Research ArticleContrast-Enhanced EUS for Differential Diagnosis of PancreaticMasses: A Meta-Analysis

Sibin Mei,1,2 Mengyu Wang,1 and Leimin Sun 1,2

1Department of Gastroenterology, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou 310016, China2Institute of Gastroenterology, Zhejiang University (IGZJU), Hangzhou 310016, China

Correspondence should be addressed to Leimin Sun; sunlm@zju.edu.cn

Received 6 October 2018; Accepted 4 February 2019; Published 6 March 2019

Academic Editor: Jose Celso Ardengh

Copyright © 2019 Sibin Mei et al. This is an open access article distributed under the Creative Commons Attribution License, whichpermits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background. Though methods for the diagnosis of pancreatic masses are various, such as ultrasonography (US), computedtomography (CT), endoscopic ultrasonography (EUS), and contrast-enhanced computed tomography (CE-CT), their sensitivity,specificity, and accuracy are not quite satisfying. Contrast-enhanced endoscopic ultrasonography (CE-EUS), as a new technique,has its own unique advantages in diagnosing pancreatic disease. However, its sensitivity, specificity, and accuracy are stillcontroversial. Objective. To evaluate the accuracy of CE-EUS for differential diagnosis between benign and malignant pancreaticmass lesions. Design. Eighteen relevant articles systemically searched from PubMed, Web of Science, Ovid, Scopus, andMEDLINE were selected. The pooled results were calculated in a fixed effects model. Main Outcome Measurement. The pooledsensitivity, specificity, positive likelihood ratio (LR), negative likelihood ratio, diagnostic odds ratio (OR), and summary receiveroperating characteristic (SROC) curve. Results. The pooled sensitivity, specificity, and diagnostic odds ratio of CE-EUS for thedifferential diagnosis of pancreatic adenocarcinomas were 0.91 (95% confidence interval (CI), 0.89-0.93), 0.86 (95% CI,0.83-0.89), and 69.50 (95% CI, 48.89-98.80), respectively. The SROC area under the curve was 0.9545. The subgroup analysisbased on excluding the outliers showed that the heterogeneity was eliminated and the pooled sensitivity and specificity were 0.92(95% CI, 0.90-0.93) and 0.87 (95% CI, 0.84-0.89), respectively. The SROC area under the curve was 0.9569. Conclusion. CE-EUSis a useful method to distinguish pancreatic adenocarcinoma from other pancreatic diseases. Compared with EUS elastography,it has higher specificity. However, it is still not superior to pathological diagnosis for the identification of pancreatic carcinomas.

1. Introduction

As the fourth leading cause of cancer-related death inWestern countries [1], pancreatic cancer remains a bigconcern and health burden for both the government andindividuals. Due to the lack of early screening methods,patients are often diagnosed with progressive pancreaticcancer on their first visit. Meanwhile, the effectivity rate ofcurrently available treatments, such as chemotherapy, radio-therapy, and surgery, is relatively low [2]. The prognosis ofpancreatic cancer is poor, with median survival time rangingfrom 6 to 10 months [3].

Therefore, to improve the early diagnostic rate of pancre-atic cancer, early detection, and differential diagnosis ofpancreatic masses are critical. Current methods available

for differential diagnosis of pancreatic masses are various,among which EUS, contrast-enhanced ultrasonography(CE-US), CE-CT, CE elastography, endoscopic ultrasonogra-phy fine-needle aspiration (EUS-FNA), and CE-EUS arecommonly used in recent years.

EUS shows quite high sensitivity due to its high resolu-tion. However, its specificity is relatively low. Tumors, suchas pancreatic neuroendocrine carcinoma and inflammatorycystadenoma, are all presented with low-echo crumb images,which are hard to distinguish on EUS.

EUS-FNA is commonly used in clinical settings as thegold standard for pancreatic mass diagnosis. The diagnosticsensitivity and specificity are 80%-85% and nearly 100%,respectively. However, related complications, such as acutepancreatitis, hemorrhage, and infection, may occur. In

HindawiGastroenterology Research and PracticeVolume 2019, Article ID 1670183, 9 pageshttps://doi.org/10.1155/2019/1670183

addition, when patients have concomitant pancreatitis, thefalse negative rate of pancreatic mass diagnosis usingEUS-FNA can increase to 20%-40% [4–6].

CE-EUS is a newly established method which combinesthe advantage of high-resolution ultrasound of internalorgans with the administration of ultrasound contrast agents[7]. It was first applied in pancreatic mass diagnosis in 1995[8]. As a noninvasive diagnostic technique combining bothadvantages of EUS and CE, it can further distinguish proper-ties of low-echo pancreatic masses by the density of bloodvessels in pancreatic tumor imaging.

