ChatGPT4 Outperforms Endoscopists for Determination of Postcolonoscopy Rescreening and Surveillance Recommendations.

BACKGROUND & AIMS: Large language models including Chat Generative Pretrained Transformers version 4 (ChatGPT4) improve access to artificial intelligence, but their impact on the clinical practice of gastroenterology is undefined. This study compared the accuracy, concordance, and reliability of ChatGPT4 colonoscopy recommendations for colorectal cancer rescreening and surveillance with contemporary guidelines and real-world gastroenterology practice. METHODS: History of present illness, colonoscopy data, and pathology reports from patients undergoing procedures at 2 large academic centers were entered into ChatGPT4 and it was queried for the next recommended colonoscopy follow-up interval. Using the McNemar test and inter-rater reliability, we compared the recommendations made by ChatGPT4 with the actual surveillance interval provided in the endoscopist's procedure report (gastroenterology practice) and the appropriate US Multisociety Task Force (USMSTF) guidance. The latter was generated for each case by an expert panel using the clinical information and guideline documents as reference. RESULTS: Text input of de-identified data into ChatGPT4 from 505 consecutive patients undergoing colonoscopy between January 1 and April 30, 2023, elicited a successful follow-up recommendation in 99.2% of the queries. ChatGPT4 recommendations were in closer agreement with the USMSTF Panel (85.7%) than gastroenterology practice recommendations with the USMSTF Panel (75.4%) (P < .001). Of the 14.3% discordant recommendations between ChatGPT4 and the USMSTF Panel, recommendations were for later screening in 26 (5.1%) and for earlier screening in 44 (8.7%) cases. The inter-rater reliability was good for ChatGPT4 vs USMSTF Panel (Fleiss κ, 0.786; 95% CI, 0.734-0.838; P < .001). CONCLUSIONS: Initial real-world results suggest that ChatGPT4 can define routine colonoscopy screening intervals accurately based on verbatim input of clinical data. Large language models have potential for clinical applications, but further training is needed for broad use.

Increasing Antibiotic-Resistant Infections With Inpatient Endoscopic Retrograde Cholangioscopies (ERCP) Is Associated With Higher Mortality in the United States: A Cross-sectional Cohort Study.

GOALS: This study aims to investigate associated mortality with inpatient endoscopic retrograde cholangiopancreatography (ERCP) with and without resistant infections. The co-primary objective compares frequencies of inpatient ERCP with resistant infections to overall hospitalizations with resistant infections. BACKGROUND: The risks of inpatient antibiotic-resistant organisms are known, but the associated mortality for inpatient ERCP is unknown. We aim to use a national database of hospitalizations and procedures to understand trends and mortality for patients with antibiotic-resistant infections during inpatient ERCP. STUDY: The largest publicly available all-payer inpatient database in the United States (National Inpatient Sample) was used to identify hospitalizations associated with ERCPs and antibiotic-resistant infections for MRSA, VRE, ESBL, and MDRO. National estimates were generated, frequencies were compared across years, and multivariate regression for mortality was performed. RESULTS: From 2017 to 2020, national weighted estimates of 835,540 inpatient ERCPs were generated, and 11,440 ERCPs had coincident resistant infections. Overall resistant infection, MRSA, VRE, and MDRO identified at the same hospitalization of inpatient ERCPs were associated with higher mortality (OR CI(95%): Overall: 2.2(1.77-2.88), MRSA: 1.90 (1.34-2.69), VRE: 3.53 (2.16-5.76), and MDRO: 2.52 (1.39-4.55)). While overall hospitalizations with resistant infections have been decreasing annually, there has been a yearly increase in admissions requiring ERCPs with simultaneous resistant infections ( P =0.001-0.013), as well as infections with VRE, ESBL, and MDRO ( P =0.001-0.016). Required Research Practices for Studies Using the NIS scoring was 0, or the most optimal. CONCLUSIONS: Inpatient ERCPs have increasing coincident resistant infections and are associated with higher mortality. These rising infections during ERCP highlight the importance of endoscopy suite protocols and endoscopic infection control devices.

