ORIGINAL RESEARCH Diagnostic Accuracy of SuperSonic Shear Imaging for Staging of Liver Fibrosis A Meta-analysis Jin-Chun Feng,MD,Jun Li,MD,Xiang-Wei Wu,MD,PhD,Xin-Yu Peng,MD,PhD Objectives-The purpose of this study was to assess the performance of SuperSonic shear imaging(SuperSonic Imagine SA,Aix-en-Provence,France)for diagnosis ofliver fibrosis. Methods-Literature databases were searched to identify relevant studies from incep- tion to February 28,2015.Sensitivity,specificity,and other information wereextracted from the studies.Pooled data were calculated by a bivariate mixed-effects binary regres- sion model.Subgroup and sensitivity analyses were performed.Publication bias was tested by funnel plots. Results-Twelve studies were included in this meta-analysis and reported on 1635 patients.The pooled sensitivity and specificity were0.78(95%confidence interval [CI], 0.69-0.85)and 0.95(95%CI,0.75-0.99),respectively,for fibrosis stages F21,0.84 (95%CL,0.81-0.86)and0.81(9s%CL,0.74-0.87)forF≥2,0.89(95%CL,0.85-0.93) and0.84(95%CL,0.77-0.89)forF≥3，and0.88(95%CL,0.82-0.91)and0.86(95% CI,0.81-0.90)for F=4.The areas under the summary receiver operating characteristic curves were0.87(95%CL,0.84-0.90)forF≥1,0.85(95%CL,0.81-0.88)forF≥2,0.93 (95%CI,0.91-0.95)for F23,and 0.93(95%CI,0.90-0.95)for F=4.No significant publication bias was found. Conclusions-SuperSonic shear imaging could be used for staging of liver fibrosis. Especially,it has high diagnostic accuracy for severe fibrosis and cirrhosis. Received March 13,2015,from the Shihezi University School of Medicine,Shihezi,China Key Words-gastrointestinal ultrasound;liver fibrosis staging;meta-analysis;shear (I.-C.F.);and Departments of Ultrasound (I.L.) wave elastography and Hepatobiliary Surgery(X-W.W,XY.P.) First Affiliated Hospital of the School Medicine, Shihezi University,Shihezi,China.Revision requested May 8,2015.Revised manuscript accepted for publication June 3 2015. epatic fibrosis is the process ofconnective tissue deposition in Drs Feng and Li contributed equally to this healing caused by chronic liver damage induced by factors work. such as viral infections,excessive alcohol use,autoimmune Address correspondence to Xin-Yu Peng, disease,drug-induced liver damage,and nonalcoholic fatty MD,PhD,Department of Hepatobiliary Surgery, liver disease,which usually leads to cirrhosis and portal hyperten- First Affiliated Hospital of the School Medicine sion.2The degree ofliver fibrosis has a positive correlation with the Shihezi University,North Second Road,832002 risk of hepatocellular carcinoma,and it is closely related to the prog- Shihezi,Xinjiang China. nosis,3-5 but clinical symptoms are unapparent in the early stage of E-mail pengxinyu0106@sina.com liver fibrosis;thus,an accurate approach for evaluation of hepatic Abbreviations fibrosis is important for treatment strategies and monitoring ofther- AUROC,area under the receiver operating apeutic efficacy for patients with chronic hepatic disease. characteristic curve;CI,confidence interval; So far,liver biopsy is still considered the reference standard for OR,odds ratio;ROC,receiver operating hepatic fibrosis staging,but it is an invasive approach,and it brings characteristic a risk of complications.In addition,many factors could influence doi:10.7863/ultra,15.03032 the accuracy ofliver biopsy,such as intraobserver and interobserver @2016 by the American Institute of Ultrasound in Medicine J Ultrasound Med 2016;35:329-339 0278-4297 www.aium.org
Diagnostic Accuracy of SuperSonic Shear Imaging for Staging of Liver Fibrosis A Meta-analysis epatic fibrosis is the process of connective tissue deposition in healing caused by chronic liver damage induced by factors such as viral infections, excessive alcohol use, autoimmune disease, drug-induced liver damage, and nonalcoholic fatty liver disease, which usually leads to cirrhosis and portal hypertension.1,2The degree of liver fibrosis has a positive correlation with the risk of hepatocellular carcinoma, and it is closely related to the prognosis,3–5 but clinical symptoms are unapparent in the early stage of liver fibrosis; thus, an accurate approach for evaluation of hepatic fibrosis is important for treatment strategies and monitoring of therapeutic efficacy for patients with chronic hepatic disease. So far, liver biopsy is still considered the reference standard for hepatic fibrosis staging,6 but it is an invasive approach, and it brings a risk of complications.6 In addition, many factors could influence the accuracy of liver biopsy, such as intraobserver and interobserver Jin-Chun Feng, MD, Jun Li, MD, Xiang-Wei Wu, MD, PhD, Xin-Yu Peng, MD, PhD Received March 13, 2015, from the Shihezi University School of Medicine, Shihezi, China (J.