Development and Validation of a Prediction Model for Admission After Endoscopic Retrograde Cholangiopancreatography

Artificial intelligence and nonoperating room anesthesia

PURPOSE OF REVIEW

The integration of artificial intelligence (AI) in nonoperating room anesthesia (NORA) represents a timely and significant advancement. As the demand for NORA services expands, the application of AI is poised to improve patient selection, perioperative care, and anesthesia delivery. This review examines AI's growing impact on NORA and how it can optimize our clinical practice in the near future.

RECENT FINDINGS

AI has already improved various aspects of anesthesia, including preoperative assessment, intraoperative management, and postoperative care. Studies highlight AI's role in patient risk stratification, real-time decision support, and predictive modeling for patient outcomes. Notably, AI applications can be used to target patients at risk of complications, alert clinicians to the upcoming occurrence of an intraoperative adverse event such as hypotension or hypoxemia, or predict their tolerance of anesthesia after the procedure. Despite these advances, challenges persist, including ethical considerations, algorithmic bias, data security, and the need for transparent decision-making processes within AI systems.

SUMMARY

The findings underscore the substantial benefits of AI in NORA, which include improved safety, efficiency, and personalized care. AI's predictive capabilities in assessing hypoxemia risk and other perioperative events, have demonstrated potential to exceed human prognostic accuracy. The implications of these findings advocate for a careful yet progressive adoption of AI in clinical practice, encouraging the development of robust ethical guidelines, continual professional training, and comprehensive data management strategies. Furthermore, AI's role in anesthesia underscores the need for multidisciplinary research to address the limitations and fully leverage AI's capabilities for patient-centered anesthesia care.

Pediatric ERCP: Evolving into an Outpatient Procedure

BACKGROUND

While most adult ERCPs are performed on an outpatient basis, pediatric ERCPs are typically performed on an inpatient basis, or with ERCP followed by at least one night inpatient admission. We have begun performing a substantial proportion of our pediatric ERCPs on an outpatient basis, using our clinical judgment to guide the decision process. In the present study, we compare patient characteristics, indications, and adverse events associated with outpatient vs. inpatient ERCP.

METHODS

Using our endoscopy database, we identified patients 18 years of age and under who underwent ERCP from 2019 to 2021. Demographics, hospitalization status, indications, findings, interventions, as well as available adverse event and clinical outcomes data were analyzed.

RESULTS

147 ERCP procedures were performed during the study period by one of two interventional endoscopists. A subset of 51 (34.7%) patients underwent outpatient ERCP. Comparison of the two groups (outpatient vs. inpatient ERCP) was notable for no statistically significant difference in patient age, range of indications, or proportion of index vs. subsequent ERCP. Overall rates of ERCP-associated adverse events were low and there was no statistically significant difference between adverse events in patients who underwent outpatient vs. inpatient ERCP.

CONCLUSION

We analyzed outpatient and inpatient pediatric ERCP patient demographics and ERCP characteristics to identify factors that guide decision to determine whether pediatric ERCPs are performed on an outpatient vs. inpatient basis. There was no significant difference in adverse events associated with outpatient vs. inpatient pediatric ERCPs, attesting to the safety of outpatient ERCP for this subset of patients in the studied context. This is an area worthy of future prospective and multi-center study.

Development and validation of a practical clinical risk prediction model for post-endoscopic retrograde cholangiopancreatography pancreatitis

Background

Pancreatitis following endoscopic retrograde cholangiopancreatography (ERCP) can lead to significant morbidity and mortality. We aimed to develop an accurate post-ERCP pancreatitis risk prediction model using easily obtainable variables.

Methods

Using prospective multi-center ERCP data, we performed logistic regression using stepwise selection on several patient-, procedure-, and endoscopist-related factors that were determined a priori. The final model was based on a combination of the Bayesian information criterion and Akaike's information criterion performance, balancing the inclusion of clinically relevant variables and model parsimony. All available data were used for model development, with subsequent internal validation performed on bootstrapped data using 10-fold cross-validation.

Results

Data from 3021 ERCPs were used to inform models. There were 151 cases of post-ERCP pancreatitis (5.0% incidence). Variables included in the final model included female sex, pancreatic duct cannulation, native papilla status, pre-cut sphincterotomy, increasing cannulation time, presence of biliary stricture, patient age, and placement of a pancreatic duct stent. The final model was discriminating, with a receiver operating characteristic curve statistic of 0.79, and well-calibrated, with a predicted risk-to-observed risk ratio of 1.003.

Conclusions

We successfully developed and internally validated a promising post-ERCP pancreatitis clinical prediction model using easily obtainable variables that are known at baseline or observed during the ERCP procedure. The model achieved an area under the curve of 0.79. External validation is planned as additional data becomes available.

