Clinical Case Summary

A 37-year-old woman with a history of heavy alcohol use (3–4 bottles of wine weekly) presented with abdominal pain and haematemesis. She had experienced intermittent abdominal pain for two years. On assessment, she was stable apart from mild tachycardia and epigastric tenderness; melaena was confirmed on rectal exam. Blood tests revealed severe anaemia (Haemoglobin "Hb" 40 g/L from a baseline of 154 g/L). Gastroscopy showed severe oesophagitis without active bleeding. Abdominal CT showed a hyperdense focus in the neck of the pancreas, raising suspicion of a neuroendocrine tumour. After blood transfusion, she was discharged (Hb 73 g/L) with plans for outpatient colonoscopy and endoscopic ultrasound (EUS) to characterise the pancreatic neck lesion.
Two days later, she was readmitted with recurrent haematemesis and a Hb drop to 56 g/L. She was stabilized and discharged again. The day after her discharge, she underwent an EUS examination in our unit. The EUS revealed signs indicative of chronic pancreatitis, as well as a thick-walled vascular structure measuring 20 mm, adjacent to the head of the pancreas. This structure appeared to originate from a posterior branch of the gastroduodenal artery, and was consistent with a pseudo-aneurysm eroding into the wall of the duodenum. A subsequent triple phase CT scan showed an increase in the size of the hyperdense focus in the neck of the pancreas, raising concern for a pseudoaneurysm near the pancreaticoduodenal artery. Following these findings, the patient underwent successful coil embolization of the pseudoaneurysm.
Conclusion: In patients with chronic alcohol use, upper gastrointestinal bleeding (UGIB) may arise from complications of pancreatitis, such as pseudoaneurysm formation, rather than solely intraluminal causes. This case highlights the need for early identification and intervention in such rare but potentially life-threatening conditions.

Introduction

Distal malignant biliary obstruction (dMBO) is a frequently occurring condition that requires immediate intervention. The primary treatment for dMBO comprises endoscopic retrograde cholangiopancreatography (ERCP) with biliary stent insertion^​. The European Society of Gastrointestinal Endoscopy recommends the use of self-expandable metal stents (SEMS) for palliative drainage of dMBO because of their superior patency and reduced incidence of adverse events (AEs), including recurrent biliary obstruction, compared to those of plastic stents^​. Three types of SEMS are available: uncovered SEMS (UCSEMS), fully covered SEMS (FCSEMS), and partially covered SEMS. However, the ideal stent type for palliative drainage of dMBO remains undetermined. Several studies have compared UCSEMS with FCSEMS to determine which type is most appropriate for dMBO^​. In terms of patency, some prospective trials have shown that FCSEMS result in better outcomes^​; however, others have observed similar results with UCSEMS and FCSEMS^​. Most studies have primarily focused on pancreatic cancer cases; however, the pathophysiology of pancreatic cancer and that of nonpancreatic cancer differ significantly, resulting in inconsistent outcomes. Post-ERCP pancreatitis (PEP) is the most common and serious AE associated with the transpapillary placement of SEMS, with an incidence rate of up to 10.8%^​. Therefore, studies have aimed to identify the risk factors for PEP after SEMS placement. Compared to UCSEMS placement, FCSEMS placement may be more strongly linked to a higher risk of PEP because they block the pancreatic duct orifice more effectively. Additionally, nonpancreatic cancer has been identified as a potential risk factor for PEP^​. Therefore, FCSEMS placement for nonpancreatic cancer cases is associated with a significantly higher risk of PEP. We examined the safety of UCSEMS for dMBO caused by nonpancreatic cancer and observed a low PEP rate, even for nonpancreatic cancer cases^​. However, a comparison of UCSEMS and FCSEMS is still required because no studies have evaluated their use for dMBO caused by nonpancreatic cancer. Therefore, this study evaluated the efficacy and safety of UCSEMS and FCSEMS for palliative drainage of dMBO caused by nonpancreatic cancer.

