3212jumonline_Layout 1 11/19/13 9:56 AM Page 2163
ORIGINAL RESEARCH
Shear Wave Elastography for Differentiation of Benign and Malignant Thyroid Nodules A Meta-analysis Binglan Zhang, MD, Xuelei Ma, MD, Nana Wu, MD, Lei Liu, MD, PhD, Xiaoxiao Liu, MD, Jing Zhang, MD, Jinrong Yang, MD, Ting Niu, MD, PhD Objectives—The purpose of this study was to assess the performance of shear wave elastography for identification of benign and malignant thyroid nodules using metaanalysis. Methods—PubMed, MEDLINE, Embase, the Cochrane Library, and the China National Knowledge Infrastructure were searched. Patients’ clinical characteristics, sensitivity, specificity, positive predictive value, and negative predictive value were extracted. The diagnostic odds ratio and summary receiver operating characteristic curve were used to examine the accuracy of shear wave elastography. A meta-analysis was performed to evaluate the clinical utility of shear wave elastography for identification of benign and malignant thyroid nodules.
Received February 26, 2013, from the Department of Medical Oncology, Cancer Center, State Key Laboratory of Biotherapy (B.Z., X.M., N.W., L.L., X.L., J.Z.), and Department of Hematology (J.Y., T.N.), West China Hospital, Sichuan University, Chengdu, China. Revision requested March 18, 2013. Revised manuscript accepted for publication April 22, 2013. Drs Zhang, Ma, Wu, and Liu contributed equally to this work. Address correspondence to Lei Liu, MD, PhD, Department of Medical Oncology, Cancer Center, State Key Laboratory of Biotherapy, or Ting Niu, MD, PhD, Department of Hematology, West China Hospital, Sichuan University, 610041 Chengdu, China. E-mail: [email protected], tingniu1@ sina.com Abbreviations
AUC, area under the curve; CI, confidence interval; LR, likelihood ratio; NPV, negative predictive value; OR, odds ratio; PPV, positive predictive value; QUADAS, Quality Assessment for Studies of Diagnostic Accuracy; ROC, receiver operating characteristic
doi:10.7863/ultra.32.12.2163
Results—The summary sensitivity and specificity for the diagnosis of malignant thyroid nodules by shear wave elastography were 0.84 (95% confidence interval [CI], 0.76–0.90) and 0.90 (95% CI, 0.87–0.92), respectively. The pooled positive likelihood ratio was 7.39 (95% CI, 4.09–13.35), and the negative likelihood ratio was 0.20 (95% CI, 0.13– 0.29). The summary diagnostic odds ratio was 41.35 (95% CI, 17.38–98.41), and the summary area under the receiver operating characteristic curve was 0.92 (Q* = 0.8538). Conclusions—Shear wave elastography has high sensitivity and specificity in the evaluation of thyroid nodules and can potentially reduce unnecessary fine-needle aspiration biopsies. Key Words—diagnosis; meta-analysis; shear wave elastography; thyroid nodules
T
he incidence rates of thyroid cancers have increased in the past decade.1 Additionally, up to 68% of people have thyroid nodules, even when the thyroid gland is normal to palpation.2 Consequently, a precise evaluation of thyroid nodules is very important and can avoid unnecessary biopsy of benign nodules. At present, fine-needle aspiration biopsy is still most commonly used as the most specific method for evaluating thyroid nodules, but it has varying sensitivity.3–5 Simultaneously, it is an invasive method with serious shortcomings. Aside from the patient discomfort and serious complications (which are extremely rare), the main drawback of fine-needle aspiration biopsy is cost, both direct and indirect. Indirect costs are primarily related to the expenses of surgical procedures that sometimes follow indeterminate fine-needle aspiration results. Therefore, there is a need for another noninvasive way to evaluate thyroid nodules. Elastography, the imaging of tissue elasticity or stiffness, may emerge as a mainstream tool for ultrasound-based
©2013 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2013; 32:2163–2169 | 0278-4297 | www.aium.org
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2164
Zhang et al—Shear Wave Elastography of Thyroid Nodules
diagnosis in the future. A meta-analysis evaluated real-time elastography for identification of benign and malignant thyroid nodules, and it was shown to be a highly sensitive and specific method for evaluation of these nodules.6 Shear wave elastography, a novel ultrasound-based elastographic method, has been used recently for diagnosis of thyroid nodules with high-frequency linear probes.7–11 It is a new real-time, quantitative, operator-independent, and reproducible technique.12,13 Briefly, shear wave elastography uses a radiation force produced by an ultrasonic beam to stress tissues and ultrafast sonographic tracking techniques to measure the speed of shear waves. Based on the Young modulus formula, tissue elasticity can be derived from the shear wave propagation speed, and a real-time color-coded elastogram can be displayed, showing softer tissue in blue and stiffer tissue in red. Quantitative information is delivered as an elasticity index expressed in kilopascals. At present, studies with small patient populations have shown promising results. In this study, we used a metaanalysis to assess the overall performance of shear wave elastography for diagnosis of benign and malignant thyroid nodules.