Currently, which method is better for pancreatic massdiagnosis remains unclear. Therefore, this meta-analysiswas conducted to evaluate the accuracy of CE-EUS and tofigure out whether CE-FNA can be replaced by CE-EUSin pancreatic mass diagnosis. The earlier version of thismeta-analysis has been presented as an abstract at the 15thCongress of Gastroenterology, China [9]. Here, we presentedwith the latest and detailed version of this study.

2. Materials and Methods

2.1. Literature Search. We searched related references inPubMed,Web of Science, Ovid, Scopus, andMEDLINE data-bases from 1980.01 to 2015.12. The search strategies are “con-trast imaging, contrast enhancement, echo enhanced, contrastenhanced, EUS, endoscopic ultrasonography, endoscopicultrasound, ultrasonography, endosonography, pancreatitis,pancreatic cancer, pancreas, pancreatic mass, pancreaticmasses.”We also searched cited references of selected studiesto expand our research database. Two authors independentlysearched and extracted the data, with disagreements beingresolved by discussion.

2.2. Study Selection. Inclusion criteria are as follows: (1) use ofCE-EUS for the diagnosis of solid pancreatic masses, (2) useof a reference standard of EUS-FNA samples, surgical histol-ogy, or clinical follow-up of at least 6 months, and (3) avail-able data to construct contingency tables for true positive,true negative, false positive, and false negative results.

Exclusion criteria are as follows: (1) no final diagnosis ofpancreatic masses, (2) incomplete data to construct contin-gency tables, (3) repeated abstracts or articles, (4) reviews,and (5) case reports.

2.3. Quality of Studies. Eighteen studies were included in thismeta-analysis. The quality of included studies was evaluatedusing QUADAS, with a total of 14 questions being answeredwith yes, no, or not clear.

2.4. Statistical Methods. We used the extracted data to con-struct a 2∗2 contingency table for each study with 0.5 beingadded to the contingency tables of the studies with 0. Weassessed the accuracy of CE-EUS by calculating the pooledsensitivity, specificity, positive likelihood ratio (LR), negativeLR, and diagnostic odds ratios (OR). Pooled results were con-structed by using both the Mantel-Haenszel method (fixedeffects model) when significant heterogeneity was absentand the DerSimonian-Laird method (random effects model)if otherwise. We used the Cochrane Q test to test the

heterogeneity among studies. Inconsistency (I2) expressesthe variability attributable to heterogeneity across the studiesin the form of a percentage. I2 > 50% was considered signifi-cant for heterogeneity. A summary receiver operating char-acteristic (SROC) curve was produced. The value of thearea under the curve close to 1 indicated a well-validateddiagnostic test.

For the studies using both diagnostic tests, sensitivityanalyses were conducted by using either one of the two orboth. Publication bias was further tested by Deeks’ asymme-try test, and a funnel plot of diagnostic log odds ratio versus1/sqrt (effective sample size) was constructed. The slope coef-ficient with P < 0 05 indicated the presence of publicationbias. The pooled sensitivity, specificity, positive LR, negativeLR, diagnostic OR, SROC curve, and meta-regression wereanalyzed by using freeware Meta-DiSc, version 1.4 (Ramony Cajal Hospital, Madrid, Spain). Publication bias wasexamined by using StataMP, version 13.0 (Stata Corporation,College Station, TX).

3. Result

3.1. Studies. According to the inclusion and exclusion cri-teria, 18 articles (4 abstracts and 14 full-text articles) wereeventually included in this meta-analysis, with a total of1497 patients involved, as shown in Figure 1 [10–27].

3.2. Quality Assessment. Qualities of the eligible studies areassessed according to the QUADAS criteria. The percentageof “yes” of the 18 studies varied from 64.29% to 92.86%.For each item, the percentage varied from 27.78% to 100%.

3.3. Meta-Analysis. CE-EUS had a pooled sensitivity (fixedeffects model) of 0.91 (95% CI, 0.89-0.93) in the differentialdiagnosis of pancreatic adenocarcinomas and other pancre-atic focal masses (Figure 2). The pooled specificity (fixedeffects model) was 0.86 (95% CI, 0.83-0.89) (Figure 3).Significant heterogeneity was not found neither in sensitivity(Cochran Q test = 25 22, df = 17, P = 0 0899, I2 = 32 6%) norin specificity (Cochran Q test = 30 93, df = 17, P = 0 0204,I2 = 45%). The AUC under the SROC was 0.9545 (standarderror SE = 0 0088) (Figure 4). CE-EUS had a pooled posi-tive LR (fixed effects model) of 6.83 (95% CI, 5.56-8.40)(Figure 5) and a pooled negative LR (fixed effects model)of 0.10 (95% CI, 0.08-0.13) in the diagnosis of pancreaticadenocarcinomas and other pancreatic focal masses(Figure 6). The pooled diagnostic OR was 69.50 (95%CI, 48.89-98.80) (Figure 7). There was no significant het-erogeneity neither in positive LR (Cochran Q test = 30 68,df = 17, P = 0 0218, I2 = 44 6%) nor in pooled negativeLR among the studies (Cochran Q test = 28 41, df = 17,P = 0 0403, I2 = 40 2%).