Aims & Methods

This retrospective multicenter study included patients who underwent UCSEMS or FCSEMS placement for nonpancreatic cancer-induced dMBO at four centers between January 2010 and April 2024. Propensity score matching (1:1) of the UCSEMS and FCSEMS groups was performed. Technical and clinical success, AEs, recurrent biliary obstruction (RBO), time to RBO (TRBO), and overall survival (OS) of these groups were compared.

Results

No significant differences in technical success (100% for both), clinical success (100% for both), and TRBO (median: 311 days [UCSEMS] vs. 317 days [FCSEMS]; p=0.42) were observed between groups. Similarly, OS did not differ significantly (median: 432 days [UCSEMS] vs. 190 days [FCSEMS]; p=0.083). The incidence of postendoscopic retrograde cholangiopancreatography pancreatitis (PEP) of the UCSEMS group was significantly lower than that of the FCSEMS group (4.2% vs. 33%; p=0.023).

Conclusion

compared to FCSEMS, UCSEMS provided sufficient patency and reduced AEs associated with dMBO caused by nonpancreatic cancer. The cause of dMBO should be carefully considered when choosing the optimal stent. UCSEMS may be safer and more effective than FCSEMS for the treatment of nonpancreatic cancer cases.

Disclosure

Authors declare no conflict of interests for this article.

Introduction

Hirschsprung’s disease is caused by the absence of ganglia in the deep intestinal wall, leading to dysfunction [1]. Resection of the affected area is the only cure, so accurate margin detection is essential. Currently, intraoperative rapid pathology is used, requiring tissue sampling and staining, but each test takes over 30 min and may need multiple samples, sometimes delaying decisions up to 3 h. A quick on-site method to detect ganglia around 200 μm deep is needed.
Various methods, including confocal microscopy [2], have been studied, but issues remain, such as dye toxicity and limited depth (>200 μm) imaging.

Aims & Methods

For non-destructive and stain-free visualization of tissues, we focused on Near-Infrared Hyperspectral Imaging (NIR-HSI)[3][4]. Near-infrared light—particularly in the 1000–1400 nm range—exhibits high tissue penetrability and is effective for detecting deep tissues. In addition, the near-infrared region provides spectroscopic information based on molecular vibrations, enabling component analysis. NIR-HSI holds promise for the identification of cancer and other biological tissues without the need for fluorescence or external marking [4][5].
To implement NIR-HSI, we developed a novel microscopic optical system capable of guiding near-infrared light. The near-infrared light source, which combines a supercontinuum light source and an acousto-optic tunable filter, allows for the selection of arbitrary wavelengths and irradiation of the target by adjusting the applied voltage [6]. This configuration enables the acquisition of near-infrared spectral images (NIR-HSI) at arbitrary wavelengths.
For performance validation, ex vivo experiments were carried out using mouse intestine in which the nerve plexus was selectively labeled with GFP (Green Fluorescent Protein). In these experiments, near-infrared light ranging from 850 to 1400 nm was irradiated in 10-nm increments, and 77 spectral images were captured. As a reference, GFP fluorescence images were acquired in the same field of view as the near-infrared images. Based on the fluorescence images, annotations were performed to delineate the enteric nerve plexus, and near-infrared spectra were obtained from both the nerve and non-nerve regions. Furthermore, using pixel-wise spectral data, segmentation of the nerve regions was performed with a machine learning model, and the accuracy of identification was evaluated.

Results

The NIR images captured by the system showed nerve distributions and mesh-like patterns similar to those in the GFP images. Spectral differences were observed between nerve and non-nerve regions. This likely reflects tissue composition differences, as muscle tissue has higher transmittance and nerve tissue higher reflectance in the NIR range [4][5][6].
Segmentation based on spectral data successfully extracted nerve contours and distribution patterns, achieving ~80% classification accuracy.

Conclusion

To enable non-invasive, label-free visualization of deep nerve distributions, we developed a NIR-HSI microscope by integrating near-infrared hyperspectral imaging with a microscopic optical system. In mouse intestine experiments, spectral differences were observed between enteric nerve plexus and surrounding tissue. The results suggest that mesh-like nerve structures can be estimated and outlined using NIR images alone. Future work will focus on imaging human samples.

