On-site Cytopathological Evaluation Research Articles

Effectiveness of data-augmentation on deep learning in evaluating rapid on-site cytopathology at endoscopic ultrasound-guided fine needle aspiration

Rapid on-site cytopathology evaluation (ROSE) has been considered an effective method to increase the diagnostic ability of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA); however, ROSE is unavailable in most institutes worldwide due to the shortage of cytopathologists. To overcome this situation, we created an artificial intelligence (AI)-based system (the ROSE-AI system), which was trained with the augmented data to evaluate the slide images acquired by EUS-FNA. This study aimed to clarify the effects of such data-augmentation on establishing an effective ROSE-AI system by comparing the efficacy of various data-augmentation techniques. The ROSE-AI system was trained with increased data obtained by the various data-augmentation techniques, including geometric transformation, color space transformation, and kernel filtering. By performing five-fold cross-validation, we compared the efficacy of each data-augmentation technique on the increasing diagnostic abilities of the ROSE-AI system. We collected 4059 divided EUS-FNA slide images from 36 patients with pancreatic cancer and nine patients with non-pancreatic cancer. The diagnostic ability of the ROSE-AI system without data augmentation had a sensitivity, specificity, and accuracy of 87.5%, 79.7%, and 83.7%, respectively. While, some data-augmentation techniques decreased diagnostic ability, the ROSE-AI system trained only with the augmented data using the geometric transformation technique had the highest diagnostic accuracy (88.2%). We successfully developed a prototype ROSE-AI system with high diagnostic ability. Each data-augmentation technique may have various compatibilities with AI-mediated diagnostics, and the geometric transformation was the most effective for the ROSE-AI system.

A Multiscale Connected UNet for the Segmentation of Lung Cancer Cells in Pathology Sections Stained Using Rapid On-Site Cytopathological Evaluation

Lung cancer is an increasingly serious health problem worldwide, and early detection and diagnosis are crucial for successful treatment. With the development of artificial intelligence and the growth of data volume, machine learning techniques can play a significant role in improving the accuracy of early detection in lung cancer.This study proposes a deep learning-based segmentation algorithm for rapid on-site cytopathological evaluation (ROSE) to enhance the diagnostic efficiency of endobronchial ultrasoundguided transbronchial needle aspiration biopsy (EBUS-TBNA) during surgery. By utilizing the CUNet3+ network model, cell clusters, including cancer cell clusters, can be accurately segmented in ROSE-stained pathological sections. The model demonstrated high accuracy, with an F1-score of 0.9604, recall of 0.9609, precision of 0.9654, and accuracy of 0.9834 on the internal testing dataset. It also achieved an AUC of 0.9972 for cancer identification. The proposed algorithm provides time savings for on-site diagnosis, improves EBUS-TBNA efficiency, and outperforms classical segmentation algorithms in accurately identifying lung cancer cell clusters in ROSE-stained images. It effectively reduces over-segmentation, decreases network parameters, and enhances computational efficiency, making it suitable for real-time patient evaluation during surgical procedures.

Application of artificial intelligence to digital-rapid on-site cytopathology evaluation during endoscopic ultrasound-guided fine needle aspiration: A proof-of-concept study.

During endoscopic ultrasound-guided fine needle aspiration (EUS-FNA), cytopathology with rapid on-site evaluation (ROSE) can improve diagnostic yield and accuracy. However, ROSE is unavailable in most Asian and European institutions because of the shortage of cytopathologists. Therefore, developing computer-assisted diagnostic tools to replace manual ROSE is crucial. Herein, we reported the validation of an artificial intelligence (AI)-based model (ROSE-AI model) to substitute manual ROSE during EUS-FNA. A total of 467 digitized images from Diff-Quik (D&F)-stained EUS-FNA slides were divided into training (3642 tiles from 367 images) and internal validation (916 tiles from 100 images) datasets. The ROSE-AI model was trained and validated using training and internal validation datasets, respectively. The specificity was emphasized while developing the model. Then, we evaluated the AI model on a 693-image external dataset. We assessed the performance of the AI model to detect cancer cells (CCs) regarding the accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The ROSE-AI model achieved an accuracy of 83.4% in the internal validation dataset and 88.7% in the external test dataset. The sensitivity and PPV were 79.1% and 71.7% in internal validation dataset and 78.0% and 60.7% in external test dataset, respectively. We provided a proof of concept that AI can be used to replace manual ROSE during EUS-FNA. The ROSE-AI model can address the shortage of cytopathologists and make ROSE available in more institutes.