-C.F.); and Departments of Ultrasound (J.L.) and Hepatobiliary Surgery (X.-W.W., X.-Y.P.), First Affiliated Hospital of the School Medicine, Shihezi University, Shihezi, China. Revision requested May 8, 2015. Revised manuscript accepted for publication June 3, 2015. Drs Feng and Li contributed equally to this work. Address correspondence to Xin-Yu Peng, MD, PhD, Department of Hepatobiliary Surgery, First Affiliated Hospital of the School Medicine, Shihezi University, North Second Road, 832002 Shihezi, Xinjiang, China. E-mail: pengxinyu0106@sina.com Abbreviations AUROC, area under the receiver operating characteristic curve; CI, confidence interval; OR, odds ratio; ROC, receiver operating characteristic H ©2016 by the American Institute of Ultrasound in Medicine | J Ultrasound Med 2016; 35:329–339 | 0278-4297 | www.aium.org ORIGINAL RESEARCH Objectives—The purpose of this study was to assess the performance of SuperSonic shear imaging (SuperSonic Imagine SA, Aix-en-Provence, France) for diagnosis of liver fibrosis. Methods—Literature databases were searched to identify relevant studies from inception to February 28, 2015. Sensitivity, specificity, and other information were extracted from the studies. Pooled data were calculated by a bivariate mixed-effects binary regression model. Subgroup and sensitivity analyses were performed. Publication bias was tested by funnel plots. Results—Twelve studies were included in this meta-analysis and reported on 1635 patients. The pooled sensitivity and specificity were 0.78 (95% confidence interval [CI], 0.69–0.85) and 0.95 (95% CI, 0.75–0.99), respectively, for fibrosis stages F≥1, 0.84 (95% CI, 0.81–0.86) and 0.81 (95% CI, 0.74–0.87) for F≥2, 0.89 (95% CI, 0.85–0.93) and 0.84 (95% CI, 0.77–0.89) for F≥3, and 0.88 (95% CI, 0.82–0.91) and 0.86 (95% CI, 0.81–0.90) for F=4. The areas under the summary receiver operating characteristic curves were 0.87 (95% CI, 0.84–0.90) for F≥1, 0.85 (95% CI, 0.81–0.88) for F≥2, 0.93 (95% CI, 0.91–0.95) for F≥3, and 0.93 (95% CI, 0.90–0.95) for F=4. No significant publication bias was found. Conclusions—SuperSonic shear imaging could be used for staging of liver fibrosis. Especially, it has high diagnostic accuracy for severe fibrosis and cirrhosis. Key Words—gastrointestinal ultrasound; liver fibrosis staging; meta-analysis; shear wave elastography doi:10.7863/ultra.15.03032 3502jum279-400 copy_Layout 1 1/20/16 12:46 PM Page 329
Feng et al-SuperSonic Shear Imaging for Staging of Liver Fibrosis diagnostic differences in pathologic evaluation and sampling tive,false-positive,false-negative,and true-negative results; errors because the size of the specimen is very small (the (3)liver biopsy was used as the reference standard for volume of the specimen is about 1/50,000 ofthe liver).6-8 assessment ofliver fibrosis;(4)the liver fibrosis staging sys- Therefore,studies have focused on noninvasive tem was similar to METAVIR;and (5)the maximum approaches for assessing liver fibrosis.In recent years,elas- interval between liver biopsy and elastography was within ticity imaging has been developed rapidly,10 and various 15 months for studies with untreated patients or within 3 ultrasound-based elasticity imaging techniques have enabled months for studies that included some or all patients who noninvasive,economical,and efficient measurements of received antiviral or antifibrotic therapy and for studies liver stiffness.11-14 SuperSonic shear imaging is a novel that did not state on this issue.Exclusion criteria were as ultrasound-based shear wave imaging technique that is per- follows:(1)reviews,editorials,letters,and comments;(2) formed with the Aixplorer ultrasound system(SuperSonic animal studies;(3)sample sizes of less than 30;and (4) Imagine SA,Aix-en-Provence,France).15 It can superim- duplicate publications(the same study group,the same pose a region ofinterest on a common B-mode image,and patient population,or an increased number ofpatients).If the probe emits a plurality of SuperSonic focused ultra- necessary data were unavailable,we tried to obtain them sonic beams inside the region of interest at increasing via e-mail from the corresponding authors. depths.The focused beams generate shear waves that propagate in tissues.The speed of the shear waves is cor- Data Extraction related with the degree oftissue stiffness and is tracked by Two researchers (J.C.F.and J.L.)performed data extrac- an ultrafast scanner so that tissue stiffness can be shown tions independently.Any discrepancies in the results ofthe by a color-coded real-time map.The average value in the researchers were solved by consultation with an addi- region of interest is calculated to show quantitative infor- tional researcher(X.