Interobserver agreement for EUS in the evaluation and diagnosis of submucosal masses

Deep learning analysis for differential diagnosis and risk classification of gastrointestinal tumors

OBJECTIVES

Recently, artificial intelligence (AI) has been applied to clinical diagnosis. Although AI has already been developed for gastrointestinal (GI) tract endoscopy, few studies have applied AI to endoscopic ultrasound (EUS) images. In this study, we used a computer-assisted diagnosis (CAD) system with deep learning analysis of EUS images (EUS-CAD) and assessed its ability to differentiate GI stromal tumors (GISTs) from other mesenchymal tumors and their risk classification performance.

MATERIALS AND METHODS

A total of 101 pathologically confirmed cases of subepithelial lesions (SELs) arising from the muscularis propria layer, including 69 GISTs, 17 leiomyomas and 15 schwannomas, were examined. A total of 3283 EUS images were used for training and five-fold-cross-validation, and 827 images were independently tested for diagnosing GISTs. For the risk classification of 69 GISTs, including very-low-, low-, intermediate- and high-risk GISTs, 2,784 EUS images were used for training and three-fold-cross-validation.

RESULTS

For the differential diagnostic performance of GIST among all SELs, the accuracy, sensitivity, specificity and area under the receiver operating characteristic (ROC) curve were 80.4%, 82.9%, 75.3% and 0.865, respectively, whereas those for intermediate- and high-risk GISTs were 71.8%, 70.2%, 72.0% and 0.771, respectively.

CONCLUSIONS

The EUS-CAD system showed a good diagnostic yield in differentiating GISTs from other mesenchymal tumors and successfully demonstrated the GIST risk classification feasibility. This system can determine whether treatment is necessary based on EUS imaging alone without the need for additional invasive examinations.

Artificial Intelligence-Based Diagnosis of Gastric Mesenchymal Tumors Using Digital Endosonography Image Analysis

Background/Objectives: Artificial intelligence (AI)-assisted endoscopic ultrasonography (EUS) diagnostic tools have shown excellent performance in diagnosing gastric mesenchymal tumors. This study aimed to assess whether incorporating clinical and endoscopic factors into AI-assisted EUS classification models based on digital image analysis could improve the diagnostic performance of AI-assisted EUS diagnostic tools. Methods: We retrospectively analyzed the data of 464 patients who underwent both EUS and surgical resection of gastric mesenchymal tumors, including 294 gastrointestinal stromal tumors (GISTs), 52 leiomyomas, and 41 schwannomas. AI-assisted classification models for GISTs and non-GIST tumors were developed utilizing clinical and endoscopic factors and digital EUS image analysis. Results: Regarding the baseline EUS classification models, the area under the receiver operating characteristic (AUC) values of the logistic regression, decision tree, random forest, K-nearest neighbor (KNN), and support vector machine (SVM) models were 0.805, 0.673, 0.781, 0.740, and 0.791, respectively. Using the new classification models incorporating clinical and endoscopic factors into the baseline classification models, the AUC values of the logistic regression, decision tree, random forest, KNN, and SVM models increased to 0.853, 0.715, 0.896, 0.825, and 0.794, respectively. In particular, the random forest and KNN models exhibited significant improvement in performance in Delong's test (both p < 0.001). Conclusion: The diagnostic performance of the AI-assisted EUS classification models improved when clinical and endoscopic factors were incorporated. Our results provided direction for developing new AI-assisted EUS models for gastric mesenchymal tumors.

Exploration of a new method for Photoshop-assisted endoscopic ultrasound to distinguish gastrointestinal stromal tumor and leiomyoma

BACKGROUND

Gastrointestinal stromal tumors (GISTs) and leiomyomas (GILs) are difficult to be distinguished by endoscopic ultrasound (EUS). Photoshop software combined with EUS has limitations in distinguishing GIST and GIL by detecting gray values. Therefore, the research aims to explore the new method by Photoshop in distinguishing the features of GISTs from GILs.

METHODS

Patients who underwent EUS and were confirmed as GIST and GIL pathologically were included. The images of EUS were analyzed by Photoshop software. The mean gray value of tumor (Tmean), muscularis propria (Mmean), submucosa (Smean), water (Wmean) and T_SD that originated from the same image, were calculated one by one. Then the ratio of the mean gray value of tumor to muscularis propria (TMratio), submucosa (TSratio), and water (TWratio) were calculated, respectively.

RESULTS

Four hundred seventy-two patients (239 GILs and 233 GISTs) were enrolled in this study retrospectively. All the tumors were located in the stomach. Tmean and T_SD were significantly higher in GISTs than in the GILs group (63.10 ± 23.29 vs. 57.70 ± 22.67, p = .011; 26.24 ± 8.99 vs. 24.30 ± 8.26, p = .015). TMratio, TSratio, and TWratio were also significantly higher in GISTs group (0.97 ± 0.37 vs. 0.81 ± 0.28, p < .001; 0.42 ± 0.14 vs. 0.38 ± 0.12, p < .001; 2.65 ± 1.36 vs. 2.16 ± 1.02, p < .001). The AUC of Tmean was 0.952 (95% CI 0.897-1.000), which can better distinguish GIST from GIL; the sensitivity was 0.900, the specificity was 0.975, and the Youden Index was 0.875, and the cutoff was 79.64. The AUCs of TMratio, TSratio, and TWratio were 0.917 (95% CI 0.844-0.991), 0.897 (95% CI 0.812-0.981), and 0.929 (95% CI 0.8870-0.987), respectively. The aforementioned data was verified in the clinical cases of known results, including 40 GISTs and 40 GILs. The sensitivity of Tmean, TMratio, TSratio, and TWratio for diagnosis of GIL was 97.5%, 82.5%, 95%, and 97.5%, respectively. And they were 62.5%, 95%, 80%, and 92.5% for GIST.

CONCLUSION

The application of Photoshop combined with EUS to detect the gray value and standard deviation has a specific value in distinguishing GIST from GIL, but with some deviation. Applying the gray value ratio also has great discrimination significance and can avoid the differences in operation from different instrument and equipment personnel. Therefore, it is worthy of clinical promotion in the future.

ACG Clinical Guideline: Diagnosis and Management of Gastrointestinal Subepithelial Lesions

Subepithelial lesions (SEL) of the GI tract represent a mix of benign and potentially malignant entities including tumors, cysts, or extraluminal structures causing extrinsic compression of the gastrointestinal wall. SEL can occur anywhere along the GI tract and are frequently incidental findings encountered during endoscopy or cross-sectional imaging. This clinical guideline of the American College of Gastroenterology was developed using the Grading of Recommendations Assessment, Development, and Evaluation process and is intended to suggest preferable approaches to a typical patient with a SEL based on the currently available published literature. Among the recommendations, we suggest endoscopic ultrasound (EUS) with tissue acquisition to improve diagnostic accuracy in the identification of solid nonlipomatous SEL and EUS fine-needle biopsy alone or EUS fine-needle aspiration with rapid on-site evaluation sampling of solid SEL. There is insufficient evidence to recommend surveillance vs resection of gastric gastrointestinal stromal tumors (GIST) <2 cm in size. Owing to their malignant potential, we suggest resection of gastric GIST >2 cm and all nongastric GIST. When exercising clinical judgment, particularly when statements are conditional suggestions and/or treatments pose significant risks, health-care providers should incorporate this guideline with patient-specific preferences, medical comorbidities, and overall health status to arrive at a patient-centered approach.

Tumor Size Differences Between Preoperative Endoscopic Ultrasound and Postoperative Pathology for Neoadjuvant-Treated Pancreatic Ductal Adenocarcinoma Predict Patient Outcome

BACKGROUND & AIMS

The assessment of therapeutic response after neoadjuvant treatment and pancreatectomy for pancreatic ductal adenocarcinoma (PDAC) has been an ongoing challenge. Several limitations have been encountered when employing current grading systems for residual tumor. Considering endoscopic ultrasound (EUS) represents a sensitive imaging technique for PDAC, differences in tumor size between preoperative EUS and postoperative pathology after neoadjuvant therapy were hypothesized to represent an improved marker of treatment response.