Materials and Methods Literature Search We searched PubMed, MEDLINE, Embase, the Cochrane Library, and the China National Knowledge Infrastructure for studies published from September 2009 to September 2012 by using the key words “elastography” and “thyroid” to evaluate the overall performance of shear wave elastography for differentiation of benign and malignant thyroid nodules. All of the ultrasound machines in our meta-analysis were Aixplorer systems (SuperSonic Imagine, Aix-enProvence, France) because no suitable publication has evaluated shear wave thyroid imaging using other vendors’ equipment. Selection Criteria All analytical studies assessing the diagnostic performance of shear wave elastography had to meet the following inclusion criteria: evaluated the performance of shear wave elastography for differentiation of benign and malignant thyroid nodules; used appropriate cytologic samples acquired by fine-needle aspiration biopsy or histologic samples acquired by surgery as reference standards for the diagnosis; reported data (sensitivity, specificity, negative predictive value [NPV], and positive predictive value [PPV]) necessary to calculate the true-positive, false-
2164
positive, true-negative, and false-negative diagnostic results of shear wave elastography for diagnosis of malignancy based on cutoff values (best thresholds for differentiation of benign and malignant thyroid nodules); and included at least 30 patients. Studies such as reviews, commentaries, editorials, case reports, and letters were excluded. Data Extraction The studies were extracted independently by 2 investigators (B.Z. and X.L.) and then verified reciprocally, with disagreements resolved by consultation with a third investigator (X.M.). For each study, the following information was recorded: author, year of publication, country, number of patients, number of thyroid nodules available for analysis, reference standard for the diagnosis, shear wave elastographic cutoff values, and study quality. True-positive, false-positive, true-negative, and false-negative rates were extracted or calculated according to the sensitivity, specificity, PPV, and NPV in each reported test. Quality Assessment The quality of the studies included in the meta-analysis was assessed with a checklist based on the Quality Assessment for Studies of Diagnostic Accuracy (QUADAS) tool, which was designed to evaluate the diagnostic accuracy of studies and investigations.14,15 The QUADAS tool has 14 items that assess the validity of each study. Two investigators (B.Z. and X.M.) performed a quality assessment of the included studies independently, and disagreements were resolved by discussion. Data Analysis Overall pooled sensitivity and specificity, positive and negative likelihood ratios (LRs), and diagnostic odds ratios with corresponding 95% confidence intervals (CIs) were used to examine the accuracy of shear wave elastography for differentiation of benign and malignant thyroid nodules. A summary receiver operating characteristic (ROC) curve, constructed as described by Moses et al,16 was also plotted to graphically present the results. The Q* index was used to define the point on the summary ROC curve where sensitivity equaled specificity.17 χ2 and I 2 tests were used to assess the heterogeneity of the sensitivity and specificity. The Cochran Q value is a percentage measure of the heterogeneity of the positive and negative LRs. A low P value (≤.05) suggests the presence of heterogeneity. If heterogeneity existed, a random-effects model was used for the meta-analysis.18 All of the above calculations (truepositive, false-positive, true-negative, and false-negative rates included) were performed with RevMan version 5.1 soft-
J Ultrasound Med 2013; 32:2163–2169
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2165
J Ultrasound Med 2013; 32:2163–2169
Unclear Unclear Yes Unclear Unclear Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Unclear Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Sebag et al7 Veyrieres et al8 Bhatia et al9 Wang et al20 Huang et al21
Item 3, Item 4, Item 5, Item 6, Item 7, Item 8, Item 9, Item 10, Item 11, Item 12, Item 13, Item 1, Item 2, Appropriate Disease Partial Differential Incorpor- Test Reference Test Diagnostic Clinical Uninter- Item 14, Spectrum Selection Reference Progression Verification Verification ation Execution Execution Review Review Review pretable WithReference Composition Criteria Standard Bias Bias Bias Bias Details Details Bias Bias Bias Results drawals
Table 2. Quality Assessment of the Included Studies Using the QUADAS Questionnaire
2010 2012 2012 2012 2011 Sebag et al7 Veyrieres et al8 Bhatia et al9 Wang et al20 Huang et al21
NAA indicates nodules available for analysis; ND, not determined; RH, reference histologic value; RC, reference cytologic value; SEN, sensitivity; and SPE, specificity.
85.20 80.00 76.90 90.34 83.30 95.90 97.10 89.07 95.77 94.86 80.00 52.80 50.13 73.22 74.16 65.00 66.00 34.50 90.34 45.00 15.00 0 74.20 ND 0 85.00 100.00 25.80 ND 100.00 29 35 17 25 24 146 297 62 88 105 146 297 81 88 105 54.5 ND 52.8 42.6 46.7 25.80 25.70 21.60 36.70 46.70 93 148 74 79 75
Cutoff, kPa RC, % RH, % Malignant Nodules NAA Nodules Mean Age, y Male, % n Country Year Reference
Quality Assessment Using the QUADAS Questionnaire The quality of the studies in our meta-analysis was evaluated by the QUADAS quality assessment tool and is shown in Table 2. The overall quality of the studies was good, with all the studies having a “yes” rating for more than 10 items. To make it more concrete, only 1 study had a negative response to item 1, concerning spectrum composition. Item 6, concerning differential verification bias, was rated “no” for 3 studies referred for surgery and fine-needle aspiration as reference standards. It was difficult to rate item 14 (withdrawals) in 4 studies, and they were rated “unclear.”
Table 1. Main Characteristics of the Included Studies Evaluating the Performance of Shear Wave Elastography for Differentiation of Thyroid Nodules
Search Results and Study Characteristics On the basis of the predefined key words “elastography” and “thyroid,” a total of 106 studies were found. Of these, 93 studies were not related to the topic, and 13 potentially relevant studies were identified for further evaluation. Eight studies were excluded because they were reviews (n = 4), had insufficient data (n = 2), had a small sample size (n = 1), and focused on Hashimoto thyroiditis (n = 1). Thus, 5 publications met the criteria for use in our meta-analysis.7– 9,20,21 A total of 698 thyroid nodules (568 benign and 130 malignant) in 469 patients were evaluated. Among the publications, although the data of 2 studies7,8 came from the same hospital, they were not repeated because 1 study7 was published online in September 2010, and the patients from the other study8 were collected between November 2010 and June 2011. The main characteristics of the studies included in our meta-analysis are summarized in Table 1, including the clinical characteristics of patients, numbers of thyroid nodules, reference standards, and cutoff values. Three studies8,20,21 included patients who were referred for histopathologic examinations only, and the remaining 2 studies7,9 included patients with fine-needle aspiration cytologic findings as reference standards; surgery was also performed in patients with suspicious or malignant fineneedle aspiration results and finally used as reference standards in those 2 studies. The cutoff shear wave speed values ranged from 34.5 to 90.34 kPa in the different studies.
PPV, %
Results
France France Hong Kong China China
NPV, %
SEN, %
SPE, %
ware (Cochrane Collaboration, Oxford, England). All statistical analyses, shown in the figures, were performed with Meta-DiSc version 1.4 software.17 P ≤ .05 was considered statistically significant. A Begg funnel plot was applied with Stata version 11.0 software (StataCorp, College Station, TX) to assess the potential publication bias. P > .05 indicated no potential publication bias.19
93.90 90.50 71.10 86.89 91.40
Zhang et al—Shear Wave Elastography of Thyroid Nodules
2165
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2166
Zhang et al—Shear Wave Elastography of Thyroid Nodules
Differentiation Between Benign and Malignant Thyroid Nodules As shown in Figure 1, the pooled sensitivity and specificity of shear wave elastography for differentiation of benign and malignant thyroid nodules were 0.84 (95% CI, 0.76–0.90) and 0.90 (95% CI, 0.87–0.92), respectively. The positive and negative LRs (Figure 2) were 7.39 (95% CI, 4.09–13.35) and 0.20 (95% CI, 0.13–0.29), demonstrating that shear wave elastography is capable of excluding malignancy within thyroid nodules. The summary diagnostic OR was 41.35 (95% CI, 17.38–98.41; Figure 3). Figure 4 shows the resulting summary ROC curve with summary operating points for sensitivity and specificity on the curve. The summary area under the curve (AUC) was 0.92 (Q* = 0.8538). Overall, a 3% false-negative rate (21 of 698) was observed. Among these, 0.45% of follicular carcinomas (2 of 443), 1.8% of papillary carcinomas (8 of 443), and 0.23% of differentiated follicular tumors of uncertain malignant potential (1 of 443) were misdiagnosed in 2 studies.7,8 The remaining 3 studies did not exactly report the spectrum compositions of false-negative results.
specificity (heterogeneity, χ2 = 16.81; P = .0021; I 2 = 76.2%) was heterogeneous. The Cochran Q values were 19.54 (P = .0006) and 3.3 (P = .5055) for the positive and negative LRs, respectively, which implied that heterogeneity was found for the positive LR.