Deeks’ funnel plot of diagnostic log odds ratio versus1/sqrt (effective sample size) did not show significant asym-metry (P = 0 17), indicating that there was no significantpublication bias in this meta-analysis (Figure 8).

2 Gastroenterology Research and Practice

4. Discussion

EUS, as a technique applied in diagnosing pancreatic tumorssince the 1980s, has attracted increasing attention due to itshigh-resolution and imaging quality [28]. A great deal ofwork has reached the conclusion that EUS is more precisecompared with traditional abdominal imaging, such as US,CT, and MRI [29–31]. Low-echo conglomeration shadow isthe characteristic manifestation of pancreatic tumors withquite high sensitivity and specificity. But not all the

low-echo conglomerations are pancreatic adenocarcinomas.Pancreatic neuroendocrine tumor, pancreatic cyst, andchronic pancreas inflammatory cystadenoma are all pre-sented with the image of low-echo conglomeration on theEUS. Therefore, further differentiation between these massesis necessary.

As one of the gold standards of diagnosing pancreatictumors, EUS-FNA has a specificity of nearly 100% throughaccessing the pathological diagnosis directly. However, itshigh false negative rate of 20%-40%, especially when it comes

287 studies acquiredfrom databases

77 potential relevant studiesfor more details

18 eligible studies(14 full text and 4 abstracts)

210 studies were excluded by titles

59 studies were excluded39 studies without EUS9 studies not included pancreatic carcinomas4 studies without contrast-enhanced agents3 studies with data available2 studies were case reports2 studies without pancreatic masses

Figure 1: Flow diagram of systematic literature search.

Sensitivity0 .2 .4 .6 .8 1

Jin-Seok Park 0.92 (0.82 - 0.97)Dan Ionut 0.94 (0.79 - 0.99)Hocke M 0.84 (0.60 - 0.97)Hiroo Imazu 0.80 (0.59 - 0.94)Masayuki Kitano 0.95 (0.91 - 0.98)Hiroshi Matsubara 0.88 (0.75 - 0.95)Joseph Romagnuolo 0.75 (0.39 - 0.96)Gincul 0.96 (0.88 - 0.99)Fusaroli 0.96 (0.87 - 1.00)Adrian S 0.91 (0.76 - 0.98)Napoleon B 0.89 (0.65 - 0.99)Hocke M 0.92 (0.85 - 0.96)Sakamoto H 0.94 (0.81 - 0.99)Christoph 0.92 (0.82 - 0.97)Michael Hocke 0.78 (0.64 - 0.88)Michael Hocke 0.91 (0.80 - 0.97)Dirk Becker 0.91 (0.67 - 0.99)Yoshiki Hirooka 0.96 (0.67 - 1.00)

Sensitivity (95% CI)

Pooled sensitivity = 0.91 (0.89 to 0.93)Chi-squared = 25.22; df = 17 (p = 0.0899)Inconsistency (I-squared) = 32.6%

Figure 2: Sensitivity of the selected studies and this meta.

3Gastroenterology Research and Practice

to diagnosis of pancreatic tumors combined with chronicpancreatitis, limits its clinical application. CE elastographyis a new diagnosis technique which has gained significant

effects in clinical settings recently. A meta-analysis investi-gating CE elastography showed that it has a high sensitivityof 0.95 in the diagnosis of benign and malignant pancreatic

Specificity0 .2 .4 .6 .8 1

Jin-SeokPark 0.68 (0.48 - 0.84)DanIonut 0.89 (0.67 - 0.99)Hocke M 0.92 (0.79 - 0.98)Hiroo Imazu 0.94 (0.59 - 1.00)Masayuki Kitano 0.88 (0.78 - 0.94)Hiros hi Matsubara 0.86 (0.72 - 0.95)Joseph Romagnuolo 0.75 (0.06 - 1.00)Gincul 0.97 (0.83 - 1.00)Fusaroli 0.77 (0.61 - 0.89)Adrian S 0.71 (0.48 - 0.89)Napoleon B 0.88 (0.64 - 0.99)Hocke M 0.96 (0.88 - 0.99)Sakamoto H 0.83 (0.75 - 0.90)Christoph 0.84 (0.66 - 0.95)Michael Hocke 0.87 (0.74 - 0.95)Michael Hocke 0.93 (0.78 - 0.99)Dirk Becker 0.94 (0.55 - 1.00)Yoshiki Hirooka 0.78 (0.61 - 0.90)

Specificity (95%CI)

Pooled specificity = 0.86 (0.83 to 0.89)Chi-squared = 30.93; df = 17 (p = 0.0204)Inconsistency (I-squared) = 45.0%

Figure 3: Specificity of the selected studies and this meta.

