Samuel Han

Background: An accurate prognostic tool is essential to aid clinical decision-making (e.g., patient triage) and to advance personalized medicine. However, such a prognostic tool is lacking for acute pancreatitis (AP). Increasingly machine learning (ML) techniques are being used to develop high-performing prognostic models in AP. However, methodologic and reporting quality has received little attention. High-quality reporting and study methodology are critical for model validity, reproducibility, and clinical implementation. In collaboration with content experts in ML methodology, we performed a systematic review critically appraising the quality of methodology and reporting of recently published ML AP prognostic models. Methods/findings: Using a validated search strategy, we identified ML AP studies from the databases MEDLINE and EMBASE published between January 2021 and December 2023. We also searched pre-print servers medRxiv, bioRxiv, and arXiv for pre-prints registered between January 2021 and December 2023. Eligibility criteria included all retrospective or prospective studies that developed or validated new or existing ML models in patients with AP that predicted an outcome following an episode of AP. Meta-analysis was considered if there was homogeneity in the study design and in the type of outcome predicted. For risk of bias (ROB) assessment, we used the Prediction Model Risk of Bias Assessment Tool. Quality of reporting was assessed using the Transparent Reporting of a Multivariable Prediction Model of Individual Prognosis or Diagnosis—Artificial Intelligence (TRIPOD+AI) statement that defines standards for 27 items that should be reported in publications using ML prognostic models. The search strategy identified 6,480 publications of which 30 met the eligibility criteria. Studies originated from China (22), the United States (4), and other (4). All 30 studies developed a new ML model and none sought to validate an existing ML model, producing a total of 39 new ML models. AP severity (23/39) or mortality (6/39) were the most common outcomes predicted. The mean area under the curve for all models and endpoints was 0.91 (SD 0.08). The ROB was high for at least one domain in all 39 models, particularly for the analysis domain (37/39 models). Steps were not taken to minimize over-optimistic model performance in 27/39 models. Due to heterogeneity in the study design and in how the outcomes were defined and determined, meta-analysis was not performed. Studies reported on only 15/27 items from TRIPOD+AI standards, with only 7/30 justifying sample size and 13/30 assessing data quality. Other reporting deficiencies included omissions regarding human–AI interaction (28/30), handling low-quality or incomplete data in practice (27/30), sharing analytical codes (25/30), study protocols (25/30), and reporting source data (19/30). Conclusions: There are significant deficiencies in the methodology and reporting of recently published ML based prognostic models in AP patients. These undermine the validity, reproducibility, and implementation of these prognostic models despite their promise of superior predictive accuracy. Registration: Research Registry (reviewregistry1727)

Background/Objectives: Performing endoscopic retrograde cholangiopancreatography (ERCP) in surgically altered gastrointestinal anatomy remains challenging, frequently necessitating the use of forward-viewing endoscopes. Given the challenge in endoscope selection based on the type of altered anatomy, the aim of this study was to examine ERCP success rates by specific endoscopes for different anatomy types. Methods: This single-center retrospective study examined ERCPs performed in patients with surgically altered gastrointestinal anatomy during an 18-year period. Enteroscopy success, cannulation success, and intervention success rates were compared between the different anatomy and endoscope types. Results: This study included a total of 334 adult patients (665 total ERCPs) with altered anatomy. The pediatric colonoscope was most frequently utilized (32.2%), and the majority of procedures were performed for biliary indications. Enteroscopy success was 82.2% in Roux-en-Y gastric bypass (RYGB), 97% in Billroth II, 91.5% in Whipple, and 93.2% in Roux-en-Y hepaticojejunostomy (RYHJ). Cannulation success was 90.5% in RYGB, 90.5% in Billroth II, 83.6% in Whipple, and 90.6% in RYHJ. Intervention success was 88.2% in Billroth II, 65.1% in RYGB, 81.6% in Whipple, and 87.5% in RYHJ. In patients with RYGB and RYHJ, SBE was utilized most frequently, with rotational enteroscopy having the highest success rates. The overall adverse event rate was 5.1%, with the majority of these being mild in severity. Conclusions: This large retrospective study found ERCP with forward-viewing endoscopes to be safe and effective for a variety of surgically altered anatomy types. Despite recent advances seen with endoscopic ultrasound-guided drainage procedures, this study advocates for ERCP as the initial approach for pancreaticobiliary access in surgically altered anatomy.