Discussion
Study Heterogeneity In the χ2 and I 2 tests, the sensitivity (heterogeneity, χ2 = 2.56; P = .6335; I 2 = 0.0%) was not heterogeneous, but the
The prevalence of thyroid nodules is quite high, which has been reported to be as high as 68% on sonography.2 In past years, B-mode sonography was usually used as a first-line procedure to differentiate benign and malignant nodules. Several sonographic features indicate the possibility of malignancy, including microcalcifications, hypoechogenicity, intranodular vascularity, and irregular margins.22–24 However, there is no single criterion or even a combination of criteria that is sensitive or specific enough to diagnose thyroid malignancy; therefore, fine-needle aspiration is currently the primary diagnostic procedure. Based on the existing reports, fine-needle aspiration has been shown to be a cost-effective method, with sensitivities of 54% to 90% and specificities of 60% to 96% for identification of benign and malignant thyroid nodules.3–5 However, it is limited by sampling difficulties with large nodules. According to a retrospective cohort study, fine-needle aspiration results were highly inaccurate, misclassifying half of all patients (155 patients) with reportedly benign lesions and thyroid nodules larger than 4 cm.25 Moreover, reliable fine-needle aspiration is dependent on the sampler’s experience and the cytologist’s expertise, particularly for distinction between benign and malignant follicular adenomas.
Figure 1. Individual study estimates for the sensitivity and specificity of shear wave elastography for diagnosis of thyroid nodules.
Figure 2. Individual estimated positive and negative LRs and pooled values with corresponding 95% CIs.
Assessment of Publication Bias Since no more than 10 studies were included in our meta-analysis, the Egger test and funnel were not used. The Begg test was used to examine the publication bias. No significant publication bias for the diagnostic OR was shown (P = .142; Figure 5).
2166
J Ultrasound Med 2013; 32:2163–2169
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2167
Zhang et al—Shear Wave Elastography of Thyroid Nodules
The results of our meta-analysis, including data from 5 published studies reporting shear wave speeds of 698 thyroid nodules, demonstrated that shear wave elastography had pooled sensitivity of 84% and specificity of 90% for differentiating malignant and benign thyroid nodules. The diagnostic accuracy of shear wave elastography quantified by the AUC was 0.92 for the diagnosis of malignant thyroid nodules. A diagnostic tool is considered as perfect if the AUC is 100%, excellent if greater than 90%, and good if greater than 80%.26 According to these results, shear wave elastography can be used in clinical practice as an excellent diagnostic tool for diagnosis of malignant thyroid nodules. This finding is of great importance because of the limitations of current approaches to identification of benign and malignant thyroid nodules. Elastography is a noninvasive imaging technique that has emerged as a useful adjunct tool to estimate tissue elasticity or stiffness.27 Shear wave elastography has been used for evaluating malignancy in several body sites.28–32 It may add a new dimension to sonographic evaluation of thyroid nodules compared with other elastographic modes, Figure 3. Summary diagnostic OR used to examine the accuracy of shear wave elastography for differentiation of benign and malignant thyroid nodules.
Figure 4. Summary ROC (SROC) curve of shear wave elastography for differentiation of thyroid nodules.
accompanied by the added advantage that it is a highly reproducible procedure and able to reveal large nodules and those with diameters of few millimeters, even in cases of multinodular goiters. A meta-analysis by Bojunga et al6 showed that real-time elastography has high sensitivity and specificity for evaluation of thyroid nodules; the percentage of false-negative findings in all undertaken studies reached 44% for follicular tumors versus 7% for papillary carcinomas. However, real-time elastography is limited in evaluating multinodular goiters, which requires clear differentiation of nodules from one another.33 Consequently, shear wave elastography seems to be the most useful method for diagnosis in the future. Heterogeneity was found for the specificity and positive LR of the 5 studies used in our meta-analysis. This finding could be explained by a spectrum composition bias and a differential verification bias of the different studies, including patients referred for surgery only in 3 studies and patients referred for fine-needle aspiration in 2 studies. Another important explanation could be the study by Bhatia et al,9 which showed a wide disparity in performance of shear wave elastography compared to the other 4 studies. That study had lower shear wave elastographic results for both benign and malignant lesions, which may have been due to a limited sample size, especially for malignant lesions. Our analysis had several limitations. First, the cutoff values for shear wave elastography ranged from 34.5 to 90.34 kPa in the different studies. However, we could not perform a subgroup analysis to obtain a single cutoff value because of the limited number of studies, and this factor needs to be confirmed by an analysis including a larger number of studies. Data were then analyzed by a randomeffects model. Because the CIs were not particularly wide, we believe that the results are valuable. Second, there are Figure 5. Publication bias summary with pseudo 95% CI for the metaanalysis of the diagnostic OR (DOR) (P = .142).
J Ultrasound Med 2013; 32:2163–2169
2167
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2168
Zhang et al—Shear Wave Elastography of Thyroid Nodules
several qualitative factors that may influence shear wave elastography, including vertical artifacts, the presence of compression artifacts in the superficial fascia, and areas devoid of a shear wave elastographic signal in some solid nodules, especially those that are hypoechoic or deeply located. In addition, the structure of the neck is very complicated, and shear wave elastographic results may also be affected by the trachea, carotid artery, and other surrounding tissues. In conclusion, our meta-analysis shows that shear wave elastography could be used to differentiate malignant and benign thyroid nodules. Large prospective international multicenter studies in various regions are needed to confirm our results and further evaluate the potential of shear wave elastography.
References 1.