Sensitivity SROC curve

1 − specificity0 .2 .4 .6 .8 1

0

.1

.2

.3

.4

.5

.6

.7

.8

.9

1

Symmetric SROCAUC = 0.9545SE (AUC) = 0.0086Q⁎ = 0.8967SE (Q⁎) = 0.0119

Figure 4: Symmetric SROC of the selected studies and this meta.

4 Gastroenterology Research and Practice

tumors but the specificity is only 0.67 [32]. Thus, CE elasto-graphy can hardly take the place of CE-FNA if its specificitycould not be improved in the future.

In comparison, CE-EUS owns different mechanisms andshows relatively higher diagnostic accuracy. Contrast agentsused in EUS are gas-containing microbubbles, which would

oscillate and generate an acoustic signal when hit by an ultra-sonicwave [33]. Therefore, apart fromassessing the echogeni-city of pancreatic masses themselves, CE-EUS could assesstheir vascularity by detecting the acoustic signal after intrave-nous bolus injection of contrast agents. Through demarcatingvascular landmarks, detecting vascular obliteration by a

Positive LR0.01 100.01

Jin-Seok Park 2.86 (1.66-4.92)Dan Ionut 8.91 (2.39-33.14)Hocke M 10.95 (3.63-33.04)Hiroo Imazu 14.48 (0.97-215.70)Masayuki Kitano 7.71 (4.18-14.23)Hiroshi Matsubara 6.27 (2.96-13.27)Joseph Romagnuolo 3.00 (0.26-33.97)Gincul 29.65 (4.31-204.03)Fusaroli 4.16 (2.34-7.40)Adrian S 3.18 (1.60-6.31)Napoleon B 7.56 (2.03-28.06)Hocke M 22.32 (7.36-67.69)Sakamoto H 5.67 (3.77-8.52)Christoph 5.70 (2.55-12.76)Michael Hocke 5.96 (2.79-12.75)Michael Hocke 13.66 (3.57-52.25)Dirk Becker 14.59 (0.99-214.45)Yoshiki Hirooka 4.28 (2.34-7.84)

Positive LR (95% CI)

Fixed effects modelPooled positive LR = 6.83 (5.56 to 8.40)Cochran Q = 30.68; df = 17 (p = 0.0218)Inconsistency (I-squared) = 44.6 %

Figure 5: Positive LR of the selected studies and this meta.

Negative LR0.01 100.01

Jin-Seok Park 0.12 (0.05 - 0.29)Dan Ionut 0.07 (0.02 - 0.27)Hocke M 0.17 (0.06 - 0.49)Hiroo Imazu 0.21 (0.09 - 0.48)Masayuki Kitano 0.06 (0.03 - 0.10)Hiroshi Matsubara 0.15 (0.07 - 0.31)Joseph Romagnuolo 0.33 (0.09 - 1.27)Gincul 0.04 (0.01 - 0.14)Fusaroli 0.05 (0.01 - 0.20)Adrian S 0.13 (0.04 - 0.39)Napoleon B 0.13 (0.03 - 0.47)Hocke M 0.09 (0.05 - 0.16)Sakamoto H 0.07 (0.02 - 0.26)Christoph 0.10 (0.04 - 0.23)Michael Hocke 0.26 (0.15 - 0.43)Michael Hocke 0.10 (0.04 - 0.22)Dirk Becker 0.09 (0.02 - 0.44)Yoshiki Hirooka 0.05 (0.00 - 0.81)

Fixed effects modelPooled negative LR = 0.10 (0.08 to 0.13)Cochran Q = 28.41; df = 17 (p = 0.0403)Inconsistency (I-squared) = 40.2 %

Negative LR (95%CI)

Figure 6: Negative LR of the selected studies and this meta.

5Gastroenterology Research and Practice

thrombus or tumor, and examining microvascular blood flowto organs and lesions [34],more information can be gained forfurther differentiation.

Based on the mechanical index (MI) displayed on themonitor to indicate the acoustic power, CE-EUS can befurther divided into contrast-enhanced high-mechanicalindex endoscopic ultrasonography (CEHMI-EUS) andcontrast-enhanced low-mechanical index endoscopic ultra-sonography (CELMI-EUS) [35]. CEHMI-EUS is com-monly used with the first-generation contrast enhancers,which requires microbubble destruction and provides onlyscattered scanning of the lesion [26], whereas CELMI-EUSis applied with the new contrast enhancers, providing cli-nicians with real-time viewing of the contrast enhancereffects [26].

Currently, ultrasound contrast agents have alreadydeveloped to the third generation. SonoVue and Sonazoidare most commonly used in clinical settings, and they bothbelong to the second generation [33]. In this meta-analysis,SonoVue was used in 7 studies and Sonazoid was used in 4studies as the contrast enhancers. Both are excreted throughthe respiratory system; therefore, CE-EUS can also beapplied to patients with liver and renal dysfunction. Severalstudies have shown that the ultrasound contrast agents aresafe and effective and have no obvious long-term adverseeffects [36, 37].