-W.W.).The following information mation in terms of the Young modulus(kilopascals)or was collected (listed in Tables 1 and2):first author,coun- meters per second.16 try of the study's origin,sample size,number of male and SuperSonic shear imaging has been used in several female patients,mean age,body mass index(BMI),study body sites such as the thyroid,breast,liver,and salivary design,maximum interval between liver biopsy and elas- gland,and a meta-analysis has shown relatively good tography,location of SuperSonic shear imaging performed performance of the technique in the evaluation of thyroid on the liver,etiology of liver disease,histologic score used, nodules.Some studies have evaluated its diagnostic accu- liver stiffness cutoff value,and sensitivity,specificity,and racy in staging liver fibrosis but to our knowledge,no meta- area under the receiver operating characteristic(ROC) analysis has been conducted before.Therefore,our study curve (AUROC)for each fibrosis stage. aimed to evaluate the overall performance of SuperSonic shear imaging for diagnosis of liver fibrosis. Quality Assessment Two researchers(I.-C.F.and X.-W.W.)assessed the quality Materials and Methods of the studies included in this analysis independently.Each study was assessed by an improved Quality Assessment of Literature Search Diagnostic Accuracy Studies tool.20 This tool comprises4 Relevant articles written in English from inception to domains(shownin Table 3).Each domain has some ques- February 28,2015,were searched in the Cochrane Library, tions to help judge the bias.The bias and applicability of PubMed,SpringerLink,Web ofScience,and EMBASE using each domain were assessed as high,unclear,or low risk the terms elastography or elasticity imaging or shear wave (assessment of applicability was not applied in the domain elastography or SuperSonic shear imaging and liver fibrosis of flow and timing).Any differences in the results of the or hepatic fibrosis or cirrhosis or hepatic stiffness or liver stiff- researchers were resolved by an additional researcher (J.L.). ness.To identify additional articles,reference lists of eligi- ble articles were also manually searched. Data Analysis All of the histologic scoring systems used in the included Selection and Exclusion Criteria studies were similar to the METAVIR system and staged Studies must have met each of the following criteria for into 5 groups.The diagnostic accuracy ofSuperSonic shear inclusion in this study:(1)they assessed the performance imaging was evaluated for the distinction offibrosis stages of SuperSonic shear imaging for staging of hepatic fibro- F21(F0 versus F1-F4),F22 (FO-F1 versus F2-F4), sis;(2)necessary data were provided to extract true-posi- F23(F0-F2 versus F3-F4),and F=4(F0-F3 versus F4). 330 J Ultrasound Med 2016:35:329-339
diagnostic differences in pathologic evaluation and sampling errors because the size of the specimen is very small (the volume of the specimen is about 1/50,000 of the liver).6–8 Therefore, studies have focused on noninvasive approaches for assessing liver fibrosis. In recent years, elasticity imaging has been developed rapidly,9,10 and various ultrasound-based elasticity imaging techniques have enabled noninvasive, economical, and efficient measurements of liver stiffness.11–14 SuperSonic shear imaging is a novel ultrasound-based shear wave imaging technique that is performed with the Aixplorer ultrasound system (SuperSonic Imagine SA, Aix-en-Provence, France).15 It can superimpose a region of interest on a common B-mode image, and the probe emits a plurality of SuperSonic focused ultrasonic beams inside the region of interest at increasing depths. The focused beams generate shear waves that propagate in tissues. The speed of the shear waves is correlated with the degree of tissue stiffness and is tracked by an ultrafast scanner so that tissue stiffness can be shown by a color-coded real-time map. The average value in the region of interest is calculated to show quantitative information in terms of the Young modulus (kilopascals) or meters per second.16 SuperSonic shear imaging has been used in several body sites such as the thyroid, breast, liver, and salivary gland,17,18 