METHODS

For 340 treatment-naïve and 365 neoadjuvant-treated PDACs, EUS and pathologic findings were analyzed and correlated with patient overall survival (OS). A separate group of 200 neoadjuvant-treated PDACs served as a validation cohort for further analysis.

RESULTS

Among treatment-naïve PDACs, there was a moderate concordance between EUS imaging and postoperative pathology for tumor size (r = 0.726, P < .001) and AJCC 8th edition T-stage (r = 0.586, P < .001). In the setting of neoadjuvant therapy, a decrease in T-stage correlated with improved 3-year OS rates (50% vs 31%, P < .001). Through recursive partitioning, a cutoff of ≥47% tumor size reduction was also found to be associated with improved OS (67% vs 32%, P < .001). Improved OS using a ≥47% threshold was validated using a separate cohort of neoadjuvant-treated PDACs (72% vs 36%, P < .001). By multivariate analysis, a reduction in tumor size by ≥47% was an independent prognostic factor for improved OS (P = .007).

CONCLUSIONS

The difference in tumor size between preoperative EUS imaging and postoperative pathology among neoadjuvant-treated PDAC patients is an important prognostic indicator and may guide subsequent chemotherapeutic management.

Artificial intelligence-based diagnosis of upper gastrointestinal subepithelial lesions on endoscopic ultrasonography images

BACKGROUND

Endoscopic ultrasonography (EUS) is useful for the differential diagnosis of subepithelial lesions (SELs); however, not all of them are easy to distinguish. Gastrointestinal stromal tumors (GISTs) are the commonest SELs, are considered potentially malignant, and differentiating them from benign SELs is important. Artificial intelligence (AI) using deep learning has developed remarkably in the medical field. This study aimed to investigate the efficacy of an AI system for classifying SELs on EUS images.

METHODS

EUS images of pathologically confirmed upper gastrointestinal SELs (GIST, leiomyoma, schwannoma, neuroendocrine tumor [NET], and ectopic pancreas) were collected from 12 hospitals. These images were divided into development and test datasets in the ratio of 4:1 using random sampling; the development dataset was divided into training and validation datasets. The same test dataset was diagnosed by two experts and two non-experts.

RESULTS

A total of 16,110 images were collected from 631 cases for the development and test datasets. The accuracy of the AI system for the five-category classification (GIST, leiomyoma, schwannoma, NET, and ectopic pancreas) was 86.1%, which was significantly higher than that of all endoscopists. The sensitivity, specificity, and accuracy of the AI system for differentiating GISTs from non-GISTs were 98.8%, 67.6%, and 89.3%, respectively. Its sensitivity and accuracy were significantly higher than those of all the endoscopists.

CONCLUSION

The AI system, classifying SELs, showed higher diagnostic performance than that of the experts and may assist in improving the diagnosis of SELs in clinical practice.

Intelligent Automatic Segmentation of Wrist Ganglion Cysts Using DBSCAN and Fuzzy C-Means

Ganglion cysts are common soft tissue masses of the hand and wrist, and small size cysts are often hypoechoic. Thus, identifying them from ultrasonography is not an easy problem. In this paper, we propose an automatic segmentation method using two artificial intelligence algorithms in sequence. A density based unsupervised learning algorithm called DBSCAN is performed as a front-end and its result determines the number of clusters used in the Fuzzy C-Means (FCM) clustering algorithm for quantification of ganglion cyst object. In an experiment using 120 images, the proposed method shows a higher extraction rate (89.2%) and lower false positive rate compared with FCM when the ground truth is set as the human expert's decision. Such human-like behavior is more apparent when the size of the ganglion cyst is small that the quality of ultrasonography is often not very high. With this fully automatic segmentation method, the operator subjectivity that is highly dependent on the experience of the ultrasound examiner can be mitigated with high reliability.

Convolutional neural network-based object detection model to identify gastrointestinal stromal tumors in endoscopic ultrasound images

BACKGROUND AND AIM

We aimed to develop a convolutional neural network (CNN)-based object detection model for the discrimination of gastric subepithelial tumors, such as gastrointestinal stromal tumors (GISTs), and leiomyomas, in endoscopic ultrasound (EUS) images.

METHODS

We used 376 images from 114 patients with histologically confirmed gastric GIST or leiomyoma to train the EUS-CNN. We constructed the EUS-CNN using an EfficientNet CNN model for feature extraction and a weighted bi-directional feature pyramid network for object detection. We assessed the performance of our EUS-CNN by calculating its accuracy, sensitivity, specificity, and area under receiver operating characteristic curve (AUC) using a validation set of 170 images from 54 patients. Four EUS experts and 15 EUS trainees were asked to judge the same validation dataset, and the diagnostic yields were compared between the EUS-CNN and human assessments.

RESULTS

In the per-image analysis, the sensitivity, specificity, accuracy, and AUC of our EUS-CNN were 95.6%, 82.1%, 91.2%, and 0.9234, respectively. In the per-patient analysis, the sensitivity, specificity, accuracy, and AUC for our object detection model were 100.0%, 85.7%, 96.3%, and 0.9929, respectively. The EUS-CNN outperformed human assessment in terms of accuracy, sensitivity, and negative predictive value.

CONCLUSIONS

We developed the EUS-CNN system, which demonstrated high diagnostic ability for gastric GIST prediction. This EUS-CNN system can be helpful not only for less-experienced endoscopists but also for experienced ones. Additional EUS image accumulation and prospective studies are required alongside validation in a large multicenter trial.

Predictive Factors for Differentiating Gastrointestinal Stromal Tumors from Leiomyomas Based on Endoscopic Ultrasonography Findings in Patients with Gastric Subepithelial Tumors: A Multicenter Retrospective Study

Background/Aims

The utility of endoscopic ultrasonography (EUS) for differentiating gastrointestinal stromal tumors (GISTs) and leiomyomas of the stomach is not well known. We aimed to evaluate the ability of EUS for differentiating gastric GISTs and leiomyomas.

Methods

We retrospectively reviewed the medical records of patients with histopathologically proven GISTs (n=274) and leiomyomas (n=87). In two consensus meetings, the inter-observer variability in the EUS image analysis was reduced. Using logistic regression analyses, we selected predictive factors and constructed a predictive model and nomogram for differentiating GISTs from leiomyomas. A receiver operating characteristic (ROC) curve analysis was performed to measure the discrimination performance in the development and internal validation sets.

Results

Multivariate analysis identified heterogeneity (odds ratio [OR], 9.48), non-cardia (OR, 19.11), and older age (OR, 1.06) as independent predictors of GISTs. The areas under the ROC curve of the predictive model using age, sex, and four EUS factors (homogeneity, location, anechoic spaces, and dimpling or ulcer) were 0.916 (sensitivity, 0.908; specificity, 0.793) and 0.904 (sensitivity, 0.908; specificity, 0.782) in the development and internal validation sets, respectively.

Conclusions

The predictive model and nomogram using age, sex and homogeneity, tumor location, presence of anechoic spaces, and presence of dimpling or ulcer on EUS may facilitate differentiation between GISTs and leiomyomas.

Diagnosis of gastric submucosal tumors and estimation of malignant risk of GIST by endoscopic ultrasound. Comparison between B mode and contrast-harmonic mode

BACKGROUND

standard B-mode EUS assessment and EUS-guided tissue acquisition present sub-optimal diagnostic yield in the differential diagnosis of gastric submucosal tumors (SMTs).

AIMS

to evaluate the performances of contrast-enhanced harmonic endoscopic ultrasonography (CH-EUS) to differentiate gastric SMTs and predict malignancy risk of gastrointestinal stromal tumors (GIST).

METHODS

a retrospective analysis was performed retrieving consecutive patients with gastric SMTs who underwent EUS between 2009 and 2014. Patients with available EUS video recordings and histological diagnosis were included. De-identified videos were presented to experts who made a diagnosis on B-mode EUS and CH-EUS.

