Features of chronic pancreatitis by endoscopic ultrasound influence the diagnostic accuracy of endoscopic ultrasound-guided fine-needle aspiration of small pancreatic lesions

Pancreatic juice cytology for diagnosing invasive pancreatic carcinoma/high-grade pancreatic intraepithelial neoplasia without visible tumors on endoscopic ultrasound

OBJECTIVES

Pancreatic ductal adenocarcinoma (PDAC) with a diameter ≤10 mm and high-grade pancreatic intraepithelial neoplasia (HG-PanIN) require pre-operative diagnosis. Most cases present only indirect imaging findings without visible tumors on endoscopic ultrasound (EUS). Therefore, EUS-guided fine-needle aspiration/biopsy is not applicable. An alternative diagnostic method is pancreatic juice cytology (PJC) via endoscopic naso-pancreatic drainage (ENPD-PJC), which is not the standard practice. This study aimed to investigate ENPD-PJC for diagnosing suspected PDAC/HG-PanIN cases without visible tumors on EUS.

METHODS

Data of patients with suspected PDAC/HG-PanIN without visible tumors who underwent PJC were retrospectively evaluated. One PJC sample was collected during endoscopic retrograde pancreatography (ERP-PJC), and 12 samples were collected during ENPD-PJC, 3-hourly for cytological analysis. ERP-PJC, ERP/ENPD-PJC, and ENPD-PJC positivity indicated cytologically positive samples. Patients with positive/negative PJC with follow-up for <4-years were excluded as undiagnosed cases. A non-malignant diagnosis was based on histopathological absence/stable imaging findings for ≥4-years. The primary endpoint was to demonstrate that ERP/ENPD-PJC has a higher diagnostic ability than ERP-PJC.

RESULTS

Twenty-two patients with histopathologically diagnosed PDAC/HG-PanIN and 31 with a non-malignant diagnosis were enrolled. ERP-PJC, ERP/ENPD-PJC, and ENPD-PJC showed sensitivities of 36.4 %, 86.4 %, and 77.3 %, specificities of 93.5 %, 87.1 %, and 93.5 %, and accuracies of 69.8 %, 86.7 %, and 86.7 %, respectively. ERP/ENPD-PJC and ENPD-PJC demonstrated superior sensitivity and accuracy compared to ERP-PJC. A greater occurrence of positive outcomes markedly distinguished true positives from false positives.

CONCLUSIONS

ERP/ENPD-PJC and ENPD-PJC had higher diagnostic accuracies for PDAC/HG-PanIN without visible tumors on EUS. ENPD-PJC is recommended for the diagnosis of these lesions.

Current status of artificial intelligence analysis for the treatment of pancreaticobiliary diseases using endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography

Pancreatic and biliary diseases encompass a range of conditions requiring accurate diagnosis for appropriate treatment strategies. This diagnosis relies heavily on imaging techniques like endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography. Artificial intelligence (AI), including machine learning and deep learning, is becoming integral in medical imaging and diagnostics, such as the detection of colorectal polyps. AI shows great potential in diagnosing pancreatobiliary diseases. Unlike machine learning, which requires feature extraction and selection, deep learning can utilize images directly as input. Accurate evaluation of AI performance is a complex task due to varied terminologies, evaluation methods, and development stages. Essential aspects of AI evaluation involve defining the AI's purpose, choosing appropriate gold standards, deciding on the validation phase, and selecting reliable validation methods. AI, particularly deep learning, is increasingly employed in endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography diagnostics, achieving high accuracy levels in detecting and classifying various pancreatobiliary diseases. The AI often performs better than doctors, even in tasks like differentiating benign from malignant pancreatic tumors, cysts, and subepithelial lesions, identifying gallbladder lesions, assessing endoscopic retrograde cholangiopancreatography difficulty, and evaluating the biliary strictures. The potential for AI in diagnosing pancreatobiliary diseases, especially where other modalities have limitations, is considerable. However, a crucial constraint is the need for extensive, high-quality annotated data for AI training. Future advances in AI, such as large language models, promise further applications in the medical field.