Diagnostic performance of endoscopic ultrasound-guided tissue acquisition by EUS-FNA versus EUS-FNB for solid pancreatic mass without ROSE: a retrospective study

BackgroundEndoscopic ultrasound-guided tissue acquisition (EUS-TA) is an established diagnostic procedure for solid pancreatic mass. However, the diagnostic yield between fine-needle aspiration (FNA) and fine-needle biopsy (FNB) remains unclear. We aimed to evaluate and compare the diagnostic yields between FNA and FNB using conventional FNA and Franseen needles of the same size 22-gauge needle, in patients with solid pancreatic mass who underwent EUS-TA without rapid onsite cytopathology evaluation (ROSE).MethodsAll cases of EUS-TA by FNA or FNB for solid pancreatic mass between January 2017 and October 2020 in a single-centre university hospital were retrospectively reviewed. All procedures were performed without an onsite cytologist. Before the endoscopist finished the procedure, macroscopic onsite evaluation (MOSE) was confirmed. The diagnostic yield and the average number of needle passes between FNB and FNA were then compared.ResultsA total of 151 patients (FNA, n = 77; FNB, n = 74) with solid pancreatic mass detected by cross-sectional imaging underwent EUS-TA. The mean age was 62.3 ± 12.8 years, with 88 (58.3%) males. Age, sex, mass location, tumour size and disease stage from imaging were not significantly different between the two groups. The diagnostic performance was higher in EUS-FNB (94.6%) than in EUS-FNA (89.6%). The mean number of needle passes was clearly fewer in FNB than in FNA (2.8 vs. 3.8, p < 0.001). The total procedure time was shorter in FNB (34.7 min) than in FNA (41 min). The adverse event rate between FNB and FNA was not significantly different.ConclusionsThe diagnostic yield of solid pancreatic mass was higher in FNB using the Franseen needle than in FNA using the conventional FNA needle in a centre where ROSE is unavailable, without serious adverse event. In addition, FNB had fewer needle passes and shorter total procedure time.

A deep learning-based segmentation system for rapid onsite cytologic pathology evaluation of pancreatic masses: A retrospective, multicenter, diagnostic study.

SummaryBackgroundWe aimed to develop a deep learning-based segmentation system for rapid on-site cytopathology evaluation (ROSE) to improve the diagnostic efficiency of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) biopsy.MethodsA retrospective, multicenter, diagnostic study was conducted using 5345 cytopathological slide images from 194 patients who underwent EUS-FNA. These patients were from Nanjing Drum Tower Hospital (109 patients), Wuxi People's Hospital (30 patients), Wuxi Second People's Hospital (25 patients), and The Second Affiliated Hospital of Soochow University (30 patients). A deep convolutional neural network (DCNN) system was developed to segment cell clusters and identify cancer cell clusters with cytopathological slide images. Internal testing, external testing, subgroup analysis, and human–machine competition were used to evaluate the performance of the system.FindingsThe DCNN system segmented stained cells from the background in cytopathological slides with an F1-score of 0·929 and 0·899–0·938 in internal and external testing, respectively. For cancer identification, the DCNN system identified images containing cancer clusters with AUCs of 0·958 and 0·948–0·976 in internal and external testing, respectively. The generalizable and robust performance of the DCNN system was validated in sensitivity analysis (AUC > 0·900) and was superior to that of trained endoscopists and comparable to cytopathologists on our testing datasets.InterpretationThe DCNN system is feasible and robust for identifying sample adequacy and pancreatic cancer cell clusters. Prospective studies are warranted to evaluate the clinical significance of the system.FundingJiangsu Natural Science Foundation; Nanjing Medical Science and Technology Development Funding; National Natural Science Foundation of China.