RESULTS

fifty-four patients (29 female, 64-year-old) were included. Final diagnoses were 40 GISTs (8 high-grade), 9 leiomyomas, 5 rare SMTs. The sensitivity, specificity, and accuracy of B-mode and CH-EUS for the differential diagnosis of GIST were 95.0% vs. 85.0%, 57.1% for both techniques, and 85.2% vs. 77.8%, respectively. The sensitivity, specificity, and accuracy of B-mode and CH-EUS for the estimation of the malignancy GISTs risk were 62.5% vs. 100%, 83.3% vs. 82.1%, and 78.9% vs. 86.1%, respectively.

CONCLUSIONS

CH-EUS showed better diagnostic performance than B-mode EUS in differentiating leiomyomas and risk stratification of GIST. When considering high-grade GISTs, the addition of CH-EUS allowed an improvement in diagnostic accuracy.

Endoscopic and Endosonographic Features of Histologically Proven Gastric Ectopic Pancreas by Endoscopic Resection

Backgrounds/Aims

Distinguishing gastric ectopic pancreas (GEP) from malignant tumors is relatively difficult. This study evaluated the endosonography findings of pathologically proven GEP.

Methods

Thirty-one patients diagnosed with GEP based on a histopathological analysis from January 2004 to July 2018 were enrolled in this study. All patients underwent EUS and an endoscopic resection.

Results

Seventeen patients were female, and the median age was 41.1 years (range, 14-74). The lesions were localized most commonly in the antrum. The mean size of the GEP was 10.6 mm (range, 7-15). Superficial type lesions, lesions with heterogeneous echogenicity, mixed pattern lesions, and lesions with indistinct borders were commonly observed on EUS. Calcification, anechoic duct-like structures, and thickening of the muscularis propria were observed in some patients. Endoscopic mucosal resection (41.9%) and endoscopic submucosal dissection (58.1%) were performed. The mean procedure time was 22.5 minutes. Complete resection was achieved for 71% of patients. No statistically significant results between the endosonography findings and complete resection rates were obtained. The mean follow-up esophagogastroduodenoscopy duration was 4.5 months. None of the patients presented with residual lesions on subsequent endoscopy.

Conclusions

EUS can help identify the features of GEP. Careful observations of the EUS findings can avoid unnecessary removal of GEP.

Efficacy of Endoscopic Ultrasound Elastography in Differential Diagnosis of Gastrointestinal Stromal Tumor Versus Gastrointestinal Leiomyoma

BACKGROUND The diagnostic efficacy of endoscopic ultrasound (EUS) elastography for alimentary tract diseases remains uncertain. The aim of this study was to evaluate the utility of EUS elastography in differential diagnosis between the 2 most common subepithelium tumors of the digestive tract - gastrointestinal stromal tumors (GISTs) and gastrointestinal leiomyomas (GILs) - which cannot be differentiated by conventional EUS imaging. MATERIAL AND METHODS Electronic records were retrospectively reviewed from Jan 2015 to Jul 2019. Patients accepting EUS elastography with histopathological diagnosis of GISTs or GILs were included. The images of EUS elastography were analyzed by hue histogram in Photoshop. Hue values of RGB, R, G, and B channels of each group were acquired. We used the t test, ROC curve analysis, and binary logistic regression analysis for data post-processing. RESULTS We included 47 patients with GISTs and 14 with GILs. The mean±standard deviations (SD) of hue values were 20.25±0.72, -0.79±0.78, 20.79±1.68, 39.72±1.30 for GISTs and 20.80±0.46, 1.80±1.05, 28.39±2.15, and 31.95±2.60 for GILs of RGB, R, G, and B channels, respectively. The t test showed statistically significant differences in mean hue values between GISTs and GILs in B and G channels, but not in RGB and R channels. The area under the ROC curve combining B and G values was 0.723. Binary logistic regression analysis suggested no statistically significant difference in ability to differentiate between GISTs and GILs with B and G values (P>0.05). CONCLUSIONS There was insufficient evidence to support the application of quantitative EUS elastography for differential diagnosis of GISTs and GILs in this study.

Interobserver Reliability of Endoscopic Ultrasonography: Literature Review

Endoscopic ultrasonography (EUS) has been applied to the diagnosis of various digestive disorders. Although it has been widely accepted and its diagnostic value is high, the dependence of EUS diagnosis on image interpretation done by the endosonographer has persisted as an important difficulty. Consequently, high interobserver reliability (IOR) in EUS diagnosis is important to demonstrate the reliability of EUS diagnosis. We reviewed the literature on the IOR of EUS diagnosis for various diseases such as chronic pancreatitis, pancreatic solid/cystic mass, lymphadenopathy, and gastrointestinal and subepithelial lesions. The IOR of EUS diagnosis differs depending on the disease; moreover, EUS findings with high IOR and those with IOR that was not necessarily high were used as diagnostic criteria. Therefore, to further increase the value of EUS diagnosis, EUS diagnostic criteria with high diagnostic characteristics based on EUS findings with high IOR must be established.

Application of A Convolutional Neural Network in The Diagnosis of Gastric Mesenchymal Tumors on Endoscopic Ultrasonography Images

Background and Aims: Endoscopic ultrasonography (EUS) is a useful diagnostic modality for evaluating gastric mesenchymal tumors; however, differentiating gastrointestinal stromal tumors (GISTs) from benign mesenchymal tumors such as leiomyomas and schwannomas remains challenging. For this reason, we developed a convolutional neural network computer-aided diagnosis (CNN-CAD) system that can analyze gastric mesenchymal tumors on EUS images. Methods: A total of 905 EUS images of gastric mesenchymal tumors (pathologically confirmed GIST, leiomyoma, and schwannoma) were used as a training dataset. Validation was performed using 212 EUS images of gastric mesenchymal tumors. This test dataset was interpreted by three experienced and three junior endoscopists. Results: The sensitivity, specificity, and accuracy of the CNN-CAD system for differentiating GISTs from non-GIST tumors were 83.0%, 75.5%, and 79.2%, respectively. Its diagnostic specificity and accuracy were significantly higher than those of two experienced and one junior endoscopists. In the further sequential analysis to differentiate leiomyoma from schwannoma in non-GIST tumors, the final diagnostic accuracy of the CNN-CAD system was 75.5%, which was significantly higher than that of two experienced and one junior endoscopists. Conclusions: Our CNN-CAD system showed high accuracy in diagnosing gastric mesenchymal tumors on EUS images. It may complement the current clinical practices in the EUS diagnosis of gastric mesenchymal tumors.

Diagnosing Gastric Mesenchymal Tumors by Digital Endoscopic Ultrasonography Image Analysis

Gastric mesenchymal tumors (GMTs) are incidentally discovered in national gastric screening programs in Korea. Endoscopic ultrasonography (EUS) is the most useful diagnostic modality for evaluating GMTs. The differentiation of gastrointestinal stromal tumors from benign mesenchymal tumors, such as schwannomas or leiomyomas, is important to ensure appropriate clinical management. However, this is difficult and operator dependent because of the subjective interpretation of EUS images. Digital image analysis computes the distribution and spatial variation of pixels using texture analysis to extract useful data, enabling the objective analysis of EUS images and decreasing interobserver and intraobserver agreement in EUS image interpretation. This review aimed to summarize the usefulness and future of digital EUS image analysis for GMTs based on published reports and our experience.

Gastrointestinal Stromal Tumors of the Small Intestine: Progress in Diagnosis and Treatment Research

In recent years, the diagnosis and treatment of gastrointestinal stromal tumors (GISTs) of the small intestine have been a hot topic due to their rarity and non-specific clinical manifestations. With the development of gene and imaging technology, surgery, and molecular targeted drugs, the diagnosis and treatment of GISTs have achieved great success. For a long time, radical resection was prioritized to treat GISTs of the small intestine. At present, preoperative tumor staging is a novel treatment for unresectable malignant tumors. In addition, karyokinesis exponent is the sole independent predictor of progression-free survival of GISTs. The DNA, miRNA, and protein of exosomes have also been found to be biomarkers with prognostic implications. The research on the treatment of GISTs has become a focus in the era of precision medicine, ushering in the use of standardized, normalized, and individualized treatment.