Simard EP, Ward EM, Siegel R, Jemal A. Cancers with increasing incidence trends in the United States: 1999 through 2008 [published online ahead of print January 4, 2012]. CA Cancer J Clin. doi:10.3322/ caac.20141. 2. Guth S Theune U, Aberle J, Galach A, Bamberger CM. Very high prevalence of thyroid nodules detected by high frequency (13 MHz) ultrasound examination. Eur J Clin Invest 2009; 39:699–706. 3. Tee YY, Lowe AJ, Brand CA, Judson RT. Fine-needle aspiration may miss a third of all malignancy in palpable thyroid nodules: a comprehensive literature review. Ann Surg 2007; 246:714–720. 4. Peng Y, Wang HH. A meta-analysis of comparing fine-needle aspiration and frozen section for evaluating thyroid nodules. Diagn Cytopathol 2008; 36:916–920. 5. Oertel YC, Miyahara-Felipe L, Mendoza MG, Yu K. Value of repeated fine needle aspirations of the thyroid: an analysis of over ten thousand FNAs. Thyroid 2007; 17:1061–1066. 6. Bojunga J, Herrmann E, Meyer G, Weber S, Zeuzem S, Friedrich-Rust M. Real-time elastography for the differentiation of benign and malignant thyroid nodules: a meta-analysis. Thyroid 2010; 20:1145–1150. 7. Sebag F, Vaillant-Lombard J, Berbis J, et al. Shear wave elastography: a new ultrasound imaging mode for the differential diagnosis of benign and malignant thyroid nodules. J Clin Endocrinol Metab 2010; 95:5281–5288. 8. Veyrieres JB, Albarel F, Lombard JV, et al. A threshold value in shear wave elastography to rule out malignant thyroid nodules: A reality? Eur J Radiol 2012; 81:3965–3972. 9. Bhatia KS, Tong CS, Cho CC, Yuen EH, Lee YY, Ahuja AT. Shear wave elastography of thyroid nodules in routine clinical practice: preliminary observations and utility for detecting malignancy. Eur Radiol 2012; 22:2397–2406. 10. Slapa RZ, Piwowonski A, Jakubowski WS, et al. Shear wave elastography may add a new dimension to ultrasound evaluation of thyroid nodules: case series with comparative evaluation. J Thyroid Res2012; 2012:657147.
2168
11. Magri F, Chytiris S, Capelli V, et al. Shear wave elastography in the diagnosis of thyroid nodules: feasibility in the case of coexistent chronic autoimmune Hashimoto’s thyroiditis. Clin Endocrinol (Oxf) 2012; 76:137–141. 12. Wells PN, Liang HD. Medical ultrasound: imaging of soft tissue strain and elasticity. J R Soc Interface 2011; 8:1521–1549. 13. Sporea I, Lie I. Shear wave elastography. Ultraschall Med 2012; 33:393– 394. 14. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 2003; 3:25. 15. Wang QB, Zhu H, Liu HL, Zhang B. Performance of magnetic resonance elastography and diffusion-weighted imaging for the staging of hepatic fibrosis: a meta-analysis. Hepatology 2012; 56:239–247. 16. Moses LE, Shapiro D, Littenberg B. Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. Stat Med 1993; 12:1293–1316. 17. Zamora J, Abraira V, Muriel A, Khan K, Coomarasamy A. Meta-DiSc: a software for meta-analysis of test accuracy data. BMC Med Res Methodol 2006; 6:31. 18. Dinnes J, Deeks J, Kirby J, Roderick P. A methodological review of how heterogeneity has been examined in systematic reviews of diagnostic test accuracy. Health Technol Assess 2005; 9:1–113, iii. 19. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994; 50:1088–1101. 20. Wang T, Zhang YX, Feng YQ, Tao CM. Quantitative assessment of realtime shear wave elastography imaging in differentiating benign and malignant thyroid nodules. Chin J Med Imaging 2012; 20:684-687. 21. Huang Y, Wang ZL, Wan WB, Su L, Tang J. Quantitative research on supersonic shear imaging in diagnosis of thyroid nodules by elastography. Chin J Med Ultrasound 2011; 8:1282–1288. 22. Frates MC, Benson CB, Charboneau JW, et al. Management of thyroid nodules detected at US: Society of Radiologists in Ultrasound consensus conference statement. Radiology 2005; 237:794–800. 23. Ivanac G, Brkljacic B, Ivanac K, Huzjan R, Skreb F, Cikara I. Vascularisation of benign and malignant thyroid nodules: CD US evaluation. Ultraschall Med 2007; 28:502–506. 24. Cappelli C, Castellano M, Pirola I, et al. Thyroid nodule shape suggests malignancy. Eur J Endocrinol 2006; 155:27–31. 25. Pinchot SN, Al-Wagih H, Schaefer S, Sippel R, Chen H. Accuracy of fineneedle aspiration biopsy for predicting neoplasm or carcinoma in thyroid nodules 4 cm or larger. Arch Surg 2009; 144:649–655. 26. Swets JA. Measuring the accuracy of diagnostic systems. Science 1988; 240:1285–1293. 27. Garra BS. Elastography: current status, future prospects, and making it work for you. Ultrasound Q 2011; 27:177–186. 28. Postnova NA, Vasil’ev A, Zykin BI, Pavlinova ES, Vykliuk MV. Shear wave elastography: possibilities of the differential diagnosis of focal and diffuse changes in various organs and tissues [in Russian]. Vestn Rentgenol Radiol 2011; 2:29–34. J Ultrasound Med 2013; 32:2163–2169
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2169
Zhang et al—Shear Wave Elastography of Thyroid Nodules
29. Arda K, Ciledag N, Aktas E, Aribas BK, Kose K. Quantitative assessment of normal soft-tissue elasticity using shear-wave ultrasound elastography. AJR Am J Roentgenol 2011; 197:532–536. 30. Chang JM, Moon WK, Cho N, et al. Clinical application of shear wave elastography (SWE) in the diagnosis of benign and malignant breast diseases. Breast Cancer Res Treat 2011; 129:89–97. 31. Bhatia KS, Cho CC, Tong CS, Lee YY, Yuen EH, Ahuja AT. Shear wave elastography of focal salivary gland lesions: preliminary experience in a routine head and neck US clinic. Eur Radiol 2012; 22:957–965. 32. Ferraioli G, Tinelli C, Dal Bello B, Zicchetti M, Filice G, Filice C; Liver fibrosis study group. Accuracy of real-time shear wave elastography for assessing liver fibrosis in chronic hepatitis C: a pilot study. Hepatology 2012; 56:2125–2133. 33. Alexander EK. Approach to the patient with a cytologically indeterminate thyroid nodule. J Clin Endocrinol Metab 2008; 93:4175–4182.