Usefulness of Macroscopic On-Site Evaluation Using a Stereomicroscope during EUS-FNB for Diagnosing Solid Pancreatic Lesions.

Methods We reviewed a total of 60 consecutive patients who underwent both S-MOSE and rapid on-site cytopathological evaluation (ROSE) during EUS-FNB between July 2019 and October 2020, and the usefulness of S-MOSE in comparison with histology was evaluated. A 22-gauge Franseen needle was used to perform EUS-FNB in all patients, and only the specimens obtained by the first pass were evaluated. The final diagnosis was based on the surgical specimen or the clinical course consistent with the EUS-FNB results. Results The final diagnoses of the 60 patients included 45 patients with pancreatic ductal adenocarcinoma, 6 with autoimmune pancreatitis, 4 with mass-forming pancreatitis, 1 with pancreatic metastasis, 2 with pancreatic neuroendocrine tumor, and 2 with intraductal papillary mucinous carcinoma. The histological diagnostic accuracy of the first pass of EUS-FNB was 83.3% (50/60). The agreement between the S-MOSE and the histological diagnosis was 90% (54/60). The positive predictive value of S-MOSE for histological diagnosis was 90.7%, which can be an indicator of when to stop the EUS-FNB procedure. There were no immediate or delayed adverse events reported after the FNB based on the chart and medical visit history review. Conclusion In the EUS-FNB of SPLs, S-MOSE can be an alternative to ROSE for specimen evaluation and has the potential to shorten the procedure time.

Artificial intelligence-based rapid on-site cytopathological evaluation for bronchoscopy examinations

Cytological rapid on-site evaluation (ROSE) is becoming an integral technique for improving the performance of bronchoscopic examinations by confirming specimenadequacy and accuracy in real-time. However, the time- and personnel-consuming nature of ROSE limits its application.We constructed an artificial intelligence (AI)-based ROSE model using deep-learning convolutional neural network (CNN) technique to assist in classifying cytologic whole-slide images (WSIs) as malignant or benign. A total of 627 patients with ROSE slides were enrolled, among whom 374 and 91 patients were included and randomly assigned into training and validation groups, respectively. Another 162 patients were selected as a testing group. The malignant-benign classification results of the test group were compared between cytopathologists' results and AI-based ROSE model results. Actual ROSE reports of the test group given on-site were considered as results of junior cytopathologists; the official cytological diagnostic reports of the test group, which were given without time pressure and with reference to more clinical and pathological information by the senior cytopathologist, were considered as results of the senior cytopathologist. The real-world comprehensive diagnosis was considered as the gold standard.The area under the ROC curve (AUC) achieved 0.9846 in the validation group at patch-level. The accuracy achieved by one senior cytopathologist, two junior cytopathologists and the AI-based ROSE model were 96.90%, 83.30%, and 84.57%, respectively.This AI-based ROSE model may have the potential to support the diagnosis and therapeutic management of patients with respiratory lesions.

Endosonographers performing on-site evaluation of solid pancreatic specimens for EUS-guided biopsy: A formal training method and learning curves.

Objectives:This study aimed to examine the effects of a training program combining formal cytological curriculum and practical assessments on endosonographers and to determine how many operations were needed for training through learning curves.Methods:A formal cytological curriculum was implemented in November 2019 for four endosonographers. The competency of endosonographers before and after the curriculum was judged by quantitative scores. From December 2019 to October 2020, trainees independently examined the adequacy and atypia grade of pancreatic specimens acquired by EUS-guided-fine-needle biopsy based on specific atypical grading criteria. The accuracy of the adequacy and atypia assessment of each trainee was calculated, and learning curves were drawn to evaluate the improvement.Results:The median posttraining score improved by 87% from 49 to 91.5 out of 100. Ninety patients were enrolled in the practical assessments. The accuracy for specimen adequacy and atypia assessment of each trainee was 91.7%, 92.8%, 91.0%, and 89.3% and 80.0%, 82.1%, 81.0%, and 78.9%. The learning curves of each trainee showed a steady and significant improvement, and ninety cases were sufficient for satisfactory adequacy assessment.Conclusions:Basic knowledge of on-site cytopathologic evaluation can be gained through standardized and systematic training. Practical assessments showed that, at the completion of ninety cases, trained endosonographers reached a satisfactory level of recognizing specimen adequacy, but continue practice was needed for atypia assessment.