Digital image analysis-based scoring system for endoscopic ultrasonography is useful in predicting gastrointestinal stromal tumors

BACKGROUND

When gastric mesenchymal tumors (GMTs) measuring 2-5 cm in size are found, whether to undergo further treatment or not is controversial. Endoscopic ultrasonography (EUS) is useful for the evaluation of malignant potential of GMTs, but has limitations, such as subjective interpretation of EUS images. Therefore, we aimed to develop a scoring system based on the digital image analysis of EUS images to predict gastrointestinal stromal tumors (GISTs).

METHODS

We included 103 patients with histopathologically proven GIST, leiomyoma or schwannoma on surgically resected specimen who underwent EUS examination between January 2007 and June 2018. After standardization of the EUS images, brightness values, including the mean (T_mean), indicative of echogenicity, and the standard deviation (T_SD), indicative of heterogeneity, in the tumors were analyzed.

RESULTS

Age, T_mean, and T_SD were significantly higher in GISTs than in non-GISTs. The sensitivity and specificity were almost optimized for differentiating GISTs from non-GISTs when the critical values of age, T_mean, and T_SD were 57.5 years, 67.0, and 25.6, respectively. A GIST-predicting scoring system was created by assigning 3 points for T_mean ≥ 67, 2 points for age ≥ 58 years, and 1 point for T_SD ≥ 26. When GMTs with 3 points or more were diagnosed as GISTs, the sensitivity, specificity, and accuracy of the scoring system were 86.5%, 75.9%, and 83.5%, respectively.

CONCLUSIONS

The scoring system based on the information of digital image analysis is useful in predicting GISTs in case of GMTs that are 2-5 cm in size.

Lesion size determines diagnostic yield of EUS-FNA with onsite cytopathologic evaluation for upper gastrointestinal subepithelial lesions

BACKGROUND/AIMS

The aim of this study was to determine the diagnostic yield and factors influencing the diagnostic yield of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) for upper gastrointestinal (GI) subepithelial lesions (SELs) with rapid onsite cytopathologic evaluation.

MATERIALS AND METHODS

This is a single-center, retrospective study.

RESULTS

Among 22 patients who underwent EUS-FNA, a cytopathological diagnosis was reached in 16 (72.7%) patients. The EUS-FNA results were as follows: seven GISTs (31.8%), six leiomyomas (27.2%), four non-diagnostics (18%), two duplication cysts (9%), two spindle cell tumor (9%), and one ectopic pancreas (4.5%). The long-axis size was > 20 mm in 12 patients (average size: 31.3 ±9.3 mm) and < 20 mm (average size: 16.6 ±2.5 mm) in 10 patients. Diagnostic accuracy of EUS-FNA from lesions < 20 mm was 50% (5/10 lesions), and of lesions > 20 mm was 91.6% (11/12 lesions) (Fisher's exact test; p=0.028). Six patients underwent surgical resection. Surgical pathology results of five lesions (four GIST, one leiomyoma) were consistent with cytopathology results (83.3%).

CONCLUSION

The diagnostic yield of EUS-FNA of the upper GI SELs with an onsite cytopathologic interpretation was 72.7%. Lesion size < 2 cm significantly reduces the diagnostic yield of EUS-FNA for the upper GI SELs.

Recognizing intrapancreatic accessory spleen via EUS: Interobserver variability

Background and Objective:

Accessory spleen (AS) may be encountered as an intrapancreatic lesion on EUS. This can look similar to other pancreatic pathologies and may lead to unnecessary interventions. The goal of this study was to evaluate the accuracy of EUS in distinguishing intrapancreatic AS (IPAS) from other pancreatic lesions.

Materials and Methods:

Twelve sets of endoscopic images of the spleen and various pancreatic lesions confirmed on histology or cytology were gathered. Ten endosonographers were asked to characterize and identify the lesions. The responses were analyzed via Excel and the interobserver agreement was analyzed using Gwet's agreement coefficient statistic via Stata I/C v15.

Results:

In our sample, the interobserver agreement was 0.37 (−1–1; 0–0.2 poor, 0.2–0.4 fair, 0.4–0.6 moderate, 0.6–0.8 substantial, and 0.8–1.0 almost perfect) for determining whether or not the pancreatic lesion is IPAS. The reviewers were able to correctly determine IPAS endosonographically with a sensitivity of 77%, specificity of 74%, and positive and negative predictive values of 50% and 92%, respectively.

Conclusion:

There is a moderate-to-substantial interobserver agreement in describing the sonographic characteristics of the pancreatic lesions, such as the shape, echogenicity compared to spleen, echotexture, and border of the lesions. However, the interobserver agreement is only fair when deciding if the pancreatic lesion is an IPAS. The similar profile of IPAS and pancreatic neuroendocrine tumor could confound the diagnosis of IPAS, thus contributing to the decreased interobserver agreement. This study demonstrates that EUS criteria alone are not accurate for IPAS diagnosis. Fine-needle aspiration (FNA) may be required for a confirmatory diagnosis.

Prospective evaluation of the performance and interobserver variation in endoscopic ultrasound staging of rectal cancer

BACKGROUND

Treatment and prognosis of patients with rectal adenocarcinoma (RAC) are dependent on accurate locoregional staging.

OBJECTIVES

The aim of this study was to measure the performance characteristics of rectal endoscopic ultrasound (EUS) compared with surgical pathology, and to assess the interobserver variation of rectal EUS in the staging of RAC.

PATIENTS AND METHODS

Patients referred for rectal EUS staging of a recently diagnosed RAC were prospectively enrolled between 2012 and 2016. Tandem EUS exams were performed by two independent endosonographers (ES1 and ES2) blinded to each other's findings.

RESULTS

Ninety-five patients were enrolled. Seventy-five (79%) underwent curative intent tumor resection, including 30 without neoadjuvant therapy. In this latter group, the sensitivity, specificity, and accuracy of transrectal ultrasonography staging were 75, 83, and 82% for uT1; 50, 65, and 58% for uT2; 56, 81, and 73% for T3; 72, 44, and 63% for N0, and 38, 75, and 63% for N1, respectively. Experienced operators rendered a more accurate N stage and were less likely to overstage compared with less experienced ones (P=0.01 and 0.02, respectively). Overall, T staging agreement between endosonographers was substantial (κ=0.61) and N stage agreement was moderate (κ=0.45).

CONCLUSION

Rectal EUS is more accurate in staging T1 and T3 tumors compared with T2 tumors. Interobserver agreement of rectal EUS in rectal cancer staging is generally good.

Ectopic pancreas in the upper gastrointestinal tract: Is endosonographic diagnosis reliable? Data from the German Endoscopic Ultrasound Registry and review of the literature

Background:

Ectopic pancreas (EP) belongs to the most frequent subepithelial lesions (SELs) of the upper gastrointestinal (GI) tract. In the majority of cases, it is detected incidentally. Differential diagnosis from mesenchymal subepithelial tumors may be difficult.

Methods:

Among 24,308 endosonographic examinations and interventions, which were prospectively enrolled in the database of the German Endoscopic Ultrasound (EUS) Registry from January 2009 to August 2013, 575 were performed for suspected SELs of the upper GI tract. Sixty three cases of EP of the upper GI tract (stomach, n = 53; duodenum, n = 10; esophagus, n = 0) were extracted and retrospectively reviewed.

Results:

In 65.1% of cases, radial echoendoscopes or radial miniprobes were used for examination. Nearly 84% of EP was found in the stomach, 16% in the duodenum, none in the esophagus. In 88.9% of cases, the EUS examination discerned the layer of origin. In 59% of cases EP was described as a heterogeneous, in 28.6% as a homogeneous-hypoechoic and in 7.9% as a homogeneous-echogenic subepithelial mass lesion. Mean diameter was 13.0 mm × 8.1 mm, the mean ratio between long and short axis diameter was 1.75. EUS-guided fine needle aspiration (EUS-FNA) was used to accomplish cytological or histological diagnosis in only 6.3% of cases.