References

1. Taguchi, T.et al., “The incidence and outcome of allied disorders of Hirschsprung's disease in Japan: Results from a nationwide survey,” Asian Journal of Surgery, vol.40-1, pp.29-34, 2017.
2. Kobayashi, M., Sumiyama, K., Shimojima, N., Ieiri, S., Okano, H., Kamba, S., Fujimura, T., Hirobe, S., Kuroda, T. and Takahashi‐Fujigasaki, J., “Technical Feasibility of Visualizing Myenteric Plexus Using Confocal Laser Endomicroscopy,” Journal of Gastroenterology and Hepatology, vol.32(9), pp.1604-10, 2017.
3. S. Hayashi et al., “Near-Infrared Multispectral Imaging Rigid Endoscope System for High-Speed Target Identification Using LED Rotating Light Source”, Advanced Biomedical Engineering , Vol.14, pp.79-88, 2025.
4. Akbari, H et al., “Cancer detection using infrared hyperspectral imaging”, Cancer Science, Vol. 102, No.4, pp.852–857, 2011
5. Afromowitz,et al., “Multispectral imaging of burn wounds: a new clinical instrument for evaluating burn depth”, IEEE Transactions on Biomedical Engineering, Vol35, No.10, pp.842-850, 1988.
6. R. Fukushima et al., “Detection of Exposed Nerves in Two Individuals In Vivo and Unexposed Nerves Ex Vivo with Near-Infrared Hyperspectral Laparoscope,” 2024IEEE/SICE International Symposium on System Integration (SII), Ha Long, Vietnam, pp. 19-24, 2024.

Introduction

Pain in chronic pancreatitis (CP) varies between patients in pattern, severity, and psychological impact. CP pain assessment has been focused on intensity, missing the complexity. To address the importance of multidimensional pain evaluations, we aimed to identify pain phenotypes based on multidimensional pain assessments and examine their association with clinical factors.

Aims & Methods

We studied 248 patients with painful CP from Asia, Europe, and the USA. A cluster analysis was performed on patients with painful CP, including five pain dimensions from the COMPAT-SF questionnaire (severity, fluctuation, provocative factors, spreading pain, and qualitative descriptors). The resulting clusters were compared based on demographics and clinical data, including patient-reported outcomes and quantitative sensory testing.

Results

		Cluster 1	Cluster 2	Cluster 3	p-value​
Questionnaires	Quality of life	66.3 (22.5)	64.8 (27.7)	51.1 (24.6)	<0.001
	Catastrophizing	10.7 (11.2)	19.1 (12.4)	24.3 (13.0)	<0.001
	Anxiety	5.3 (4.1)	6.2 (4.2)	8.7 (4.5)	<0.001
	Depression	4.3 (4.3)	5.5 (4.6)	7.3 (4.7)	<0.001
	Sleep	5.0 (3.5)	5.9 (3.8)	7.5 (4.5)	<0.001
QST	PPDT sum, kPa	2284 (997)	1811.5 (574.6)	1640.1 (600.4)	<0.001
	PPTT sum, kPa	2909.6 (1247.3)	2227.1 (691.3)	2184.9 (903.2)	<0.001
	Temporal summation, abdomen	1.7 (1.8)	1.9 (1.5)	1.4 (1.8)	0.178
	CPM (%)	2.3 (18.3)	5.1 (23.7)	7.2 (17.7)	0.135

Three distinct pain phenotypes were identified in the cluster analysis. Cluster 1 (n=73) included patients with low-severity pain; Cluster 2 (n=80) was characterized by severe, fluctuating pain; and Cluster 3 (n=95) comprised patients reporting constant, severe pain. Patients in Clusters 1 and 2 reported low scores across the remaining dimensions, whereas Cluster 3 consistently scored higher. Quality of life and sleep were best in Cluster 1, declined in Cluster 2, and were lowest in Cluster 3 (both p<0.001). Scores for anxiety, depression, and catastrophizing increased progressively across clusters (all p<0.001). Sensory testing revealed a pattern of increasing hyperalgesia, with Cluster 3 showing the most pronounced responses (p<0.001).

