario Miguel Servet, Zaragoza,

Spainb Centro Universitario de la Defensa, Zaragoza, Spain

Received 4 January 2016; accepted 6 May 2016

KEYWORDSThyroid nodule;Elastography;Biopsy

Abstract

Objective: To retrospectively review the diagnostic capacity of semiquantitative elastographyin differentiating between benign and malignant thyroid nodules.Patients and methods: We analyzed 314 thyroid nodules in 295 consecutive patients referredto the endocrinology department for cytological study, studying all by conventional ultra-sonography, elastography, and fine-needle aspiration cytology (FNAC). Using a semiquantitativeelastography system that portrays tissue stiffness through a color map, we designed our ownclassification system for thyroid nodules based on their characteristics on elastography. We clas-sified nodules into three groups: predominantly soft, predominantly stiff, and mosaic patterned.We used logistic regression analysis to investigate the relation between elastography and thyroidcancer.Results: We obtained a definite diagnosis of malignancy after surgery in 19 nodules, of whichon elastography 8 had the mosaic pattern, 6 were predominantly stiff, and 5 were predomi-nantly soft. We found no significant association between the pattern on elastography and theprobability of malignancy in any of the models.Conclusion: According to our study, a probability of malignancy in a thyroid nodule is not relatedto the findings at elastography. Therefore, semiquantitative elastography as used in this studycannot obviate FNAC.© 2016 SERAM. Published by Elsevier Espana, S.L.U. All rights reserved.

! Please cite this article as: Franco Uliaque C, Pardo Berdún FJ, Laborda Herrero R, Pérez Lórenz C. Utilidad de la elastografía semicuan-titativa para predecir la malignidad de los nódulos tiroideos. Radiología. 2016;58:366---372.

∗ Corresponding author.E-mail address: [email protected] (C. Franco Uliaque).

2173-5107/© 2016 SERAM. Published by Elsevier Espana, S.L.U. All rights reserved.

Document downloaded from http://www.elsevier.es, day 04/11/2018. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited.Document downloaded from http://www.elsevier.es, day 04/11/2018. This copy is for personal use. Any transmission of this document by any media or format is strictly prohibited.

Usefulness of semiquantitative elastography in thyroid nodules 367

PALABRAS CLAVENódulo tiroideo;Elastografía;Biopsia

Utilidad de la elastografía semicuantitativa para predecir la malignidad de los nódulostiroideos

Resumen

Objetivo: Evaluar retrospectivamente la capacidad diagnóstica de la elastografía semicuanti-tativa para diferenciar entre nódulos tiroideos benignos y malignos.Pacientes y métodos: Se analizaron 314 nódulos tiroideos de 295 pacientes consecutivos remi-tidos por el Servicio de Endocrinología del Centro Médico de Especialidades para estudiocitológico; a todos ellos se les realizó ecografía convencional, elastografía y punción aspira-tiva con aguja fina. Disponemos de una elastosonografía semicuantitativa que mediante unmapa de colores refleja la elasticidad de los tejidos. Elaboramos una clasificación propia de losnódulos tiroideos basándonos en sus características de elastografía, distinguiendo tres grupos:predominantemente blandos, predominantemente rígidos y en mosaico. Se realizó un análisisestadístico mediante regresión logística para investigar la relación entre la elastografía y elcáncer de tiroides.Resultados: Obtuvimos un resultado definitivo de malignidad después de la cirugía en 19 nódu-los, de los cuales 8 presentaron un patrón en mosaico en la elastografía, 6 fueron predo-minantemente rígidos y 5 predominantemente blandos. En ninguno de los modelos estimadosobtuvimos que el patrón de elastografía analizado estuviera significativamente relacionado conla probabilidad de malignidad del nódulo tiroideo.Conclusión: La probabilidad de malignidad de un nódulo tiroideo, según nuestro estudio, no estárelacionada con los resultados de la elastografía. Por consiguiente, la elastografía semicuan-titativa, tal y como nosotros la hemos definido, no permite reducir el número de puncionesaspirativas con aguja fina realizadas.© 2016 SERAM. Publicado por Elsevier Espana, S.L.U. Todos los derechos reservados.