and a meta-analysis has shown relatively good performance of the technique in the evaluation of thyroid nodules.19 Some studies have evaluated its diagnostic accuracy in staging liver fibrosis but to our knowledge, no metaanalysis has been conducted before. Therefore, our study aimed to evaluate the overall performance of SuperSonic shear imaging for diagnosis of liver fibrosis. Materials and Methods Literature Search Relevant articles written in English from inception to February 28, 2015, were searched in the Cochrane Library, PubMed, SpringerLink, Web of Science, and EMBASE using the terms elastography or elasticity imaging or shear wave elastography or SuperSonic shear imaging and liver fibrosis or hepatic fibrosis orcirrhosis or hepatic stiffness or liver stiffness. To identify additional articles, reference lists of eligible articles were also manually searched. Selection and Exclusion Criteria Studies must have met each of the following criteria for inclusion in this study: (1) they assessed the performance of SuperSonic shear imaging for staging of hepatic fibrosis; (2) necessary data were provided to extract true-positive, false-positive, false-negative, and true-negative results; (3) liver biopsy was used as the reference standard for assessment of liver fibrosis; (4) the liver fibrosis staging system was similar to METAVIR; and (5) the maximum interval between liver biopsy and elastography was within 15 months for studies with untreated patients or within 3 months for studies that included some or all patients who received antiviral or antifibrotic therapy and for studies that did not state on this issue. Exclusion criteria were as follows: (1) reviews, editorials, letters, and comments; (2) animal studies; (3) sample sizes of less than 30; and (4) duplicate publications (the same study group, the same patient population, or an increased number of patients). If necessary data were unavailable, we tried to obtain them via e-mail from the corresponding authors. Data Extraction Two researchers (J.C.F. and J.L.) performed data extractions independently. Any discrepancies in the results of the researchers were solved by consultation with an additional researcher (X.-W.W.). The following information was collected (listed in Tables 1 and 2): first author, country of the study’s origin, sample size, number of male and female patients, mean age, body mass index (BMI), study design, maximum interval between liver biopsy and elastography, location of SuperSonic shear imaging performed on the liver, etiology of liver disease, histologic score used, liver stiffness cutoff value, and sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUROC) for each fibrosis stage. Quality Assessment Two researchers (J.-C.F. and X.-W.W.) assessed the quality of the studies included in this analysis independently. Each study was assessed by an improved Quality Assessment of Diagnostic Accuracy Studies tool.20This tool comprises 4 domains (shown in Table 3). Each domain has some questions to help judge the bias. The bias and applicability of each domain were assessed as high, unclear, or low risk (assessment of applicability was not applied in the domain of flow and timing). Any differences in the results of the researchers were resolved by an additional researcher (J.L.). Data Analysis All of the histologic scoring systems used in the included studies were similar to the METAVIR system and staged into 5 groups. The diagnostic accuracy of SuperSonic shear imaging was evaluated for the distinction of fibrosis stages F≥1 (F0 versus F1–F4), F≥2 (F0–F1 versus F2–F4), F≥3 (F0–F2 versus F3–F4), and F=4 (F0–F3 versus F4). Feng et al—SuperSonic Shear Imaging for Staging of Liver Fibrosis 330 J Ultrasound Med 2016; 35:329–339 3502jum279-400 copy_Layout 1 1/20/16 12:46 PM Page 330
Feng et al-SuperSonic Shear Imaging for Staging of Liver Fibrosis dIAV3W dIAVL3W biMpn pue sneg 8IAVW 3 乏 2.251 aoejns ulxs 227 22741 220m wnwxeW p awes p May p awes pPZ 品 88 尉 网 的对 丙月 Apms 是 是 是 是 器 K'a6v 品器 零 等 罕 065 品 /W 697192 11146 男 6100550 6810060 闲品 (81276 692627 6100500 292861 色 euly euyo JUltrasound Med 2016:35:329-339 331
