Recent Advances in Endoscopic Ultrasound (EUS) for Pancreatic Cystic Lesions

Abstract

Pancreatic cystic lesions (PCLs), increasingly being detected via advanced imaging, pose diagnostic and management challenges due to their varying malignant potential. This review explores recent advances in endoscopic ultrasound (EUS) for PCL evaluation. PCLs are classified as neoplastic (e.g., intraductal papillary mucinous neoplasms [IPMNs], mucinous cystic neoplasms [MCNs]) or nonneoplastic (e.g., serous cystic neoplasms), with IPMNs and MCNs carrying higher malignancy risks (16–60% and 10–17%, respectively). Conventional EUS offers high-resolution imaging, outperforming computed tomography/magnetic resonance imaging in detecting high-risk features, though operator dependency limits reproducibility. Contrast-enhanced EUS enhances vascularity assessment, achieving 97% sensitivity for identifying high-grade dysplasia. Detective flow imaging EUS detects microvasculature without contrast, showing promise but requiring further validation. EUS-guided needle-based confocal laser endomicroscopy provides real-time histopathology, with 98% sensitivity for mucinous PCLs. EUS-guided sulfur hexafluoride pancreatography differentiates IPMNs with 96.6% accuracy. EUS-guided fine-needle aspiration and biopsy (FNB) improve diagnostic yield, with FNB offering 87% accuracy. Through-the-needle biopsy achieves 80 to 90% sensitivity for mucinous cysts, enhanced by molecular analysis (e.g., KRAS mutations). Artificial intelligence (AI) boosts EUS accuracy to 98.5% for cyst differentiation, reducing operator variability. Therapeutically, EUS-guided chemoablation and radiofrequency ablation offer minimally invasive options, with alcohol-free protocols improving safety (67% resolution). Challenges include complication risks and nonstandardized protocols and surveillance. Future directions involve AI integration, multiomics, and standardized protocols to optimize personalized PCL management, minimizing overtreatment while prioritizing high-risk lesions.

Keywords

endoscopic ultrasound - EUS - mucinous cysts - pancreas - pancreatic cysts - pseudocyst

Introduction

Pancreatic cysts are commonly encountered entities, often detected on imaging scans with increasing incidence, around 2 to 20% over the past few decades.[1] With increased availability of imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI) scans, the incidence of pancreatic cysts noted on unrelated imaging has also increased.[2] The highest incidence rates are noted in the high-risk populations, with family history of pancreatic cysts and patients presenting with pancreatitis.[1] [3]

Classification and Malignant Potential

Pancreatic cysts can be broadly divided into neoplastic and nonneoplastic cysts. Neoplastic cysts are predominantly mucinous or nonmucinous. Intraductal papillary mucinous neoplasms (IPMNs), mucinous cystic neoplasms (MCNs), solid pseudopapillary neoplasms (SPNs), and cystic pancreatic endocrine neoplasms constitute neoplastic cysts. Nonneoplastic cysts include simple cysts, serous cystic neoplasms (SCNs), mucinous nonneoplastic cysts, and lymphoepithelial cysts. Majority are usually benign and do not require any treatment or surveillance.[1] [3]

IPMNs are the most common pancreatic cystic lesions (PCLs) comprising more than 50% of all the pancreatic cysts. They arise from pancreatic duct (PD) epithelial cell lining. These are further classified as main-duct (MD-IPMN), branch-duct (BD-IPMN), or mixed-type.[4] MD-IPMNs and mixed-type IPMNs carry a high malignancy risk, while BD-IPMNs have a modest risk.[3] [5] High-risk features for IPMNs include main PD (MPD) dilation > 10 mm, cyst size > 4 cm, and mural nodules, indicating a need for closer monitoring or intervention.[6]

MCNs are solitary, macrocystic lesions in the pancreatic body or tail carrying a 10 to 17% risk of progressing to pancreatic ductal adenocarcinoma (PDAC). This risk is lower for lesions < 3 cm without other high-risk features.[4] [7] [8] SCNs are commonly seen in women aged 40 to 60. They are usually benign with a low malignancy risk and a characteristic microcystic “honeycomb” pattern with a central scar.[4] [9] SPNs, typically in young women, are low-grade malignant with a fibrous pseudocapsule and excellent postresection outcomes despite moderate malignancy risk.[4] [10] [11] Cystic pancreatic neuroendocrine tumors (PNETs), making up 15% of PNETs, carry a moderate malignancy risk, especially for lesions > 2 cm, often necessitating surgery.[4]

As most of these cysts are discovered incidentally, accurate diagnosis is important for management of these lesions as management changes drastically depending on the type of the cystic lesions and if the cyst has any malignant potential and risk of progression.[1] [12] [13] [14] MD-IPMNs and mixed-type IPMNs have the highest risk of PDAC progression, followed by MCNs. SPNs and cNETs have lower but notable malignancy risks, while SCNs and pseudocysts are generally benign. Endoscopic ultrasound (EUS) is an excellent diagnostic tool playing a crucial role in identifying high-risk features, sampling and guiding management decisions ([Table 1]).[15]

Table 1 Pancreatic cysts and characteristics

Cyst type	Imaging findings	Malignant potential
Serous cystadenoma	Microcystic/oligocystic; central scar; no duct connection	0%
Intraductal papillary mucinous neoplasm	Duct communication; multiplicity; MPD dilatation;	1–38% (branch); 33–85% (main)
Mucinous cystic neoplasm	Unilocular/oligolocular Thickened wall; calcifications (25%)	10–34%
Solid pseudopapillary tumor	Heterogeneous; calcifications	10–15%
Cystic neuroendocrine tumor	Enhancing; thickened wall	5–10%
Pseudocyst	Unilocular or multilocular; may connect to MPD	0%
Cyst type	EUS findings
Serous cystadenoma[102]
Intraductal papillary mucinous neoplasm[103] [104] ▓a. Branch duct IPMN ▓b. Main duct IPMN
Mucinous cystic neoplasm[100]
Solid pseudopapillary tumor[105]
Cystic neuroendocrine tumor[106]
Pseudocyst[107]

Abbreviations: EUS, endoscopic ultrasound; IPMN, intraductal papillary mucinous neoplasm; MPD, main pancreatic duct.

Note: Adapted from Gonda et.al.[108]

Role of EUS in Diagnosis

Conventional EUS

EUS is the gold standard for pancreatic imaging and has been a pivotal diagnostic tool for PCLs, providing high-resolution imaging to characterize cyst morphology, size, and high-risk features with superior accuracy compared with CT or MRI. It also has the additional utility of tissue and fluid sampling, which is crucial in the management of these cysts.[16] [17]

Conventional EUS provides detailed imaging of PCL morphology, with characteristic findings such as grape-like clusters in BD-IPMNs or diffuse MPD dilation in MD-IPMNs, solitary multilocular cysts without MPD communication in MCNs, and spongy honeycomb-like polycystic/microcystic appearance in SCNs.[6] [16]

EUS showed improved sensitivity and specificity for detecting pancreatic cancer in IPMN patients, significantly outperforming CT (36%) and MRI (54%) in a large retrospective study by Khashab et al.[18] [19] However, its diagnostic accuracy is operator-dependent, with modest interobserver agreement (K = 0.24) for differentiating mucinous versus nonmucinous cysts and neoplastic versus nonneoplastic cysts.[3] [20] [21]

Despite its efficacy, EUS requires expertise and carries a 1 to 2% risk of complications like pancreatitis post-fine-needle aspiration (FNA). EUS remains essential for accurate PCL diagnosis and risk stratification.