Introduction

The increase in the number of ultrasound exams of the neckhas led to an important rise in the number of diagnosed thy-roid nodules1,2 and consequently it has given rise to the needfor performing numerous fine needle aspirations (FNA) inorder to differentiate benign from malignant nodules sinceit is the modality of choice to this end. However, FNA is aninvasive procedure that does not provide immediate infor-mation and it is subjected to uncertainties or sample andanalysis errors,1,3,4 so tests are needed to determine whatnodules should be biopsied and which are susceptible offollow-up. The elastography could be that tool that wouldhelp us reduce the number of patients undergoing FNA.

Elastosonography is a non-invasive diagnostic modalitythat uses ultrasounds to provide an estimation of tissueconsistency while providing information about the internalstructure of the tissue by measuring the degree of deformitythat appears when an external force is applied.5 It is basedon the principle that soft tissues are deformed more easilythan hard tissue, and these differences can be reflected byimages, called elastograms, which represent the distribu-tion of the deformation in the tissue by means of a coloredmap. In general, malignant tumor tissues are more rigidthan normal surrounding tissue. This way, when mechani-cal compression is applied, tension in the tumor is lowerthan in adjacent healthy tissues, and consequently it is lessdeformed.3,5

The goal of this study is to analyze the diagnostic capac-ity of elastographies to distinguish benign from malignant

thyroid nodules. To this end, the anatomopathologic analy-sis of the piece after surgical exeresis has been used as thestandard reference test, and in non-operated cases we usedthe cytological study obtained through FNA.

Material and methods

We expressly declare our adherence to the Helsinki dec-laration. The necessary written informed consents wereobtained in order to be able to examine all the patients.

Patients

Two hundred and ninety-five (295) consecutive patients wereincluded in the study, with 314 thyroid nodules, of which 257corresponded to women, with an average age of 57 years(age range 21---87), and 57 to men, with a average age of56 years (age range 23---79 years), referred by theEndocrinology Unit of the Specialty Medical Center to per-form ultrasound-guided FNA for a period of 3 years, fromJanuary, 2012 to December, 2014. To determine the inclu-sion criteria, we followed the recommendations set bythe Guidelines from the American Association of ClinicalEndocrinology and the American Thyroid Association.6---8

Seven (7) patients were excluded from the study becausetheir elastography showed artifacts due to the followingcauses: the thyroid nodule was very superficial or very deepand it could not be properly compared to the surrounding tis-sue, the nodule to be examined was not clearly distinguished


368 C. Franco Uliaque et al.

Figure 1 Elastogram: lesion with predominance of elastic tissues.

from the rest of the nodules in the context of a MNG, or thelesion was so small or large that it would not allow any com-parisons to adjacent tissues.9 In these situations, it is notpossible to define the nodule through elastography, that iswhy we ruled out these observations in our study.

Conventional ultrasound, elastography and ultrasound-guided FNA were performed on all the patients. Theanatomopathologist used the Bethesda-2007 classification10

to perform the cytological diagnosis. In the cases where atype I or non-diagnostic cytology was obtained, the FNA wasrepeated and if the same results were obtained, ultrasoundcontrol was performed at 6 months with a new assessment.All patients whose second FNA was null after repeating itwere excluded from the study. When benign or type II cytol-ogy was obtained, they were eligible to clinical follow-up. Incategories III or uncertain, IV or proliferation/follicular neo-plasms, V or suspicious of malignancy and VI or malignant,surgery was performed undergoing the patients partial ortotal thyroidectomy.