A Preliminary Controlled Trial of Endoscopic Ultrasound-guided Fiducial Markers to Guide Pancreas Surgery.

OBJECTIVE: Endoscopic ultrasound (EUS) is routinely used for fiducial marker placement (FMP) to guide stereotactic radiation of pancreatic tumors, but EUS-FMP explicitly to guide surgery has not been studied in a prospective, controlled manner. Multipurpose EUS systems have been developed that facilitate simultaneous EUS-FMP at the time of biopsy. We aimed to evaluate the feasibility of EUS-FMP to guide pancreatic resection. METHODS: In this prospective trial, we enrolled patients with resectable pancreas masses undergoing tissue sampling and placed preloaded fiducials immediately after biopsy. Intraprocedure confirmation of carcinoma, neuroendocrine, and nonlymphomatous neoplasia by rapid on-site evaluation and lesion size <4 cm was required. The main outcomes were the feasibility and ease of preoperative placement and intraoperative detection of the markers using predefined Likert scales. RESULTS: In 20 patients, EUS-FMP was successful before planned surgery and placement was technically straightforward (Likert Scale: 9.1 ± 1.3; range: 1, most challenging to 10, most facile). Intraoperative detection was feasible and improved when compared with a pre-established comparator of 5 representing an equivalent lesion without a marker (Likert Scale: 7.8 ± 2.2; range: 1, most difficult to 10, most facile; P = 0.011). The mean tumor size on EUS was 1.7 ± 0.9 (range: 0.5 to 3.6) cm. CONCLUSION: EUS-FMP is feasible and safe for resectable pancreatic tumors before surgery and may assist in perioperative detection. Preloaded fiducials may be considered for placement at the time of initial referral for EUS-fine needle biopsy.

Qualitative and Quantitative Contrast-enhanced Endoscopic Ultrasound Improves Evaluation of Focal Pancreatic Lesions.

BACKGROUND & AIMS: Endoscopic ultrasound (EUS) is a sensitive method to evaluate the pancreas but its diagnostic capability for several diseases is limited. We compared the diagnostic yield of unenhanced EUS with that of contrast-enhanced EUS for focal pancreas lesions and identified and tested quantitative parameters of contrast enhancement. METHODS: We performed a prospective tandem-controlled trial in which 101 patients with focal pancreas lesions (48 with masses, 28 with cysts, and 25 with pancreatitis) underwent conventional EUS followed by contrast EUS using intravenous perflutren microspheres. The diagnosis at each stage was scored and compared with a standard (findings from surgical pathology analysis, cytologic, and/or 6-month clinical follow-up evaluations). Quantitative parameters were generated by time-intensity curve analysis. Solid lesions were divided into derivation and testing cohorts for a crossover validation analysis of the quantitative parameters. The primary outcome was diagnostic yield of unenhanced vs contrast EUS in analysis of focal pancreas lesions. RESULTS: Contrast increased the diagnostic yield of EUS from 64% (65/101 lesions accurately assessed) to 91% (92/101 lesions accurately assessed); the odds ratio [OR] was 7.8 (95% CI, 2.7-30.2) for accurate analysis of lesions by contrast-enhanced EUS relative to unenhanced EUS. The contrast increased accuracy of analysis of pancreas masses (OR, 6.0; 95% CI, 1.8-31.8), improving assessment of neuroendocrine and other (non-carcinoma) tumors. Contrast increased the diagnostic yield for pancreas cysts to 96% (27/28) compared with 71.4% (20/28) for unenhanced EUS (P = .02), due to improved differentiation of mural nodules vs debris. Time-intensity curve analysis revealed distinct patterns of relative peak enhancement (rPE) and in-slope (rIS) for different lesions following contrast injection: for adenocarcinomas, values were low rPE and low rIS; for neuroendocrine masses, values were high rPE and normal IS; and for chronic pancreatitis foci, values were normal rPE and low rIS. In the validation cohort, these parameters correctly characterized 91% of lesions and improved yield relative to unenhanced EUS (OR, 10; 95% CI, 1.4-34.0). CONCLUSIONS: Contrast-enhanced EUS improves the accuracy of analysis of focal pancreas lesions, compared with unenhanced EUS. Integration of practical quantitative parameters, specifically relative peak enhancement and in-slope, might increase the diagnostic accuracy of contrast EUS. ClinicalTrials.gov no: 02863770.

