Introduction

Computer-Aided Diagnosis (CADx) is emerging as a tool to assist endoscopists in differentiating neoplastic from non-neoplastic small polyps during colonoscopy. In clinical practice, the application of optical diagnosis strategies, such as ‘’the resect and discard’’ and the ‘’leave-in-situ’’, vary according to the polyp location. This study aimed to evaluate the impact of polyp location (proximal colon vs. distal colon) on the diagnostic performance of CADx systems for small polyps (<5mm) identified during colonoscopy.

Aims & Methods

A comprehensive literature search was conducted in MEDLINE, Scopus, and Embase databases, with the last search performed on March 12, 2024. Studies evaluating the stand-alone performance of real-time CADx for predicting the histology of colorectal polyps ≤5mm in size were included. Studies without pathological confirmation or with fewer than 50 patients were excluded. The primary endpoints were the sensitivity and specificity of the CADx in the proximal and distal colon. Secondary outcomes were the negative predictive value (NPV), positive predictive value (PPV) and accuracy of the CADx alone in the proximal and distal colon. The distal colon comprised the rectum and sigmoid, while the proximal colon comprised the portion extending from the descending colon to the cecum.

Results

We included 11 studies for analysis with a total of 7,782 polyps (<5mm). CADx accuracy was significantly lower in the proximal colon compared to the rectosigmoid in terms of specificity (62% versus 84%; risk ratio (RR): 0.78 [95% confidence interval (CI): 0.72-0.84], I²=47.08%), NPV (64% versus 93%; RR: 0.71 [95% CI: 0.64-0.79], I²=74.69%), and accuracy (81% vs 86%; RR: 0.95 [95% CI: 0.91-0.99], I²=72.50%). On the other hand, sensitivity (87% vs 87%; RR: 1.00 [95% CI: 0.98-1.02], I²=0%) and PPV (87% vs 76%; RR: 1.11 [95% CI: 1.06-1.17], I²=53.85%) were comparable.

Conclusion

The diagnostic performance of CADx appears to depend on the location of the polyp, being substantially lower for proximal than distal lesions, mainly due to a lower specificity in the proximal colon.

Disclosure

Cesare Hassan: Fujifilm Co. (consultancy); Medtronic Co. (consultancy), Yuichi Mori: Olympus Corp (consultancy, speaking honorarium, equipment loan); Cybernet System (ownership interest), Alessandro Repici: Fujifilm Co. (consultancy); Olympus Corp (consultancy); Medtronic Co. (consultancy).

Introduction

Acute variceal bleeding is a life-threatening decompensating event in the natural history of cirrhosis. Endoscopic variceal ligation (EVL) along with non-selective beta blockers or carvedilol are current standard of care for patients with acute esophageal variceal bleeding. Use of proton pump inhibitors (PPI) in post EVL period is more of a personal choice as various society guidelines do not recommend continuation of PPI after EVL owing to low quality of evidence to support their use. The literature regarding the benefits of PPI in post EVL re-bleeding and mortality is heterogeneous without any clear answer particularly for mortality.

Aims & Methods

We conducted a single centre, outcome assessor blinded prospective randomized control trial where patients of liver cirrhosis with suspected or documented acute variceal bleeding were screened and those who underwent EVL were randomized into PPI or No-PPI group. Patients in the PPI group received Tab esomeprazole 40 mg once a day in addition to standard of care for all the patients. Both groups were followed up at 2 weeks post index EVL after randomization for occurrence of rebleeding and a repeat upper gastrointestinal endoscopy (UGIE) to document presence or absence of post EVL ulcers, their number and size. All patients were again followed up at 3 months for development of recurrent bleeding, mortality and adverse effect like pain and dysphagia post EVL. Primary outcome was to compare the rate of recurrent bleeding after EVL between the 2 groups. Secondary outcomes were to assess the difference in mortality, number and size of post EVL ulcers and adverse effects in the form of pain and dysphagia post EVL. Association between categorical outcomes was assessed by cross tabulation and comparison of percentages. Chi square test was used to test statistical significance. The association between quantitative variables was assessed by independent sample t-test to assess statistical significance. A p value of <0.05 was considered significant.