Ultrasound and elastography

The studies were conducted by two radiologists with 35and 12 years of experience respectively in ultrasound, usingSiemens Acuson Antares (Siemens Medical Solutions, Moun-tain View, CA, USA) ultrasound machine equipped withan 8---12 MHz multifrequency linear transducer with colorDoppler and elastography. All FNPA (fine needle punctureaspirations) were performed by the same radiologist, with25 years of experience in this modality. A 21---22 gauge nee-dle was used connected to a 20 cm3 syringe and an aspirationmechanism (Cameco Medical Ltd London NW6 2BP). Theywere performed under ultrasound control because this wayis more effective than FNPA guided by palpation since itreduces the non-diagnostic results and the false negativesthrough a better location of the nodule.2

Two types of elastography are distinguished:5,11,12

semiquantitative (strain elastography) and quantitativeelastography (shear-wave elastography). Semiquantitativeelastography gathers information from tissue elasticity, ini-tially acquiring the data corresponding to tissue compositionbefore the deformation, in a neutral situation, and thenanother map is obtained after exerting a slight compres-sion with the transducer or with the patient’s respiratorymovements. The displacement of the deformed tissue is

calculated in real time, by comparing these two maps andthen an elastogram is obtained. Quantitative elastographymeasures tissue elasticity calculating shear wave propaga-tion speed. It does not require compression to be exerted,measures displacement regardless of the pressure appliedby sending acoustic microimpulses with minimum levels ofenergy to the different tissues. In this case, elasticity isexpressed in speed units in m/s or as pressure in kilopascals(kPa).5

We have a semiquantitative elastography that shows usa color map of the different tissues based on whether theyare more or less elastic. We have four shades on the colorscale, which are blue, green, yellow and red in decreasingorder of elasticity. Blue corresponds to very elastic tissue,usually one of liquid nature that translates substances suchas colloid material, blood, necrosis. The next level --- thecolor green, corresponds to elastic tissue. Yellow translatesrigid tissues and red translates very rigid ones.

We have used our own classification, which is very simple,differentiating three groups: lesions with a predominance ofsoft or elastic tissues (Fig. 1); lesions with a predominanceof rigid or hard tissues (Fig. 2) and lesions with mixed tissues(Fig. 3), in which we find soft and hard tissue areas in similarproportions showing a mosaic-like pattern.

The radiologist performs the elastography in real time,after performing the ultrasound in mode B and color Doppler.During the test, the transducer is placed perpendicular tothe skin, and it must come softly in contact with it, becauseif we were to exert a strong initial compression, the risk offalse negatives would increase. The region of interest (ROI)is centered on the lesion and enough surrounding thyroidtissue is included so that the upper edge includes the subcu-taneous fat tissue and the lower edge reaches the muscularplane, avoiding, as much as possible, the inclusion of thelarge cervical vessels.4,13,14 Repeated, slight compressionsare exerted on the nodule. The screen displays an image ofthe mode-B ultrasound and another image of the elastogra-phy.

Methods of statistical analysis

We analyzed the probability that a thyroid nodule is malig-nant based on the patterns obtained in the analysis ofthe elastography, in conformity with the classification men-tioned in the previous section. To this end we estimate a



Figure 2 Elastogram: lesion with predominance of rigid tissues.

logistic regression model that provides a coefficient thatrelates each variable with the probability that the noduleanalyzed is malignant, as well as the value of p. In addition,the odds ratios are calculated as well as a measure of thetotal model fitting (pseudo R2).

In our empirical analysis we put special emphasis in thecapacity of the models estimated to classify correctlythe nodules into malignant or benign. To this end, we calcu-lated the sensitivity and specificity of each model.

Results

The study includes 314 thyroid nodules, of which 194obtained cytological results of benignity (Bethesda II) inthe FNPA, and no other diagnostic test was performed. On32 occasions we received a non-diagnostic result (Bethesda I)in the two aspirations performed, which were eliminatedfrom the analysis so the results would not be altered.In either one of them significant changes were observed inthe ultrasound control done at 6 months, and subsequentlyone evolutionary control was performed one year later. Theremaining 88 nodules underwent to surgery. Among them3 nodules were classified as Bethesda III in the FNA, and theirfinal diagnosis after surgery was thyroid nodular hyperpla-sia; 67 nodules were classified as Bethesda IV after the FNPA,of which only 6 were cancer in the final diagnosis; 9 nodules

obtained a result of malignancy suspicion in the FNPA andamong them surgery confirmed malignancy in 4 cases; and9 nodules were classified as Bethesda VI or malignant in theFNPA and all of them were confirmed after surgery.