Contrast-Enhanced EUS and Contrast-Harmonic EUS (CH-EUS)

Tissue harmonic imaging (THI) improves EUS image quality by using harmonic frequencies generated as ultrasound waves pass through tissue, reducing noise and artifacts. It provides clearer visualization of pancreatic structures, vascularity, and small lesions compared with conventional B-mode imaging. In EUS, THI enhances diagnostic accuracy for detecting mural nodules, vascular invasion, and subtle parenchymal changes in pancreatic diseases.[22]

Contrast-enhanced EUS (CE-EUS) and contrast harmonic EUS (CH-EUS) have become critical tools in the diagnostic evaluation of pancreatic cystic neoplasms (PCNs), particularly IPMNs, due to its superior ability to assess vascularity and differentiate malignant from benign lesions.[23] It uses intravenous contrast agents like Sonazoid or SonoVue to enhance microvasculature visualization, aiding differentiation between hypo-, hyper-, and isovascular lesions, offering higher sensitivity than conventional imaging modalities.[24]

Yamashita et al evaluated CH-EUS in 17 IPMN patients, which showed 100% sensitivity, 80% specificity, 92% positive predictive value (PPV), 100% negative predictive value, and 94% accuracy in distinguishing mural nodules from mucous clots.[24] A meta-analysis by Lisotti et al, covering eight studies with 320 patients, reported CH-EUS pooled sensitivity of 97.0%, specificity of 90.4, and diagnostic accuracy of 95.6% for identifying high-grade dysplasia (HGD) or invasive carcinoma (IC) in mural nodules.[25]

CH-EUS leverages harmonic imaging to detect microbubble signals from contrast agents, avoiding Doppler-related artifacts like blooming, which limits Doppler EUS's utility.[26] Despite its strengths, CH-EUS has limitations. Operator dependency can affect reproducibility, though good interobserver (K = 0.66) and excellent intraobserver agreement (K = 0.82–0.83) have been reported ([Fig. 1]).[26] [27]

Fig. 1 Contrast-enhanced endoscopic ultrasound (EUS) shows hyperechoic mural nodules.[103]

Detective Flow Imaging EUS

Another novel tool that is being studied is detective flow imaging EUS (DFI-EUS). It is an emerging, noncontrast imaging modality designed to detect low-velocity blood flow and fine vessels in pancreatic lesions, offering a promising alternative to CE-EUS. DFI-EUS uses a multidimensional filter to remove motion artifacts while preserving microvascular signals, making it suitable for patients with contrast contraindications.[25]

A prospective study by Yamashita et al showed DFI-EUS achieved a 91% vessel detection rate and distinguishing mural nodules from mucous clots in IPMNs, outperforming e-FLOW EUS (53%, p < 0.001).[28] For solid pancreatic lesions, a retrospective study of 104 patients reported 96% vessel visualization, 99% sensitivity, and 88% accuracy for pancreatic cancer.[29] Despite promising data with DFI-EUS, CH-EUS remains the standard due to its robust evidence base. The limited specificity and small PCN sample sizes necessitate further research.

EUS-Guided Needle-Based Confocal Laser Endomicroscopy

EUS-guided needle-based confocal laser endomicroscopy (EUS-nCLE) is an advanced diagnostic tool that offers real-time, high-resolution (1–3.5 μm) microscopic imaging of PCLs and solid masses. Utilizing a submillimeter miniprobe (e.g., Cellvizio AQ-Flex 19, Mauna Kea Technologies) that is passed through a 19-gauge EUS-FNA needle, nCLE delivers “optical biopsies” with up to 1000-fold magnification with intravenous fluorescein injection, enabling in vivo histopathological assessment. It overall enhances the diagnostic yield of EUS-FNA.[30]

A prospective study by Krishna et al involving 144 patients, with 65 undergoing surgical resection, demonstrated EUS-nCLE's superior performance in differentiating mucinous from nonmucinous PCLs, achieving 98% sensitivity, 94% specificity, and 97% accuracy compared with standard-of-care methods (carcinoembryonic antigen [CEA] and cytology: 74% sensitivity, 61% specificity, 71% accuracy; p < 0.001).[31] Specific nCLE patterns include finger-like papillary projections for IPMNs, superficial vascular networks for serous cystadenomas (SCAs), and trabecular patterns for PNETs.[32]

In a retrospective study of 30 patients with suspected PNETs, Yamada et al reported a 70% accuracy for nCLE, successfully diagnosing one of two cases with inconclusive EUS-FNA, though diagnostic yield was limited by stromal fibrosis.[33] For complications such as mild postprocedure pancreatitis, intracystic bleeding can be reduced by preloading the miniprobe and limiting needle dwell time to 10 minutes.[34] [35] Despite challenges like high costs and a learning curve, nCLE's integration with artificial intelligence (AI), as shown by Machicado et al, enhances risk stratification of IPMNs with high sensitivity and specificity. EUS-nCLE's ability to reduce unnecessary surgeries and surveillance underscores its potential, though further multicenter studies are needed for broader adoption ([Fig. 2]).

Fig. 2 Needle-based confocal laser endomicroscopy (nCLE) findings in various pancreatic cystic lesions (PCLs).[111] (A) Serous cystadenoma: superficial vascular network. (B) Intraductal papillary mucinous neoplasm (IPMN): multiple papillae (arrows) with epithelial border in dark gray. (C) Cystic neuroendocrine tumor (NEN) irregular clusters of tumoral cells (arrows).

EUS-Guided Sulfur Hexafluoride Pancreatography

EUS-guided sulfur hexafluoride (SF6) pancreatography (ESP) is an innovative technique to assess PCL communication with the PD, aiding differentiation of IPMNs from other cysts like MCNs and SCNs.

A retrospective study by Li et al evaluated ESP in 29 patients with PCLs, using 2 to 5 mL of diluted SF6 injected into the cyst postaspiration via a 19- or 22-gauge needle. ESP demonstrated 96.6% accuracy, 100% specificity, and PPV, for detecting duct communication, correctly identifying all eight IPMNs with positive pancreatography and 20 of 21 noncommunicating cysts. No ESP-related complications were reported, contrasting with endoscopic retrograde cholangiopancreatography's (ERCP's) higher pancreatitis risk.[36] [37] ESP's safety leverages SF6's established intravenous safety profile. Limitations include its retrospective single-center design and underrepresentation of certain PCLs (e.g., pseudocysts).[36] ESP offers a safer alternative to ERCP, but multicenter prospective studies are needed to validate its role in PCL diagnosis ([Fig. 3]).[36] [38]

Fig. 3 Sulfur hexafluoride (SF6) injection, flow from cyst to pancreatic duct (PD)—positive SF6 pancreaticography.[36]

EUS-Guided Tissue Acquisition

EUS-FNA and fine-needle biopsy (EUS-FNB) are essential tools for diagnosing PCLs, critical for distinguishing mucinous from nonmucinous cysts and assessing malignancy risk. EUS-FNA, the traditional standard, samples cyst fluid for cytology and biomarkers like CEA. Although simple and reliable, its diagnostic yield is hampered by low cellularity (31–49%) and inability to preserve histological architecture.[39] [40]

EUS-FNB, using Franseen or fork-tip needles, acquires intact tissue cores, enhancing histological evaluation and immunohistochemistry for high-risk markers like SMAD4. A retrospective study by Moussavi et al of 130 PCLs found EUS-FNB's diagnostic yield significantly higher than FNA (81% vs. 62%) with better IPMN grade identification (relative risk = 1.92).[41] EUS-FNB requires fewer passes (mean 2.3 vs. 3.0) and shorter procedure times, with comparable safety (1.3% adverse events). However, EUS-FNB's higher cost and need for specialized needles limit widespread adoption. While EUS-FNB excels in histological yield, its role in molecular profiling remains unclear, as FNA may yield better deoxyribonucleic acid for sequencing.[42] A systematic review of 18 randomized controlled trials by van Riet et al showed that EUS-FNB offers 87% diagnostic accuracy and an 80% tissue core sampling rate, superior to EUS-FNA's 80 and 62%, with fewer needle passes.[42] [43]

EUS-Through-the-Needle Biopsy

EUS-guided through-the-needle biopsy (EUS-TTNB) has emerged as a pivotal tool for diagnosing PCLs, offering superior diagnostic yield compared with traditional EUS-FNA. Studies report EUS-TTNB achieves a diagnostic yield of up to almost 80%, with high sensitivity and specificity for mucinous cysts, significantly outperforming EUS-FNA.[44] [45] [46] By sampling cyst wall tissue, TTNB provides histological and molecular insights, enabling accurate cyst subtyping (e.g., IPMNs, MCNs, SCAs) and risk stratification, which is critical for reducing unnecessary surgeries.[47] [48] A 2020 meta-analysis by Facciorusso et al reported EUS-FNB with TTNB achieved 85.3% sample adequacy, significantly outperforming FNA (p = 0.004).[49]

Notably, combining TTNB with next-generation sequencing (NGS) enhances diagnostic accuracy, identifying mutations like KRAS and GNAS for IPMNs with high sensitivity.[50] However, TTNB carries a higher adverse event rate (up to 11%) including intracystic bleeding and pancreatitis, compared with EUS-FNA.[46] [50] Patient selection is crucial, with indications focusing on indeterminate cysts where cyst type alters management.[48] Technical standardization and prophylactic measures require further optimization to improve safety.

