Petersen, Bret

Abstract Background The authors previously developed an artificial intelligence (AI) to assist cytologists in the evaluation of digital whole‐slide images (WSIs) generated from bile duct brushing specimens. The aim of this trial was to assess the efficiency and accuracy of cytologists using a novel application with this AI tool. Methods Consecutive bile duct brushing WSIs from indeterminate strictures were obtained. A multidisciplinary panel reviewed all relevant information and provided a central interpretation for each WSI as being “positive,” “negative,” or “indeterminate.” The WSIs were then uploaded to the AI application. The AI scored each WSI as positive or negative for malignancy (i.e., computer‐aided diagnosis [CADx]). For each WSI, the AI prioritized cytologic tiles by the likelihood that malignant material was present in the tile. Via the AI, blinded cytologists reviewed all WSIs and provided interpretations (i.e., computer‐aided detection [CADe]). The diagnostic accuracies of the WSI evaluation via CADx, CADe, and the original clinical cytologic interpretation (official cytologic interpretation [OCI]) were compared. Results Of the 84 WSIs, 15 were positive, 42 were negative, and 27 were indeterminate after central review. The WSIs generated on average 141,950 tiles each. Cytologists using the AI evaluated 10.5 tiles per WSI before making an interpretation. Additionally, cytologists required an average of 84.1 s of total WSI evaluation. WSI interpretation accuracies for CADx (0.754; 95% CI, 0.622–0.859), CADe (0.807; 95% CI, 0.750–0.856), and OCI (0.807; 95% CI, 0.671–0.900) were similar. Conclusions This trial demonstrates that an AI application allows cytologists to perform a triaged review of WSIs while maintaining accuracy.

INTRODUCTION: Prophylactic pancreatic stent placement (PSP) is effective for preventing pancreatitis after endoscopic retrograde cholangiopancreatography (ERCP) in high-risk cases, but the optimal technical approach to this intervention remains uncertain. METHODS: In this secondary analysis of 787 clinical trial patients who underwent successful stent placement, we studied the impact of (i) whether pancreatic wire access was achieved for the sole purpose of PSP or naturally during the conduct of the case, (ii) the amount of effort expended on PSP, (iii) stent length, (iv) stent diameter, and (v) guidewire caliber. We used logistic regression models to examine the adjusted association between each technical factor and post-ERCP pancreatitis (PEP). RESULTS: Ninety-one of the 787 patients experienced PEP. There was no clear association between PEP and whether pancreatic wire access was achieved for the sole purpose of PSP (vs occurring naturally; odds ratio [OR] 0.82, 95% confidence interval [CI] 0.37–1.84), whether substantial effort expended on stent placement (vs nonsubstantial effort; OR 1.58, 95% CI 0.73–3.45), stent length (>5 vs ≤5 cm; OR 1.01, 95% CI 0.63–1.61), stent diameter (≥5 vs <5 Fr; OR 1.13, 95% CI 0.65–1.96), or guidewire caliber (0.035 vs 0.025 in; 0.83, 95% CI 0.49–1.41). DISCUSSION: The 5 modifiable technical factors studied in this secondary analysis of large-scale randomized trial data did not appear to have a strong impact on the benefit of prophylactic PSP in preventing PEP after high-risk ERCP. Within the limitations of post hoc subgroup analysis, these findings may have important implications in procedural decision making and suggest that the benefit of PSP is robust to variations in technical approach.

Abstract Background and study aims Obtaining quality tissue during ERCP biliary stricture sampling is of paramount importance for a timely diagnosis. While single-operator cholangioscopy (SOC)-guided biopsies have been suggested to be the superior biliary tissue acquisition modality given direct tissue visualization, less is known about the specimen histological quality. We aimed to analyze the specimen quality of SOC biopsies and compare the new generation forceps with prior “legacy” forceps. Patients and methods Patients who underwent SOC from January 2017-August 2021 for biliary sampling were reviewed. In February 2020, the SOC-guided biopsy forceps were changed from legacy SpyBite to the SpyBite Max forceps (max). Specimens were assessed by blinded pathologists for crush artifact (none, mild, or severe) and gross size (greatest dimension in mm). Crush artifact and gross size were compared between the two groups. The diagnostic performance characteristics for cholangiocarcinoma (CCA), were assessed in an exploratory fashion. Results Eighty-one patients (max = 27, legacy = 54) with similar baseline characteristics were included in this study. On blinded pathological assessment, 58 % had crush artifact, without significant differences between the two groups (Max 63 % vs. Legacy 56 %; P = 0.64). A similar mean specimen size was found (max 3 mm vs. legacy 3.2 mm; P = 0.24). The overall prevalence of CCA was 40 %. The sensitivity, specificity, positive predictive value, and negative predictive value of the entire cohort using a combination of cytology, fluorescence in situ hybridization, and SOC-guided biopsies were 78.1 %, 91.8 %, 86.2 %, and 86.5 %, respectively. No difference between legacy or max groups was found. Conclusions A high rate of crush artifact was found in SOC-guided biopsy specimens. Further investigation regarding proper biopsy technique and handling is necessary to increase the diagnostic yield with SOC-guided biopsies.

Abstract Background and study aims The preferred management of bleeding esophageal varices includes endoscopic band ligation. Endoscopic ligation devices (ELDs) are expensive and designed for single use, limiting their uptake in developing countries. We aimed to assess the efficacy of reprocessing ELDs using terminal microbial cultures and adenosine triphosphate (ATP) testing. Materials and methods ELDs were recovered after clinical use and their components (cap, handle, and cord) were subjected to reprocessing. This included manual cleaning, automated high-level disinfection (HLD), and drying with forced air. Using sterile technique, ELD components were sampled for ATP at three stages: before manual cleaning, after manual cleaning, and after HLD. Components were sent to an external laboratory for culturing. Cultures were interpreted as positive upon identification of Gram-negative bacilli. Results A total of 14 clinically used ELDs were studied, and 189 ATP tests and 41 cultures were evaluated. Overall, 95 % (39/41) of components and 86 % (12/14) of ELDs were culture-negative or did not yield Gram-negative bacilli. Two components (5 %; one handle and one cord) harbored Gram-negative bacilli in quantities of 1 CFU per component. There was no apparent correlation between ATP at any juncture of reprocessing and terminal cultures. Conclusions Reprocessing of ELDs is effective, resulting in infrequent and minimal microbial contamination. Microbial culturing can be used to ensure adequacy of ELD reprocessing if pursued. Until reusable ELDs are commercially available, continued efforts to better define the adequacy and long-term effects of reprocessing ELDs are needed.