Development and external validation of a nomogram for prediction of post-endoscopic retrograde cholangiopancreatography pancreatitis

BACKGROUND AND AIMS

Endoscopic retrograde cholangiopancreatography (ERCP) is widely performed for management of pancreatobiliary diseases; however, post-ERCP pancreatitis (PEP) remains as an unsolved problem. Although various risk factors for PEP have been reported, the prediction of PEP remains controversial. This study aimed to develop a predictive model for PEP.

METHODS

Consecutive patients undergoing ERCP for biliary indications at two centers were retrospectively studied. Using data from a training cohort, we utilized a multivariable model to select five variables to construct a nomogram. The predictive model was internally and externally validated. Based on the nomogram, the patients were categorized into low-, moderate-, and high-risk groups.

RESULTS

Using the data of 2224 patients in the training cohort, five variables were selected to generate a nomogram: 1) sex, 2) indication for ERCP, 3) difficult cannulation, 4) guidewire insertion into the pancreatic duct, and 5) endoscopic sphincterotomy or sphincteroplasty. The most significant risk factor was endoscopic papillary balloon dilation such as endoscopic sphincterotomy or sphincteroplasty. The bias-corrected concordance index was 0.72 in the training cohort and 0.72 in the validation cohort. Calibration curves for both cohorts demonstrated good agreement between the predicted and observed frequencies of the actual outcome. In the validation cohort, PEP developed in 5.0% and 14% of patients in the moderate- and high-risk groups, respectively.

CONCLUSIONS

We successfully developed a good predictive model for PEP. The prevention of PEP in high risk patients should be investigated further.

[Structural, procedural, and personnel requirements for cross-sectoral provision of endoscopic gastroenterological procedures]

Gastroenterology has made crucial advances in diagnostic and interventional endoscopic procedures, opening up improvements in the treatment of many patients. Thus, organ-preserving treatments are increasingly being made possible, replacing more invasive organ resecting surgical procedures. At the same time, the degree of complexity and risks varies widely between different endoscopic procedures. In many cases, simpler endoscopic procedures are now offered on an outpatient basis. Further potential for cross-sectoral performance of endoscopic procedures exists in the case of complex endoscopic procedures, which, however, require special structural, procedural and personnel requirements in order to provide quality-assured treatment, enable post-interventional monitoring and, if necessary, take measures to ensure the success of the treatment. We summarize the essential prerequisites and limitations for cross-sector performance of endoscopic procedures in gastroenterology.

Reducing the environmental footprint of gastrointestinal endoscopy: European Society of Gastrointestinal Endoscopy (ESGE) and European Society of Gastroenterology and Endoscopy Nurses and Associates (ESGENA) Position Statement

Climate change and the destruction of ecosystems by human activities are among the greatest challenges of the 21st century and require urgent action. Health care activities significantly contribute to the emission of greenhouse gases and waste production, with gastrointestinal (GI) endoscopy being one of the largest contributors. This Position Statement aims to raise awareness of the ecological footprint of GI endoscopy and provides guidance to reduce its environmental impact. The European Society of Gastrointestinal Endoscopy (ESGE) and the European Society of Gastroenterology and Endoscopy Nurses and Associates (ESGENA) outline suggestions and recommendations for health care providers, patients, governments, and industry. MAIN STATEMENTS 1: GI endoscopy is a resource-intensive activity with a significant yet poorly assessed environmental impact. 2: ESGE-ESGENA recommend adopting immediate actions to reduce the environmental impact of GI endoscopy. 3: ESGE-ESGENA recommend adherence to guidelines and implementation of audit strategies on the appropriateness of GI endoscopy to avoid the environmental impact of unnecessary procedures. 4: ESGE-ESGENA recommend the embedding of reduce, reuse, and recycle programs in the GI endoscopy unit. 5: ESGE-ESGENA suggest that there is an urgent need to reassess and reduce the environmental and economic impact of single-use GI endoscopic devices. 6: ESGE-ESGENA suggest against routine use of single-use GI endoscopes. However, their use could be considered in highly selected patients on a case-by-case basis. 7: ESGE-ESGENA recommend inclusion of sustainability in the training curricula of GI endoscopy and as a quality domain. 8: ESGE-ESGENA recommend conducting high quality research to quantify and minimize the environmental impact of GI endoscopy. 9: ESGE-ESGENA recommend that GI endoscopy companies assess, disclose, and audit the environmental impact of their value chain. 10:  ESGE-ESGENA recommend that GI endoscopy should become a net-zero greenhouse gas emissions practice by 2050.