A systematic review by Gopakumar and Puli, of 11 studies involving 575 patients found that EUS-TTNB offers high diagnostic performance for PCLs, achieving a pooled sensitivity of 76.6% and specificity of 98.9% for differentiating neoplastic from nonneoplastic cysts. It demonstrated a diagnostic odds ratio of 41.34, with positive and negative likelihood ratios of 10.28 and 0.26, respectively, confirming strong diagnostic reliability. However, it was associated with notable procedure-related risks, including pancreatitis (3%) and intracystic bleeding (4%).[51]

Other Diagnostic Advances

Tridimensional (3D) EUS enhances anatomical visualization within the pancreatobiliary region, improving interpretation of complex structures and vascular landmarks crucial for tumor staging and surgical planning. This technique enables detailed evaluation of venous invasion and compression in focal pancreatic masses, including those arising in chronic pancreatitis or pancreatic cancer. By capturing volumetric data, 3D-EUS allows retrospective reconstruction and multiplanar slicing, facilitating precise localization and characterization of lesions, even those missed during real-time imaging. While current results are promising, further technological refinements are needed to fully realize its diagnostic and staging potential.[52] [53]

Fluid Analysis

Cyst fluid analysis via EUS-FNA or EUS-TTNB provides biochemical, cytological, and molecular insights into PCL type and malignant potential.[54] Cytology via EUS-FNA sampling has 54 to 65% sensitivity but 88 to 93% specificity for identifying mucinous cysts, with high accuracy for malignant cysts.[55] [56] EUS-FNA cytology achieves 64.8% sensitivity and 90.6% specificity for benign versus malignant IPMNs.[56] Repeating EUS-FNA in surveillance can increase neoplasm diagnosis from 83% to almost 96%.[57]

Common biochemical markers that are tested include levels of CEA, amylase and glucose, and cytology, which enhance diagnosis. CEA levels are low in SCAs, higher in mucinous lesions, and markedly elevated in mucinous cystadenocarcinomas, improving diagnostic sensitivity. A higher level of CEA > 192 ng/mL showed up to 77% sensitivity and 96% specificity for mucinous cysts.[20] [55] [58] [59] Low glucose levels, especially < 50 mg/dL, showed 95% sensitivity and up to 96% specificity, often surpassing CEA in mucinous lesions.[55] [60] [61] [62] Amylase > 250 U/L is specific for pseudocysts but less useful for neoplastic cysts.[63] CA 19–9 < 37 U/mL indicates benign cysts, and vascular endothelial growth factor A > 5000 pg/mL with CEA > 10 ng/mL offers near 100% sensitivity for SCNs.[64]

Glucose measurement with EUS-guided aspirated pancreatic cyst fluid has emerged as a simple, cost-effective, and highly accurate biomarker to distinguish mucinous from nonmucinous cystic lesions. Several studies demonstrate that intracystic glucose levels are significantly lower in mucinous PCNs compared with nonmucinous cysts like SCAs or pseudocysts. In a large French multicenter study, a glucose threshold of < 41.8 mg/dL identified mucinous cysts with sensitivity of 95% and specificity of 91%, clearly outperforming CEA, which showed CEA > 192 ng/mL had a much lower overall accuracy of 67.6%.[65] Similarly, in a prospective Italian cohort, a cutoff of ≤ 58 mg/dL yielded an accuracy of 93.5%, surpassing CEA across thresholds, and providing greater diagnostic utility.[66]

A meta-analysis by Thornton et al reported EUS-FNA's sensitivity for mucinous cysts at 54% and specificity at 93%, with CEA >192 ng/mL yielding 63% sensitivity and 88% specificity.[67] Combining CEA with molecular markers (e.g., KRAS) improves sensitivity to 100%.[40] However, only 49% of FNA samples provide sufficient fluid for CEA analysis, and complications like pancreatitis occur.[39]

Beyond its accuracy, glucose testing is inexpensive and reproducible across laboratories, making it an attractive adjunct to cytology and imaging in EUS evaluation of pancreatic cysts. Its reliability supports consideration of glucose as a frontline fluid biomarker, complementing CEA in future clinical guidelines ([Tables 2] and [3]).

Table 2 Cyst types and fluid analysis[52] [109]

Cyst type	CEA	Amylase	Glucose	Viscosity	Mucin	Cytology
Serous cystadenoma	Low	Low (< 250 U/L)	High (> 50 mg/dL)	Low	Low	Negative or glycogen-containing cuboid cells
Mucinous cystic neoplasm	High	Low (< 250 U/L)	Low (< 50 mg/dL)	High	High	Mucin-containing columnar cells
IPMN	High	High	Low (< 50 mg/dL)	High	High	Papillary clusters of mucinous columnar cells, atypia
Solid pseudopapillary tumor	Low	Low	Variable	NA	NA	Branching papillae, cuboid/cylindric cells, myxoid stroma
Pseudocyst	Low	High	High (> 50 mg/dL)	Low	Low	Dirty material, macrophages, inflammatory cells

Abbreviations: CEA, carcinoembryonic antigen; IPMN, intraductal papillary mucinous neoplasm; NA, not available.

Table 3 Diagnostic modalities and specifications

Test	Sensitivity	Specificity	Accuracy
CE-EUS	97%	90%	95.6%
nCLE	98%	94%	97%
DFI-EUS	96%	99%	88%
EUS with cyst fluid	51%	94%	95%
1. CEA	56%	96%	85%
2. Glucose	91%	86%	94%
EUS – TTNB
1. Mucinous	87%	83%	–
2. Malignant PCL	97%	95%	–
NGS	90%	100%	–
Cytology	89%	98%	–
AI augmented	100%	99.7%	–

Abbreviations: AI, artificial intelligence; CE, contrast-enhanced; CEA, carcinoembryonic antigen; DFI, detective flow imaging; EUS, endoscopic ultrasound; nCLE, needle-based confocal laser endomicroscopy; NGS, next-generation sequencing; PCL, pancreatic cystic lesion; TTNB, through-the-needle biopsy.

Molecular and Genetic Analysis

Molecular analysis using NGS of pancreatic cyst fluid, obtained via EUS-guided FNA or TTNB, enhances diagnostic accuracy for PCLs. NGS identifies key genetic alterations, such as KRAS and GNAS mutations, which are highly specific for mucinous cysts like IPMNs and MCNs, with a sensitivity of 94.7% and specificity near 100%.[68] Additionally, mutations in TP53, SMAD4, or PIK3CA indicate HGD or PDAC, guiding decisions on surveillance versus surgical resection.[68] [69] Tests like PancreaSeq, a 22-gene NGS panel, have demonstrated superior performance over traditional markers like CEA, correctly classifying cysts in 48% of cases with nonelevated CEA and detecting advanced neoplasia with 93% sensitivity.[69] [70] This molecular approach enhances early detection and personalized management, reducing unnecessary surgeries for benign cysts.

Artificial Intelligence Integration

Conventional EUS with FNA achieves diagnostic accuracy of 65 to 75% for mucinous PCLs, limited by interobserver variability and sampling errors. AI, particularly deep learning (DL) and convolutional neural networks (CNNs), significantly improves accuracy, with studies reporting up to 98.5% for differentiating mucinous from nonmucinous cysts.[4] [71]

A 2023 review highlights AI's ability to analyze EUS images, reducing operator dependency and enhancing lesion detection.[1] Similarly, a 2024 study demonstrated a CNN model with 82.9 to 85.7% accuracy depending on the algorithm used, in detecting HGD in IPMNs using EUS-guided nCLE.[34] [72]

Automatic pancreatic segmentation efforts using various machine language models include Tang and colleague's use of UNet + +-based model for pancreatic mass segmentation (92.3% accuracy), Machicado et al's mask region-based CNN model for EUS video segmentation, and Zhong et al's UNet + + application on EUS videos (Dice score 83.6[34] [73]). Oh et al specifically targeted PCL segmentation with an Attention U-net model, achieving a Dice score of 0.794 and high pixel accuracy (0.983[74]).