J Ultrasound Med 2016; 35:329–339 331 Feng et al—SuperSonic Shear Imaging for Staging of Liver Fibrosis Table 1. Main Characteristics of the Included Studies Depth of Target Area Histologic Score Sample Mean Study BMI, Maximum From Liver Used for Fibrosis Reference Country Size M/F Age, y Design kg/m2 Intervala Capsule, cmb Etiology and Patients Assessment Beland27 United States 50 25/25 52.0 Pro NA Same d <6.0 CHC, 21; ELF, 15; NASH, 5; other, 9 METAVIR Cassinotto28 France 349 (301c) 188/161 54.8 Pro NA 2 wk 1.5 Alcoholic/NASH, 145; viral hepatitis, 127; METAVIR unexplained chronic cytolysis, 31; other, 46 Deffieux29 France 120 (118c) 86/34 46.2 Pro 24.2 Few d NA CHC, 44; CHB, 24; healthy, 11; NASH, 11; alcoholic, METAVIR 10; AI, 10; other, 10 Ferraioli14 Italy 121 (118c) 87/34 44.8 Pro 25.4 Same d 1.5–2.0 CHC, 121 METAVIR Jeong30 Korea 70 45/25 45.9 Ret 23.8 2 dd >2.0 CHB, 23; CHC, 18; alcoholic, 12; NAFLD, 4; other, 13 Batts and Ludwig Leung31 China 226 146/80 48.8 Pro 24.2 12 mo 1.0–3.0 CHB, 226 METAVIR Samir32 United States 136 70/66 49.0 Pro NA Same d <5.0 from ELF, 60; CHC, 43; AI, 18; CHB, 8; other, 7 METAVIR skin surface Suh33 Korea 196 (218e) 130/66 29.2 Ret 22.8 Same d >2.0 Healthy, 123; hepatic steatosis, 73; other, 22 METAVIR Tada34 Japan 55 23/32 61.0 Pro 21.3 2 d 1.0–2.0 CHC, 55 METAVIR Yoon35 Korea 129 (75c) 87/42 51.0 Pro 23.7 67 d 1.5–2.0 Chronic hepatitis, 67; liver donor, 18; Korean Study Group hepatic steatosis, 16; other, 28 for the Pathology of Digestive Diseasesf Zeng37 China 104 (101c) 251/59 36.6 Pro 22.1 3 d >1.0 CHB, 104 METAVIR Zheng36 China 198 (167c) 145/53 37.7 Pro 21.6 3 d 1.0–2.0 CHB, 164; CHC, 9; alcoholic, 7; AI, 7; other, 11 METAVIR AI indicates autoimmune disease; CHB, chronic hepatitis B; CHC, chronic hepatitis C; ELF, elevated liver function values of unknown cause; F, female; M, male; NA, not available; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; Pro, prospective; and Ret, retrospective. aMaximum interval between liver biopsy and elastography. bAll liver stiffness values were obtained from the right lobe of the liver. cPatients suitable for analysis, excluding patients with failed elastographic measurements or liver biopsies. dObtained from the corresponding author via e-mail. eTwenty-two more patients were added to the statistical analysis. fYoon et al stated that the Korean Study Group for the Pathology of Digestive Diseases staging system is similar to the METAVIR scoring system. 3502jum279-400 copy_Layout 1 1/20/16 12:46 PM Page 331
Feng et al-SuperSonic Shear Imaging for Staging of Liver Fibrosis The Midas module of Stata version 12.0 software(StataCorp, College Station,TX)and SPSS version 17.0 software(IBM Corporation,Armonk,NY)was used for this meta-analysis. 常品是架站品耐乏乏乏留时 This module is a specialized program for diagnostic meta- analysis,which uses a bivariate mixed-effects binary regres- 昌品品品卡乏乏乏能 sion model to obtain the summary sensitivity,specificity, likelihood ratio,diagnostic odds ratio (OR),and summary ROC curve for each hepatic fibrosis stage.2 图S目昌目目8乏乏乏娟至 Inevitable differing will exist when studies are combined in a meta-analysis.The term heterogeneity means any clinical or methodological diversity amongincluded studies.Usually, 思器思袋乏乏乏g substantial heterogeneity in a meta-analysis leads to lower- quality results.Toidentifyheterogeneity,a statisticaltestis usually conducted.The I2 statistic of Higgins et al23 is an 89 品品朵恩品常受芝 improved approach for quantifying heterogeneity.23 It ranges from 0%to 100%,and a previous guideline showed that Iof greater than 50%signified substantial heterogeneity.24 ” 品景品8品醫世乏乏芝品乏 We assessed potential heterogeneity at every fibrosis stage. We used the funnel plot asymmetry test of Deeks et al2s to investigate publication bias at every fibrosis stage. 目厨品品昌是品乏乏乏易乏 This test is a modified method for diagnostic meta-analysis derived from randomized meta-analysis trials.Deeks et al demonstrated that the log diagnostic OR is more helpful 县8每8乏乏乏8乏 for detecting sample size effects,and the effective sample size(ESS)is more appropriate than the total sample size. Therefore,in the Deeks funnel plot,the horizontal axis rep- 尽 品是品88份乏8器周 resents measure of the effect(log diagnostic OR),and the vertical axis represents the measure ofprecision(3/ESS). In diagnostic studies,numbers of nondiseased patients 员品尽景使苏京乏贸哭始的 =n,and numbers of diseased patients =n2;ESS (4nn2)/(n+n2).The random variation of the effect decreases as the sample size increases.Therefore,if there is 品是品果图乏8月乏 no publication bias,data from the studies on the diagram will form a symmetric funnel shape and can be formally tested by a linear regression of log diagnostic OR against 西88器6乏超 3/ESS.P<.10 for the slope coefficient of the regression line would suggest significant asymmetry of the funnel plot 9 and a significant publication bias. B昌乏乏乏乏8乏乏乏乏 To assess the clinical utility of SuperSonic shear imag- ing,Fagan plot analysis was conducted.26 We assumed 'SN3S N⑧乏受乏品乏品受乏受受 pretest probabilities of 25%,50%,and 75%to calculate corresponding posttest probabilities after positive and negative SuperSonic shear imaging results on the basis of 园乏乏乏日乏品乏乏乏乏 pooled sensitivity and specificity.Positive and negative results were defined as all results from each study that were S8乏乏乏思乏乏乏乏乏乏 above and below the liver stiffness cutoff value at a given stage,respectively.The following prespecified subgroup analyses were conducted:different etiologies,different BMI ranges (<18.50,18.50-24.99,and 225.00 kg/m2),different geographic areas,and different histologic scores used. 332 J Ultrasound Med 2016:35:329-339