Bleeding from Small Intestine: No Man's Land.

Bleedings from small intestine account for 5% of all gastrointestinal bleeding. With advanced endoscopic tools, such as video capsule endoscopy and deep enteroscopy, accurate diagnosis and treatment is possible in majority of cases with low mortality and morbidity. Nonoperative management includes endoscopic hemostasis and angiographic embolization. Recurrence after initial treatment is relatively common. Surgery is reserved for the cases that are refractory to endoscopic or angiographic treatment, bleeding from tumor or mass lesions, or hemodynamic instability. At the time of surgical exploration, unless the lesion has been marked by endoscopic tattoo or clip, intraoperative enteroscopy is often necessary to localize the lesion.

Metallic coil and N-butyl-2-cyanoacrylate for closure of pancreatic duct leak (with video).

BACKGROUND AND AIMS: Pancreatic fistula is a challenging yet common adverse event of partial pancreatectomy. Our objective is to determine the feasibility of endoscopic closure of a pancreatic fistula using a combination of a metallic coil and N-butyl-2-cyanoacrylate (NBCA) glue. METHODS: A patient with a postoperative pancreatic stump leak recalcitrant to conservative management and pancreatic duct stent placement underwent endoscopic/fluoroscopic placement of a metallic coil in the pancreatic duct followed by injection of .5 mL NBCA and lipiodol mixture directed at the coil. The patient's clinical condition, Jackson-Pratt (JP) drain output, and pancreatic enzyme content were monitored daily after the procedure. RESULTS: The patient's clinical condition improved. JP drain output and amylase/lipase levels progressively decreased to resolution within 7 days of the procedure. No adverse events occurred as a result of the procedure. CONCLUSIONS: Endoscopic closure of pancreatic fistula with a metallic coil and NBCA glue is feasible and may be a useful modality for treatment of refractory postpancreatectomy-related fistula.

Randomized trial of cholangioscopy-guided laser lithotripsy versus conventional therapy for large bile duct stones (with videos).

BACKGROUND AND AIMS: Bile duct stones >1 cm have a decreased incidence of successful endoscopic extraction and often require lithotripsy. Although previous guidelines suggested mechanical lithotripsy for large common bile duct stones, current guidelines suggest cholangioscopy-guided lithotripsy as an adjunct with or without balloon dilation or mechanical lithotripsy. However, no randomized trials have assessed the usefulness of this practice. METHODS: Patients with bile duct stones >1 cm in diameter were randomized in a 2:1 ratio to cholangioscopy-guided laser lithotripsy versus conventional therapy only. Conventional therapies such as mechanical lithotripsy or balloon dilation were also allowed in the laser lithotripsy group. Randomization was stratified by history of ERCP in the past 3 months. The primary outcome was endoscopic clearance of the bile duct stones. RESULTS: Endoscopic clearance was achieved in 39 (93%) of 42 patients treated with cholangioscopy-guided laser lithotripsy and 12 (67%) of 18 treated with conventional therapy only (P = .009). The 9 patients in whom ERCP was unsuccessful underwent surgical common duct exploration with stone removal. Mean procedure time was 120.7 ± 40.2 minutes for the cholangioscopy-guided laser lithotripsy group compared with 81.2 ± 49.3 minutes for the conventional therapy group (P = .0008). There was no significant difference in fluoroscopy time, number of procedures, or adverse events (cholangitis) (cholangioscopy, 2; conventional, 1) and post-ERCP pancreatitis (cholangioscopy, 2; conventional, 1). CONCLUSION: Cholangioscopy-guided laser lithotripsy increases the incidence of endoscopic clearance of large bile duct stones and decreases the need for surgery compared with conventional therapy alone. However, it is associated with longer procedure times. (Clinical trial registration number: NCT0175997.).