EUS-based classification studies are more prevalent, focusing on differentiating benign from malignant cysts. Kuwahara et al used a ResNet-50 model to classify IPMNs, achieving patient-level accuracy of 94.0% in patient level predictions.[75] Machicado et al employed a three-stage CNN-based approach with VGG-16 for identifying HGD in IPMNs, reaching an area under the curve (AUC) of 87.3%.[34] Nguon et al developed a ResNet-50-based model to distinguish MCNs from SCNs, with the ResNet-Conv + FC method yielding an AUC of 0.88.[76] Vilas-Boas et al achieved an AUC of 1.0 for mucinous versus nonmucinous cyst classification using a Xception model, while Schulz et al reported a 99.6% accuracy in predicting IPMN histologic outcomes.[77] [78]

EUS-AI also enhances procedural efficiency by optimizing FNA/FNB sampling and reducing complications through precise lesion localization.[74] [79] It shortens the learning curve for novice endoscopists, improving EUS accessibility.[80] However, challenges of these studies include small sample sizes, lack of standardized data sets, and high nCLE costs, limiting adoption.[71] [72] [81] The “black box” nature of DL/machine learning (ML) where there is only access to input and output layers but not the process and operation layer raises ethical concerns and room for potential biases, errors, and unintended consequences necessitating explainable AI tools like Grad-CAM (Gradient-weighted Class Activation Mapping). These advancements demonstrate high diagnostic potential with increased sensitivity, specificity, and accuracy, but the challenge in AI-enhanced EUS lies in real-time image, “black box” problem, risk of data breaches, and responsibility and risk of misdiagnosis.[81]

Therapeutic Advances

It is well established that EUS-guided management of PCLs offers a minimally invasive alternative to surgery.

EUS-guided chemoablation has advanced significantly over the past two decades. Initial ethanol lavage studies reported a 33% complete resolution rate but noted adverse events like pancreatitis.[82] Combining ethanol with paclitaxel improved outcomes, achieving 62% resolution.[83] The ChARM trial demonstrated that alcohol-free ablation (paclitaxel + gemcitabine) was noninferior to alcohol-based methods with 74% versus 62.5% resolution at 3 months and 67% versus 61% resolution at 12 months with fewer complications (0% vs. 6%).[84] [85] A novel microparticle paclitaxel enhances efficacy by sustaining drug concentrations, making chemoablation a preferred approach for mucinous cysts with volume and surface area reduction, morphological changes, and loss of pathogenic mutations.[86]

EUS-guided cryothermal ablation involves use of hybrid bipolar probe that uses thermal energy of radiofrequency ablation (RFA) and cryogenic gas, this has been an emerging technique for high-risk PCLs unsuitable for chemoablation.[87] Barthet et al reported a 50% resolution rate in small cohorts, but pancreatitis occurred in 10% of cases.[88] RFA targets neoplastic tissue with thermal energy, offering potential for cysts with malignant features. However, its higher complication rate and limited long-term data necessitate careful patient selection.[13] Large human studies are needed to determine the role, efficacy, and safety of cryothermal ablation of pancreatic cysts given its potential benefits and safety profile as well as a multidisciplinary evaluation at high-volume centers is critical to balance efficacy and safety.[20]

EUS-guided lauromacrogol ablation uses lauromacrogol, a sclerosing agent, which is gaining attention for PCL ablation. Linghu et al demonstrated a 37.9 and 36.4% resolution rate with lauromacrogol in the pancreatic head/uncinate and in the body/tail, respectively, with adverse events (e.g., fever, mild pancreatitis) in < 5% of cases.[89] Its mechanism involves inducing fibrosis to collapse cysts, offering a cost-effective alternative to chemoablation. However, data remain limited, and larger trials are needed to establish its role relative to established methods.

Laser ablation with neodymium:yttrium aluminum (Nd:YAG) uses light with infrared spectrum at energy output sufficient for the induction of tissue necrosis. Di Matteo et al showed feasibility and promising results in a prospective animal study, but human studies are needed to validate and confirm these findings.[90] Karaca et al studied a porcine model for feasibility and safety of EUS-guided drug-eluting beads into the pancreas and was noted to be feasible and safe. As with the Nd:YAG laser therapy, outcomes and long-term effects on humans are unknown.[91]

Finally, postablation surveillance remains unstandardized, contributing to patient burden. Recent data suggest reduced imaging intervals for low-risk cysts postablation, with annual EUS or MRI proposed for stable cases.[92] Elta et al noted that current biomarkers (e.g., CEA) lack specificity, complicating risk stratification.[14] Molecular profiling of cyst fluid could enable tailored surveillance, reducing costs and patient anxiety while maintaining safety.[93] Postablation, reduced radiographic surveillance protocols (e.g., ChARM protocol) have been evaluated for cost-effectiveness and patient outcomes. These protocols minimize the need for frequent MRI/CT, reducing economic and psychological burdens while maintaining efficacy.[94]

Guidelines for Pancreatic Cyst Management

The updated Kyoto 2024 International Evidence-Based Guidelines elevate the role of EUS as a central diagnostic and risk-stratifying modality in the management of IPMNs. While acknowledging that EUS availability and operator expertise vary by institution, the guidelines explicitly integrate EUS, CE-EUS, and EUS-FNA findings into the definitions of high-risk stigmata (HRS) and worrisome features (WF). EUS is recommended for lesions suspicious for HGD or IC, particularly to confirm the presence and size of mural nodules and to obtain cyst fluid or tissue for cytological and molecular analysis.

The guidelines reaffirm enhancing mural nodule ≥ 5 mm and MPD ≥ 10 mm as HRS, with smaller mural nodules and MPD dilation (5–9 mm) categorized as WF. Cytology “suspicious” or “positive” for HGD/IC is itself now considered a HRS, underscoring the diagnostic value of EUS-FNA when performed safely. Additionally, the guidelines endorse molecular profiling of cyst fluid with KRAS/GNAS mutations for mucinous differentiation and TP53, SMAD4, CDKN2A, and PIK3CA alterations as markers of advanced neoplasia, thereby highlighting EUS-FNA as a gateway for precision diagnostics. Overall, the Kyoto update positions EUS as an indispensable, though operator-dependent, tool that bridges morphological imaging with cytologic and molecular characterization, guiding both surgical and surveillance decisions in IPMN management.[95]

The Fukuoka (2017) and the newer revised Kyoto (2024) guidelines, both developed under the International Association of Pancreatology, share a foundation in risk stratification using HRS and WF. The Kyoto revision builds upon the Fukuoka guidelines by incorporating EUS and cytology results into HRS/WF assessment, simplifying surveillance algorithms, and introducing molecular profiling for precision diagnostics. Kyoto uniquely allows for discontinuation of surveillance for BD-IPMNs < 20 mm that remain unchanged for 5 years, reflecting an evidence-based approach balancing risk and health care burden.[94] [95]

In contrast, the U.S.-based American Gastroenterological Association (AGA) (2015) and American College of Gastroenterology (ACG) (2018) guidelines are more conservative in surgical indications and emphasize patient comorbidity and shared decision making. The AGA advocates EUS-FNA only for cysts with ≥ 2 high-risk features (size ≥ 3 cm, mural nodule, or MPD dilation) and suggests stopping surveillance after 5 years if stable. The ACG provides a more granular framework, recommending MRI/magnetic resonance cholangiopancreatography as first-line imaging and EUS-FNA when results would alter management, while cautioning against overuse of invasive testing and surgery due to morbidity ([Table 4]).[14] [92]