Conclusions:

EP accounts for 11% of all EUS examinations performed for subepithelial lesions of the upper GI tract and prospectively enrolled in the German EUS registry. Rather than being an eyecatcher, EP is a chameleon with numerous differential diagnoses. In selected cases, EUS-FNA may help clarifying the diagnosis.

Present status and perspectives of endosonography 2017 in gastroenterology

Endoscopic ultrasound has become an essential tool in modern gastroenterology and abdominal surgery. Compared with all other endoscopic methods, it has the most potential for innovation and its future looks bright. Thus, we compiled this summary of established and novel applications of endoscopic ultrasound methods to inform the reader about what is already possible and where future developments will lead in improving patient care further. This review is structured in four parts. The first section reports on developments in diagnostic endoscopic ultrasound, the second looks at semi-invasive endoscopic ultrasound, and the third discusses advances in therapeutic endoscopic ultrasound. An overview on the future prospects of endoscopic ultrasound methods concludes this article.

Evaluation and Endoscopic Management of Esophageal Submucosal Tumor

Submucosal tumors (SMTs) originate from tissues that constitute the submucosal layer and muscularis propria, and are covered by normal mucosa. Esophageal SMTs are rare, accounting for <1% of all esophageal tumors. However, the recent widespread use of endoscopy has led to a rapid increase in incidental detection of SMTs in Korea. Esophageal SMTs are benign in ≥90% of cases, but the possibility of malignancies such as gastrointestinal stromal tumor and malignant leiomyosarcoma still exists. Therefore, patients undergo resection in the presence of symptoms or the possibility of a malignant tumor. For resection of esophageal SMTs, surgical resection was the only option available in case of possible malignancy, but minimally invasive surgery by endoscopic resection is becoming more preferable to surgical resection with the development of endoscopic ultrasonography, endoscopic techniques, and other devices.

Endoscopic ultrasound without tissue acquisition has poor accuracy for diagnosing gastric subepithelial tumors

Incidental gastric subepithelial tumor (SET) is frequently found during endoscopy. Although endoscopic ultrasonography (EUS) can provide useful information, its diagnostic accuracy varies. Most of the potentially malignant tumors observed on EUS are hypoechoic lesions. Therefore, we aimed to investigate the diagnostic accuracy of EUS for hypoechoic lesions located in the submucosa or proper muscle layer. We also evaluated various characteristics for potential associations with diagnostic accuracy.A retrospective review was conducted of the medical records of 99 patients who were diagnosed with gastric SET and who underwent EUS with pathologic confirmation between March 2008 and April 2015. After reviewing the endoscopic and pathologic findings, we attempted to analyze factors that were associated with the diagnostic accuracy of EUS.The mean ± standard deviation size of the lesions was 20.0 ± 12.7 mm. The most common location was the upper third of the stomach (43.4%). The overall accuracy of EUS was 66.7%. No statistically significant difference in EUS accuracy was observed according to the location, size, or layer of the lesion. The following pathologic diagnostic methods were used: EUS-guided fine needle aspiration (3.0%), forceps biopsy (16.2%), deep tissue biopsy using cap-assisted mucosal resection (8.1%), endoscopic submucosal dissection (25.2%), and operation (47.5%). The accuracy of EUS according to the expected diagnosis of the lesion was 77.1% for gastrointestinal stromal tumor, 50% for neuroendocrine tumor, and 50% for ectopic pancreas.Although EUS is a useful tool for gastric SET in clinical practice, the accuracy of diagnostic EUS is suboptimal. When considering whether to treat gastric SET, the decision should be made based on the pathologic diagnosis.

Digital image analysis of endoscopic ultrasonography is helpful in diagnosing gastric mesenchymal tumors

BACKGROUND

Endoscopic ultrasonography (EUS) is a valuable imaging tool for evaluating subepithelial lesions in the stomach. However, there are few studies on differentiation between gastrointestinal stromal tumors (GISTs) and benign mesenchymal tumors, such as leiomyoma or schwannoma, with the use of EUS. In addition, there are limitations in the analysis of the characteristic features of such tumors due to poor interobserver agreement as a result of subjective interpretation of EUS images. Therefore, the aim of this study was to evaluate the role of digital image analysis in distinguishing the features of GISTs from those of benign mesenchymal tumors on EUS.

METHODS

We enrolled 65 patients with histopathologically proven gastric GIST, leiomyoma or schwannoma on surgically resected specimens who underwent EUS examination at our endoscopic unit from January 2007 to September 2010. After standardization of the EUS images, brightness values including the mean (Tmean), indicative of echogenicity, and the standard deviation (TSD), indicative of heterogeneity, in the tumors were analyzed.

RESULTS

The Tmean and TSD were significantly higher in GIST than in leiomyoma and schwannoma (p < 0.001). However, there was no significant difference in the Tmean or TSD between benign and malignant GISTs. The sensitivity and specificity were almost optimized for differentiating GIST from leiomyoma or schwannoma when the critical values of Tmean and TSD were 65 and 75, respectively. The presence of at least 1 of these 2 findings in a given tumor resulted in a sensitivity of 94%, specificity of 80%, positive predictive value of 94%, negative predictive value of 80%, and accuracy of 90.8% for predicting GIST.

CONCLUSIONS

Digital image analysis provides objective information on EUS images; thus, it can be useful in diagnosing gastric mesenchymal tumors.

Image analysis of endosocopic ultrasonography in submucosal tumor using fuzzy inference

Endoscopists usually make a diagnosis in the submucosal tumor depending on the subjective evaluation about general images obtained by endoscopic ultrasonography. In this paper, we propose a method to extract areas of gastrointestinal stromal tumor (GIST) and lipoma automatically from the ultrasonic image to assist those specialists. We also propose an algorithm to differentiate GIST from non-GIST by fuzzy inference from such images after applying ROC curve with mean and standard deviation of brightness information. In experiments using real images that medical specialists use, we verify that our method is sufficiently helpful for such specialists for efficient classification of submucosal tumors.

Gastric subepithelial masses: evaluation of multidetector CT (multiplanar reconstruction and virtual gastroscopy) versus endoscopic ultrasonography

OBJECTIVES

To assess the role of multidetector computed tomography (MDCT) with multiplanar reconstruction (MPR) and virtual gastroscopy (VG) for detection and differentiation of gastric subepithelial masses (SEMs) by comparison with endoscopic ultrasonography (EUS).

METHODS

Forty-one patients with a suspected SEM were evaluated using EUS and MDCT. MDCT findings were analyzed based on the consensus of two radiologists who were blinded to the EUS findings. The analysis of the CT features included the location, size, and contours of the tumor, the presence of central dimpling, as well as the growth pattern, enhancement pattern, and enhancement degree. The long diameter (LD) and the short diameter (SD) of each lesion were measured and the LD/SD ratios were calculated. EUS and MDCT results were compared with histopathology for the pathologically proven lesions. For the non-pathologically proven lesions, MDCT results were compared with EUS.

RESULTS

Among the 41 patients, 34 SEMs were detected using EUS. For the detection of SEMs with MDCT, a sensitivity of 85.3%, a specificity of 85.7%, a positive predictive value of 96.7%, and a negative predictive value of 54.5% were calculated. The overall accuracy of MDCT for detecting and classifying the SEMs was 85.3 and 78.8%, respectively.

CONCLUSIONS

MDCT with MPR and VG is a valuable method for the evaluation of SEMs. Specific MDCT criteria for various SEMs may be helpful in making an accurate diagnosis.