Differential diagnosis of pancreatic masses can beachieved based on their CE-EUS imaging findings. InCE-EUS, diffuse homogeneous enhancement is presentedin normal pancreatic parenchyma and the normal bileduct and pancreatic duct are depicted as nonenhancedductal structures [38]. Pancreatic adenocarcinomas aretypically hypoenhanced [39]. Within the tumor, arterial

vessel architecture is irregular and venous vessels are absent[40]. In comparison, focal chronic pancreatitis is com-monly hyperenhanced with regular arterial microvasculararchitecture [40]. Besides, both venous and arterial vesselsare presented in chronic pancreatitis [40]. Small singularwell-demarcated lesion with hyperenhancement stronglyindicates neuroendocrine tumors of the pancreas [40].For pancreatic cystic lesions, in general, wall and nodulevascularization indicates a cystic tumor which cannot beseen in dysontogenetic cysts or pseudocysts [41]. Withincystic tumors, typical serous cystadenoma shows the intrale-sional septation enhancement [42]. In comparison, mucin-ous cystadenoma is characterized by intralesional irregularseptation and parietal nodule enhancement [42]. Malignancyof intraductal papillary mucinous neoplasms (IPMNs) can beassessed according to the enhancement pattern of muralnodules. Papillary module and invasive nodule are evidencesof invasive IPMNs while low papillary nodule and polypoidnodule indicate benign IPMNs [38]. Solid pseudopapillaryneoplasms (SPNs) are presented with inhomogeneousenhancement of the thickened peripheral capsule and solidcomponents surrounding the cystic and necrotic avascularareas [42].

This meta-analysis shows that CE-EUS has a highsensitivity of 0.92 (95% CI, 0.90-0.93) and a relatively highspecificity of 0.86 (95% CI, 0.84-0.89) in the diagnosis ofbenign and malignant pancreatic tumors. There is noobvious heterogeneity between studies. The Spearman cor-relation coefficient between the log of sensitivity and log of1−specificity was 0.089 (P = 0 751), which shows no obvi-ous threshold effect between studies. Meta-regression anal-ysis also finds no possible sources of heterogeneity. Thefunnel chart shows no publication bias. We exclude the

Diagnostic odds ratio0.01 100.01

Jin-Seok Park 24.07 (7.17 - 80.73)Dan Ionut 127.50 (16.44 - 988.72)Hocke M 64.00 (11.63 - 352.17)Hiroo Imazu 69.89 (3.37 - 1 450.17)Masayuki Kitano 137.96 (53.68 - 354.52)Hiroshi Matsubara 43.17 (12.81 - 145.46)Joseph Romagnuolo 9.00 (0.27 - 299.86)Gincul 660.00 (65.91 - 6 609.08)Fusaroli 81.67 (16.52 - 403.74)Adrian S 25.00 (5.48 - 114.10)Napoleon B 60.00 (7.48 - 481.57)Hocke M 259.00 (68.88 - 973.86)Sakamoto H 85.00 (18.88 - 382.75)Christoph 59.28 (15.78 - 222.70)Michael Hocke 23.33 (7.99 - 68.12)Michael Hocke 142.80 (26.00 - 784.34)Dirk Becker 155.00 (5.62 - 4 275.48)Yoshiki Hirooka 79.82 (4.25 - 1 498.55)

Diagnostic OR (95%CI)

Fixed effects modelPooled diagnostic odds ratio = 69.50 (48.89 to 98.80)Cochran Q = 21.51; df = 17 (p = 0.2044)Inconsistency (I-square) = 21.0 %

Figure 7: Diagnostic odds ratio of the selected studies and this meta.

6 Gastroenterology Research and Practice

articles from both Park et al. [14] and Romagnuolo et al. [27]according to the summary of the forest figure. Subgroupanalysis shows that pooled calculation values increase.

5. Conclusions

Taken together, it is still difficult for CE-EUS to replaceCE-FNA in making a definite diagnosis. However, consider-ing its higher accuracy compared with CT and noninvasivecharacteristic compared with CE-FNA, CE-EUS should bean important part in the pancreatic mass diagnosis algo-rithm. For patients clinically suspicious of pancreatic mass,helical CT should be firstly conducted. If patients are pre-sented with negative CT results but still strongly suspiciousof pancreatic mass or patients are presented with ambiguousCT results and need further confirmation, CE-EUS should bethe choice. Based on the result of CE-EUS, clinicians candecide whether CE-FNA is needed or not.

In the future, the accumulation of operating experience,the establishment of a more accurate diagnosis standard,the development of CE-EUS technologies, and a largeramount of summary data analysis may help to improve thediagnostic accuracy in the diagnosis between benign andmalignant pancreatic tumors using CE-EUS. Which is better

between CE-EUS and CE-FNA for pancreatic mass diagnosiswill be an interesting debate at that time.