Table 4 Comparison of EUS indications by guidelines

Guidelines	High-risk features		EUS indication
	High-risk stigmata (HRS)	Worrisome features (WF)
Kyoto 2024[95]	● Obstructive jaundice ● Enhancing mural nodule ≥ 5 mm ● MPD ≥ 10 mm ● Cytology	● Pancreatitis ● Cyst ≥ 3 cm ● Enhancing mural nodule < 5 mm ● Thickened cyst wall ● MPD 5–9 mm ● Abrupt PD caliber change ● Lymphadenopathy ● Increased CA 19–9 ● Cyst growth ≥ 5 mm / 2 years ● New onset/worsening diabetes	MRI is preferred Use MDCT and EUS if changes on MRI
Fukuoka 2017[110]	● Obstructive jaundice ● Enhancing mural nodule ≥ 5 mm ● MPD ≥ 10 mm	● Pancreatitis ● Cyst ≥ 3 cm ● Enhancing mural nodule < 5 mm ● Thickened cyst wall ● MPD 5–9 mm ● Abrupt PD caliber change ● Lymphadenopathy ● Increased CA 19–9  Cyst growth ≥ 5 mm / 2 years	Presence of WF Surveillance: 2–3 cm cyst – EUS 3–6 months → MRI/EUS yearly 3 cm cyst – MRI/EUS 3–6 monthly
ACG 2018[14]	Obstructive jaundice Associated solid mass MPD ≥ 5 mm Cyst ≥ 3 cm Change in MPD caliber / Upstream dilation		Diagnostic: - Presence of high-risk features - History of pancreatitis, cyst as cause - 2–3 cm cyst, diagnostic EUS, MRI or EUS 6–12 monthly Surveillance: - Increase in cyst size - 2–3 cm cyst - MRI or EUS q6–12 monthly for 3 y ->3 cm – MRI/EUS alternating 6 monthly for 3 years
AGA 2015[92]	Size ≥ 3 cm Dilated MPD Solid component		Presence/development of high-risk features MRI for surveillance

Abbreviations: ACG, American College of Gastroenterology; AGA, American Gastroenterological Association; EUS, endoscopic ultrasound; MPD, main pancreatic duct; MRI, magnetic resonance imaging; PD, pancreatic duct; MDCT, multidetector computed tomography.

Challenges and Future Directions

Despite its utility, EUS faces significant challenges in accurately diagnosing and managing PCLs. The diagnostic accuracy of EUS morphology for distinguishing mucinous from nonmucinous cysts ranges from 48 to 94%, with sensitivity of 48 to 95% and specificity of 83 to 100%, but lower sensitivity of 27 to 48% in nonmucinous cysts. These wide ranges in accuracy and sensitivity are influenced by operator experience and cyst characteristics.[96] EUS-FNA improves diagnostic yield but often has nondiagnostic results due to scant cellular material. The procedure carries risks, including infection (0.4% with antibiotics), bleeding, and pancreatitis, necessitating cautious patient selection.[96] [97]

Cyst fluid analysis, particularly CEA levels (> 192 ng/mL for mucinous cysts) and glucose, enhances diagnostic precision. However, variability in fluid markers and the need for molecular testing, such as KRAS and GNAS mutations, complicate accurate diagnosis.[98] Differentiating mural nodules from mucin balls remains challenging, often requiring advanced techniques like through-the-needle forceps or CE-EUS. Guidelines, such as those from the European Study Group, advocate EUS-FNA for cysts with high-risk features, but its low sensitivity (27–48%) for mucinous cysts risks misdiagnosis.[96] Emerging AI applications promise to enhance EUS accuracy by improving image analysis and risk stratification, potentially reducing unnecessary interventions. Future advancements, including EUS-guided elastography and NGS, may address these limitations, but standardized protocols and larger cohort studies are essential to optimize PCL management.[96] [99]

EUS-guided ablation, including radiofrequency and ethanol injection, offers a minimally invasive approach for managing high-risk PCLs; however, long-term outcomes, including recurrence and survival, remain underexplored. Complications, such as pancreatitis and bleeding, underscore the need for rigorous validation.[100] Multicenter trials are essential to establish standardized outcome metrics—cyst regression, malignancy-free intervals—and compare ablation with surgery, particularly for nonsurgical candidates.

Use of AI, particularly DL models like CNNs, enhances real-time EUS, achieving 96% accuracy in distinguishing malignant from benign PCLs. Future integration will enable real-time lesion characterization and biopsy guidance, minimizing unnecessary procedures. Challenges include nonstandardized data and opaque AI algorithms, limiting generalizability. Transparent, multicenter validation studies and standardized protocols for developing AI models for real-time assistance with EUS procedures are critical to ensure consistent diagnostic performance across diverse populations, fostering smooth clinical adoption and integration.[79]

Integration of the novel EUS-guided diagnostic tools such as the CH-EUS and nCLE, with the AI-driven models could enhance malignancy prediction. Future research should combine multiomics data and advanced imaging to develop personalized management, reducing overtreatment of low-risk PCLs while prioritizing high-risk lesions.[100] [101]

Conclusion

Recent advances in EUS for PCLs include enhanced imaging (CE-EUS, nCLE), AI integration, and improved tissue/fluid acquisition techniques (EUS-FNA, EUS-TTNB), which refine diagnostic accuracy and risk stratification. Therapeutically, EUS-guided chemoablation and RFA offer minimally invasive options for high-risk patients, with alcohol-free protocols improving safety and efficacy with decreased recurrence and need for surveillance reducing patient and health care burden. Ongoing trials and AI advancements promise further improvements in personalized management of PCLs.

Conflict of Interest

None declared.