Prospective Analysis of a Novel Endobronchial Augmented Fluoroscopic Navigation System for Diagnosis of Peripheral Pulmonary Lesions.

Background:Navigational bronchoscopy has improved upon traditional bronchoscopy to identify suspicious pulmonary lesions, but wide variability exists in the diagnostic yield of various modalities. The aim of this study was to measure localization accuracy and diagnostic yield of a novel endobronchial augmented fluoroscopic navigation system (first-generation LungVision system) for peripheral pulmonary lesions (PPLs).Methods:This prospective single-center study included adults undergoing guided bronchoscopy to evaluate PPLs. Preprocedure computed tomography (CT) images were obtained, and planning software calculated a pathway to the lesion. A flexible bronchoscope was used to navigate along the pathway overlaid on the intraprocedural fluoroscopic image. When real-time display indicated the catheter tip had reached the lesion, cone-beam computed tomography (CBCT) was used to measure the actual location of the tip. Biopsy and rapid on-site cytopathologic evaluation were performed.Results:Fifty-one patients were included in the analysis. The median lesion diameter was 18.0 mm (range: 7.0 to 48.0 mm). Localization success was 96.1%. The average distance between lesion location as shown by LungVision augmented fluoroscopy and actual location measured by CBCT was 5.9 mm (range: 2.1 to 10.0 mm). Diagnostic yield at the index procedure was 78.4%. Diagnostic accuracy assessed at 12 months follow-up was 88.2%. Average CT-to-body divergence was 14.5 mm (range: 2.6 to 33.0 mm) from preprocedure CT to intraprocedural CBCT images.Conclusion:Augmented fluoroscopy for navigation and biopsy of PPLs with the LungVision system showed a high localization success rate and corresponding high diagnostic yield. Navigation and biopsy with real-time visualization can improve diagnostic yield for PPLs.

Endobronchial ultrasound needles: does size matter?

Endobronchial ultrasound guided sampling is a commonly performed endoscopy procedure. Despite its widespread use, the technical aspects of endobronchial ultrasound-guided transbronchial needle aspiration, including the selection of needle size, have not been standardized among operators or institutions. Furthermore, professional guidelines are vague regarding their recommendations in choosing a needle size for optimal sampling. Various factors such as operator training, type of needle, use of suction, needle-tissue interaction, sample handling, processing, and presence of rapid on-site cytopathologic evaluation may impact the diagnostic yield. This review focuses on the various needle sizes used in endobronchial ultrasound-guided transbronchial needle aspiration, specifically comparing diagnostic accuracy and adequacy of tissue sampling, with a focus on use in lung malignancy, sarcoidosis, and lymphoproliferative disorders. Additionally, the advantages and limitations of different needle sizes will be addressed.

Endoscopic Ultrasound-Guided Tissue Acquisition Using Shark Core and Fork-Tip Needles Improves Histological Yield, Reduces Needle Passes, Without On-Site Cytopathological Evaluation. A Comparative Study between Two Tertiary Medical Centers

Endoscopic Ultrasound-Guided Tissue Acquisition Using Shark Core and Fork-Tip Needles Improves Histological Yield, Reduces Needle Passes, Without On-Site Cytopathological Evaluation. A Comparative Study between Two Tertiary Medical Centers

Endoscopic Ultrasound-Guided Tissue Acquisition Using Fork-Tip Needle Improves Histological Yield, Reduces Needle Passes, Without On-Site Cytopathological Evaluation.