J Ultrasound Med 2013; 32:2163–2169
2169
ORIGINAL RESEARCH
Shear Wave Elastography for Differentiation of Benign and Malignant Thyroid Nodules A Meta-analysis Binglan Zhang, MD, Xuelei Ma, MD, Nana Wu, MD, Lei Liu, MD, PhD, Xiaoxiao Liu, MD, Jing Zhang, MD, Jinrong Yang, MD, Ting Niu, MD, PhD Objectives—The purpose of this study was to assess the performance of shear wave elastography for identification of benign and malignant thyroid nodules using metaanalysis. Methods—PubMed, MEDLINE, Embase, the Cochrane Library, and the China National Knowledge Infrastructure were searched. Patients’ clinical characteristics, sensitivity, specificity, positive predictive value, and negative predictive value were extracted. The diagnostic odds ratio and summary receiver operating characteristic curve were used to examine the accuracy of shear wave elastography. A meta-analysis was performed to evaluate the clinical utility of shear wave elastography for identification of benign and malignant thyroid nodules.
Received February 26, 2013, from the Department of Medical Oncology, Cancer Center, State Key Laboratory of Biotherapy (B.Z., X.M., N.W., L.L., X.L., J.Z.), and Department of Hematology (J.Y., T.N.), West China Hospital, Sichuan University, Chengdu, China. Revision requested March 18, 2013. Revised manuscript accepted for publication April 22, 2013. Drs Zhang, Ma, Wu, and Liu contributed equally to this work. Address correspondence to Lei Liu, MD, PhD, Department of Medical Oncology, Cancer Center, State Key Laboratory of Biotherapy, or Ting Niu, MD, PhD, Department of Hematology, West China Hospital, Sichuan University, 610041 Chengdu, China. E-mail: [email protected], tingniu1@ sina.com Abbreviations
AUC, area under the curve; CI, confidence interval; LR, likelihood ratio; NPV, negative predictive value; OR, odds ratio; PPV, positive predictive value; QUADAS, Quality Assessment for Studies of Diagnostic Accuracy; ROC, receiver operating characteristic
doi:10.7863/ultra.32.12.2163
Results—The summary sensitivity and specificity for the diagnosis of malignant thyroid nodules by shear wave elastography were 0.84 (95% confidence interval [CI], 0.76–0.90) and 0.90 (95% CI, 0.87–0.92), respectively. The pooled positive likelihood ratio was 7.39 (95% CI, 4.09–13.35), and the negative likelihood ratio was 0.20 (95% CI, 0.13– 0.29). The summary diagnostic odds ratio was 41.35 (95% CI, 17.38–98.41), and the summary area under the receiver operating characteristic curve was 0.92 (Q* = 0.8538). Conclusions—Shear wave elastography has high sensitivity and specificity in the evaluation of thyroid nodules and can potentially reduce unnecessary fine-needle aspiration biopsies. Key Words—diagnosis; meta-analysis; shear wave elastography; thyroid nodules
T
he incidence rates of thyroid cancers have increased in the past decade.1 Additionally, up to 68% of people have thyroid nodules, even when the thyroid gland is normal to palpation.2 Consequently, a precise evaluation of thyroid nodules is very important and can avoid unnecessary biopsy of benign nodules. At present, fine-needle aspiration biopsy is still most commonly used as the most specific method for evaluating thyroid nodules, but it has varying sensitivity.3–5 Simultaneously, it is an invasive method with serious shortcomings. Aside from the patient discomfort and serious complications (which are extremely rare), the main drawback of fine-needle aspiration biopsy is cost, both direct and indirect. Indirect costs are primarily related to the expenses of surgical procedures that sometimes follow indeterminate fine-needle aspiration results. Therefore, there is a need for another noninvasive way to evaluate thyroid nodules. Elastography, the imaging of tissue elasticity or stiffness, may emerge as a mainstream tool for ultrasound-based
©2013 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2013; 32:2163–2169 | 0278-4297 | www.aium.org
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2164
Zhang et al—Shear Wave Elastography of Thyroid Nodules
diagnosis in the future. A meta-analysis evaluated real-time elastography for identification of benign and malignant thyroid nodules, and it was shown to be a highly sensitive and specific method for evaluation of these nodules.6 Shear wave elastography, a novel ultrasound-based elastographic method, has been used recently for diagnosis of thyroid nodules with high-frequency linear probes.7–11 It is a new real-time, quantitative, operator-independent, and reproducible technique.12,13 Briefly, shear wave elastography uses a radiation force produced by an ultrasonic beam to stress tissues and ultrafast sonographic tracking techniques to measure the speed of shear waves. Based on the Young modulus formula, tissue elasticity can be derived from the shear wave propagation speed, and a real-time color-coded elastogram can be displayed, showing softer tissue in blue and stiffer tissue in red. Quantitative information is delivered as an elasticity index expressed in kilopascals. At present, studies with small patient populations have shown promising results. In this study, we used a metaanalysis to assess the overall performance of shear wave elastography for diagnosis of benign and malignant thyroid nodules.