• References

• 1 Patel N, Asafo-Agyei KO, Osueni A, Mukherjee S. Pancreatic Cysts. In: StatPearls. Treasure Island, FL: StatPearls Publishing; 2023
• 2 Moris M, Bridges MD, Pooley RA. et al. Association between advances in high-resolution cross-section imaging technologies and increase in prevalence of pancreatic cysts from 2005 to 2014. Clin Gastroenterol Hepatol 2016; 14 (04) 585-593.e3
• 3 Singh RR, Gopakumar H, Sharma NR. Diagnosis and management of pancreatic cysts: a comprehensive review of the literature. Diagnostics (Basel) 2023; 13 (03) 13
• 4 Seyithanoglu D, Durak G, Keles E. et al. Advances for managing pancreatic cystic lesions: integrating imaging and AI innovations. Cancers (Basel) 2024; 16 (24) 16
• 5 Lévy P, Jouannaud V, O'Toole D. et al. Natural history of intraductal papillary mucinous tumors of the pancreas: actuarial risk of malignancy. Clin Gastroenterol Hepatol 2006; 4 (04) 460-468
• 6 Zerboni G, Signoretti M, Crippa S, Falconi M, Arcidiacono PG, Capurso G. Systematic review and meta-analysis: prevalence of incidentally detected pancreatic cystic lesions in asymptomatic individuals. Pancreatology 2019; 19 (01) 2-9
• 7 Yamao K, Yanagisawa A, Takahashi K. et al. Clinicopathological features and prognosis of mucinous cystic neoplasm with ovarian-type stroma: a multi-institutional study of the Japan Pancreas Society. Pancreas 2011; 40 (01) 67-71
• 8 Crippa S, Salvia R, Warshaw AL. et al. Mucinous cystic neoplasm of the pancreas is not an aggressive entity: lessons from 163 resected patients. Ann Surg 2008; 247 (04) 571-579
• 9 Caglià P, Cannizzaro MT, Tracia A. et al. Cystic pancreatic neuroendocrine tumors: to date a diagnostic challenge. Int J Surg 2015; 21 (Suppl. 01) S44-S49
• 10 Xu Y, Zhao G, Pu N. et al. One hundred twenty-one resected solid pseudopapillary tumors of the pancreas: an 8-year single-institution experience at Zhongshan Hospital, Shanghai, China. Pancreas 2017; 46 (08) 1023-1028
• 11 Park JK, Cho EJ, Ryu JK, Kim Y-T, Yoon Y-B. Natural history and malignant risk factors of solid pseudopapillary tumors of the pancreas. Postgrad Med 2013; 125 (02) 92-99
• 12 Gardner TB, Park WG, Allen PJ. Diagnosis and management of pancreatic cysts. Gastroenterology 2024; 167 (03) 454-468
• 13 Tanaka M, Fernández-del Castillo C, Adsay V. et al; International Association of Pancreatology. International consensus guidelines 2012 for the management of IPMN and MCN of the pancreas. Pancreatology 2012; 12 (03) 183-197
• 14 Elta GH, Enestvedt BK, Sauer BG, Lennon AM. ACG clinical guideline: diagnosis and management of pancreatic cysts. Am J Gastroenterol 2018; 113 (04) 464-479
• 15 Ohno E, Kuzuya T, Kawabe N. et al. Current status of endoscopic ultrasound in the diagnosis of intraductal papillary mucinous neoplasms. DEN Open 2024; 5 (01) e413
• 16 Yamashita Y, Yamazaki H, Nakahata A. et al. Advances in endoscopic ultrasonography-based diagnosis of pancreatic lesions: narrative review. Cancers (Basel) 2025; 17 (02) 17
• 17 Iwashita T, Uemura S, Shimizu M. Endoscopic ultrasound-guided fine-needle aspiration for pancreatic cystic lesions: a comprehensive review. J Med Ultrason 2024; 51 (02) 219-226
• 18 Kamata K, Kitano M, Kudo M. et al. Value of EUS in early detection of pancreatic ductal adenocarcinomas in patients with intraductal papillary mucinous neoplasms. Endoscopy 2014; 46 (01) 22-29
• 19 Khashab MA, Kim K, Lennon AM. et al. Should we do EUS/FNA on patients with pancreatic cysts? The incremental diagnostic yield of EUS over CT/MRI for prediction of cystic neoplasms. Pancreas 2013; 42 (04) 717-721
• 20 Brugge WR, Lewandrowski K, Lee-Lewandrowski E. et al. Diagnosis of pancreatic cystic neoplasms: a report of the cooperative pancreatic cyst study. Gastroenterology 2004; 126 (05) 1330-1336
• 21 Ahmad NA, Kochman ML, Brensinger C. et al. Interobserver agreement among endosonographers for the diagnosis of neoplastic versus non-neoplastic pancreatic cystic lesions. Gastrointest Endosc 2003; 58 (01) 59-64
• 22 Spadaccini M, Koleth G, Emmanuel J. et al. Enhanced endoscopic ultrasound imaging for pancreatic lesions: the road to artificial intelligence. World J Gastroenterol 2022; 28 (29) 3814-3824
• 23 Serrani M, Lisotti A, Caletti G, Fusaroli P. Role of contrast harmonic-endoscopic ultrasound in pancreatic cystic lesions. Endosc Ultrasound 2017; 6 (01) 25-30
• 24 Yamashita Y, Ueda K, Itonaga M. et al. Usefulness of contrast-enhanced endoscopic sonography for discriminating mural nodules from mucous clots in intraductal papillary mucinous neoplasms: a single-center prospective study. J Ultrasound Med 2013; 32 (01) 61-68
• 25 Lisotti A, Napoleon B, Facciorusso A. et al. Contrast-enhanced EUS for the characterization of mural nodules within pancreatic cystic neoplasms: systematic review and meta-analysis. Gastrointest Endosc 2021; 94 (05) 881-889.e5
• 26 Ahmad SA, Pryor AD. Shackelford's Surgery of the Alimentary Tract. Elsevier Health Sciences; Philadelphia, PA, USA: 2025
• 27 Palazzo M, Napoléon B, Gincul R. et al. Contrast harmonic EUS for the prediction of pancreatic neuroendocrine tumor aggressiveness (with videos). Gastrointest Endosc 2018; 87 (06) 1481-1488
• 28 Yamashita Y, Yoshikawa T, Yamazaki H. et al. A novel endoscopic ultrasonography imaging technique for depicting microcirculation in pancreatobiliary lesions without the need for contrast-enhancement: a prospective exploratory study. Diagnostics (Basel) 2021; 11 (11) 11
• 29 Miwa H, Sugimori K, Yonei S. et al. Differential diagnosis of solid pancreatic lesions using detective flow imaging endoscopic ultrasonography. Diagnostics (Basel) 2024; 14 (09) 14
• 30 Wu CCH, Lim SJM, Tan DMY. Endoscopic ultrasound-guided needle-based confocal laser endomicroscopy for pancreatic cystic lesions: current status and future prospects. Clin Endosc 2024; 57 (04) 434-445
• 31 Krishna SG, Hart PA, Malli A. et al. Endoscopic ultrasound-guided confocal laser endomicroscopy increases accuracy of differentiation of pancreatic cystic lesions. Clin Gastroenterol Hepatol 2020; 18 (02) 432-440.e6
• 32 Bhutani MS, Koduru P, Joshi V. et al. EUS-guided needle-based confocal laser endomicroscopy: a novel technique with emerging applications. Gastroenterol Hepatol (N Y) 2015; 11 (04) 235-240
• 33 Yamada M, Hara K, Mizuno N. et al. The role of needle-based confocal laser endomicroscopy in the diagnosis of pancreatic neuroendocrine tumors. Clin Endosc 2024; 57 (03) 393-401
• 34 Machicado JD, Chao W-L, Carlyn DE. et al. High performance in risk stratification of intraductal papillary mucinous neoplasms by confocal laser endomicroscopy image analysis with convolutional neural networks (with video). Gastrointest Endosc 2021; 94 (01) 78-87.e2
• 35 Konda VJA, Meining A, Jamil LH. et al. A pilot study of in vivo identification of pancreatic cystic neoplasms with needle-based confocal laser endomicroscopy under endosonographic guidance. Endoscopy 2013; 45 (12) 1006-1013
• 36 Li H, Feng X, Gao F, Chen Q, Linghu E. Diagnostic value of EUS-guided SF6 pancreatography for pancreatic cystic lesions on cyst communication with the pancreatic duct. Endosc Ultrasound 2023; 12 (02) 245-250
• 37 Singh S, Chandan S, Vinayek R. et al. Endoscopic techniques for the diagnosis of pancreatic cystic lesions. World J Gastroenterol 2025; 31 (01) 101082
• 38 European Study Group on Cystic Tumours of the Pancreas. European evidence-based guidelines on pancreatic cystic neoplasms. Gut 2018; 67 (05) 789-804