Ecoendoscopia nas lesões subepiteliais do trato digestório: artigo de revisão

Lesões subepiteliais ou abaulamentos da mucosa são recobertas por mucosa normal e geralmente assintomáticas. Sua maioria é diagnosticada em exames radiológicos ou de endoscopia digestiva e podem corresponder a qualquer camada da parede do órgão (intramural) ou serem extramurais. Este artigo descreve estudos para análise da ultrassonografia após ecoendoscopia (USE) como método diagnóstico de elevada acurácia diante do achado de lesão subepitelial. Trata-se de trabalho de revisão de literatura sobre as características ecoendoscópicas das lesões subepiteliais e diferenciação em intra ou extramurais, camada de origem, ecogenicidade, vascularização, margens e dimensões, punção aspirativa por agulha fina (PAAF) ou biópsias com agulha do tipo trucut. Ambas as formas são aceitáveis) para análise histológica. A ultrassonografia endoscópica tem melhores índices de acurácia no diagnóstico da camada da parede gastrointestinal comprometida por lesões ou massas, além de estudar a ecogenicidade da lesão. A ultrassonografia endoscópica é um método seguro e detalhado, considerado o melhor exame de imagem para diagnóstico definitivo e programação terapêutica das lesões subepiteliais.

Endoscopic ultrasonography-guided diagnosis of subepithelial tumors

Subepithelial lesions are frequently discovered during routine endoscopic examinations. These lesions represent a wide spectrum of heterogeneous benign to malignant conditions. Most of these lesions are asymptomatic. There is no consensus regarding how to manage these lesions. Over the last 2 decades, the approach to these lesions has significantly improved owing to the introduction of endoscopic ultrasonography, fine-needle aspiration, immunohistochemical staining methods, and different treatment options. This article discusses the nature of subepithelial lesions, focusing on the most recent developments that use endoscopic ultrasonography to diagnose and manage these lesions.

Small intestine gastrointestinal stromal tumors

PURPOSE OF REVIEW

To review the contemporary management of gastrointestinal stromal tumor (GIST), including endoscopy, surgery, and systemic therapy, highlighting the aspects unique to small intestinal tumors.

RECENT FINDINGS

Tumor size, mitotic count, and site of origin are the three key prognostic factors, with mitotic count being the single strongest predictor of recurrence. Tumors arising in the small bowel have worse prognosis than those of comparable size and mitotic count arising in other organs. Endoscopy and endoscopic ultrasound-guided, fine-needle aspiration are key components in the diagnosis of GIST. The role of endoscopy in surveillance and resection remain investigational. Surgery, either open or laparoscopic, remains the only curative option, but recurrence rates are high. Adjuvant therapy with imatinib mesylate improves recurrence-free survival rates and may improve overall survival (OS) with longer duration of treatment. Neoadjuvant imatinib may play an important role in the management of patients with locally advanced disease. For patients with advanced disease, first-line imatinib and second-line sunitinib malate have improved progression-free and OS rates. Systemic treatment should be continued life-long or until treatment failure.

SUMMARY

Advances in the last decade have dramatically changed the management and prognosis of patients with primary and advanced GIST.

Endoscopic management of gastrointestinal stromal tumors

Subepithelial lesions are increasingly being identified with improved endoscopic imaging technologies. Many of these lesions are now recognized as gastrointestinal stromal tumors (GISTs). Recent advances in immunohistochemistry have allowed for reliable differentiation of GISTs from other subepithelial tumors, thereby significantly improving our understanding of these lesions. The wealth of recent information and continual evolution in our understanding of GISTs has exposed some knowledge gaps pertaining to the optimal management of these lesions. In this article, we review the endoscopic management of GISTs as it relates to the identification, diagnosis and management of these lesions based on the best available literature and our own clinical experience to date.

Pediatric gastrointestinal stromal tumor

Pediatric gastrointestinal stromal tumor (GIST) is a rare entity that can be quite different from its adult counterpart. This report provides a comprehensive review on the diagnosis and management of this tumor in children and adolescents, including its oncogenesis and associated syndromes. Surgery remains a mainstay of treatment, but there are no standard guidelines available at this time regarding the best practice for multimodality therapy as our understanding of the biology of GIST is still in evolution. Therefore, pediatric patients with GIST should be ideally treated in the context of clinical trials at specialized, multidisciplinary centers throughout the course of their disease, especially because these patients may live for years after diagnosis.

Are endoscopic ultrasonography imaging characteristics reliable for the diagnosis of small upper gastrointestinal subepithelial lesions?

PURPOSE OF THE STUDY

To compare the accuracy of endoscopic ultrasonography (EUS) imaging with histopathology in the diagnosis of upper gastrointestinal subepithelial lesions.

METHODS

Thirty-seven patients (21 female; mean age: 55 y) underwent endoscopic submucosal resection (ESMR) of upper gastro intestinal subepithelial lesions at a tertiary care facility. All patients underwent EUS before ESMR of the lesion. Information regarding location, size, echogenecity, layer of origin, presumptive diagnosis based on EUS imaging, and histopathology diagnosis after ESMR of the subepithelial lesion was recorded.

RESULTS

Twenty-seven subepithelial lesions were resected from the stomach, 5 from the esophagus, and 5 from the duodenum. The mean size of the lesions was 9 mm (range, 6-18 mm). Thirty-six lesions originated from the submucosa, and 1 from the muscularis propria. Using histopathology as the gold standard, the overall diagnostic accuracy of EUS imaging was 49% (18 out of 37). The accuracy of EUS imaging for the diagnosis of esophageal, gastric, and duodenal subepithelial lesions was 20%, 56%, and 40%, respectively. One patient developed a microperforation, and 1 developed bleeding during the ESMR procedure. No complications were reported with the EUS procedure.

CONCLUSIONS

The diagnostic accuracy of EUS imaging is inferior to histopathology in the diagnosis of small upper gastrointestinal subepithelial lesions. Endoluminal resection is a relatively safe and noninvasive modality that not only provides tissue sample for accurate diagnostic interpretation, but also aids in the complete removal of small subepithelial lesions of the upper gastrointestinal tract.

Estimation of malignant potential of GI stromal tumors by contrast-enhanced harmonic EUS (with videos)

BACKGROUND

Contrast-enhanced harmonic EUS (CEH-EUS) is a new sonographic technique that uses US contrast agents and depicts intratumoral vessels in real time.

OBJECTIVE

To evaluate whether assessment of tumor vascularity by CEH-EUS can predict the preoperative malignancy risk of GI stromal tumors (GISTs).

DESIGN

Prospective study to observe GIST vascularity.

SETTING

Kinki University School of Medicine.

PATIENTS

Between June 2007 and September 2009, 76 consecutive patients suspected of having subepithelial lesions underwent CEH-EUS.

INTERVENTION

CEH-EUS was performed by using a prototype echoendoscope in an extended pure harmonic detection mode.

MAIN OUTCOME MEASUREMENTS

Resected GIST specimens in 29 patients who underwent surgical resection were divided into high-grade (n=16) and low-grade (n=13) malignancy groups based on mitotic activity. The abilities of EUS-guided FNA and CEH-EUS to diagnose the malignant potential were compared. The sensitivities with which contrast-enhanced multidetector CT, power-Doppler EUS, and CEH-EUS detected intratumoral vessels in high-grade malignancy GISTs also were compared.

RESULTS

CEH-EUS identified irregular vessels and thereby predicted GIST malignancies with a sensitivity, specificity, and accuracy of 100%, 63%, and 83%, respectively. Diagnosis of high-grade malignancy GISTs by EUS-guided FNA had a sensitivity, specificity, and accuracy of 63%, 92%, and 81%, respectively. Contrast-enhanced multidetector CT, power-Doppler EUS, and CEH-EUS detected intratumoral vessels in high-grade malignancy GISTs with sensitivities of 31%, 63%, and 100%, respectively (P<.05).

LIMITATIONS

A single center was involved in this study.

CONCLUSIONS

CEH-EUS successfully visualized intratumoral vessels and may play an important role in predicting the malignancy risk of GISTs.