Data Availability

The data used to support the findings of this study areincluded within the article.

Conflicts of Interest

The authors declare that there is no conflict of interestregarding the publication of this paper.

References

[1] M. Del Chiaro, R. Segersvärd, M. Lohr, and C. Verbeke, “Earlydetection and prevention of pancreatic cancer: is it really pos-sible today?,” World Journal of Gastroenterology, vol. 20,no. 34, pp. 12118–12131, 2014.

[2] J. R. Hecht, J. J. Farrell, N. Senzer et al., “EUS or percutane-ously guided intratumoral TNFerade biologic with5-fluorouracil and radiotherapy for first-line treatment oflocally advanced pancreatic cancer: a phase I/II study,” Gastro-intestinal Endoscopy, vol. 75, no. 2, pp. 332–338, 2012.

1

23

4

5

6

7

89

10

11

12

1314

1516

17

18

0

.1

.2

.3

.4

1/ro

ot (E

SS)

1 10 100 1000Diagnostic odds ratio

StudyRegression line

Deek's funnel plot asymmetry testP value = 0.17

Figure 8: Deeks’ funnel plot asymmetry test of the selected studies.

7Gastroenterology Research and Practice

[3] G. Schneider, J. T. Siveke, F. Eckel, and R. M. Schmid, “Pancre-atic cancer: basic and clinical aspects,” Gastroenterology,vol. 128, no. 6, pp. 1606–1625, 2005.

[4] J. DeWitt, K. McGreevy, S. Sherman, and J. LeBlanc, “Utility ofa repeated EUS at a tertiary-referral center,” GastrointestinalEndoscopy, vol. 67, no. 4, pp. 610–619, 2008.

[5] J. Iglesias Garcia and J. E. Dominguez-Munoz, “Endoscopicultrasound-guided biopsy for the evaluation of pancreatictumors,” Gastroenterología y Hepatología, vol. 30, no. 10,pp. 597–601, 2007.

[6] S. Varadarajulu, A. Tamhane, and M. A. Eloubeidi, “Yield ofEUS-guided FNA of pancreatic masses in the presence or theabsence of chronic pancreatitis,” Gastrointestinal Endoscopy,vol. 62, no. 5, pp. 728–736, 2005.

[7] C. F. Dietrich, M. Sharma, and M. Hocke, “Contrast-enhancedendoscopic ultrasound,” Endosopic Ultrasound, vol. 1, no. 3,pp. 130–136, 2012.

[8] T. Kato, Y. Tsukamoto, Y. Naitoh, Y. Hirooka, T. Furukawa,and T. Hayakawa, “Ultrasonographic and endoscopic ultraso-nographic angiography in pancreatic mass lesions,” Acta Radi-ologica, vol. 36, no. 4-6, pp. 381–387, 1995.

[9] L. M. Sun, S. B. Mei, and M. Y. Wang, “Contrast-enhancedEUS for differential diagnosis of pancreatic masses: ameta-analysis,” Journal of Digestive Diseases, vol. 16, no. S1,p. 47, 2015.

[10] H. Sakamoto, M. Kitano, Y. Suetomi, K. Maekawa,Y. Takeyama, and M. Kudo, “Utility of contrast-enhancedendoscopic ultrasonography for diagnosis of small pancreaticcarcinomas,” Ultrasound in Medicine & Biology, vol. 34,no. 4, pp. 525–532, 2008.

[11] D. I. Gheonea, C. T. Streba, T. Ciurea, and A. Săftoiu, “Quan-titative low mechanical index contrast-enhanced endoscopicultrasound for the differential diagnosis of chronic pseudotu-moral pancreatitis and pancreatic cancer,” BMC Gastroenter-ology, vol. 13, no. 1, p. 2, 2013.

[12] H. Imazu, K. Kanazawa, N. Mori et al., “Novel quantitativeperfusion analysis with contrast-enhanced harmonic EUS fordifferentiation of autoimmune pancreatitis from pancreaticcarcinoma,” Scandinavian Journal of Gastroenterology,vol. 47, no. 7, pp. 853–860, 2012.

[13] C. F. Dietrich, A. Ignee, B. Braden, A. P. Barreiros, M. Ott, andM. Hocke, “Improved differentiation of pancreatic tumorsusing contrast-enhanced endoscopic ultrasound,” ClinicalGastroenterology and Hepatology, vol. 6, no. 5, pp. 590–597.e1, 2008.

[14] J. S. Park, H. K. Kim, B. W. Bang, S. G. Kim, S. Jeong, and D. H.Lee, “Effectiveness of contrast-enhanced harmonic endoscopicultrasound for the evaluation of solid pancreatic masses,”World Journal of Gastroenterology, vol. 20, no. 2, pp. 518–524, 2014.