Conclusion

The COMPAT-SF questionnaire identifies three pain phenotypes in CP. Greater pain severity and persistence correlate with increased hyperalgesia, higher psychological distress, and worse overall well-being.

Introduction

Cardiovascular diseases (CVDs) are often a limiting factor in cancer treatment. In our systematic review and meta-analysis we evaluated the prevalence and incidence of CVDs in pancreatic ductal adenocarcinoma (PDAC).

Aims & Methods

We conducted the systematic search on 5 February 2024 in three databases (PubMed, EMBASE, CENTRAL). Studies reporting the prevalence or incidence of CVDs in PDAC were included. Subgroup analyses based on cancer stage and treatment type were performed. Pooled proportions with the 95% confidence interval (CI) were calculated using a random effect model. We evaluated the risk of bias using the Joanna Briggs Institute Prevalence Critical Appraisal. (PROSPERO CRD42023482295).

Results

We included 198 articles. The prevalence of CVDs other than thrombotic events at PDAC diagnosis was similar to the one in the general population: hypertension 33 % (95%CI:27-40%), ischemic heart disease 6% (95%CI:3-12%), heart failure, arrhythmia, and stroke affecting each 2-3% of cases. Their incidence during treatment was low (1%-10% overall). Thrombotic events other than embolism affected 11% (95%CI: 7%-18%) of cases at diagnosis and the incidence during treatment was between 8-10% regardless of cancer stage or treatment type. Pulmonary embolism affected 3% of cases at diagnosis and occurred in a similar proportion during treatment. The included studies had an overall low-moderate risk of bias.

Conclusion

Thrombotic events affect around 10% of PDAC patients at baseline. A similar proportion develop thrombotic events during follow-up regardless of disease stage or treatment type. The prevalence of CVDs other than thrombotic events is equivalent to the one in the general population.

Clinical Case Summary

A 48-year-old woman with locally advanced squamous cell carcinoma of the cervix received chemoradiotherapy with cisplatin, followed by brachytherapy achieving a radiologic complete response. Then, she was enrolled for PD-1 immune checkpoint inhibitor (ICI) Dostarlimab clinical trial.
After 2 years into immunotherapy, she developed severe epigastric pain, nausea and vomiting. Laboratory tests showed elevated pancreatic enzymes, cholestasis, and inflammatory markers. Magnetic resonance cholangiopancreatography (MRCP) revealed acute interstitial edematous pancreatitis, narrowing of a short segment of the common bile duct (CBD), and proximal biliary dilatation (Figure 1). Initial conservative treatment led to partial clinical improvement.
She was re-admitted with worsening jaundice. Laboratory workup ruled out common causes (e.g. IgG4-related disease) and imaging showed persistent findings with 18 mm CBD dilatation without evidence of choledocholithiasis or radiologic features of autoimmune pancreatitis (Figure 2). These findings were consistent with mild acute pancreatitis per the revised Atlanta criteria.
Given the temporal relationship with PD-1 blockade and lack of alternative causes, immune-related pancreatitis with secondary biliary involvement was strongly suspected. Endoscopic retrograde cholangiopancreatography (ERCP) with biliary plastic stenting was performed (Figure 3), and so corticosteroids therapy (1 mg/kg tapered). The patient rapidly improved clinically, biochemically, and radiologically (figure 4), supporting the diagnosis of immune-mediated pancreatitis.
In light of this immune-related adverse event and sustained oncologic remission, Dostarlimab therapy was permanently discontinued. This case highlights the need to consider immune-related pancreaticobiliary toxicity in patients on ICI, particularly when imaging is subtle and diagnostic criteria are not yet well established. It also emphasizes the importance of early multidisciplinary intervention.

References

Figure 1: CT scan (left) and MRCP (right) images showing diffusely edematous pancreatitis with no evidence of main pancreatic duct dilatation.
Figure 2: MRCP revealing a distal common bile duct stricture, with upstream dilatation of both intrahepatic and extrahepatic bile ducts.
Figure 3: ERCP fluoroscopic view demonstrating placement of a plastic pancreatic stent within the main pancreatic duct (Wirsung).
Figure 4: CT scan (left) and MRCP (right) images showing significant improvement of the inflammatory changes of the pancreatic gland, along with reduced dilatation of the intrahepatic bile ducts, after steroids initiation.