• 39 Hawes RH, Clancy J, Hasan MK. Endoscopic ultrasound-guided fine needle aspiration in cystic pancreatic lesions. Clin Endosc 2012; 45 (02) 128-131
• 40 Pan S, Liu J, Guo J, Zhu Q, Wang L, Shi X. The clinical impact of endoscopic ultrasound-guided fine-needle aspiration on the patients with low-risk pancreatic cystic lesions. Front Oncol 2022; 12: 961293
• 41 Moussavi GH, Gomez MF, Permuth J. et al. Endoscopic ultrasound–guided fine-needle biopsy as a novel approach for risk stratification of pancreatic cystic lesions. iGIE 2025; 4: 42-47
• 42 Chen Y-I, Chatterjee A, Berger R. et al. Endoscopic ultrasound (EUS)-guided fine needle biopsy alone vs. EUS-guided fine needle aspiration with rapid onsite evaluation in pancreatic lesions: a multicenter randomized trial. Endoscopy 2022; 54 (01) 4-12
• 43 van Riet PA, Erler NS, Bruno MJ, Cahen DL. Comparison of fine-needle aspiration and fine-needle biopsy devices for endoscopic ultrasound-guided sampling of solid lesions: a systemic review and meta-analysis. Endoscopy 2021; 53 (04) 411-423
• 44 Vilas-Boas F, Costa-Moreira P, Moutinho-Ribeiro P, Martins D, Lopes J, Macedo G. Through-the-needle biopsy sampling may allow preoperative intraductal papillary mucinous neoplasia subtyping. Gastrointest Endosc 2020; 92 (01) 235-236
• 45 Tacelli M, Celsa C, Magro B. et al. Diagnostic performance of endoscopic ultrasound through-the-needle microforceps biopsy of pancreatic cystic lesions: systematic review with meta-analysis. Dig Endosc 2020; 32 (07) 1018-1030
• 46 McCarty T, Rustagi T. Endoscopic ultrasound-guided through-the-needle microforceps biopsy improves diagnostic yield for pancreatic cystic lesions: a systematic review and meta-analysis. Endosc Int Open 2020; 8 (10) E1280-E1290
• 47 Vilas-Boas F, Ribeiro T, Macedo G. et al. Endoscopic ultrasound-guided through-the-needle biopsy: a narrative review of the technique and its emerging role in pancreatic cyst diagnosis. Diagnostics (Basel) 2024; 14 (15) 14
• 48 Kovacevic B, Klausen P, Rift CV. et al. Clinical impact of endoscopic ultrasound-guided through-the-needle microbiopsy in patients with pancreatic cysts. Endoscopy 2021; 53 (01) 44-52
• 49 Facciorusso A, Del Prete V, Buccino VR, Purohit P, Setia P, Muscatiello N. Diagnostic yield of Franseen and Fork-Tip biopsy needles for endoscopic ultrasound-guided tissue acquisition: a meta-analysis. Endosc Int Open 2019; 7 (10) E1221-E1230
• 50 Rift CV, Melchior LC, Kovacevic B. et al. Targeted next-generation sequencing of EUS-guided through-the-needle-biopsy sampling from pancreatic cystic lesions. Gastrointest Endosc 2023; 97 (01) 50-58.e4
• 51 Gopakumar H, Puli SR. Value of endoscopic ultrasound-guided through-the-needle biopsy in pancreatic cystic lesions. a systematic review and meta-analysis. J Gastrointest Cancer 2024; 55 (01) 15-25
• 52 Sakamoto H, Kitano M, Kamata K, El-Masry M, Kudo M. Diagnosis of pancreatic tumors by endoscopic ultrasonography. World J Radiol 2010; 2 (04) 122-134
• 53 Saftoiu A, Gheonea DI. Tridimensional (3D) endoscopic ultrasound - a pictorial review. J Gastrointestin Liver Dis 2009; 18 (04) 501-505
• 54 Cocieru A, Brandwein S, Saldinger PF. The role of endoscopic ultrasound and cyst fluid analysis in the initial evaluation and follow-up of incidental pancreatic cystic lesions. HPB (Oxford) 2011; 13 (07) 459-462
• 55 McCarty TR, Garg R, Rustagi T. Pancreatic cyst fluid glucose in differentiating mucinous from nonmucinous pancreatic cysts: a systematic review and meta-analysis. Gastrointest Endosc 2021; 94 (04) 698-712.e6
• 56 Serinelli S, Khurana KK. Intraductal papillary mucinous neoplasms of the pancreas: cytologic-histologic correlation study and evaluation of the cytologic accuracy in identifying high-grade dysplasia/invasive adenocarcinoma. Cytojournal 2024; 21: 6
• 57 Kim EY. Role of repeated endoscopic ultrasound-guided fine needle aspiration for inconclusive initial cytology result. Clin Endosc 2013; 46 (05) 540-542
• 58 Nagashio Y, Hijioka S, Mizuno N. et al. Combination of cyst fluid CEA and CA 125 is an accurate diagnostic tool for differentiating mucinous cystic neoplasms from intraductal papillary mucinous neoplasms. Pancreatology 2014; 14 (06) 503-509
• 59 Cizginer S, Turner BG, Bilge AR, Karaca C, Pitman MB, Brugge WR. Cyst fluid carcinoembryonic antigen is an accurate diagnostic marker of pancreatic mucinous cysts. Pancreas 2011; 40 (07) 1024-1028
• 60 Faias S, Pereira L, Roque R. et al. Excellent accuracy of glucose level in cystic fluid for diagnosis of pancreatic mucinous cysts. Dig Dis Sci 2020; 65 (07) 2071-2078
• 61 Zikos T, Pham K, Bowen R. et al. Cyst fluid glucose is rapidly feasible and accurate in diagnosing mucinous pancreatic cysts. Am J Gastroenterol 2015; 110 (06) 909-914
• 62 Smith ZL, Satyavada S, Simons-Linares R. et al. Intracystic glucose and carcinoembryonic antigen in differentiating histologically confirmed pancreatic mucinous neoplastic cysts. Am J Gastroenterol 2022; 117 (03) 478-485
• 63 ASGE Standards of Practice Committee. Muthusamy VR, Chandrasekhara V, Acosta RD, Bruining DH, Chathadi KV, Eloubeidi MA, Faulx AL, Fonkalsrud L, Gurudu SR, Khashab MA, Kothari S, Lightdale JR, Pasha SF, Saltzman JR, Shaukat A, Wang A, Yang J, Cash BD, DeWitt JM. The role of endoscopy in the diagnosis and treatment of cystic pancreatic neoplasms. Gastrointest Endosc 2016; 84: 1-9
• 64 Carr RA, Yip-Schneider MT, Dolejs S. et al. Pancreatic cyst fluid vascular endothelial growth factor a and carcinoembryonic antigen: a highly accurate test for the diagnosis of serous cystic neoplasm. J Am Coll Surg 2017; 225 (01) 93-100
• 65 Williet N, Caillol F, Karsenti D. et al. The level of glucose in pancreatic cyst fluid is more accurate than carcinoembryonic antigen to identify mucinous tumors: a French multicenter study. Endosc Ultrasound 2023; 12 (04) 377-381
• 66 Rossi G, Petrone MC, Tacelli M. et al. Glucose and lactate levels are lower in EUS-aspirated cyst fluid of mucinous vs non-mucinous pancreatic cystic lesions. Dig Liver Dis 2024; 56 (05) 836-840
• 67 Thornton GD, McPhail MJW, Nayagam S, Hewitt MJ, Vlavianos P, Monahan KJ. Endoscopic ultrasound guided fine needle aspiration for the diagnosis of pancreatic cystic neoplasms: a meta-analysis. Pancreatology 2013; 13 (01) 48-57
• 68 Springer S, Masica DL, Dal Molin M. et al. A multimodality test to guide the management of patients with a pancreatic cyst. Sci Transl Med 2019; 11 (501) 11
• 69 Singhi AD, McGrath K, Brand RE. et al. Preoperative next-generation sequencing of pancreatic cyst fluid is highly accurate in cyst classification and detection of advanced neoplasia. Gut 2018; 67 (12) 2131-2141
• 70 Paniccia A, Polanco PM, Boone BA. et al. Prospective, multi-institutional, real-time next-generation sequencing of pancreatic cyst fluid reveals diverse genomic alterations that improve the clinical management of pancreatic cysts. Gastroenterology 2023; 164 (01) 117-133.e7
• 71 Rangwani S, Ardeshna DR, Rodgers B. et al. Application of artificial intelligence in the management of pancreatic cystic lesions. Biomimetics (Basel) 2022; 7 (02) 7
• 72 Jiang J, Chao W-L, Cao T. et al. Improving pancreatic cyst management: artificial intelligence-powered prediction of advanced neoplasms through endoscopic ultrasound-guided confocal endomicroscopy. Biomimetics (Basel) 2023; 8 (06) 8
• 73 Zhong L, Chai N, Linghu E, Li H, Yang J, Tang P. A prospective study on contrast-enhanced endoscopic ultrasound for differential diagnosis of pancreatic cystic neoplasms. Dig Dis Sci 2019; 64 (12) 3616-3622