The Midas module of Stata version 12.0 software (StataCorp, College Station, TX) and SPSS version 17.0 software (IBM Corporation, Armonk, NY) was used for this meta-analysis. This module is a specialized program for diagnostic metaanalysis, which uses a bivariate mixed-effects binary regression model to obtain the summary sensitivity, specificity, likelihood ratio, diagnostic odds ratio (OR), and summary ROC curve for each hepatic fibrosis stage.21 Inevitable differing will exist when studies are combined in a meta-analysis. The term heterogeneitymeans any clinical or methodological diversity among included studies. Usually, substantial heterogeneity in a meta-analysis leads to lowerquality results.22To identify heterogeneity, a statistical test is usually conducted. The I 2 statistic of Higgins et al23 is an improved approach for quantifying heterogeneity.23 It ranges from 0% to 100%, and a previous guideline showed that I 2of greater than 50% signified substantial heterogeneity.24 We assessed potential heterogeneity at every fibrosis stage. We used the funnel plot asymmetry test of Deeks et al25 to investigate publication bias at every fibrosis stage. This test is a modified method for diagnostic meta-analysis derived from randomized meta-analysis trials. Deeks et al demonstrated that the log diagnostic OR is more helpful for detecting sample size effects, and the effective sample size (ESS) is more appropriate than the total sample size. Therefore, in the Deeks funnel plot, the horizontal axis represents measure of the effect (log diagnostic OR), and the vertical axis represents the measure of precision ( ). In diagnostic studies, numbers of nondiseased patients = n1, and numbers of diseased patients = n2; ESS = (4n1n2)/(n1 + n2). The random variation of the effect decreases as the sample size increases. Therefore, if there is no publication bias, data from the studies on the diagram will form a symmetric funnel shape and can be formally tested by a linear regression of log diagnostic OR against . P < .10 for the slope coefficient of the regression line would suggest significant asymmetry of the funnel plot and a significant publication bias. To assess the clinical utility of SuperSonic shear imaging, Fagan plot analysis was conducted.26 We assumed pretest probabilities of 25%, 50%, and 75% to calculate corresponding posttest probabilities after positive and negative SuperSonic shear imaging results on the basis of pooled sensitivity and specificity. Positive and negative results were defined as all results from each study that were above and below the liver stiffness cutoff value at a given stage, respectively.11The following prespecified subgroup analyses were conducted: different etiologies, different BMI ranges (<18.50, 18.50–24.99, and ≥25.00 kg/m2 ), different geographic areas, and different histologic scores used. 3/√ESS 3/√ESS Feng et al—SuperSonic Shear Imaging for Staging of Liver Fibrosis 332 J Ultrasound Med 2016; 35:329–339 Table 2. Results of Studies Evaluating the Performance of SuperSonic Shear Imaging for Diagnosis of Liver Fibrosis F≥1 F≥2 F≥3 F=4 Cutoff, SENS, SPEC, Cutoff, SENS, SPEC, Cutoff, SENS, SPEC, Cutoff, SENS, SPEC, Reference AUROC kPa % % AUROC kPa % % AUROC kPa % % AUROC kPa % % Beland27 0.71 9.51 72.2 64.3 0.85 10.72 75.0 75.0 0.94 11.12 88.9 87.5 0.93 12.24 100.0 80.5 Cassinotto28 0.72 7.80 68.0 100.0 0.82 8.00 83.0 82.0 0.84 8.90 90.0 81.0 0.83 10.70 85.0 83.0 Deffieux29 NA NA NA NA 0.81 8.90 77.0 79.0 0.80 9.10 85.0 72.0 0.85 10.20 83.0 76.0 Ferraioli14 NA NA NA NA 0.92 7.10 90.0 87.5 0.98 8.70 97.3 85.1 0.98 10.40 87.5 96.8 Jeong30 NA NA NA NA 0.92 8.60 78.2 93.3 0.91 10.46 88.6 80.0 0.88 14.00 77.3 85.4 Leung31 0.86 6.50 83.5 91.2 0.88 7.10 84.7 92.1 0.93 7.90 89.8 90.3 0.98 10.10 97.4 93.0 Samir32 NA NA NA NA 0.77 7.29 91.4 52.5 0.82 8.90 76.5 76.5 0.82 9.59 71.4 82.2 Suh33 NA 6.20 91.0 95.9 NA NA NA NA NA NA NA NA NA NA NA NA Tada34 NA NA NA NA 0.94 8.80 88.9 91.9 NA NA NA NA NA NA NA NA Yoon35 NA NA NA NA 0.85 6.65 78.8 75.6 NA NA NA NA NA NA NA NA Zeng37 NA NA NA NA 0.91 7.20 85.2 80.9 0.93 9.10 89.7 80.6 0.97 11.70 88.2 88.1 Zheng36 NA NA NA NA 0.86 NA 85.7 73.9 NA NA NA NA 0.93 NA 91.2 79.7 NA indicates not available; SENS, sensitivity; and SPEC, specificity. 3502jum279-400 copy_Layout 1 1/20/16 12:46 PM Page 332
