Background and Aims: We performed a meta-analysis of all published studies to determine the diagnostic accuracy of artificial intelligence (AI) on histology prediction and detection of colorectal polyps. Method: We searched Embase, PubMed, Medline, Web of Science, and Cochrane library databases to identify studies using AI for colorectal polyp histology prediction and detection. The quality of included studies was measured by the Quality Assessment of Diagnostic Accuracy Studies tool. We used a bivariate meta-analysis following a random-effects model to summarize the data and plotted hierarchical summary receiver operating characteristic curves. The area under the hierarchical summary receiver operating characteristic curve (AUC) served as an indicator of the diagnostic accuracy and during head-to-head comparisons. Results: A total of 7680 images of colorectal polyps from 18 studies were included in the analysis of histology prediction. The accuracy of the AI (AUC) was .96 (95% confidence interval [CI], .95-.98), with a corresponding pooled sensitivity of 92.3% (95% CI, 88.8%-94.9%) and specificity of 89.8% (95% CI, 85.3%-93.0%). The AUC of AI using narrow-band imaging (NBI) was significantly higher than the AUC using non-NBI (.98 vs .84, P < .01). The performance of AI was superior to nonexpert endoscopists (.97 vs .90, P < .01). For characterization of diminutive polyps using a deep learning model with nonmagnifying NBI, the pooled negative predictive value was 95.1% (95% CI, 87.7%-98.1%). For polyp detection, the pooled AUC was .90 (95% CI, .67-1.00) with a sensitivity of 95.0% (95% CI, 91.0%-97.0%) and a specificity of 88.0% (95% CI, 58.0%-99.0%). Conclusions: AI was accurate in histology prediction and detection of colorectal polyps, including diminutive polyps. The performance of AI was better under NBI and was superior to nonexpert endoscopists. Despite the difference in AI models and study designs, AI performances are rather consistent, which could serve as a reference for future AI studies.
Background and Aims: Propofol-based sedation is widely used in ERCP procedures, but adverse respiratory or cardiovascular events commonly occur. Intravenous injection of lidocaine has an analgesic effect and can reduce the requirements of fentanyl and propofol during abdominal surgery. The objective of this study was to assess the efficacy and safety of intravenous lidocaine on propofol requirements during ERCP procedures. Methods: Forty-eight patients scheduled for ERCP were randomly divided into 2 groups, the lidocaine group and the control group. All patients received.02 mg/kg midazolam and .1 mg/kg sufentanil intravenously as premedication. A bolus of propofol was applied for induction of sedation, and perfusion of propofol was applied for maintenance. Patients in the lidocaine group received a bolus of 1.5 mg/kg lidocaine intravenously followed by continuous infusion of 2 mg/kg/h, whereas the control group received the same volumes of saline solution. The primary outcome was the propofol requirement during ERCP. Results: Compared with the control group, propofol requirements were reduced by 33.8% in the lidocaine group (212.0 +/- 118.2 mg vs 320.0 +/- 189.6 mg, P = .023). Involuntary movement was less common in the lidocaine group than in the control group (12.5% vs 41.7%, P = .049). In the lidocaine group, postprocedure pain and fatigue, as measured by the visual analog scale, were significantly reduced (0 [range, 0-4] vs 3 [range, 0-5], P=.005; 2 [range, 0-4] vs 5 [range, 2-8], P < .001).The incidence of oxygen desaturation, hypotension, and bradycardia tended to be lower in the lidocaine group. Conclusions: Intravenous lidocaine can significantly decrease propofol requirements during ERCP, with higher sedation quality and endoscopist satisfaction.
Background and Aims: One of the main reasons for failed endoscopic transpapillary gallbladder stenting (ETGS) under fluoroscopic guidance is the inability to cannulate the cystic duct. Single-operator peroral cholan-gioscopy (SOC)-assisted ETGS is an adjunct technique to facilitate ETGS. We aimed to demonstrate its efficacy. Methods: Between 2015 and 2019, 104 patients with acute cholecystitis at moderate to high surgical risk underwent ETGS, which involved 3 steps: (1) cystic duct cannulation under fluoroscopic guidance with or without additional SOC guidance; (2) guidewire placement; and (3) stent placement in the gallbladder. The technical success rate was determined when stent placement was confirmed endoscopically and radiographically. Results: Of 104 patients, 55 (53%) patients had successful ETGS under fluoroscopic guidance. Of 49 patients who had failed fluoroscopy-guided ETGS, 41 patients underwent additional SOC-assisted ETGS and 5 patients proceeded to other interventions. Of patients who underwent SOC-assisted ETGS (n = 41), 23 (56%) cystic cannulation followed by stent placement were successful; cystic duct cannulations, guidewire, and stent placement failed in 8, 9, and 1 patients, respectively. The overall technical success rate of ETGS increased from 53% (55 of 104) to 75% (78 of 104) after additional SOC assistance. Adverse events and recurrence were not different between patients who underwent ETGS under fluoroscopic guidance and those who underwent SOC-assisted ETGS. Conclusions: In patients with acute cholecystitis who are not surgical candidates, SOC-assisted ETGS can increase the technical success rate after failed fluoroscopic guidance. SOC can help for the cystic duct cannulation and guidewire placement steps but not for the stent placement step.
Background and Aims: The prevalence of gastroparesis (Gp), a chronic debilitating disorder, and resulting hospitalizations are increasing. Gastric peroral endoscopic pyloromyotomy (POP or GPOEM) is a novel technique in the treatment of refractory Gp. Despite the initial promising results of GPOEM, one-third of patients do not exhibit any clinical response. Furthermore, loss of clinical response was reported in several studies. No response or loss of response after GPOEM may be related to inadequate myotomy. The aim of our study is to examine whether double pyloromyotomy at GPOEM is superior to single pyloromyotomy. Method: A retrospective case-controlled study of patients who underwent GPOEM for refractory Gp at our tertiary care institution between June 2015 and March 2018 was performed. Because the follow-up time for the single myotomy group was much longer than that of the double myotomy group, we matched the length of follow-up for the single myotomy group to that of the double myotomy group. The outcomes were measured by the changes in the Gastroparesis Cardinal Symptom Index (GCSI) before and 3 to 6 months after the procedure. Adverse events and other procedural and clinical parameters were also compared. Results: Ninety patients underwent GPOEM (55 single and 35 double pyloromyotomy). The mean age was 47 +/- 14 years, and the mean duration of symptoms was 5.3 +/- 4.4 years. The average GCSI score was 3.8 before the GPOEM, and the average GCSI score 6 months after procedure was 1.8. Thirty-seven of 55 (67%) patients who underwent single pyloromyotomy achieved clinical response compared with 30 of 35 (86%) patients who underwent double pyloromyotomy. There were no significant differences for procedure time, postoperative pain, or length of hospital stay between the 2 groups. There was no difference in adverse events in the 2 pyloromyotomy groups. Conclusion: Double pyloromyotomy is a safe and feasible technique during GPOEM. Clinical success was higher in patients undergoing double pyloromyotomy compared with single pyloromyotomy in this nonrandomized, short-term follow-up study. Long-term studies are needed to further confirm our results.