Background and Aim: Endoscopic ultrasound (EUS)-guided fine needle biopsy (FNB) and fine needle aspiration (FNA) are established methods in tissue acquisition. A new fork-tip FNB needle has been used to obtain core tissue samples. We compared the performance of the FNB using fork-tip needles with that of the FNA using conventional needles in patients who had solid neoplastic lesions within and around the upper gastrointestinal (GI) tract.Methods: In this retrospective single-center study, patients who underwent EUS examinations for solid neoplastic lesions between October 2013 and February 2017 were included. The procedures were performed in the absence of an on-site cytologist. The main objectives were to compare the diagnostic yield and average number of passes of FNB using fork-tip needles versus those of FNA using conventional needles.Results: EUS/FNA and EUS/FNB were performed on 181 solid neoplastic lesions primarily in the pancreas and GI tract walls. There was no significant difference in patient's age, gender, tumor location, or tumor size. The mean number of needle passes was significantly lower in the fork-tip needle group than in the conventional needle group (3.8 vs. 5.9; p < 0.0001). There was a trend toward higher sensitivity (89.9% vs. 81%) using the fork-tip needles than when using the conventional needles (p = 0.119). No significant difference in rates of adverse events between two groups was found.Conclusions: Our study demonstrates that, compared with FNA using conventional needles, FNB using fork-tip needles required significantly fewer needle passes while achieving a relatively higher diagnostic yield due to its superior capacity in tissue acquisition from solid neoplastic lesions in and around GI tract walls without on-site cytological assessment.

Additive Benefit of Confocal Laser Endomicroscopy and Micro-Forceps Needle Biopsy in the Evaluation of a Mucinous Cystic Neoplasm

A 76 F presented with 6 months of abdominal pain and CT images of a pancreatic cyst. Multiple inconclusive EUS FNA procedures were performed at an outside hospital. We used a linear array echoendoscope to visualize the pancreatic parenchyma at our facility and then passed a confocal probe (AQ Flex) and micro-forceps (Moray needle) through a 19g needle (Cook Medical) into the cystic lesion. Initial on-site cytopathology evaluation of cyst fluid suggested atypical cells. Cyst fluid analysis subsequently demonstrated elevated CEA 8537 ng/mL and amylase 1788 U/L. The AQ flex probe was advanced along the cyst wall for CLE and the pattern demonstrated a single thick band suggestive of a mucinous cystic neoplasm (MCN). The micro forceps were advanced to that region and biopsy the cyst wall. Touch prep evaluation of the forceps biopsy and cell block analysis were positive for mucinous lesion. The patient received Cipro 400 mg IV during the procedure then discharged on 3-day oral course. She developed intense abdominal pain with syncope before outpatient resection. An outside hospital CT showed the pancreatic cyst with moderate amount of peripancreatic fluid with an intralesional linear density, possibly a stent or remnant of a biopsy apparatus. A distal pancreatectomy, splenectomy, left nephrectomy and left adrenalectomy was performed and revealed MCN with dysplasia measuring 8 cm with no invasive carcinoma. Margins were negative for tumor and no malignancy was identified in the lymph node sampled. Also obtained was the foreign body. At 2 week follow up, patient was doing well with no abdominal pain. We examined a pancreatic cyst case using EUS-FNA, CLE, micro-forceps needle biopsy and surgical pathology. This unique case report demonstrates the additive efficacy of EUS-FNA, CLE and micro-forceps biopsy for diagnosis. The case also describes a complication of retained foreign material within the cyst cavity; 19 g needle from the EUS device. Future prospective studies evaluating the added benefit of pancreatic cyst confocal and micro-biopsy are warranted. Additionally, it is important to note that many products utilized in clinical practice have not been designed for that specific purpose. The passage of the CLE probe and moray needle through the 19g needle at an appropriate echoendoscope position for cyst sampling may have resulted in breakage of the needle. Our institution now performs evaluations on devices after use as part of a post-procedure checklist.1382_A.tif Figure 1: CT of pancreas tail cystic lesion.1382_B.tif Figure 2: The dysplastic images show the mucinous cystic neoplasm with multilayered cells and nuclei that are coming off the basement membrane. There is also some nuclear pleomorphism as well. This is in contrast to the non-dysplastic image attached which shows a single layer of uniform cells. Note in both instances the ovarian type stroma underneath the epithelial layer, characteristic of a mucinous cystic neoplasm1382_C.tif Figure 3: FNA needle in the surgically resected pancreatic cyst specimen.

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

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. This is a single-center, retrospective study. 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%). 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.