Endoscopic ultrasonography-guided fine-needle aspiration for diagnosing upper gastrointestinal submucosal lesions: a prospective study of 50 cases

The objective was to assess EUS-FNA for diagnosing intramural upper GI tract lesions. The subjects were 50 patients (21M/29F) with upper GI submucosal lesions who underwent EUS-FNA at a referral center for GI system over a 12-month period. All cases were followed for 1 year after initial EUS-FNA. Cytologic diagnoses were categorized as benign, malignant, suspicious for malignancy, mesenchymal tumor, endocrine tumor, or nondiagnostic. All tumors were assessed for various cytomorphologic features. The accuracy of the initial FNA diagnoses was evaluated for each patient who also underwent subsequent histopathological examination of a core biopsy and/or surgical biopsy/resection material of the same lesion. According to the site of the lesions; while 84% of all esophageal lesions were diagnosed as mesenchymal; 67% of all gastric lesions were mesenchymal. The sole lesion was nonmesenchymal (benign cyst) in duodenum. The sensitivity, specificity, positive and negative predictive values, and accuracy of EUS-FNA for diagnosing submucosal mesenchymal tumors of the upper GI tract were 82.9, 73.3, 87.9, 64.7, and 80%, respectively. The corresponding values for nonmesenchymal lesions were 100, 85.7, 80, 100, and 90.9%. Our experience confirms that EUS-FNA is an extremely valuable tool for diagnosing submucosal lesions of the upper GI, and is particularly useful in cases where endoscopic forceps biopsy does not lead to diagnosis. Optimal results can be yielded by a close working relationship between the gastroenterologist and pathologist.

Digital image analysis is a useful adjunct to endoscopic ultrasonographic diagnosis of subepithelial lesions of the gastrointestinal tract

OBJECTIVE

The purpose of this study was to explore the role of digital image analysis in differentiating endoscopic ultrasonographic (EUS) features of potentially malignant gastrointestinal subepithelial lesions (SELs) from those of benign lesions.

METHODS

Forty-six patients with histopathologically confirmed gastrointestinal stromal tumors (GISTs), carcinoids, and lipomas who had undergone EUS evaluation were identified from our database. Representative regions of interest (ROIs) were selected from the EUS images, and features were extracted by texture analysis. On the basis of these features, an artificial neural network (ANN) was built, trained, and internally validated by unsupervised learning followed by supervised learning. Outcomes were the performance characteristics of the ANN.

RESULTS

A total of 106, 111, and 124 ROIs were selected from EUS images of 8, 10, and 28 patients with lipomas, carcinoids, and GISTs, respectively. For each ROI, 228 statistical parameters were extracted and later reduced to the 11 most informative features by principal component analysis. After training with 50% of the data, the remainder of the data were used to validate the ANN. The model was "good" in differentiating carcinoids and GISTs, with area under the receiver operating characteristic curve (AUC) values of 0.86 and 0.89, respectively. The model was "excellent" in identifying lipomas correctly, with an AUC of 0.92.

CONCLUSIONS

Digital image analysis of EUS images is a useful noninvasive adjunct to EUS evaluation of SELs.

Diagnosis of subepithelial tumors in the upper gastrointestinal tract by endoscopic ultrasonography

Endoscopic ultrasonography (EUS) is the most accurate procedure for detecting and diagnosing subepithelial tumors, due to its higher sensitivity and specificity than other imaging modalities. EUS can characterize lesions by providing information on echogenic origin, size, borders, homogeneity, and the presence of echogenic or anechoic foci. Linear echoendoscopes, and recently also electronic radial echoendoscopes, can be used with color Doppler or power Doppler to assess the vascular signals from subepithelial masses, and thus permit the differentiation of vascular structures from cysts, as well as the assessment of the tumor blood supply. However, the diagnostic accuracy of EUS imaging alone has been shown to be low in subepithelial lesions with 3rd and 4th layers. It is also difficult to differentiate exactly between benign and malignant tumors and to gain an accurate picture of histology using EUS. On the other hands, EUS guided fine needle aspiration (EUS-FNA) can provide samples for cytologic or histologic analysis. Hypoechoic lesions of the 3rd and the 4th EUS layers, more than in 1 cm diameter are recommended, and histologic confirmation using endoscopic submucosal resection or EUS-FNA should be obtained when possible. Therefore, EUS-FNA plays an important role in the clinical management of subepithelial tumors. Furthermore improvements in endoscopic technology are expected to be more useful modalities in differential diagnosis and discrimination between benign and malignant subepithelial tumors.

Accuracy of EUS in the evaluation of small gastric subepithelial lesions

BACKGROUND

EUS combined with endoluminal resection techniques is increasingly used to provide a definitive diagnosis of small gastric subepithelial lesions seen on standard upper endoscopy.

OBJECTIVE

To evaluate the accuracy of EUS in diagnosing small gastric subepithelial lesions by using histology as the criterion standard.

DESIGN

A retrospective study.

SETTING

Academic tertiary care center.

PATIENTS

A total of 22 patients (15 women, mean age 62.2 years) with an endoscopically resected gastric subepithelial lesion were included in this 3-year retrospective study.

MAIN OUTCOME MEASUREMENTS

The size, echogenicity, the layer of origin, and presumptive diagnosis were determined by EUS. The diagnostic accuracy of EUS was determined by using histology as the criterion standard.

RESULTS

The mean size of the 22 lesions was 13.6 mm (range 8-20 mm). An endoscopic cap band mucosectomy device was used to resect 16 (72.7%) lesions, whereas 6 (27.3%) were resected with a saline solution-assisted and snare technique. Using histology as a criterion standard, we found that the accuracy of the EUS diagnosis was 10 of 22 (45.5%). EUS alone had an accuracy rate of 30.8% and 66.7%, respectively, in the diagnosis of neoplastic and non-neoplastic lesions.

LIMITATIONS

A single-center, retrospective analysis.

CONCLUSION

EUS imaging had a low accuracy rate in the diagnosis of gastric subepithelial lesions, and endoscopic submucosal resection should be performed to provide a histologic diagnosis. Resection of small subepithelial lesions of 20 mm or less can be accomplished en bloc with an endoscopic cap band mucosectomy device.

The natural history of gastrointestinal subepithelial tumors arising from muscularis propria: an endoscopic ultrasound survey

AIM

The majority of lesions originating from muscularis propria of stomach, duodenum, and colon are gastrointestinal stromal tumors (GISTs). Surgery is indicated when endosonographic criteria of malignancy are met, but little is known about the natural history of lesions with benign endosonographic features. Aim of this study was to evaluate the natural course of benign-appearing lesions originating from muscularis propria in organs where GISTs significantly overcome leiomyomas.

MATERIALS AND METHODS

A total of 49 asymptomatic patients with hypoechoic lesions originating from the fourth layer of the gastrointestinal tract entered a follow-up program by means of endoscopic ultrasonography. All lesions were nonulcerated, <3 cm in maximal diameter, with regular margins, and cystic spaces of <3 mm.

RESULTS

After a mean follow-up of 31+/-20.8 months and a median of 2 (range, 1 to 5) endosonographies/patient, no change in echostructure or dimensions was seen in 44 subjects whereas in 5, an increase of at least 25% in 1 diameter occurred. Surgical removal was proposed to all: 1 patient refused (she is still alive and symptom-free after 4 y), 3 of the 4 lesions removed proved to be GISTs with very low or low risk of malignancy and 1 lesion was classified as a glomus tumor with no malignant appearance.

CONCLUSIONS

Even small and benign-appearing lesions from muscularis propria may increase in size over time but this increase cannot be considered as an index of malignancy. As most of these lesions are GISTs, a policy of surveillance is advisable.

The role of endosonography in submucosal tumours

Submucosal tumours (SMTs) are relatively common findings in patients undergoing endoscopy, especially in upper gastrointestinal (GI) tract. This term includes various non-neoplastic and neoplastic conditions. Endoscopic ultrasonography (EUS) is the best imaging procedure to characterise SMT and to determine the need for further explorations. In this review, the following issues will be addressed: the endosonographic features of SMT; the diagnostic accuracy of EUS; the ability of EUS to distinguish benign and malignant SMTs; the value of EUS-guided fine-needle aspiration (FNA); and the influence of EUS on clinical management.

Diagnostic yield of tissue sampling using a bite-on-bite technique for incidental subepithelial lesions

BACKGROUND/AIMS

Techniques for endoscopic evaluation of gastrointestinal subepithelial lesions include conventional endoscopy, jumbo biopsy, endoscopic ultrasonogrphy (EUS), EUS-guided fine needle aspiration, and endoscopic submucosal resection. However, these procedures have many limitations, such as low diagnostic yields and high complication rates. We therefore evaluated the diagnostic yield for tissue sampling of incidental subepithelial lesions using the bite-on-bite technique.