Introduction

Tuberculous lymphadenopathy represents a challenging diagnostic scenario, particularly in regions with high TB prevalence. Its diagnosis is challenging due to the nonspecific clinical and radiological findings. Blood and skin tests for TB have variable sensitivities and specificities, so histopathologic diagnosis and molecular tests for diagnosis of TB in tissue samples may be more accurate than blood tests. Obtaining tissue samples under ultrasonography or CT guidance is sometimes very difficult and risky especially in the intra-abdominal and mediastinal regions. Endoscopic ultrasound (EUS) with EUS fine needle biopsy (EUS-FNB) is a safe technique in obtaining tissue samples for diagnosis of abdominal and mediastinal TB. EUS introduces a critical advantage of real-time imaging and precise tissue sampling. Many studies found that EUS-FNB yielded significantly higher diagnostic yield in complex TB presentations, with detection rates ranging from 78-92% across various anatomical sites. Comparative analysis with Gene X-pert MTB/RIF assay and Polymerase Chain Reaction (PCR) techniques demonstrated nuanced diagnostic capabilities. Gene X-pert provides rapid molecular detection with high specificity for rifampicin resistance while PCR offers molecular amplification of bacterial DNA.

Aims & Methods

In this prospective cross-sectional study, we explored the diagnostic efficacy of molecular techniques, specifically PCR and Gene X-pert MTB/RIF assay, in detecting Mycobacterium tuberculosis in tissue samples obtained through EUS-FNB. We also aim to evaluate the sensitivity, specificity, and rapid detection capabilities of these molecular methods compared to traditional diagnostic techniques. Patients and Methods: This study included 56 patients, 28 males and 28 females with mean age of 50 ±16 years. TB PCR or Gene X-pert was done in tissue samples including mediastinal and intra-abdominal lymphadenopathy in 75% of cases, obtained by EUS-FNB. Results: PCR or Gene X-pert for TB were true-positive in 7 cases and false negative in one case with sensitivity of 88%, specificity of 100%, positive predictive value of 100% and negative predictive value of 98%.

Results

PCR or Gene X-pert for TB were true-positive in 7 cases and false negative in one case with sensitivity of 88%, specificity of 100%, positive predictive value of 100% and negative predictive value of 98%.

Conclusion

We concluded that PCR and Gene X-pert of TB in tissues obtained by EUS-FNB are of great value in diagnosis of extrapulmonary TB.

References

Mukhida S, Vyawahare CR, Mirza SB, Gandham NR, Khan S, Kannuri S, Bhaumi S. Role of GeneXpert MTB/RIF assay for the diagnosis of cervical lymph node tuberculosis and rifampicin resistance. Tzu Chi Med J. 2022 Sep 5;34(4):418-422. doi: 10.4103/tcmj.tcmj_86_22. PMID: 36578636; PMCID: PMC9791852.
Tadesse M, Abebe G, Abdissa K, Aragaw D, Abdella K, Bekele A, et al. GeneXpert MTB/RIF assay for the diagnosis of tuberculous lymphadenitis on concentrated fine needle aspirates in high tuberculosis burden settings. PLoS One. 2015;10:e0137471. doi: 10.1371/journal.pone.0137471.
Penz E, Boffa J, Roberts DJ, Fisher D, Cooper R, Ronksley PE, et al. Diagnostic accuracy of the Xpert® MTB/RIF assay for extra-pulmonary tuberculosis: A meta-analysis. Int J Tuberc Lung Dis. 2015;19:278–84. doi: 10.5588/ijtld.14.0262. i-iii.
Ligthelm LJ, Nicol MP, Hoek KG, Jacobson R, van Helden PD, Marais BJ, et al. Xpert MTB/RIF for rapid diagnosis of tuberculous lymphadenitis from fine-needle-aspiration biopsy specimens. J Clin Microbiol. 2011;49:3967–70. doi: 10.1128/JCM.01310-11.