True diagnostic ability of EUS-guided fine-needle aspiration/biopsy sampling for small pancreatic lesions ≤10 mm and salvage diagnosis by pancreatic juice cytology: a multicenter study

BACKGROUND AND AIMS

The diagnostic performance of EUS-guided fine-needle aspiration/biopsy sampling (EUS-FNAB) for pancreatic ductal adenocarcinoma (PDAC) ≤10 mm in diameter is relatively low. Pancreatic juice cytology (PJC) has gained attention because of its high sensitivity for small PDACs. We aimed to clarify the diagnostic ability of EUS-FNAB and the salvage ability of PJC for PDAC ≤10 mm.

METHODS

Data obtained from attempted EUS-FNAB for patients with EUS-confirmed pancreatic tumors ≤10 mm (excluding pancreatic metastases/malignant lymphomas) were retrospectively analyzed. Patients who experienced technical failure or had a negative EUS-FNAB result and had a strong likelihood of PDAC based on imaging characteristics underwent PJC. PDAC was diagnosed using resected histologic specimens, EUS-FNAB-positive tumor growth on the imaging examination, or additional EUS-FNAB-positive results after increase in tumor size. The primary endpoint was the diagnostic ability of EUS-FNAB for PDAC ≤10 mm. The salvage ability of PJC was also assessed.

RESULTS

Overall, 86 of 271 patients with pancreatic tumors ≤10 mm who underwent attempted EUS-FNAB were diagnosed with PDAC. The technical success rate, sensitivity, specificity, and accuracy of EUS-FNAB for PDAC ≤10 mm were 80.8%, 82.3%, 94.9%, and 91.3%, respectively. Among the 35 PDAC patients who experienced technical failure or false-negative results of EUS-FNAB, 26 (74.3%) were correctly diagnosed using salvage PJC.

CONCLUSIONS

The true success rate and sensitivity of EUS-FNAB for PDAC ≤10 mm were relatively low. When EUS-FNAB for a pancreatic lesion ≤10 mm strongly suspected to be PDAC is unsuccessful or yields a negative result, PJC is recommended. (Clinical trial registration number: UMIN000049965.).

Sensitivity of endoscopic ultrasound-guided fine-needle aspiration cytology and biopsy for a diagnosis of pancreatic ductal adenocarcinoma: A comparative analysis

The utility of endoscopic ultrasound fine-needle aspiration cytology (EUS-FNAC) or endoscopic ultrasound fine-needle aspiration biopsy (EUS-FNAB) for diagnosis of small and large pancreatic ductal adenocarcinomas (PDACs) remains in question. We addressed this by analyzing 97 definitively diagnosed cases of PDAC, for which both EUS-FNAC and EUS-FNAB had been performed. We subclassified the 97 solid masses into small (n = 35) or large (n = 62) according to the maximum tumor diameter (<24 mm or ≥24 mm) and compared the diagnostic sensitivity (truly positive rate) of EUS-FNAC and of EUS-FNAB for small and large masses. Diagnostic sensitivity of EUS-FNAC did not differ between large and small masses (79.0% vs. 60.0%; p = 0.0763). However, the diagnostic sensitivity of EUS-FNAB was significantly higher for large masses (85.5% vs. 62.9%; p = 0.0213). Accurate EUS-FNAC-based diagnosis appeared to depend on the degree of cytological atypia of cancer cells, which was not associated with quantity of cancer cells. The accuracy of EUS-FNAB-based diagnosis appeared to depend on cancer cell viability in large masses and cancer volume in small masses. Based on the advantages or disadvantages in each modality, both modalities play an important role in the qualitative diagnosis of PDAC as a complementary procedure.