• 74 Oh S, Kim Y-J, Park Y-T, Kim K-G. Automatic pancreatic cyst lesion segmentation on EUS Images using a deep-learning approach. Sensors (Basel) 2021; 22 (01) 22
• 75 Kuwahara T, Hara K, Mizuno N. et al. Usefulness of deep learning analysis for the diagnosis of malignancy in intraductal papillary mucinous neoplasms of the pancreas. Clin Transl Gastroenterol 2019; 10 (05) 1-8
• 76 Nguon LS, Seo K, Lim J-H. et al. Deep learning-based differentiation between mucinous cystic neoplasm and serous cystic neoplasm in the pancreas using endoscopic ultrasonography. Diagnostics (Basel) 2021; 11 (06) 11
• 77 Vilas-Boas F, Ribeiro T, Afonso J. et al. Deep learning for automatic differentiation of mucinous versus non-mucinous pancreatic cystic lesions: a pilot study. Diagnostics (Basel) 2022; 12 (09) 12
• 78 Schulz D, Heilmaier M, Phillip V. et al. Accurate prediction of histological grading of intraductal papillary mucinous neoplasia using deep learning. Endoscopy 2023; 55 (05) 415-422
• 79 Tacelli M, Lauri G, Tabacelia D, Tieranu CG, Arcidiacono PG, Săftoiu A. Integrating artificial intelligence with endoscopic ultrasound in the early detection of bilio-pancreatic lesions: current advances and future prospects. Best Pract Res Clin Gastroenterol 2025; 74: 101975
• 80 Lv B, Wang K, Wei N, Yu F, Tao T, Shi Y. Diagnostic value of deep learning-assisted endoscopic ultrasound for pancreatic tumors: a systematic review and meta-analysis. Front Oncol 2023; 13: 1191008
• 81 Huang J, Fan X, Liu W. Applications and prospects of artificial intelligence-assisted endoscopic ultrasound in digestive system diseases. Diagnostics (Basel) 2023; 13 (17) 13
• 82 Gan SI, Thompson CC, Lauwers GY, Bounds BC, Brugge WR. Ethanol lavage of pancreatic cystic lesions: initial pilot study. Gastrointest Endosc 2005; 61 (06) 746-752
• 83 Oh H-C, Seo DW, Song TJ. et al. Endoscopic ultrasonography-guided ethanol lavage with paclitaxel injection treats patients with pancreatic cysts. Gastroenterology 2011; 140 (01) 172-179
• 84 Moyer M, Sharzehi S, Dye C, El-Deiry W, Mathew A, Gusani N. The CHARM Trial (CHemotherapy for Ablation and Resolution of Mucinous Pancreatic Cysts): the initial pilot study for the prospective, randomized, double-blind, single-center clinical trial: 294. Off J Am Coll Gastroenterol 2014; 109: S89
• 85 Moyer MT, Sharzehi S, Mathew A. et al. The safety and efficacy of an alcohol-free pancreatic cyst ablation protocol. Gastroenterology 2017; 153 (05) 1295-1303
• 86 Krishna SG, Ardeshna DR, Shah ZK. et al. Intracystic injection of large surface area microparticle paclitaxel for chemoablation of intraductal papillary mucinous neoplasms: Insights from an expanded access protocol. Pancreatology 2024; 24 (02) 289-297
• 87 Ofosu A, Ramai D, Adler DG. Endoscopic ultrasound-guided ablation of pancreatic cystic neoplasms: ready for prime time?. Ann Gastroenterol 2019; 32 (01) 39-45
• 88 Barthet M, Giovannini M, Lesavre N. et al. Endoscopic ultrasound-guided radiofrequency ablation for pancreatic neuroendocrine tumors and pancreatic cystic neoplasms: a prospective multicenter study. Endoscopy 2019; 51 (09) 836-842
• 89 Linghu E, Du C, Chai N. et al. A prospective study on the safety and effectiveness of using lauromacrogol for ablation of pancreatic cystic neoplasms with the aid of EUS. Gastrointest Endosc 2017; 86 (05) 872-880
• 90 Di Matteo F, Grasso R, Pacella CM. et al. EUS-guided Nd:YAG laser ablation of a hepatocellular carcinoma in the caudate lobe. Gastrointest Endosc 2011; 73 (03) 632-636
• 91 Karaca C, Cizginer S, Konuk Y. et al. Feasibility of EUS-guided injection of irinotecan-loaded microspheres into the swine pancreas. Gastrointest Endosc 2011; 73 (03) 603-606
• 92 Vege SS, Ziring B, Jain R, Moayyedi P. Clinical Guidelines Committee, American Gastroenterology Association. American Gastroenterological Association Institute guideline on the diagnosis and management of asymptomatic neoplastic pancreatic cysts. Gastroenterology 2015; 148 (04) 819-822 , e12–e13
• 93 Farrell JJ. Prevalence, diagnosis and management of pancreatic cystic neoplasms: current status and future directions. Gut Liver 2015; 9 (05) 571-589
• 94 Moyer MT, Heinle JW, Rhoades SE, Birkholz JH, Peng JS. Successful EUS-guided pancreatic cyst chemoablation safely allows reduction in the frequency of radiographic surveillance: long-term follow-up of randomized prospective data. Gastrointest Endosc 2024; 99 (06) 962-970
• 95 Ohtsuka T, Fernandez-Del Castillo C, Furukawa T. et al. International evidence-based Kyoto guidelines for the management of intraductal papillary mucinous neoplasm of the pancreas. Pancreatology 2024; 24 (02) 255-270
• 96 Rogowska J, Semeradt J, Durko Ł, Małecka-Wojciesko E. Diagnostics and management of pancreatic cystic lesions-new techniques and guidelines. J Clin Med 2024; 13 (16) 13
• 97 Taylor JR. Top tips for EUS-guided evaluation and sampling of pancreatic cystic lesions (with video). Gastrointest Endosc 2023; 98 (01) 113-115
• 98 McCarty TR, Paleti S, Rustagi T. Molecular analysis of EUS-acquired pancreatic cyst fluid for KRAS and GNAS mutations for diagnosis of intraductal papillary mucinous neoplasia and mucinous cystic lesions: a systematic review and meta-analysis. Gastrointest Endosc 2021; 93 (05) 1019-1033.e5
• 99 Oh CH, Song TJ, Lee JK. et al. Clinical practice guidelines for the endoscopic management of peripancreatic fluid collections. Gut Liver 2021; 15 (05) 677-693
• 100 Rangwani S, Juakiem W, Krishna SG, El-Dika S. Role of endoscopic ultrasound in the evaluation of pancreatic cystic neoplasms: a concise review. Diagnostics (Basel) 2023; 13 (04) 13
• 101 Ohno E, Hirooka Y, Kawashima H, Ishikawa T, Fujishiro M. Endoscopic ultrasonography for the evaluation of pancreatic cystic neoplasms. J Med Ultrason 2020; 47 (03) 401-411
• 102 Kimura W, Moriya T, Hirai I. et al. Multicenter study of serous cystic neoplasm of the Japan Pancreas Society. Pancreas 2012; 41 (03) 380-387
• 103 Jenssen C, Kahl S. Management of incidental pancreatic cystic lesions. Viszeralmedizin 2015; 31 (01) 14-24
• 104 Al-Haddad M, El Hajj II, Eloubeidi MA. Endoscopic ultrasound for the evaluation of cystic lesions of the pancreas. JOP 2010; 11 (04) 299-309
• 105 Adil H, Semedo A, Fenni JEL. et al. Solid-pseudopapillary neoplasms of the pancreas: a rare cause of abdominal pain. J Med Case Rep Case Ser 2021; 5 (05) 1-4
• 106 Canakis A, Lee LS. Current updates and future directions in diagnosis and management of gastroenteropancreatic neuroendocrine neoplasms. World J Gastrointest Endosc 2022; 14 (05) 267-290
• 107 Nelsen EM, Johnson EA, Walker AJ, Pfau P, Gopal DV. Endoscopic ultrasound-guided pancreatic pseudocyst cystogastrostomy using a novel self-expandable metal stent with antimigration system: a case series. Endosc Ultrasound 2015; 4 (03) 229-234
• 108 Gonda TA, Cahen DL, Farrell JJ. Pancreatic cysts. N Engl J Med 2024; 391 (09) 832-843
• 109 Kim E, Telford JJ. Endoscopic ultrasound advances, part 1: diagnosis. Can J Gastroenterol 2009; 23 (09) 594-601
• 110 Tanaka M, Fernández-Del Castillo C, Kamisawa T. et al. Revisions of international consensus Fukuoka guidelines for the management of IPMN of the pancreas. Pancreatology 2017; 17 (05) 738-753
• 111 Palazzo M, Sauvanet A, Gincul R. et al. Impact of needle-based confocal laser endomicroscopy on the therapeutic management of single pancreatic cystic lesions. Surg Endosc 2020; 34 (06) 2532-2540

Publication History

Article published online:
06 January 2026