In total, 295 thyroid nodules were benign, of which, 134showed a rigid pattern through the elastography, 83 a softpattern and 78 a mosaic pattern. Of the 19 nodules diag-nosed as malignant (12 papillary carcinomas, 3 follicularvariants of papillary carcinoma, 2 follicular carcinomas and1 poorly differentiated carcinoma), 8 showed a mosaic pat-tern in the elastography, 6 a rigid pattern and 5 a softpattern. The application of elastography reveals some dif-ferences in the comparison of malignant thyroid nodules asopposed to benign ones, as showed by the increase in thepercentage of cases classified under the mosaic pattern.

Table 1 details the absolute and relative frequency ofthe patients’ characteristics (sex and age) and those of theelastography results for the total of the sample, as well asbased on the malignancy or benignity of the thyroid nodulesanalyzed.

Univariant analysis of malignancy of thyroidnodules

Table 2 shows the results obtained for the single-variablelogistic regression analysis.

Figure 3 Elastogram: mixed or mosaic lesion with rigid elastic areas without a clear predominance.



Table 1 Absolute and relative frequencies of explanatory variables based on the diagnosis of needles.

Explanatory variable Category Malignant thyroid nodule Benign thyroid nodule Total

Sex Male 4 (1.27%) 53 (16.73%) 57 (18%)Woman 15 (4.77%) 242 (75.23%) 257 (80%)

Age Average 50.42 57.15 56.74

Elastography Soft 5 (1.59%) 83 (26.43%) 88 (28.02%)Rigid 6 (1.91%) 134 (42.67%) 140 (44.58%)Mosaic 8 (2.54%) 78 (24.84%) 86 (27.38%)

Table 2 Results from the logistic regression analysis for the detection of malignant thyroid nodule.a

Elastography Odds ratio P Value Likelihood ratio model !2 (1) P Value Pseudo R2

Soft −0.29 0.74 0.61 0.26 0.61 0.00Rigid −0.33 0.72 0.50 0.46 0.50 0.00Mosaic 0.67 1.96 0.16 1.86 0.17 0.01

a It shows the results of the logistic regression analysis that associates the probability of malignancy of a thyroid nodule and thecategorization of thyroid nodules based on the elastography test: (1) lesions with predominance of soft or elastic tissues (soft), (2)lesions with predominance of rigid or hard tissues (rigid) and (3) lesions with mixed tissues and a mosaic-like pattern (mosaic).

In either one of the estimations performed we obtain thatthe pattern analyzed (soft tissue predominance, rigid tis-sue predominance or mosaic pattern) is significantly relatedwith the likelihood of thyroid nodule malignancy. This way,the pseudo R2 of the models estimated is 0 and the oddsratios are not significant either. The study of the predic-tive ability of the models estimated also shows that theelastography is not capable of properly predicting malig-nancy of a thyroid nodule. All the models estimated show0% sensitivity and 100% specificity for a cut-off value ofmalignancy probability of 50%. Also the possibility of reach-ing a 100% sensitivity happens at the expense of obtaining a0% specificity.

Discussion

The results obtained in our study and subsequent statisticanalysis coincide with some of the empirical evidence fromprevious studies13,15,16 though they also show discrepancieswith some of the previous medical literature.3,4,11,17

Under the hypothesis that thyroid cancer is more rigidthan benign nodules, elastography was suggested as a pos-sible tool for differential diagnosis of benign and malignantthyroid nodules. Many studies3,4,11,17 support elastographyas a reliable test with great potential in thyroid cancerdiagnosis. Several authors claim4,9,18 that its greater diag-nostic utility lies in nodules with undetermined cytologyin the FNPA. In our study, and based on the negativeresults obtained in the statistic analysis, we did not findany empirical evidences which would support the reliableuse of semiquantitative elastography to differentiate benignfrom malignant nodules, just as it happened with otherauthors.13,15,16 In their paper, Kagoya et al.1 claim that elas-tographies play an important role exclusively in the diagnosisof benign lesions, and always in combination with conven-tional ultrasounds.