Background: An accurate prognostic tool is essential to aid clinical decision making (e.g., patient triage) and to advance personalized medicine. However, such prognostic tool is lacking for acute pancreatitis (AP). Increasingly machine learning (ML) techniques are being used to develop high-performing prognostic models in AP. However, methodologic and reporting quality has received little attention . High-quality reporting and study methodology are critical to model validity, reproducibility, and clinical implementation. In collaboration with content experts in ML methodology, we performed a systematic review critically appraising the quality of methodology and reporting of recently published ML AP prognostic models. Methods: Using a validated search strategy, we identified ML AP studies from the databases MEDLINE, PubMed, and EMBASE published between January 2021 and December 2023. Eligibility criteria included all retrospective or prospective studies that developed or validated new or existing ML models in patients with AP that predicted an outcome following an episode of AP. Meta-analysis was considered if there was homogeneity in the study design and in the type of outcome predicted. For risk of bias (ROB) assessment, we used the Prediction Model Risk of Bias Assessment Tool (PROBAST). Quality of reporting was assessed using the Transparent Reporting of a Multivariable Prediction Model of Individual Prognosis or Diagnosis – Artificial Intelligence (TRIPOD+AI) statement that defines standards for 27 items that should be reported in publications using ML prognostic models. Results: The search strategy identified 6480 publications of which 30 met the eligibility criteria. Studies originated from China (22), U.S (4), and other (4). All 30 studies developed a new ML model and none sought to validate an existing ML model, producing a total of 39 new ML models. AP severity (23/39) or mortality (6/39) were the most common outcomes predicted. The mean area-under-the-curve for all models and endpoints was 0.91 (SD 0.08). The ROB was high for at least one domain in all 39 models, particularly for the analysis domain (37/39 models). Steps were not taken to minimize over-optimistic model performance in 27/39 models. Due to heterogeneity in the study design and in how the outcomes were defined and determined, meta-analysis was not performed. Studies reported on only 15/27 items from TRIPOD+AI standards, with only 7/30 justifying sample size and 13/30 assessing data quality. Other reporting deficiencies included omissions regarding human-AI interaction (28/30), handling low-quality or incomplete data in practice (27/30), sharing analytical codes (25/30), study protocols (25/30) and reporting source data (19/30),. Discussion: There are significant deficiencies in the methodology and reporting of recently published ML based prognostic models in AP patients. These undermine the validity, reproducibility and implementation of these prognostic models despite their promise of superior predictive accuracy. Funding: none Registration: Research Registry ( reviewregistry1727 )

Background: Chronic abdominal pain is the hallmark symptom of chronic pancreatitis (CP), with 50% to 80% of patients seeking medical attention for pain control. Although several management options are available, outcomes are often disappointing, and opioids remain a mainstay of therapy. Opioid-induced hyperalgesia is a phenomenon resulting in dose escalation, which may occur partly because of the effects of opioids on voltage-gated sodium channels associated with pain. Preclinical observations demonstrate that the combination of an opioid and the antiseizure drug lacosamide diminishes opioid-induced hyperalgesia and improves pain control. Objective: In this phase 1 trial, we aim to determine the safety, tolerability, and dose-limiting toxicity of adding lacosamide to opioids for the treatment of painful CP and assess the feasibility of performance of a pilot study of adding lacosamide to opioid therapy in patients with CP. As an exploratory aim, we will assess the efficacy of adding lacosamide to opioid therapy in patients with painful CP. Methods: Using the Bayesian optimal interval design, we will conduct a dose-escalation trial of adding lacosamide to opioid therapy in patients with painful CP enrolled in cohorts of size 3. The initial dose will be 50 mg taken orally twice a day, followed by incremental increases to a maximum dose of 400 mg/day, with lacosamide administered for 7 days at each dose level. Adverse events will be documented according to Common Terminology Criteria for Adverse Events (version 5.0). Results: As of December 2023, we have currently enrolled 6 participants. The minimum number of participants to be enrolled is 12 with a maximum of 24. We expect to publish the results by March 2025. Conclusions: This trial will test the feasibility of the study design and provide reassurance regarding the tolerability and safety of opioids in treating painful CP. It is anticipated that lacosamide will prove to be safe and well tolerated, supporting a subsequent phase 2 trial assessing the efficacy of lacosamide+opioid therapy in patients with painful CP, and that lacosamide combined with opiates will lower the opioid dose necessary for pain relief and improve the safety profile of opioid use in treating painful CP. Trial Registration: Clinicaltrials.gov NCT05603702; https://clinicaltrials.gov/study/NCT05603702 International Registered Report Identifier (IRRID): PRR1-10.2196/50513

Background and Objectives: While studies have demonstrated the efficacy of cholangioscopy-guided Holmium-Yttrium aluminum garnet (Ho:YAG) laser lithotripsy for the treatment of refractory bile duct stones, data regarding the safety of the operating parameters for laser lithotripsy are lacking. The aim of this study was to determine safe, yet effective, energy settings for Ho:YAG laser in the ex-vivo model. Materials and Methods: This ex vivo experimental study utilized the Ho:YAG laser on porcine bile duct epithelium and human gallstones. Ho:YAG laser lithotripsy was applied in different power settings from 8 to 15 Watts (W) to six explanted porcine bile ducts. Settings that appeared safe were then utilized to fragment seventy-three human gallstones. Results: The median bile duct perforation times with the Ho:YAG laser between 8–15 W were: >60 s (8 W); 23 s (9 W); 29 s (10 W); 27 s (12 W); 12 s (14 W); and 8 s (15 W). Statistically significant differences in the median perforation times were noted between 8 W vs. 15 W, 9 W vs. 15 W, 10 W vs. 15 W, and 12 W vs. 15 W (p < 0.05). When using a 365 µm Ho:YAG laser probe at 8–12 W, the fragmentation rates on various size stones were: 100% (2.0 cm). Optimal fragmentation was seen utilizing 12 W with high energy (2.4 J) and low frequency (5 Hz) settings. Using a larger 550 µm probe at these settings resulted in 100% fragmentation of stones larger than 2 cm. Conclusions: The Ho:YAG laser appears to be safe and effective in the treatment of large bile duct stones when used between 8–12 W in 5 s bursts in an ex vivo model utilizing porcine bile ducts and human gallstones.