[15] D. Becker, D. Strobel, T. Bernatik, and E. G. Hahn,“Echo-enhanced color- and power-Doppler EUS for thediscrimination between focal pancreatitis and pancreaticcarcinoma,” Gastrointestinal Endoscopy, vol. 53, no. 7,pp. 784–789, 2001.

[16] H. Matsubara, A. Itoh, H. Kawashima et al., “Dynamic quanti-tative evaluation of contrast-enhanced endoscopic ultrasonog-raphy in the diagnosis of pancreatic diseases,” Pancreas,vol. 40, no. 7, pp. 1073–1079, 2011.

[17] R. Gincul, M. Palazzo, B. Pujol et al., “Contrast-harmonicendoscopic ultrasound for the diagnosis of pancreatic

adenocarcinoma: a prospective multicenter trial,” Endoscopy,vol. 46, no. 5, pp. 373–379, 2014.

[18] B. Napoleon, M. Alvarez-Sanchez, R. Gincoul et al., “Contras-t-enhanced harmonic endoscopic ultrasound in solid lesions ofthe pancreas: results of a pilot study,” Endoscopy, vol. 42,no. 07, pp. 564–570, 2010.

[19] M. Hocke, A. Ignee, T. Topalidis, A. Stallmach, and C. F.Dietrich, “Contrast-enhanced endosonographic Doppler spec-trum analysis is helpful in discrimination between focal chronicpancreatitis and pancreatic cancer,” Pancreas, vol. 35, no. 3,pp. 286–288, 2007.

[20] M. Hocke, E. Schulze, P. Gottschalk, T. Topalidis, and C. F.Dietrich, “Contrast-enhanced endoscopic ultrasound in dis-crimination between focal pancreatitis and pancreatic cancer,”World Journal of Gastroenterology, vol. 12, no. 2, pp. 246–250,2006.

[21] Y. Hirooka, H. Goto, A. Ito et al., “Contrast-enhanced endo-scopic ultrasonography in pancreatic diseases: a preliminarystudy,” The American Journal of Gastroenterology, vol. 93,no. 4, pp. 632–635, 1998.

[22] P. Fusaroli, A. Spada, M. G. Mancino, and G. Caletti, “Contrastharmonic echo-endoscopic ultrasound improves accuracy indiagnosis of solid pancreatic masses,” Clinical Gastroen-terology and Hepatology, vol. 8, no. 7, pp. 629–634.e2,2010.

[23] M. Hocke, C. Schmidt, B. Zimmer, T. Topalidis, C. Dietrich,and A. Stallmach, “Kontrastverstärkte Endosonographie zurVerbesserung der Differenzialdiagnose zwischen chron-ischer Pankreatitis und Pankreaskarzinom,” Deutsche Med-izinische Wochenschrift, vol. 133, no. 38, pp. 1888–1892,2008.

[24] A. Săftoiu, S. Iordache, D. I. Gheonea et al., “Combinedcontrast-enhanced power Doppler and real-time sonoelasto-graphy performed during EUS, used in the differential diagno-sis of focal pancreatic masses (with videos),” GastrointestinalEndoscopy, vol. 72, no. 4, pp. 739–747, 2010.

[25] M. Kitano, M. Kudo, K. Yamao et al., “Characterization ofsmall solid tumors in the pancreas: the value ofcontrast-enhanced harmonic endoscopic ultrasonography,”The American Journal of Gastroenterology, vol. 107, no. 2,pp. 303–310, 2012.

[26] M. Hocke, A. Ignee, and C. F. Dietrich, “Advanced endosono-graphic diagnostic tools for discrimination of focal chronicpancreatitis and pancreatic carcinoma–elastography, contrastenhanced high mechanical index (CEHMI) and low mechani-cal index (CELMI) endosonography in direct comparison,”Zeitschrift für Gastroenterologie, vol. 50, no. 2, pp. 199–203,2012.

[27] J. Romagnuolo, B. Hoffman, S. Vela, R. Hawes, and S. Vignesh,“Accuracy of contrast-enhanced harmonic EUS with asecond-generation perflutren lipid microsphere contrast agent(with video),” Gastrointestinal Endoscopy, vol. 73, no. 1,pp. 52–63, 2011.

[28] T. T. Gong, D. M. Hu, and Q. Zhu, “Contrast-enhancedEUS for differential diagnosis of pancreatic mass lesions: ameta-analysis,” Gastrointestinal Endoscopy, vol. 76, no. 2,pp. 301–309, 2012.

[29] B. Agarwal, E. Abu-Hamda, K. L. Molke, A. M. Correa, andL. Ho, “Endoscopic ultrasound-guided fine needle aspirationand multidetector spiral CT in the diagnosis of pancreaticcancer,” The American Journal of Gastroenterology, vol. 99,no. 5, pp. 844–850, 2004.