Artificial intelligence using deep learning analysis of endoscopic ultrasonography images for the differential diagnosis of pancreatic masses

BACKGROUND : There are several types of pancreatic mass, so it is important to distinguish between them before treatment. Artificial intelligence (AI) is a mathematical technique that automates learning and recognition of data patterns. This study aimed to investigate the efficacy of our AI model using endoscopic ultrasonography (EUS) images of multiple types of pancreatic mass (pancreatic ductal adenocarcinoma [PDAC], pancreatic adenosquamous carcinoma [PASC], acinar cell carcinoma [ACC], metastatic pancreatic tumor [MPT], neuroendocrine carcinoma [NEC], neuroendocrine tumor [NET], solid pseudopapillary neoplasm [SPN], chronic pancreatitis, and autoimmune pancreatitis [AIP]). METHODS : Patients who underwent EUS were included in this retrospective study. The included patients were divided into training, validation, and test cohorts. Using these cohorts, an AI model that can distinguish pancreatic carcinomas from noncarcinomatous pancreatic lesions was developed using a deep-learning architecture and the diagnostic performance of the AI model was evaluated. RESULTS : 22 000 images were generated from 933 patients. The area under the curve, sensitivity, specificity, and accuracy (95 %CI) of the AI model for the diagnosis of pancreatic carcinomas in the test cohort were 0.90 (0.84-0.97), 0.94 (0.88-0.98), 0.82 (0.68-0.92), and 0.91 (0.85-0.95), respectively. The per-category sensitivities (95 %CI) of each disease were PDAC 0.96 (0.90-0.99), PASC 1.00 (0.05-1.00), ACC 1.00 (0.22-1.00), MPT 0.33 (0.01-0.91), NEC 1.00 (0.22-1.00), NET 0.93 (0.66-1.00), SPN 1.00 (0.22-1.00), chronic pancreatitis 0.78 (0.52-0.94), and AIP 0.73 (0.39-0.94). CONCLUSIONS : Our developed AI model can distinguish pancreatic carcinomas from noncarcinomatous pancreatic lesions, but external validation is needed.

Early detection of pancreatic tumors by advanced EUS imaging

The early detection of pancreatic ductal adenocarcinoma (PDAC) dramatically improves outcome. All available state-of-the-art imaging methods allow early detection with EUS being the best technique for exclusion of PDAC and detection of very early PDAC. Etiological differentiation of small SPL is important to guide individually tailored patients' management including radical surgery in resectable PDAC, medical (neoadjuvant or palliative intended) treatment in patients with non-resectable malignancy, pancreatic parenchyma saving strategies in some non-PDAC, and follow-up in particular in low-grade PanNEN or other small benign lesions. Multimodality EUS imaging including B-Mode assessment, elastography, contrast-enhancement and EUS-guided sampling is the most appropriate technique for diagnosis and risk assessment of small SPL. We present a review discussing modern (endoscopic) ultrasound imaging techniques including contrast enhanced ultrasound and elastography for the early detection and characterization of solid pancreatic lesions.

Efficacy of endoscopic samplings during initial biliary drainage for cases of pancreatic head cancer: United diagnostic yields of multiple pathological samplings

BACKGROUND/OBJECTIVES

The diagnostic ability of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) has been fully studied; however, the efficacy of other endoscopic samplings (OESs) is less clear. The aim of this study was to examine the diagnostic efficacies of OESs for pancreatic head cancer (PHC).

METHODS

The diagnostic efficacies of endoscopic samplings were retrospectively analyzed in 448 PHC cases and 63 cases of mass-forming pancreatitis (MFP) during initial transpapillary biliary drainage. The OESs included duodenal biopsy (118 PHCs and 50 MFPs), biliary biopsy (218 and 51) with cytology (368 and 53), and pancreatic duct biopsy (23 and 13) with cytology (56 and 43). EUS-FNA was conducted in a different session (149 and 62). Factors associated with OES sensitivity were analyzed. The sensitivity of biliary biopsy was compared between 1.95 mm and 1.8 mm forceps.

RESULTS

Cancer cells were confirmed in 87.9% of the EUS-FNA samplings and in 64.1% (268/418) obtained by combined OESs (average 1.7 OES types per case): 68.6% by duodenal biopsy, 59.6% by biliary biopsy, 32.6% by biliary cytology, 73.9% by pancreatic duct biopsy, and 33.9% by pancreatic duct cytology. No MFP cases revealed cancer by any sampling. OESs did not increase adverse events. Duodenal stenosis, serum bilirubin, tumor size, and pancreatic juice amounts were associated with OES sensitivity. Biliary biopsy had the same sensitivity with different forceps.