Materials and Methods Literature Search We searched PubMed, MEDLINE, Embase, the Cochrane Library, and the China National Knowledge Infrastructure for studies published from September 2009 to September 2012 by using the key words “elastography” and “thyroid” to evaluate the overall performance of shear wave elastography for differentiation of benign and malignant thyroid nodules. All of the ultrasound machines in our meta-analysis were Aixplorer systems (SuperSonic Imagine, Aix-enProvence, France) because no suitable publication has evaluated shear wave thyroid imaging using other vendors’ equipment. Selection Criteria All analytical studies assessing the diagnostic performance of shear wave elastography had to meet the following inclusion criteria: evaluated the performance of shear wave elastography for differentiation of benign and malignant thyroid nodules; used appropriate cytologic samples acquired by fine-needle aspiration biopsy or histologic samples acquired by surgery as reference standards for the diagnosis; reported data (sensitivity, specificity, negative predictive value [NPV], and positive predictive value [PPV]) necessary to calculate the true-positive, false-
2164
positive, true-negative, and false-negative diagnostic results of shear wave elastography for diagnosis of malignancy based on cutoff values (best thresholds for differentiation of benign and malignant thyroid nodules); and included at least 30 patients. Studies such as reviews, commentaries, editorials, case reports, and letters were excluded. Data Extraction The studies were extracted independently by 2 investigators (B.Z. and X.L.) and then verified reciprocally, with disagreements resolved by consultation with a third investigator (X.M.). For each study, the following information was recorded: author, year of publication, country, number of patients, number of thyroid nodules available for analysis, reference standard for the diagnosis, shear wave elastographic cutoff values, and study quality. True-positive, false-positive, true-negative, and false-negative rates were extracted or calculated according to the sensitivity, specificity, PPV, and NPV in each reported test. Quality Assessment The quality of the studies included in the meta-analysis was assessed with a checklist based on the Quality Assessment for Studies of Diagnostic Accuracy (QUADAS) tool, which was designed to evaluate the diagnostic accuracy of studies and investigations.14,15 The QUADAS tool has 14 items that assess the validity of each study. Two investigators (B.Z. and X.M.) performed a quality assessment of the included studies independently, and disagreements were resolved by discussion. Data Analysis Overall pooled sensitivity and specificity, positive and negative likelihood ratios (LRs), and diagnostic odds ratios with corresponding 95% confidence intervals (CIs) were used to examine the accuracy of shear wave elastography for differentiation of benign and malignant thyroid nodules. A summary receiver operating characteristic (ROC) curve, constructed as described by Moses et al,16 was also plotted to graphically present the results. The Q* index was used to define the point on the summary ROC curve where sensitivity equaled specificity.17 χ2 and I 2 tests were used to assess the heterogeneity of the sensitivity and specificity. The Cochran Q value is a percentage measure of the heterogeneity of the positive and negative LRs. A low P value (≤.05) suggests the presence of heterogeneity. If heterogeneity existed, a random-effects model was used for the meta-analysis.18 All of the above calculations (truepositive, false-positive, true-negative, and false-negative rates included) were performed with RevMan version 5.1 soft-
J Ultrasound Med 2013; 32:2163–2169
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2165
J Ultrasound Med 2013; 32:2163–2169
Unclear Unclear Yes Unclear Unclear Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Unclear Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No No Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Sebag et al7 Veyrieres et al8 Bhatia et al9 Wang et al20 Huang et al21
Item 3, Item 4, Item 5, Item 6, Item 7, Item 8, Item 9, Item 10, Item 11, Item 12, Item 13, Item 1, Item 2, Appropriate Disease Partial Differential Incorpor- Test Reference Test Diagnostic Clinical Uninter- Item 14, Spectrum Selection Reference Progression Verification Verification ation Execution Execution Review Review Review pretable WithReference Composition Criteria Standard Bias Bias Bias Bias Details Details Bias Bias Bias Results drawals
Table 2. Quality Assessment of the Included Studies Using the QUADAS Questionnaire
2010 2012 2012 2012 2011 Sebag et al7 Veyrieres et al8 Bhatia et al9 Wang et al20 Huang et al21
NAA indicates nodules available for analysis; ND, not determined; RH, reference histologic value; RC, reference cytologic value; SEN, sensitivity; and SPE, specificity.
85.20 80.00 76.90 90.34 83.30 95.90 97.10 89.07 95.77 94.86 80.00 52.80 50.13 73.22 74.16 65.00 66.00 34.50 90.34 45.00 15.00 0 74.20 ND 0 85.00 100.00 25.80 ND 100.00 29 35 17 25 24 146 297 62 88 105 146 297 81 88 105 54.5 ND 52.8 42.6 46.7 25.80 25.70 21.60 36.70 46.70 93 148 74 79 75
Cutoff, kPa RC, % RH, % Malignant Nodules NAA Nodules Mean Age, y Male, % n Country Year Reference
Quality Assessment Using the QUADAS Questionnaire The quality of the studies in our meta-analysis was evaluated by the QUADAS quality assessment tool and is shown in Table 2. The overall quality of the studies was good, with all the studies having a “yes” rating for more than 10 items. To make it more concrete, only 1 study had a negative response to item 1, concerning spectrum composition. Item 6, concerning differential verification bias, was rated “no” for 3 studies referred for surgery and fine-needle aspiration as reference standards. It was difficult to rate item 14 (withdrawals) in 4 studies, and they were rated “unclear.”
Table 1. Main Characteristics of the Included Studies Evaluating the Performance of Shear Wave Elastography for Differentiation of Thyroid Nodules
Search Results and Study Characteristics On the basis of the predefined key words “elastography” and “thyroid,” a total of 106 studies were found. Of these, 93 studies were not related to the topic, and 13 potentially relevant studies were identified for further evaluation. Eight studies were excluded because they were reviews (n = 4), had insufficient data (n = 2), had a small sample size (n = 1), and focused on Hashimoto thyroiditis (n = 1). Thus, 5 publications met the criteria for use in our meta-analysis.7– 9,20,21 A total of 698 thyroid nodules (568 benign and 130 malignant) in 469 patients were evaluated. Among the publications, although the data of 2 studies7,8 came from the same hospital, they were not repeated because 1 study7 was published online in September 2010, and the patients from the other study8 were collected between November 2010 and June 2011. The main characteristics of the studies included in our meta-analysis are summarized in Table 1, including the clinical characteristics of patients, numbers of thyroid nodules, reference standards, and cutoff values. Three studies8,20,21 included patients who were referred for histopathologic examinations only, and the remaining 2 studies7,9 included patients with fine-needle aspiration cytologic findings as reference standards; surgery was also performed in patients with suspicious or malignant fineneedle aspiration results and finally used as reference standards in those 2 studies. The cutoff shear wave speed values ranged from 34.5 to 90.34 kPa in the different studies.
PPV, %
Results
France France Hong Kong China China
NPV, %
SEN, %
SPE, %
ware (Cochrane Collaboration, Oxford, England). All statistical analyses, shown in the figures, were performed with Meta-DiSc version 1.4 software.17 P ≤ .05 was considered statistically significant. A Begg funnel plot was applied with Stata version 11.0 software (StataCorp, College Station, TX) to assess the potential publication bias. P > .05 indicated no potential publication bias.19
93.90 90.50 71.10 86.89 91.40
Zhang et al—Shear Wave Elastography of Thyroid Nodules
2165
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2166
Zhang et al—Shear Wave Elastography of Thyroid Nodules
Differentiation Between Benign and Malignant Thyroid Nodules As shown in Figure 1, the pooled sensitivity and specificity of shear wave elastography for differentiation of benign and malignant thyroid nodules were 0.84 (95% CI, 0.76–0.90) and 0.90 (95% CI, 0.87–0.92), respectively. The positive and negative LRs (Figure 2) were 7.39 (95% CI, 4.09–13.35) and 0.20 (95% CI, 0.13–0.29), demonstrating that shear wave elastography is capable of excluding malignancy within thyroid nodules. The summary diagnostic OR was 41.35 (95% CI, 17.38–98.41; Figure 3). Figure 4 shows the resulting summary ROC curve with summary operating points for sensitivity and specificity on the curve. The summary area under the curve (AUC) was 0.92 (Q* = 0.8538). Overall, a 3% false-negative rate (21 of 698) was observed. Among these, 0.45% of follicular carcinomas (2 of 443), 1.8% of papillary carcinomas (8 of 443), and 0.23% of differentiated follicular tumors of uncertain malignant potential (1 of 443) were misdiagnosed in 2 studies.7,8 The remaining 3 studies did not exactly report the spectrum compositions of false-negative results.
specificity (heterogeneity, χ2 = 16.81; P = .0021; I 2 = 76.2%) was heterogeneous. The Cochran Q values were 19.54 (P = .0006) and 3.3 (P = .5055) for the positive and negative LRs, respectively, which implied that heterogeneity was found for the positive LR.