We have used our own categorization of elastographypatterns which is characterized by being more simple,

reproducible and faster than other classifications describedin the specialized literature and considered by other authorsin their studies.4,13 The most frequently used classificationsare those by: (a) Ueno et al.19 distinguishing 5 degrees:degree 1 if there is elasticity in the entire nodule, degree2 if the elasticity predominates in the center and there areperipheral areas of rigidity, degree 3 if elasticity is in thecenter and rigidity in the periphery, degree 4 if the entirenodule is rigid and degree 5 if there is also rigidity area sur-rounding the nodule; (b) Rago et al.4 categorizes nodules infour groups: group 1 if there is elasticity in the entire nod-ule, group 2 if there is elasticity in most of the nodule, group3 when elasticity is only located in the periphery, degree 4 ifit is a nodule without elasticity and degree 5 when rigidity isadded around the nodule and (c) Asteria et al.20 describingfour levels: levels 1, 2 and 4 just like the ones described byRago et al.,4 and level 3 when there is rigidity in most of thenodule.

There are also differences in the results obtaineddescribed by different authors based on the type of elas-tography used, given that both present advantages andshortcomings. Quantitative elastography shows as an advan-tage in the reduction of inter- and intraobserver variabilitythough some articles indicate that though its specificityis greater, its sensitivity is lower than that of the semi-quantitative one.11 In their paper, Samir et al.21 claimthe quantitative or shear-wave elastography is a usefultool inpreoperative diagnosis to detect malignancy in fol-licular lesions. Slapa et al.22 compare semiquantitative(strain elastography) elastography with the new genera-tion of quantitative elastographies (supersonic shear waveelastography, SSWE) for the differential diagnosis of thy-roid nodules. These authors claim that SSWE is a greatimprovement in the elastographic evaluation of thyroid nod-ules given their high reproducibility regardless of the skillsof the radiologists performing the test, so that itovercomes the limitations of semiquantitative elastog-raphy. They even claim that it allow us to identify



microcalcifications that are not visible through B-modeultrasound.

Among the causes that justify the differences observedbetween our study and those by other authors we could con-sider the size of the sample which is much greater in ourstudy --- we have 314 thyroid nodules as opposed to 52 nod-ules analyzed in the study by Lyschchilk et al.3 and the 92nodules examined by Rago et al.’s paper;4 therefore, ouranalysis sample is more representative and less exposedto isolated observations or measurement errors, above allconsidering the low prevalence observed. Another source ofdiscrepancy could be due to the type of elastography used.Certain empirical evidence17,21,22 would support that elas-tography based on analytical quantitative methods insteadof elastography based on a color scale is more objective andreproducible.

In our study we faced some limitations inherent toelastography, which we describe below. There are severalcircumstances that alter tissue elasticity, and therefore theycan modify the elastogram of the thyroid nodules to bestudied, such as the presence of calcifications that canbe present in a benign nodule and increase its rigidity consid-erably leading to an interpretation of malignant finding. Thestiffness of thyroid nodules could be diminished if they showbleeding areas or colloid degeneration areas. In patientswith thyroiditis and/or multinodular goiter (MNG), the thy-roid parenchyma shows greater stiffness than normal condi-tions, so a malignant nodule in the core of these two patholo-gies might not present erroneously a rigid pattern when com-pared to the adjacent thyroid parenchyma.11 On the otherhand, in their paper, Xu et al.23 describe a technical limita-tion to bear in mind-aware of the fact that common carotidartery pulsations affect the measurement of elasticity whenthe nodule is in its proximity. Another negative factor in ourstudy is inter- and intraobserver variability when obtainingthe elastogram which reduces reproducibility --- a limitationdescribed by other authors too.16,22---24 This variability is dueto the fact that in the semiquantitative elastography used,it is necessary to perform an extrinsic compression. Basedon the force we exert when compressing the thyroid nod-ule, we can obtain different elastograms, so it is necessaryto maintain a constant level of pressure during the exam-ination of the same patient, and at the same time, exertpressure with similar intensity in all patients, which is ratherdifficult. This limitation could be minimized if there was asystem assigning a numerical value to the compressions anda graphic scale in real time, as described by Kura et al.’spaper.11 Another possibility to avoid this problem is to usea quantitative elastography that does not require compres-sion with the transducer therefore reducing the operator’sdependence and increasing its reproducibility. Another limi-tation of elastography is due to the fact that the peak depthrecorded depends on mode-B ultrasound penetration depth.If image quality in mode B is low then the calculation of theimage through elastography will be almost impossible.