3:09 PM Abstract No. 92 Impact of rapid onsite pathology evaluation (ROSE) on biopsy adequacy: a retrospective review of 610 image-guided biopsies

Rapid onsite evaluation (ROSE) for percutaneous biopsy is a system in which a cytopathologist is available at the time of procedure to evaluate sample adequacy. The aim of our study is to evaluate sample adequacy rates at two academic hospitals staffed by the same group of interventional radiologists, one which uses ROSE and one which does not. We hypothesize ROSE improves the rate of diagnostic specimens obtained and decreases both the number of needle passes and the need for repeat biopsy. IRB approval was obtained at center 1 (with ROSE) and waived at center 2 (without ROSE) for this retrospective review of 610 liver, lung and thyroid biopsies. Center 1 included 96 thyroid, 110 liver, and 114 lung biopsies; center 2 included 117 thyroid, 60 liver, and 112 lung biopsies. The two groups were compared using the t-test and ANOVA for age and sex, respectively. Comparisons of the number of needle passes were done using least squares regression, adjusting for age, sex, and target site. Comparison of whether a specimen was diagnostic was done using a multivariate logistic regression model. After performing multivariate analysis adjusting for age, sex, and tissue type, the odds of having a diagnostic specimen were almost 4 times greater at center 1 than at center 2 (OR 3.97, p<0.01). The rebiopsy rate was significantly lower at center 1 when compared to center 2 (p<0.01). There was no significant difference in the number of needle passes taken per biopsy site when compared between the two centers. ROSE improves the diagnostic yield of image-guided liver, lung, and thyroid biopsies. Our large study presents evidence that on-site cytopathology is a valuable resource; it improves the time to diagnosis by increasing the diagnostic yield at the time of procedure, and it decreases the need for rebiopsy. Thus, centers that utilize image-guided percutaneous biopsy should allocate resources to cover on-site cytopathologic evaluation.

High-resolution endoscopic ultrasound imaging and the number of needle passages are significant factors predicting high yield of endoscopic ultrasound-guided fine needle aspiration for pancreatic solid masses without an on-site cytopathologist.

Endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) is the accurate diagnostic method for pancreatic masses and its accuracy is affected by various FNA methods and EUS equipment. Therefore, we aimed to elucidate the instrumental and methodologic factors for determining the diagnostic yield of EUS-FNA for pancreatic solid masses without an on-site cytopathology evaluation.We retrospectively reviewed the medical records of 260 patients (265 pancreatic solid masses) who underwent EUS-FNA. We compared historical conventional EUS groups with high-resolution imaging devices and finally analyzed various factors affecting EUS-FNA accuracy.In total, 265 pancreatic solid masses of 260 patients were included in this study. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of EUS-FNA for pancreatic solid masses without on-site cytopathology evaluation were 83.4%, 81.8%, 100.0%, 100.0%, and 34.3%, respectively. In comparison with conventional image group, high-resolution image group showed the increased accuracy, sensitivity and specificity of EUS-FNA (71.3% vs 92.7%, 68.9% vs 91.9%, and 100% vs 100%, respectively). On the multivariate analysis with various instrumental and methodologic factors, high-resolution imaging (P = 0.040, odds ratio = 3.28) and 3 or more needle passes (P = 0.039, odds ratio = 2.41) were important factors affecting diagnostic yield of pancreatic solid masses.High-resolution imaging and 3 or more passes were the most significant factors influencing diagnostic yield of EUS-FNA in patients with pancreatic solid masses without an on-site cytopathologist.

P1.03-044 EUS-Guided Sampling of Mediastinal Lymph Nodes and Abdominal Lesions in Lung Cancer