Discussion
Study Heterogeneity In the χ2 and I 2 tests, the sensitivity (heterogeneity, χ2 = 2.56; P = .6335; I 2 = 0.0%) was not heterogeneous, but the
The prevalence of thyroid nodules is quite high, which has been reported to be as high as 68% on sonography.2 In past years, B-mode sonography was usually used as a first-line procedure to differentiate benign and malignant nodules. Several sonographic features indicate the possibility of malignancy, including microcalcifications, hypoechogenicity, intranodular vascularity, and irregular margins.22–24 However, there is no single criterion or even a combination of criteria that is sensitive or specific enough to diagnose thyroid malignancy; therefore, fine-needle aspiration is currently the primary diagnostic procedure. Based on the existing reports, fine-needle aspiration has been shown to be a cost-effective method, with sensitivities of 54% to 90% and specificities of 60% to 96% for identification of benign and malignant thyroid nodules.3–5 However, it is limited by sampling difficulties with large nodules. According to a retrospective cohort study, fine-needle aspiration results were highly inaccurate, misclassifying half of all patients (155 patients) with reportedly benign lesions and thyroid nodules larger than 4 cm.25 Moreover, reliable fine-needle aspiration is dependent on the sampler’s experience and the cytologist’s expertise, particularly for distinction between benign and malignant follicular adenomas.
Figure 1. Individual study estimates for the sensitivity and specificity of shear wave elastography for diagnosis of thyroid nodules.
Figure 2. Individual estimated positive and negative LRs and pooled values with corresponding 95% CIs.
Assessment of Publication Bias Since no more than 10 studies were included in our meta-analysis, the Egger test and funnel were not used. The Begg test was used to examine the publication bias. No significant publication bias for the diagnostic OR was shown (P = .142; Figure 5).
2166
J Ultrasound Med 2013; 32:2163–2169
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2167
Zhang et al—Shear Wave Elastography of Thyroid Nodules
The results of our meta-analysis, including data from 5 published studies reporting shear wave speeds of 698 thyroid nodules, demonstrated that shear wave elastography had pooled sensitivity of 84% and specificity of 90% for differentiating malignant and benign thyroid nodules. The diagnostic accuracy of shear wave elastography quantified by the AUC was 0.92 for the diagnosis of malignant thyroid nodules. A diagnostic tool is considered as perfect if the AUC is 100%, excellent if greater than 90%, and good if greater than 80%.26 According to these results, shear wave elastography can be used in clinical practice as an excellent diagnostic tool for diagnosis of malignant thyroid nodules. This finding is of great importance because of the limitations of current approaches to identification of benign and malignant thyroid nodules. Elastography is a noninvasive imaging technique that has emerged as a useful adjunct tool to estimate tissue elasticity or stiffness.27 Shear wave elastography has been used for evaluating malignancy in several body sites.28–32 It may add a new dimension to sonographic evaluation of thyroid nodules compared with other elastographic modes, Figure 3. Summary diagnostic OR used to examine the accuracy of shear wave elastography for differentiation of benign and malignant thyroid nodules.
Figure 4. Summary ROC (SROC) curve of shear wave elastography for differentiation of thyroid nodules.
accompanied by the added advantage that it is a highly reproducible procedure and able to reveal large nodules and those with diameters of few millimeters, even in cases of multinodular goiters. A meta-analysis by Bojunga et al6 showed that real-time elastography has high sensitivity and specificity for evaluation of thyroid nodules; the percentage of false-negative findings in all undertaken studies reached 44% for follicular tumors versus 7% for papillary carcinomas. However, real-time elastography is limited in evaluating multinodular goiters, which requires clear differentiation of nodules from one another.33 Consequently, shear wave elastography seems to be the most useful method for diagnosis in the future. Heterogeneity was found for the specificity and positive LR of the 5 studies used in our meta-analysis. This finding could be explained by a spectrum composition bias and a differential verification bias of the different studies, including patients referred for surgery only in 3 studies and patients referred for fine-needle aspiration in 2 studies. Another important explanation could be the study by Bhatia et al,9 which showed a wide disparity in performance of shear wave elastography compared to the other 4 studies. That study had lower shear wave elastographic results for both benign and malignant lesions, which may have been due to a limited sample size, especially for malignant lesions. Our analysis had several limitations. First, the cutoff values for shear wave elastography ranged from 34.5 to 90.34 kPa in the different studies. However, we could not perform a subgroup analysis to obtain a single cutoff value because of the limited number of studies, and this factor needs to be confirmed by an analysis including a larger number of studies. Data were then analyzed by a randomeffects model. Because the CIs were not particularly wide, we believe that the results are valuable. Second, there are Figure 5. Publication bias summary with pseudo 95% CI for the metaanalysis of the diagnostic OR (DOR) (P = .142).
J Ultrasound Med 2013; 32:2163–2169
2167
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2168
Zhang et al—Shear Wave Elastography of Thyroid Nodules
several qualitative factors that may influence shear wave elastography, including vertical artifacts, the presence of compression artifacts in the superficial fascia, and areas devoid of a shear wave elastographic signal in some solid nodules, especially those that are hypoechoic or deeply located. In addition, the structure of the neck is very complicated, and shear wave elastographic results may also be affected by the trachea, carotid artery, and other surrounding tissues. In conclusion, our meta-analysis shows that shear wave elastography could be used to differentiate malignant and benign thyroid nodules. Large prospective international multicenter studies in various regions are needed to confirm our results and further evaluate the potential of shear wave elastography.
References 1.