Another limitation of the study that could affect thesolidity of our conclusions is the low prevalence observed,which determines predictive values. In this sense, an exten-sion of the sample would allow us to have more trust in theestimations obtained.

Conclusions

Our statistic analysis determines that THE semiquantitativeelastography, just as applied in our study, is not useful asan instrument to recommend or advise against FNPA sinceno statistically significant association was found betweenthe elastography patterns and the malignancy/benignity ofthyroid nodules.

Ethical responsibilities

Protection of people and animals. The authors declare thatno experiments with human beings or animals have beenperformed while conducting this investigation.

Confidentiality of data. The authors confirm that in thisarticle there are no data from patients.

Right to privacy and informed consent. The authors con-firm that in this article there are no data from patients.

Author’s contribution

1. Manager of the integrity of the study: CFU.2. Study Idea: CFU, JPB.3. Study Design: CFU.4. Data Mining: CFU, JPB.5. Data Analysis and Interpretation: CFU, JPB, RLH.6. Statistical Analysis: CFU, RLH.7. Reference: CFU, JPB, CPL.8. Writing: CFU, JPB, RLH, CPL.9. Critical review of the manuscript with intellectually rel-

evant remarks: JPB, CPL.10. Approval of final version: CFU, JPB, RLH, CPL.

Conflict of interests

The authors declare no conflict of interests

References

1. Kagoya R, Monobe H, Tojima H. Utility of elastography for dif-ferential diagnosis of benign and malignant thyroid nodules.Otolaryngol Head Neck Surg. 2010;143:230---4.

2. Manso S, Velasco MJ. Valor actual de la ecografía en la carac-terización de los nódulos tiroideos. Revisión de las últimas guíasclínicas de actuación. Radiología. 2015;57:248---58.

3. Lyshchik A, Higashi T, Asato R, Tanaka S, Ito J, Mai JJ, et al.Thyroid gland tumor diagnosis at US elastography. Radiology.2005;237:202---11.

4. Rago T, Santini F, Scutari M, Pinchera A, Vitti P. Elastography:new developments in ultrasound for predicting malignancy inthyroid nodules. J Clin Endocrinol Metab. 2007;92:2917---22.

5. Guzman F, Abellán D, Reus M. La elastografía: una nueva apli-cación de la ecografía. ¿Cuál es su utilidad clínica? Radiología.2014;56:290---4.

6. Gharib H, Papini E, Valcavi R, Baskin HJ, Crescenzi A, DottoriniME, et al. American Association of Clinical Endocrinologistsand Associazione Medici Endocrinologi medical guidelines for



clinical practice for the diagnosis and management of thyroidnodules. Endocrinol Pract. 2006;12:63---102.

7. Cooper DS, Doherty GM, Haugen BR, Kloos RT, Lee SL, MandelSJ, et al. Revised American Thyroid Association managementguidelines for patients with thyroid nodules and differentiatedthyroid cancer. Thyroid. 2009;19:1167---214.

8. 2015 American Thyroid Association Management Guidelines foradult patients with thyroid nodules and differentiated thyroidcancer. Thyroid. 2016;26:1---13, http://dx.doi.org/10.1089/thy.20150020.