Narrow-band imaging versus white light versus mapping biopsy for gastric intestinal metaplasia: a prospective blinded trial.

BACKGROUND AND AIMS: Gastric intestinal metaplasia (GIM) is a gastric cancer precursor. Narrow-band imaging (NBI) may improve detection of GIM. We compared detection of GIM with high-definition white-light (HD-WL) endoscopy, NBI, and mapping biopsies in a population with increased gastric cancer risk. METHODS: Patients undergoing upper endoscopy had HD-WL examination by 1 endoscopist, followed by an NBI examination by a second endoscopist blinded to HD-WL findings. The location of abnormalities detected by HD-WL and NBI were recorded by a research coordinator, and targeted biopsies of abnormal areas were performed after NBI. Subsequently, 5 mapping biopsies were performed per patient. Biopsy specimens were read by a pathologist blinded to mode of acquisition. The primary outcome was the proportion of patients with GIM. RESULTS: We enrolled 112 patients: 107 (96%) were Hispanic or Asian, and 34 (30%) had GIM. Higher proportions of patients with GIM were detected by NBI (22/34 [65%]) and mapping (26/34 [76%]) versus HD-WL (10/34 [29%]) (P < .005 for both comparisons). GIM was detected by NBI in only 6 patients and only by mapping biopsy in 10 patients; no patient had GIM detected solely by HD-WL. Higher proportions of sites with GIM also were detected with NBI (30/57 [53%]) and mapping biopsies (38/57 [67%]) than HD-WL (16/57 [28%]) (P < .005 for both comparisons). The median number of biopsies per patient with mapping biopsies (5) was significantly higher than with NBI (2) or HD-WL (1). CONCLUSIONS: HD-WL endoscopy is insufficient for detection of GIM in patients at increased risk for gastric cancer. NBI-targeted biopsies plus mapping biopsies should be used. (Clinical trial registration number: NCT02197351.).

Optimal Timing of Endoscopic Retrograde Cholangiopancreatography in Acute Cholangitis.

OBJECTIVES: Acute cholangitis mandates resuscitation, antibiotic therapy, and biliary decompression. Our aim was to define the optimal timing of endoscopic retrograde cholangiopancreatography (ERCP) for patients with acute cholangitis. METHODS: Clinical data on all cases of cholangitis managed by ERCP were prospectively collected from September 2010 to July 2013. The clinical impact of the time to ERCP, defined as the time from presentation in the emergency department to the commencement of the ERCP, was determined. The primary outcome was length of hospitalization. Secondary outcomes included vasopressor use, endotracheal intubation, intensive care unit admission, and death. RESULTS: ERCP was successful in 182 (92%) of 199 patients with cholangitis. Length of hospitalization was significantly longer for patients undergoing ERCP at ≥48 versus <48 hours (median 9.1 vs. 6.5 d, P=0.004) even though patients having ERCP at ≥48 hours were less sick as indicated by less frequent intensive care unit admission [odds ratio,0.3; 95% confidence interval (CI), 0.2-0.6]. Multivariate analysis revealed that hospitalization increased by 1.44 days for every day ERCP was delayed (P<0.001). Comparison of ERCP≥72 versus <72 hours revealed odds ratios of 2.6 (95% CI, 1.0-7.0) for vasopressor requirement and 3.6 (95% CI, 0.8-15.9) for mortality. Time to ERCP did not impact technical success or procedural adverse events. CONCLUSIONS: ERCP should be performed within 2 days of presentation as a delay of 48 or more hours is associated with disproportionate increase in hospital stay. Delay>72 hours is associated with additional adverse outcomes including hypotension requiring vasopressor support.