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS–TBNA) was introduced to provide access to mediastinal and hilar lymph nodes. However, it is difficult to use EBUS to approach the aortopulmonary window and paraesophagaeal stations. Transesophageal endoscopic ultrasound (EUS) was introduced to provide access to this area. In addition, transgastroduodenal endoscopic ultrasound can evaluate abdominal lesions. Endoscopic ultrasound fine needle aspiration (EUS-FNA) was performed under conscious sedation with the administration of intravenous midazolam and pethidine hydrochloride. It was performed with a convex array echoendoscope connected to an ultrasound scanning system. Lymph nodes of paraesophageal, subcarinal, lower paratracheal, subaortic, and upper paratracheal regions were evaluated from esophagus. Left adrenal gland and right adrenal gland were evaluated from stomach and duodenum, respectively. Abdominal lesions were also evaluated from stomach and duodenum. After obtaining tissue via EUS-FNA, the tissue was reviewed immediately (rapid on-site cytopathological evaluation: ROSE) by a cytopathologist. Subsequent punctures in the same patient were not performed before confirming the results of ROSE so as to minimize the complications. As to the lymph node level, the lower mediastinum and the aortopulmonary window are particularly important for detection by transesophageal EUS, whereas pretracheal and hilar lymph nodes are out of reach because of the interposition of air from the trachea and bronchi. EUS was chosen to assess the posterior mediastinum nodes (#5, 7, 8, or 9) but not the anterior ones. A final diagnosis was obtained by EUS-FNA in 76 patients. The lesions sampled were mediastinal lymph nodes (n=64; #5, 7, 8, or 9), abdominal lymph nodes (n=8), and adrenal gland (n=4). Repeat tumor biopsies from patients with acquired resistance were initially obtained through research efforts to ascertain mechanisms of resistance, but are now recommended to help select second-line therapies. However, such biopsies are associated with both risk and discomfort and may not always supply enough tumor tissue for genetic analyses. Although EUS–FNA does not provide access to pretracheal and hilar lymph nodes, EUS-FNA is an accurate, safe, and minimally invasive modality for evaluating mediastinal lymphadenopathy and abdominal lesions in patients suspected of having lung metastases.

Increasing Number of Passes Beyond 4 Does Not Increase Sensitivity of Detection of Pancreatic Malignancy by Endoscopic Ultrasound-Guided Fine-Needle Aspiration.

It is not clear exactly how many passes are required to determine whether pancreatic masses are malignant using endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA). We aimed to define the per-pass diagnostic yield of EUS-FNA for establishing the malignancy of a pancreatic mass, and identify factors associated with detection of malignancies. In a prospective study, 239 patients with solid pancreatic masses were randomly assigned to groups that underwent EUS-FNA, with the number of passes determined by an on-site cytopathology evaluation or set at 7 passes, at 3 tertiary referral centers. A final diagnosis of pancreatic malignancy was made based on findings from cytology, surgery, or a follow-up evaluation at least 1 year after EUS-FNA. The cumulative sensitivity of detection of malignancy by EUS-FNA was calculated after each pass; in the primary analysis, lesions categorized as malignant or suspicious were considered as positive findings. Pancreatic malignancies were found in 202 patients (84.5% of the study population). EUS-FNA detected malignancies with 96% sensitivity (95% confidence interval [CI], 92%-98%); 4 passes of EUS-FNA detected malignancies with 92% sensitivity (95% CI, 87%-95%). Tumor size greater than 2cm was the only variable associated with positive results from cytology analysis (odds ratio, 7.8; 95% CI, 1.9-31.6). In masses larger than 2 cm, 4 passes of EUS-FNA detected malignancies with 93% sensitivity (95% CI, 89%-96%) and in masses ≤2 cm, 6 passes was associated with 82% sensitivity (95% CI, 61%-93%). Sensitivity of detection did not increase with increasing number of passes. In a prospective study, we found 4 passes of EUS-FNA to be sufficient to detect malignant pancreatic masses; increasing the number of passes did not increase the sensitivity of detection. Tumor size greater than 2 cm was associated with malignancy, and a greater number of passes may be required to evaluate masses 2 cm or less. ClinicalTrials.gov number, NCT01386931.

Su1989 Optimal Sample Smear for Rapid on-site Cytopathological Evaluation for EUS-guided FNA in Solid Pancreatic Masses

Su1989 Optimal Sample Smear for Rapid on-site Cytopathological Evaluation for EUS-guided FNA in Solid Pancreatic Masses

Endoscopic ultrasound fine-needle aspiration vs. fine-needle biopsy: tissue is always the issue.

Endoscopic ultrasound fine-needle aspiration vs. fine-needle biopsy: tissue is always the issue.