Results

A total of 246 patients were screened from June 2023 to April 2024 and 170 patients were randomized to both the groups. Baseline parameters were comparable between the groups. In the primary outcome, number of patients who developed recurrent bleeding episode were 23 (28.05%) in the no-PPI group, which were significantly more in comparison to 9 (11.54%) in the PPI group (p=0.009). Among the secondary outcomes, incidence of mortality was higher in the no-PPI group [16 (19.51%) vs 13 (16.67%) respectively], although statistically non-significant (p=0.640). All 16 (100%) deaths in the no-PPI group were related to rebleeding episode in comparison to only 5 (38.46%) deaths in the PPI group (p=0.035). Incidence of post EVL ulcers was significantly more in the no-PPI group [65 (86.67%) vs 22 (30.99%) respectively; p<0.001]. Median post EVL ulcer number and size were also significantly higher in the no-PPI group [1 (IQR 1,2) vs 0 (IQR 0,1) (p=<0.001); 6 mm (IQR 5,8) vs 5 mm (IQR 5,6) (p=0.0061) respectively]. At 2 weeks, chest pain was reported by 7 (9.33%) patients in the no PPI group vs 2 (2.82%) patients in the PPI group (p=0.167). No patient reported dysphagia at 2 weeks. No patient reported chest pain or dysphagia at 3 months.

Conclusion

Prescription of PPI after EVL in patients of liver cirrhosis with acute esophageal variceal bleeding is beneficial as it significantly decreases the recurrent bleeding episodes as well as mortality in such patients.

Introduction

The vacuum-assisted closure stent (VAC-stent) is a novel approach for managing esophageal defects, combining endoscopic vacuum therapy (EVT) and metal stent support to promote effective tissue healing [1].
We present a case of esophageal perforation following circumferential Endoscopic Submucosal Dissection (ESD), successfully managed with a VAC-Stent (video 1).

Aims & Methods

We report a case of a 59-year-old man with a 40mm lesion involving 75% of the esophageal circumference (Paris 0-IIa + Is), arising in Barrett’s esophagus (Prague C4M5) and consistent with adenocarcinoma, who was referred to our unit. CT scans excluded metastatic disease, and after multidisciplinary consultation, ESD was planned.

Results

Circumferential ESD was performed using a triple-tunnel approach. After the resection, muscular tears were closed with through-the-scope (TTS) clips, and pre-emptive Endoscopic Vacuum Therapy (EVT) was applied to enhance granulation and healing.
Due to rising inflammatory markers and fever onset, an esophagogastroduodenoscopy (EGD) was performed on the 3rd Post-Operative Day (POD). The EGD revealed two <5mm perforations 4cm apart, and a VAC-stent was placed to bridge the defects. At the follow-up EGD on the 9th POD, a single defect remained, and a new VAC-Stent was placed. The patient showed progressive clinical improvement and successfully resumed a liquid diet. By the 15th POD, the perforations had healed.
The patient was discharged in good clinical condition on the 18th POD.
At the EGD on the 30th POD, homogeneous healing tissue was observed, with no stricture detected at the resection site.
Histological analysis of the ESD specimen revealed moderately differentiated adenocarcinoma with submucosal invasion: pT1b (sm1) pNX G2, with R0 resection achieved and no high-risk pathological features. An oncological and endoscopic follow-up was planned.

Conclusion

Our case demonstrates that the VAC-Stent is a safe and effective method for treating perforations following esophageal ESD. In cases of circumferential ESD, VAC-stent use may also help prevent strictures.

References

[1]. Dell’Anna G, Fanti L, Fanizza J, et al. VAC-Stent in the Treatment of Post-Esophagectomy Anastomotic Leaks: A New “Kid on the Block” Who Marries the Best of Old Techniques—A Review. Journal of Clinical Medicine 2024; 13: 3805. doi:10.3390/jcm13133805