Randomized Trial of Endoscopist-Controlled vs. Assistant-Controlled Wire-Guided Cannulation of the Bile Duct.

OBJECTIVES: Biliary cannulation is frequently the most difficult component of endoscopic retrograde cholangiopancreatography (ERCP). Techniques employed to improve safety and efficacy include wire-guided access and the use of sphincterotomes. However, a variety of options for these techniques are available and optimum strategies are not defined. We assessed whether the use of endoscopist- vs. assistant-controlled wire guidance and small vs. standard-diameter sphincterotomes improves safety and/or efficacy of bile duct cannulation. METHODS: Patients were randomized using a 2 × 2 factorial design to initial cannulation attempt with endoscopist- vs. assistant-controlled wire systems (1:1 ratio) and small (3.9Fr tip) vs. standard (4.4Fr tip) sphincterotomes (1:1 ratio). The primary efficacy outcome was successful deep bile duct cannulation within 8 attempts. Sample size of 498 was planned to demonstrate a significant increase in cannulation of 10%. Interim analysis was planned after 200 patients-with a stopping rule pre-defined for a significant difference in the composite safety end point (pancreatitis, cholangitis, bleeding, and perforation). RESULTS: The study was stopped after the interim analysis, with 216 patients randomized, due to a significant difference in the safety end point with endoscopist- vs. assistant-controlled wire guidance (3/109 (2.8%) vs. 12/107 (11.2%), P=0.016), primarily due to a lower rate of post-ERCP pancreatitis (3/109 (2.8%) vs. 10/107 (9.3%), P=0.049). The difference in successful biliary cannulation for endoscopist- vs. assistant-controlled wire guidance was -0.5% (95% CI-12.0 to 11.1%) and for small vs. standard sphincerotome -0.9% (95% CI-12.5 to 10.6%). CONCLUSIONS: Use of the endoscopist- rather than assistant-controlled wire guidance for bile duct cannulation reduces complications of ERCP such as pancreatitis.

Can a Computerized Simulator Assess Skill Level and Improvement in Performance of ERCP?

BACKGROUND: Endoscopic retrograde cholangiography (ERCP) is a challenging procedure with considerable risk. Computerized simulators are valuable in training for flexible endoscopy, but little data exist for their use in ERCP training. AIM: To determine a simulator's ability to assess the level of ERCP skill and its responsiveness over time to increasing trainee experience. MATERIALS AND METHODS: In this prospective parallel-arm cohort study, six novice gastroenterology fellows and four gastroenterology faculty with expertise in ERCP completed four simulated baseline cases and the same four cases at a later date. This study took place at a surgical skills center at an academic tertiary referral center. The primary outcome was the total time to complete the ERCP procedure. RESULTS: For the baseline session, experts had a shorter total procedure time than novices (444.0 vs. 616.9 s; least squares mean; p = 0.026). There was no significant difference between experts and novices in the difference of total procedure time between session 1 and session 2 (-200.3 vs. -164.4; least squares mean; p = 0.402). CONCLUSIONS: The simulator was able to differentiate experts from novices for the primary outcome of total procedure time. The simulator was not responsive to an increase in trainee experience over time.

Prevention and management of post-ERCP pancreatitis.

Pancreatitis remains as one of the most frequent and serious complications of ERCP. Research has identified several patient-related and procedural risk factors, which help guide the endoscopist in prophylaxis and management of pancreatitis. Recent studies have had a major impact on both procedural techniques and pharmacological methods for prophylaxis of post-ERCP pancreatitis. The purpose of this article is to review the relevant literature and describe the most recent and effective approaches in prevention and management of post-ERCP pancreatitis.

The utility of esophagogastroduodenoscopy before endoscopic ultrasonography in patients undergoing endoscopic ultrasonography for pancreatico-biliary and mediastinal indications.