CONCLUSION

EUS-FNA was the most diagnostic protocol; however, OESs can be safely applied during the initial biliary drainage to reduce the demand for EUS-FNA while providing good diagnostic yields.

Automatic Segmentation of Pancreatic Tumors Using Deep Learning on a Video Image of Contrast-Enhanced Endoscopic Ultrasound

Background: Contrast-enhanced endoscopic ultrasound (CE-EUS) is useful for the differentiation of pancreatic tumors. Using deep learning for the segmentation and classification of pancreatic tumors might further improve the diagnostic capability of CE-EUS. Aims: The aim of this study was to evaluate the capability of deep learning for the automatic segmentation of pancreatic tumors on CE-EUS video images and possible factors affecting the automatic segmentation. Methods: This retrospective study included 100 patients who underwent CE-EUS for pancreatic tumors. The CE-EUS video images were converted from the originals to 90-s segments with six frames per second. Manual segmentation of pancreatic tumors from B-mode images was performed as ground truth. Automatic segmentation was performed using U-Net with 100 epochs and was evaluated with 4-fold cross-validation. The degree of respiratory movement (RM) and tumor boundary (TB) were divided into 3-degree intervals in each patient and evaluated as possible factors affecting the segmentation. The concordance rate was calculated using the intersection over union (IoU). Results: The median IoU of all cases was 0.77. The median IoUs in TB-1 (clear around), TB-2, and TB-3 (unclear more than half) were 0.80, 0.76, and 0.69, respectively. The IoU for TB-1 was significantly higher than that of TB-3 (p < 0.01). However, there was no significant difference between the degrees of RM. Conclusions: Automatic segmentation of pancreatic tumors using U-Net on CE-EUS video images showed a decent concordance rate. The concordance rate was lowered by an unclear TB but was not affected by RM.

Current status of artificial intelligence analysis for endoscopic ultrasonography

Endoscopic ultrasonography (EUS) is an essential diagnostic tool for various types of pancreatic diseases such as pancreatic tumors and chronic pancreatitis; however, EUS imaging has low specificity for the diagnosis of pancreatic diseases. Artificial intelligence (AI) is a mathematical prediction technique that automates learning and recognizes patterns in data. This review describes the details and principles of AI and deep learning algorithms. The term AI does not have any definite definition; almost all AI systems fall under narrow AI, which can handle single or limited tasks. Deep learning is based on neural networks, which is a machine learning technique that is widely used in the medical field. Deep learning involves three phases: data collection and annotation, building the deep learning architecture, and training and ability validation. For medical image diagnosis, image classification, object detection, and semantic segmentation are performed. In EUS, AI is used for detecting anatomical features, differential pancreatic tumors, and cysts. For this, conventional machine learning architectures are used, and deep learning architecture has been used in only two reports. Although the diagnostic abilities in these reports were about 85-95%, these were exploratory research and very few reports have included substantial evidence. AI is increasingly being used for medical image diagnosis due to its high performance and will soon become an essential technique for medical diagnosis.

Diagnostic Fine-Needle Biopsy of Small Solid Pancreatic Lesions Using a Franseen Needle during Endoscopic Ultrasound Examination

BACKGROUND AND AIM

During endoscopic ultrasound-guided fine needle aspiration biopsy (EUS-FNB), Franseen needles can help collect sufficient tissue to permit histopathological assessment. However, its efficacy might be limited by the size of the targeted lesion. This study aimed to evaluate the feasibility of histopathological assessment of small solid pancreatic lesions using a 22-gauge Franseen needle during EUS-FNB.

METHODS

This retrospective study evaluated data from all patients who underwent EUS-FNB using a Franseen needle for solid pancreatic lesions at the University of Toyama Hospital between June 2018 and April 2020.