8 Gastroenterology Research and Practice

[30] M. A. Eloubeidi, S. Varadarajulu, S. Desai et al., “A prospectiveevaluation of an algorithm incorporating routine preoperativeendoscopic ultrasound-guided fine needle aspiration in sus-pected pancreatic cancer,” Journal of GastrointestinalSurgery, vol. 11, no. 7, pp. 813–819, 2007.

[31] A. Soriano, A. Castells, C. Ayuso et al., “Preoperative stagingand tumor resectability assessment of pancreatic cancer: pro-spective study comparing endoscopic ultrasonography, helicalcomputed tomography, magnetic resonance imaging, andangiography,” The American Journal of Gastroenterology,vol. 99, no. 3, pp. 492–501, 2004.

[32] M. Mei, J. Ni, D. Liu, P. Jin, and L. Sun, “EUS elastography fordiagnosis of solid pancreatic masses: a meta-analysis,” Gastro-intestinal Endoscopy, vol. 77, no. 4, pp. 578–589, 2013.

[33] H. C. Yip, A. Y. Teoh, C. C. Chong, and J. Y. Lau, “Current sta-tus and future applications of contrast-enhanced endoscopicultrasonography,” World Journal of Gastrointestinal Endos-copy, vol. 6, no. 4, pp. 121–127, 2014.

[34] N. K. Reddy, A. M. Ioncică, A. Săftoiu, P. Vilmann, and M. S.Bhutani, “Contrast-enhanced endoscopic ultrasonography,”World Journal of Gastroenterology, vol. 17, no. 1, pp. 42–48,2011.

[35] M. V. Alvarez-Sanchez and B. Napoleon, “Contrast-enhancedharmonic endoscopic ultrasound imaging: basic principles,present situation and future perspectives,” World Journal ofGastroenterology, vol. 20, no. 42, pp. 15549–15563, 2014.

[36] M. Matsumura and H. Sugihara, “Basic and clinical profile ofperflubutane (Sonazoid powder for injection),” Folia Pharma-cologica Japonica, vol. 130, no. 5, pp. 413–420, 2007.

[37] K. G. Toft, S. O. Hustvedt, P. A. Hals et al., “Disposition ofperfluorobutane in rats after intravenous injection ofSonazoid,” Ultrasound in Medicine & Biology, vol. 32, no. 1,pp. 107–114, 2006.

[38] M. Kitano, K. Kamata, H. Imai et al., “Contrast-enhancedharmonic endoscopic ultrasonography for pancreatobiliarydiseases,” Digestive Endoscopy, vol. 27, Supplement 1, pp. 60–67, 2015.

[39] C. F. Dietrich, B. Braden, M. Hocke, M. Ott, and A. Ignee,“Improved characterisation of solitary solid pancreatictumours using contrast enhanced transabdominal ultra-sound,” Journal of Cancer Research and Clinical Oncology,vol. 134, no. 6, pp. 635–643, 2008.

[40] R. M. Mohamed and B. M. Yan, “Contrast enhanced endo-scopic ultrasound: more than just a fancy Doppler,” WorldJournal of Gastrointestinal Endoscopy, vol. 2, no. 7, pp. 237–243, 2010.

[41] M. Hocke, X. W. Cui, D. Domagk, A. Ignee, and C. F. Dietrich,“Pancreatic cystic lesions: the value of contrast-enhancedendoscopic ultrasound to influence the clinical pathway,”Endoscopic Ultrasound, vol. 3, no. 2, pp. 123–130, 2014.

[42] M. D’Onofrio, A. Gallotti, and R. Pozzi Mucelli, “Imagingtechniques in pancreatic tumors,” Expert Review of MedicalDevices, vol. 7, no. 2, pp. 257–273, 2014.

9Gastroenterology Research and Practice

Stem Cells International

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

MEDIATORSINFLAMMATION

of

EndocrinologyInternational Journal of

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Disease Markers

Hindawiwww.hindawi.com Volume 2018

BioMed Research International

OncologyJournal of

Hindawiwww.hindawi.com Volume 2013

Hindawiwww.hindawi.com Volume 2018

Oxidative Medicine and Cellular Longevity

Hindawiwww.hindawi.com Volume 2018

PPAR Research

Hindawi Publishing Corporation http://www.hindawi.com Volume 2013Hindawiwww.hindawi.com

The Scientific World Journal

Volume 2018

Immunology ResearchHindawiwww.hindawi.com Volume 2018

Journal of

ObesityJournal of

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Computational and Mathematical Methods in Medicine

Hindawiwww.hindawi.com Volume 2018

Behavioural Neurology

OphthalmologyJournal of

Hindawiwww.hindawi.com Volume 2018

Diabetes ResearchJournal of

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Research and TreatmentAIDS

Hindawiwww.hindawi.com Volume 2018

Gastroenterology Research and Practice

Hindawiwww.hindawi.com Volume 2018

Parkinson’s Disease

Evidence-Based Complementary andAlternative Medicine

Volume 2018Hindawiwww.hindawi.com

Submit your manuscripts atwww.hindawi.com