BACKGROUND: Oblique-viewing echoendoscopes may miss luminal lesions. There is no consensus on whether to routinely perform esophagogastroduodenoscopy (EGD) before endoscopic ultrasonography (EUS). Currently, practice patterns are variable and prospective data are needed. AIM: : To determine the proportion of clinically meaningful lesions detected when EGD is performed routinely before EUS. STUDY: This was a multicenter prospective cohort study conducted at tertiary referral center and large community practice. Patients undergoing EUS for pancreatico-biliary and mediastinal indications were enrolled. MAIN OUTCOMES: The primary outcome was the proportion of patients with a clinically meaningful lesion found on EGD. This was a combined outcome defined as any lesion that would alter medical management, or impact the subsequent EUS examination. RESULTS: Two hundred four patients were included in the final analysis. Clinically meaningful lesions were found on EGD in 45 patients [22.1%; 95% confidence interval (CI), 16.4-27.8]. Lesions that altered medical management were found in 32 patients (15.7%; 95% CI, 10.7-20.7). Lesions impacting the subsequent EUS examination were found in 20 patients (9.8%; 95% CI, 5.7-13.9). Clinically meaningful lesions found were (number of patients): esophagitis (14), ulcer (9), ring/stricture (7), large hiatal hernia (6), hyperplastic gastric polyp (5), Barrett esophagus (3), surgically altered anatomy (2), neoplastic lesion (2), subepithelial mass/GIST (1), stenosis (1), diverticulum (1), and fistula (1). CONCLUSIONS: EGD before EUS may detect enough clinically meaningful lesions to support the routine performance of EGD before EUS.

Hybrid resection of GI stromal tumor with endoscopic submucosal dissection and the full-thickness resection device.

Video 1Hybrid resection of gastric GI stromal tumor with endoscopic submucosal dissection and the Full-Thickness Resection Device.

Underwater versus conventional endoscopic mucosal resection for colorectal lesions: An updated meta-analysis of randomized controlled trials.

Background and study aims Colorectal malignancy is a leading cause of death. Conventional endoscopic mucosal resection (CEMR) is a strategy used to resect precancerous lesions that involves injecting fluid beneath a polyp to create a gap for resection. Underwater endoscopic mucosal resection (UEMR) is a newer method that forgoes injection, instead filling the intestinal cavity with water to facilitate polyp resection. Our aim was to compare the safety and efficacy of these approaches by synthesizing the most contemporary evidence. Methods PubMed, Embase, and Cochrane libraries were searched from inception through November 11, 2022 for randomized controlled trials (RCTs) comparing UEMR and CEMR for resection of colorectal lesions. The primary outcome was the rate of en bloc resection and secondary outcomes included recurrence, procedure time, and adverse events (AEs). Results A total of 2539 studies were identified through our systematic literature search. After screening, seven RCTs with a total of 1581 polyps were included. UEMR was associated with significantly increased rates of en bloc resection (RR 1.18 [1.03, 1.35]; I 2 = 76.6%) versus conventional approaches. No significant differences were found in procedure time, recurrence, or AEs. Conclusions UEMR is a promising effective technique for removal of colorectal lesions. The most contemporary literature indicates that it improves en bloc resection rate without increasing procedure time, recurrence, or AEs (PROSPERO ID CRD42022374935).

Pancreatic schwannoma: Case report, clinico-pathologic correlation, and review of the literature.

Schwannomas are common peripheral nerve sheath tumors that typically occur on the head, neck, trunk, or extremities. Intra-abdominal schwannomas, however, are rare. We describe a young woman who presented for imaging evaluation of suspected nephrolithiasis and was incidentally found to have a schwannoma centered within the pancreatic parenchyma. In addition, we detail the clinical, imaging, and histopathologic features of pancreatic schwannoma and summarize diagnosis and management of this rare clinical entity.