RESULTS

The study included 159 patients who had 152 malignant lesions and 7 benign lesions. The malignant lesions included pancreatic cancers (n = 134), neuroendocrine neoplasms (n = 15), metastatic tumors (n = 2), and a solid pseudopapillary neoplasm (n = 1). The diagnostic accuracy of EUS-FNB (combining histology and cytology) was 98.7%. However, the histopathological diagnosis was only confirmed for 64.3% of small lesions (<10 mm), relative to 97.2% for larger lesions. Multivariate analysis also revealed that lesion size of <10 mm predicted a less accurate histopathological diagnosis (odds ratio: 6.97, 95% confidence interval: 1.02-47.67; p = 0.041). Further analyses revealed a failed histological diagnosis in 4 patients with lesions of <5 mm in size and accurate diagnoses in 9 out of 10 patients with lesions of 5-10 mm in size.

CONCLUSIONS

The diagnostic accuracy for small lesions (<10 mm), especially for lesions of <5 mm, based on histological examination alone, was significantly lower than that for others (>10 mm). Furthermore, multivariate analysis revealed that only lesion size was an independent predictor of histopathological diagnosis accuracy.

Chronic pancreatitis: an overview of diagnosis and management

INTRODUCTION

Chronic pancreatitis entails a heavy burden on the healthcare system because of its often protracted evolution, requiring complex diagnostic and therapeutic procedures.

AREAS COVERED

This review focuses on novel imaging and endoscopic diagnostic and therapeutic interventions that have changed the management of patients with chronic pancreatitis. We have conducted an extensive search of original papers and guidelines, in order to provide a comprehensive and up to date review of available evidence in these areas of interest.

EXPERT OPINION

The traditional challenges in managing chronic pancreatitis patients stemmed from the limitations of diagnostic modalities, which could not correctly identify patients in an early stage of the disease, as well as from the scarcity of therapeutic options available. Advances in imaging of CT-scan, MRI, and EUS have opened the way for early diagnosis and staging. This has allowed more aggressive and tailored therapeutic modalities, particularly in endoscopic therapy and minimally invasive surgical interventions. Although high-quality data from large RCTs is still scarce, evidence-based algorithms for diagnosis and therapy are now changing the way we address this chronic disease. In the near future, we can expect a tailored approach based on patient and disease-related predictive factors, relying on a vast armamentarium of endoscopic and surgical solutions.

Differential diagnosis of solid pancreatic masses

Solid pancreatic lesions include mainly adenocarcinoma, neuroendocrine tumors pancreatic cystic neoplasms with solid component, solid pseudopapillary tumor, pancreatoblastoma, pancreatic lymphoma, and pancreatic metastasis. The most frequent pancreatic lesion is the adenocarcinoma, representing between 70% and 95% of all solid pancreatic neoplasm. The diagnosis of these lesions can be a challenge and currently, there are different imaging techniques such as CT scan, EUS and MRI with high sensitivity and specificity. The most widely used technique for the initial evaluation is the CT scan with a sensitivity between 76% and 92% for the diagnosis of pancreatic cancer. The EUS has a sensitivity for the detection of pancreatic lesions of around 98% and is accepted to be the most sensitive technique for the detection of small pancreatic tumors (<2 cm). The MRI, with a very high soft-tissue contrast resolution, provides an accuracy in the detection and staging of adenocarcinoma of 90-100%. A multimodality approach is usually necessary in patients with clinical suspicion of pancreatic lesion. The EUS is required for the local evaluation of the relation of the lesion with vessels and for tissue acquisition and the CT scan and/or MRI is usually required for the local and distance staging in case of pancreatic cancer. The purpose of this review is to provide an overview of solid pancreatic lesions and the role of the different imaging techniques in their evaluation.

Utility of endoscopic ultrasound and endoscopic ultrasound-guided fine-needle aspiration for the diagnosis and management of pancreatic cystic lesions: Differences between the guidelines

Recent advances and frequent use of cross-sectional imaging studies have increased opportunities for incidental diagnoses of pancreatic cystic lesions (PCL). In the management of PCL, distinguishing between mucinous versus non-mucinous and malignant versus benign cysts is important to diagnose pancreatic cancer in its early stage. For this reason, there have been several guidelines to manage PCL. Endoscopic ultrasound (EUS) and EUS-guided fine-needle aspiration (FNA) play important roles in the guidelines, although there are some differences in their roles. In this review, we aimed to evaluate the current status of EUS and EUS-FNA in the management of PCL and the status of these procedures in the guidelines.