METHODS

One hundred and forty subepithelial lesions were found in 129 patients during conventional diagnostic esophagogastroduodenoscopy by one examiner from October 2003 to November 2004. Bite-on-bite biopsies with conventional-sized forceps were taken from 36 patients having 37 lesions that did not appear to be hypervascular or to have a thick overlying epithelium. Two to eight bites were performed to obtain submucosal tissue for one lesion.

RESULTS

The bite-on-bite technique was diagnostic in 14 of the 37 lesions (38%). Blood oozing for more than 30 seconds occurred in five cases, but was easily controlled by epinephrine injection (2 cases) or hemoclip (3 cases). The diagnostic yield tended to be higher in the esophagus than in the stomach and duodenum (54% vs. 28%, p=0.109).

CONCLUSIONS

The bite-on-bite technique for subepithelial lesions is an effective and safe method in selected cases. This technique may be useful for incidental subepithelial lesions, especially those of the esophagus, except for ones with a high risk of bleeding or thick overlying epithelium.

A guide for the diagnosis and management of gastrointestinal stromal cell tumors

Gastrointestinal stromal cell tumors (GISTs) are the most common mesenchymal neoplasm of the gastrointestinal tract and are frequently detected on routine endoscopy. Although only approximately 10-30% of GISTs are clinically malignant, all may have some degree of malignant potential. Preoperative determination of malignancy risk can be estimated from tumor size and location, but reliable histopathologic criteria are not currently available. Given such biological uncertainty, accurate diagnosis is essential to differentiate these lesions from other truly benign, subepithelial tumors. Endoscopic ultrasound-guided fine-needle aspiration has emerged as an important procedure to secure a tissue diagnosis of a GIST. When encountering GISTs, gastroenterologists are faced with challenging management decisions, especially in the face of small, incidentally discovered lesions. The majority of localized GISTs are managed via surgical resection, although a select few may be observed using serial endoscopic ultrasound examinations. This Review provides a general overview of GISTs, with an emphasis on their endoscopic diagnosis, the management of localized disease, and the management of incidentally discovered GISTs.

Interobserver and intraobserver variation of two-dimensional and three-dimensional anal endosonography in the evaluation of recurrent anal cancer

PURPOSE

This study was designed to evaluate the interobserver and intraobserver agreement of two-dimensional (2-D) and three-dimensional (3-D) anal endosonography for the detection of local recurrence anal carcinoma.

METHODS

Thirty-six patients were treated for anal carcinoma, and seven had recurrent disease. They were investigated by using 3-D endosonography at Rigshospitalet from July 2001 to January 2005 because of suspected local recurrence. The 3-D endosonographic examinations were reviewed from the hard disc by two observers who twice reviewed all 2-D examinations (the axial projection) as well as all 3-D examinations (the axial, as well as the reconstructed coronal and sagittal projections). The observers scored each examination according to the following scale regarding presence of local recurrence: 1 = no finding/benign findings; 2 = properly benign findings; 3 = suspicious findings/malignant findings. Kappa statistic-statistic was used to evaluate interobserver and intraobserver variation.

RESULTS

Three-dimensional endosonography achieved better interobserver agreement than 2-D endosonography: kappa 3-D 0.34-047 vs. kappa 2-D 0.15-0.28 and better intraobserver agreement: kappa 3-D 0.34-0.62 vs. kappa 2-D 0.22-0.28. These differences showed P < 0.05.

CONCLUSIONS

Three-dimensional endosonography proved to have significantly better interobserver and intraobserver agreement than 2-D endosonography concerning detection of recurrent anal cancer. Three-dimensional endosonography seems to be less dependent of the individual examiner than 2-D endosonography.

Reproducibility of bowel ultrasonography in the evaluation of Crohn's disease

BACKGROUND

Bowel ultrasonography is increasingly used in the detection and follow-up of patients with Crohn's disease, but a limitation to its further diffusion is the lack of standardisation of ultrasonography parameters.

AIMS

This study aimed to standardise the most common bowel ultrasonography parameters in order to develop an unequivocal imaging interpretation and to assess bowel ultrasonography reproducibility.

PATIENTS

Twenty patients with Crohn's disease were examined.

METHODS

Six ultrasonographers (mean bowel ultrasonography experience=16 years) performed the study. They chose and discussed a common assessment methodology concerning eight ultrasonography parameters: bowel wall thickness, bowel wall pattern, bowel wall blood flow, enlarged mesenteric lymph nodes, mesenteric hypertrophy, abdominal free fluid, and stenosis or fistulae at four preliminary meetings. The day of the study operators were randomised to two rooms where they independently and in turn performed ultrasonography scans. Interobserver agreement was scored by kappa statistics.

RESULTS

Excellent k values were observed for bowel wall thickness (0.72-1). k Values were poor for bowel wall pattern (-0.22-0.85) and good for bowel wall blood flow (0.53-0.89). The presence of lymph nodes was reproducible (0.56-0.90) except in one case (0.25). Concordance on free fluid was excellent (0.85-1), whereas that on mesenteric hypertrophy was generally poor (0.14-0.69). Agreement was excellent for stenosis (0.81-1) whereas that for fistula was fair in room abscesses (0.31-0.48) and very good in room B (0.87-1).

CONCLUSION

Bowel ultrasonography signs used in Crohn's disease can be standardised as most of them showed a fair to good reproducibility. In particular, bowel wall thickness, the most relevant parameter for Crohn's disease detection, showed an excellent reproducibility.

The value of endoscopic ultrasonography on diagnosis and treatment of esophageal hamartoma

OBJECTIVE

To examine the values of endoscopic ultrasonography (EUS) on diagnosis and treatment of esophageal hamartoma.

METHODS

We compared and analyzed various kinds of imaging examinations such as barium esophagram, contrast-enhanced computed tomography (CT) and conventional gastroscopy in retrospectively reviewing the clinical data of an esophageal hamartoma patient seen in our clinic in the First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China. Having received various imaging examinations, this patient was finally diagnosed with esophageal hamartoma and underwent gastroscopic resection of hamartoma with the diagnostic information obtained from EUS. The patient had been regularly followed up for 13 months after treatment.

RESULTS

Barium esophagram, CT and conventional gastroscopy detected the lesion, but were unable to distinguish it from common esophagopolypus and other submucosal lesions, and unable to determine etiopathogenisis. EUS detected the hamartoma and identified its internal structure, echo, exact size, depth of invasion, origin and the relationship between adjacent tissues and organs, differentiating the lesion from other submucosal tumors and clearly defining the diagnosis. EUS-guided fine needle aspiration (FNA) also helped to identify the etiological diagnosis.

CONCLUSION

EUS was superior to other imaging means in diagnosis and treatment of hamartoma.

Interobserver agreement in diagnosing liver involvement in hereditary hemorrhagic telangiectasia by Doppler ultrasound

The purpose was to evaluate interobserver agreement on Doppler ultrasonographic (US) diagnosis of liver vascular malformations (VMs) in hereditary hemorrhagic teleangiectasia (HHT) and on their severity grading. During the interobserver agreement study, three observers with at least seven years of specific experience using Doppler US for the diagnosis of liver VMs, judged about the presence/absence of liver VMs and their severity on a set of images and videoclips. Interobserver agreement was estimated with kappa statistics. One-hundred ten cases were reviewed during interobserver study (80 cases with liver VMs, 30 without). Very good kappa values of the interobserver agreement were found for all pairs concerning the distinction between presence and absence of hepatic VMs. All observers demonstrated excellent sensitivity and specificity in identifying hepatic VMs, with their respective areas under the curve ranging from 0.97 to 0.99. Interobserver agreement among the three investigators in staging the hepatic VMs in HHT patients was moderate (Kendall's coefficient of concordance = 0.26). Study results indicate that Doppler US diagnosis of liver VMs in HHT has a high degree of agreement among ultrasonographers; a moderate agreement was found regarding severity staging.