Feng et al-SuperSonic Shear Imaging for Staging of Liver Fibrosis Results Diagnostic Accuracy and Heterogeneity The numbers of studies that reported data for F21,F22, Search Results and Quality of Studies F23,and F=4 were 4,11,8,and 9,respectively.Mean cut- Initially,the titles and abstracts of 44 primary studies were off values for the 4 stages were 7.50+1.51 (range,6.20- identified by the search strategy.Ofthese,14 studies were 9.51),8.04±1.24(range,6.65-10.72),9.27±1.03(range unrelated to the topic;9 studies evaluated liver fibrosis 7.90-11.12),and11.12±1.45(range,9.59-14.00)kPa, without a liver biopsy as the reference standard;1 article respectively. was a review article;2 conference abstracts had insufficient The forest plots of summary sensitivities and specifici- data to extract true-positive,false-positive,false-negative, ties are shown in Figure 1.The forest plots showed that the and true-negative results,and we failed to obtain them pooled sensitivity and specificity were 0.78(95%confidence from the corresponding authors via e-mail;3 studies were interval[CI],0.69-0.85)and0.95(95%CL,0.75-0.99) redundant publications;1 study was excluded because the respectively,for F21,0.84(95%CI,0.81-0.86)and 0.81 maximum interval between liver biopsy and elastography 95%CL,0.74-0.87)forF≥2,0.89(95%CL.0.85-0.93) was greater than 15 months;1 study has less than 30 patients; and0.84(95%CL0.77-0.89)forF≥3，and0.88(95%CL, and 1 study met all of the planned selection criteria but was 0.82-0.91)and0.86(95%CL,0.81-0.90)forF=4,respec- then excluded after our discussion because it was a pedi- tively.The summary ROC curves for the 4 stages are atric study (median age SD,92.71 66.65 months). shown in Figure 2.Other details are shown in Table 4. Finally,12 studies including 1635 patients were retrieved Subgroup analyses were performed and are shown in for the meta-analysis.427-37 All of them were full articles, Table 5.Heterogeneity was reduced significantly when and we did not ask for raw liver stiffness data from corre- differentiating etiologies and geographic areas.We were sponding authors.The results and main characteristics of restricted by the number ofstudies(24 studies are needed the included studies are presented in Tables 1 and 2. in the Midas model)and the data types(most ofthe studies One of the included studies(Beland et al,27 with 50 integrated different etiologies in their statistical analyses), patients)reported the optimized cutoff value,sensitivity, so we could only perform different etiologies(viral hepa- and specificity only for F22,but the raw data were avail- titis versus mixed)for F22 and different geographic areas able in the article.In our study,we took the Young modu- (Europe and North America versus Asia)for F22 and F=4. lus as the unit and calculated optimized cutoff values, A sensitivity analysis was conducted for stages F22, sensitivities,and specificities for the other stages (F21, F23,and F=4.We removed 2 studies that used different F23,and F=4).The quality of the included studies is histologic scores,1 study with BMIs of25 or higher,and 1 shown in Table 3.Most of the studies had a potential risk study with a 12-month maximum interval between biopsy of bias for the index test. and elastography,but no significant results were found. Table 3.Quality Assessment of the Included Studies Using the Quality Assessment of Diagnostic Accuracy Studies Questionnaire Risk of Bias Applicability Concerns Patient Index Reference Flow and Patient Index Reference Reference Selection Test Standard Timing Selection Test Standard Beland27 Unclear High Unclear Low Unclear Low LOw Cassinotto28 Low High LOW High LOw LOw Low Deffieux29 Low High Low Low Low Low Low Ferraioli14 Low High Low Low Low Low Low Jeong30 Low High Unclear Unclear LOW Low Low Leung31 LOw High Unclear LOW LOW Low Low Samir32 Unclear High Low LOW Low Low Low Suh33 High High Unclear High High Low Low Tada34 LOw High LOW Unclear LOW Low Low Yoon35 Unclear High Low High Unclear Low Low Zeng37 Low LOw Low LOW Low Low Low Zheng36 LOw High Low Unclear LOW Low Low High indicates high risk of bias;low,low risk of bias;and unclear,unclear risk of bias,which means there was insufficient information to permit judgment. JUltrasound Med 2016:35:329-339 333