Simard EP, Ward EM, Siegel R, Jemal A. Cancers with increasing incidence trends in the United States: 1999 through 2008 [published online ahead of print January 4, 2012]. CA Cancer J Clin. doi:10.3322/ caac.20141. 2. Guth S Theune U, Aberle J, Galach A, Bamberger CM. Very high prevalence of thyroid nodules detected by high frequency (13 MHz) ultrasound examination. Eur J Clin Invest 2009; 39:699–706. 3. Tee YY, Lowe AJ, Brand CA, Judson RT. Fine-needle aspiration may miss a third of all malignancy in palpable thyroid nodules: a comprehensive literature review. Ann Surg 2007; 246:714–720. 4. Peng Y, Wang HH. A meta-analysis of comparing fine-needle aspiration and frozen section for evaluating thyroid nodules. Diagn Cytopathol 2008; 36:916–920. 5. Oertel YC, Miyahara-Felipe L, Mendoza MG, Yu K. Value of repeated fine needle aspirations of the thyroid: an analysis of over ten thousand FNAs. Thyroid 2007; 17:1061–1066. 6. Bojunga J, Herrmann E, Meyer G, Weber S, Zeuzem S, Friedrich-Rust M. Real-time elastography for the differentiation of benign and malignant thyroid nodules: a meta-analysis. Thyroid 2010; 20:1145–1150. 7. Sebag F, Vaillant-Lombard J, Berbis J, et al. Shear wave elastography: a new ultrasound imaging mode for the differential diagnosis of benign and malignant thyroid nodules. J Clin Endocrinol Metab 2010; 95:5281–5288. 8. Veyrieres JB, Albarel F, Lombard JV, et al. A threshold value in shear wave elastography to rule out malignant thyroid nodules: A reality? Eur J Radiol 2012; 81:3965–3972. 9. Bhatia KS, Tong CS, Cho CC, Yuen EH, Lee YY, Ahuja AT. Shear wave elastography of thyroid nodules in routine clinical practice: preliminary observations and utility for detecting malignancy. Eur Radiol 2012; 22:2397–2406. 10. Slapa RZ, Piwowonski A, Jakubowski WS, et al. Shear wave elastography may add a new dimension to ultrasound evaluation of thyroid nodules: case series with comparative evaluation. J Thyroid Res2012; 2012:657147.
2168
11. Magri F, Chytiris S, Capelli V, et al. Shear wave elastography in the diagnosis of thyroid nodules: feasibility in the case of coexistent chronic autoimmune Hashimoto’s thyroiditis. Clin Endocrinol (Oxf) 2012; 76:137–141. 12. Wells PN, Liang HD. Medical ultrasound: imaging of soft tissue strain and elasticity. J R Soc Interface 2011; 8:1521–1549. 13. Sporea I, Lie I. Shear wave elastography. Ultraschall Med 2012; 33:393– 394. 14. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 2003; 3:25. 15. Wang QB, Zhu H, Liu HL, Zhang B. Performance of magnetic resonance elastography and diffusion-weighted imaging for the staging of hepatic fibrosis: a meta-analysis. Hepatology 2012; 56:239–247. 16. Moses LE, Shapiro D, Littenberg B. Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. Stat Med 1993; 12:1293–1316. 17. Zamora J, Abraira V, Muriel A, Khan K, Coomarasamy A. Meta-DiSc: a software for meta-analysis of test accuracy data. BMC Med Res Methodol 2006; 6:31. 18. Dinnes J, Deeks J, Kirby J, Roderick P. A methodological review of how heterogeneity has been examined in systematic reviews of diagnostic test accuracy. Health Technol Assess 2005; 9:1–113, iii. 19. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994; 50:1088–1101. 20. Wang T, Zhang YX, Feng YQ, Tao CM. Quantitative assessment of realtime shear wave elastography imaging in differentiating benign and malignant thyroid nodules. Chin J Med Imaging 2012; 20:684-687. 21. Huang Y, Wang ZL, Wan WB, Su L, Tang J. Quantitative research on supersonic shear imaging in diagnosis of thyroid nodules by elastography. Chin J Med Ultrasound 2011; 8:1282–1288. 22. Frates MC, Benson CB, Charboneau JW, et al. Management of thyroid nodules detected at US: Society of Radiologists in Ultrasound consensus conference statement. Radiology 2005; 237:794–800. 23. Ivanac G, Brkljacic B, Ivanac K, Huzjan R, Skreb F, Cikara I. Vascularisation of benign and malignant thyroid nodules: CD US evaluation. Ultraschall Med 2007; 28:502–506. 24. Cappelli C, Castellano M, Pirola I, et al. Thyroid nodule shape suggests malignancy. Eur J Endocrinol 2006; 155:27–31. 25. Pinchot SN, Al-Wagih H, Schaefer S, Sippel R, Chen H. Accuracy of fineneedle aspiration biopsy for predicting neoplasm or carcinoma in thyroid nodules 4 cm or larger. Arch Surg 2009; 144:649–655. 26. Swets JA. Measuring the accuracy of diagnostic systems. Science 1988; 240:1285–1293. 27. Garra BS. Elastography: current status, future prospects, and making it work for you. Ultrasound Q 2011; 27:177–186. 28. Postnova NA, Vasil’ev A, Zykin BI, Pavlinova ES, Vykliuk MV. Shear wave elastography: possibilities of the differential diagnosis of focal and diffuse changes in various organs and tissues [in Russian]. Vestn Rentgenol Radiol 2011; 2:29–34. J Ultrasound Med 2013; 32:2163–2169
3212jumonline_Layout 1 11/19/13 9:56 AM Page 2169
Zhang et al—Shear Wave Elastography of Thyroid Nodules
29. Arda K, Ciledag N, Aktas E, Aribas BK, Kose K. Quantitative assessment of normal soft-tissue elasticity using shear-wave ultrasound elastography. AJR Am J Roentgenol 2011; 197:532–536. 30. Chang JM, Moon WK, Cho N, et al. Clinical application of shear wave elastography (SWE) in the diagnosis of benign and malignant breast diseases. Breast Cancer Res Treat 2011; 129:89–97. 31. Bhatia KS, Cho CC, Tong CS, Lee YY, Yuen EH, Ahuja AT. Shear wave elastography of focal salivary gland lesions: preliminary experience in a routine head and neck US clinic. Eur Radiol 2012; 22:957–965. 32. Ferraioli G, Tinelli C, Dal Bello B, Zicchetti M, Filice G, Filice C; Liver fibrosis study group. Accuracy of real-time shear wave elastography for assessing liver fibrosis in chronic hepatitis C: a pilot study. Hepatology 2012; 56:2125–2133. 33. Alexander EK. Approach to the patient with a cytologically indeterminate thyroid nodule. J Clin Endocrinol Metab 2008; 93:4175–4182.
J Ultrasound Med 2013; 32:2163–2169
2169