9. Rago T, Vitti P. Role of thyroid ultrasound in the diagnostic eval-uation of thyroid nodules. Best Pract Res Clin Endocrinol Metab.2008;22:913---28.

10. Cibas ES, Ali SZ. The Bethesda System for reporting thyroidcytopathology. Am J Clin Pathol. 2009;132:658---65.

11. Kura M, Ballarino C, Tamagnone F, Campagno B, Bertini K,Gómez J, et al. Relación entre el valor del ratio elastográfico yla clasificación citológica de Bethesda en la patología tiroidea.Rev Argent Radiol. 2014;78:128---37.

12. Klauser AS, Miyamoto H, Bellmann-Wiler, Feuchtner GM, WickMC, Jaschke WR. Sonoelastography: musculoskeletal applica-tions. Radiology. 2014;272:622---33.

13. Moon HJ, Sung JM, Kim EK, Yoon JH, Youk JH, Kwak JY. Diagnosticperformance of gray-scale US and elastography in solid thyroidnodules. Radiology. 2012;262:1002---13.

14. Wang HL, Zhang S, Xin XJ, Zhao LH, Li CX, Mu JL, et al.Application of real-time ultrasound elastography in diagnosingbenign and malignant thyroid solid nodules. Cancer Biol Med.2012;9:124---7.

15. Rivo Vázquez A, Rodríguez Lorenzo A, Rivo Vázquez JE, PáramoFernández C., García Lorenzo F, Pardellas Rivera H, et al. Theuse of ultrasound elastography in the assessment of malig-nancy risk in the thyroid nodules and multinodular goitres. ClinEndocrinol. 2013;79:887---91.

16. Lippolis PV, Tognini S, Materazzi G, Polini A, Mancini R,Ambrosini CE, et al. Is elastography actually useful in the

presurgical selection of thyroid nodules with indeterminatecytology? J Clin Endocrinol Metab. 2011;96:1826---30.

17. Luo S, Kim EH, Dighe M, Kim Y. Thyroid nodule classificationusing ultrasound elastography via linear discriminant analysis.Ultrasonics. 2011;51:425---31.

18. Mehrotra P, McQueen A, Kolla S, Johnson SJ, Richardson DL. Doeselastography reduce the need for thyroid FNAs? Clin Endocrinol.2013;78:942---9.

19. Itoh A, Ueno E, Tohno E, Kamma H, Takahashi H, Shiina T, et al.Breast disease: clinical application of US elastography for diag-nosis. Radiology. 2006;239:341---50.

20. Asteria C, Giovanardi A, Pizzocaro A, Cozzaglio L, MorabitoA, Somalvico F, et al. US-elastography in the differentialdiagnosis of benign and malignant thyroid nodules. Thyroid.2008;18:523---31.

21. Samir AE, Dhyani M, Anvari A, Prescott J, Halperm EF, FaquinWC, et al. Shear-Wave elastography for the preoperative riskstratification of follicular-patterned lesions of the thyroid: diag-nostic accuracy and optimal measurement plane. Radiology.2015;277:565---73.

22. Slapa RZ, Piwowonski A, Jakubowski WS, Bierca J, Szopinki KT,Slowinska-Srzednick J, et al. Shear wave elastography may adda new dimension to ultrasound evaluation of the thyroid nod-ules: case series with comparative evaluation. J Thyroid Res.2012;2012:657147, http://dx.doi.org/10.1155/2012/657147.

23. Xu JM, Xu XH, Xu HX, Zhang YF, Zhang J, Guo LH, et al. Con-ventional US, US elasticity imaging, and acoustic radiation forceimpulse imaging for prediction of malignancy in thyroid nodules.Radiology. 2014;272:577---86.

24. Park SH, Kim SJ, Kim EK, Kim MJ, Son EJ, Kwak JY. Interobserveragreement in assessing the sonographic and elastographic fea-tures of malignant thyroid nodules. AJR Am J Roentgenol.2009;193:416---23.


Usefulness of semiquantitative elastography in predicting ...

Documents

Transcript of Usefulness of semiquantitative elastography in predicting ...