Results Search Results and Quality of Studies Initially, the titles and abstracts of 44 primary studies were identified by the search strategy. Of these, 14 studies were unrelated to the topic; 9 studies evaluated liver fibrosis without a liver biopsy as the reference standard; 1 article was a review article; 2 conference abstracts had insufficient data to extract true-positive, false-positive, false-negative, and true-negative results, and we failed to obtain them from the corresponding authors via e-mail; 3 studies were redundant publications; 1 study was excluded because the maximum interval between liver biopsy and elastography was greater than 15 months; 1 study has less than 30 patients; and 1 study met all of the planned selection criteria but was then excluded after our discussion because it was a pediatric study (median age ± SD, 92.71 ± 66.65 months). Finally, 12 studies including 1635 patients were retrieved for the meta-analysis.14,27–37 All of them were full articles, and we did not ask for raw liver stiffness data from corresponding authors. The results and main characteristics of the included studies are presented in Tables 1 and 2. One of the included studies (Beland et al,27 with 50 patients) reported the optimized cutoff value, sensitivity, and specificity only for F≥2, but the raw data were available in the article. In our study, we took the Young modulus as the unit and calculated optimized cutoff values, sensitivities, and specificities for the other stages (F≥1, F≥3, and F=4). The quality of the included studies is shown in Table 3. Most of the studies had a potential risk of bias for the index test. Diagnostic Accuracy and Heterogeneity The numbers of studies that reported data for F≥1, F≥2, F≥3, and F=4 were 4, 11, 8, and 9, respectively. Mean cutoff values for the 4 stages were 7.50 ± 1.51 (range, 6.20– 9.51), 8.04 ± 1.24 (range, 6.65–10.72), 9.27 ± 1.03 (range, 7.90–11.12), and 11.12 ± 1.45 (range, 9.59–14.00) kPa, respectively. The forest plots of summary sensitivities and specificities are shown in Figure 1. The forest plots showed that the pooled sensitivity and specificity were 0.78 (95% confidence interval [CI], 0.69–0.85) and 0.95 (95% CI, 0.75–0.99), respectively, for F≥1, 0.84 (95% CI, 0.81–0.86) and 0.81 (95% CI, 0.74–0.87) for F≥2, 0.89 (95% CI, 0.85–0.93) and 0.84 (95% CI, 0.77–0.89) for F≥3, and 0.88 (95% CI, 0.82–0.91) and 0.86 (95% CI, 0.81–0.90) for F=4, respectively. The summary ROC curves for the 4 stages are shown in Figure 2. Other details are shown in Table 4. Subgroup analyses were performed and are shown in Table 5. Heterogeneity was reduced significantly when differentiating etiologies and geographic areas. We were restricted by the number of studies (≥4 studies are needed in the Midas model) and the data types (most of the studies integrated different etiologies in their statistical analyses), so we could only perform different etiologies (viral hepatitis versus mixed) for F≥2 and different geographic areas (Europe and North America versus Asia) for F≥2 and F=4. A sensitivity analysis was conducted for stages F≥2, F≥3, and F=4. We removed 2 studies that used different histologic scores, 1 study with BMIs of 25 or higher, and 1 study with a 12-month maximum interval between biopsy and elastography, but no significant results were found. J Ultrasound Med 2016; 35:329–339 333 Feng et al—SuperSonic Shear Imaging for Staging of Liver Fibrosis Table 3. Quality Assessment of the Included Studies Using the Quality Assessment of Diagnostic Accuracy Studies Questionnaire Risk of Bias Applicability Concerns Patient Index Reference Flow and Patient Index Reference Reference Selection Test Standard Timing Selection Test Standard Beland27 Unclear High Unclear Low Unclear Low Low Cassinotto28 Low High Low High Low Low Low Deffieux29 Low High Low Low Low Low Low Ferraioli14 Low High Low Low Low Low Low Jeong30 Low High Unclear Unclear Low Low Low Leung31 Low High Unclear Low Low Low Low Samir32 Unclear High Low Low Low Low Low Suh33 High High Unclear High High Low Low Tada34 Low High Low Unclear Low Low Low Yoon35 Unclear High Low High Unclear Low Low Zeng37 Low Low Low Low Low Low Low Zheng36 Low High Low Unclear Low Low Low High indicates high risk of bias; low, low risk of bias; and unclear, unclear risk of bias, which means there was insufficient information to permit judgment. 3502jum279-400 copy_Layout 1 1/20/16 12:46 PM Page 333