The milestone for preventing post-ERCP pancreatitis using novel simplified predictive scoring system: a propensity score analysis

BACKGROUND

Post-ERCP pancreatitis (PEP) with trans-papillary approach remains a major issue, and the multi-factorial etiology can lead to the development of unpredictable PEP. Therefore, the early identification of PEP is highly desirable to assist with the health cost containment, the reduction in unnecessary admissions, earlier appropriate primary care, and intensive care for preventing progression of severe pancreatitis. This study aimed to establish a simplified predictive scoring system for PEP.

METHODS

Between January 1, 2012, and December 31, 2019, 3362 consecutive trans-papillary ERCP procedures were retrospectively analyzed. Significant risk factors were extracted by univariate, multivariate, and propensity score analyses, and the probability of PEP in the combinations of each factor were quantified using propensity score analysis. The results were internally validated using bootstrapping resampling.

RESULTS

In the scoring system with four stratifications using combinations of only five extracted risk factors, the very high-risk group showed 28.79% (95% confidence interval [CI], 18.30%-41.25%; P < 0.001) in the predicted incidence rate of PEP, and 9.09% (95% CI, 3.41%-18.74%; P < 0.001) in that of severe PEP; although the adjusted prevalence revealed 3.74% in PEP and 0.90% in severe PEP, respectively. The prediction model had an area under the curve of 0.86 (95% CI, 0.82-0.89) and the optimism-corrected model as an internal validation had an area under the curve of 0.81 (95% CI, 0.77-0.86).

CONCLUSIONS

We established and validated a simplified predictive scoring system for PEP using five risk factors immediately after ERCP to assist with the early identification of PEP.

Development and validation of a risk prediction model and scoring system for post-endoscopic retrograde cholangiopancreatography pancreatitis

Background

A few models have been proposed for the prediction of the risk of post-endoscopic retrograde cholangiopancreatography pancreatitis (PEP), but many include factors that are not assessed routinely. Herein, we intend to develop and validate a predictive model for the occurrence of PEP.

Methods

Data of patients who underwent endoscopic retrograde cholangiopancreatography (ERCP) from January 01, 2016 to May 16, 2019 (training set and internal test set) and from May 17, 2019 to December 25, 2019 (external test set) were retrospectively collected. The performance of the model was validated in the two validation cohorts.

Results

A total of 342 patients were included for the external test set, and 47 (13.7%) developed PEP. The variables included in the scoring system were gastrectomy history, high direct bilirubin (DBIL), high albumin (ALB), villous type of papillary orifice, nodular type of papillary orifice, pancreatic guidewire passages, precut sphincterotomy, and high operator experience. A total score >5 indicated high risk. In the external test set, the area under the curve (AUC) was 0.718, the sensitivity was 0.723, and the specificity was 0.676. In the external test set, the probability of PEP was 6.1%, 17.0%, and 37.5% in patients with low (<0), moderate (0-5), and high (>5) risk scores, respectively.

Conclusions

This study established a scoring system for predicting the risk of PEP using routinely measured clinical variables. Its application in routine work warrants further investigation.

Three-hour post-ERCP amylase level: a useful indicator for early prediction of post-ERCP pancreatitis

Background

To evaluate the value of the 3-h post-ERCP serum amylase level for early prediction of post-ERCP pancreatitis (PEP).

Method

A study of 206 patients performed ERCP (Encoscopic Retrograde Cholangio-Pancreatography) at a single centre was done from Jan. 2011 to Nov. 2016. The serum amylase or lipase level was measured at 3 h after ERCP. The patients with PEP were recorded. ROC curves were used to statistically analyze the data: The enrolled patients were divided into two groups according to gender, then we analyzed the data respectively. We comprehensively evaluated the predictive value of PEP by 3-h post-ERCP serum amylase level based on the results above.

Results

Two hundred six patients (92 males, 114 females) were enrolled. PEP occurred in 21 patients (10.19%) among them. The median time to discharge was 7 days (min = 1d, max = 13d) after the procedure. In the 206 patients, the 3-h post-ERCP pancreatic amylase level was used as the test variable, and the PEP occurrence as the state variable to plot the ROC curve. The area under the curve (AUC) was 0.816, and was statistically significant (P < 0.001). The standard error (SE) was 0.0507, the 95% confidence interval (CI) was 0.756–0.866, and the optimal cut-off value was 351 U/L (sensitivity 76.19%, specificity 83.24%, positive likelihood ratio 4.55, negative likelihood ratio 0.29, Youden index 59.43%). Of the 206 patients, there were 83 patients with both 3-h post-ERCP amylase level and lipase level detected, and the ROC curves were plotted for both serum amylase and lipase respectively. The ROC curve matched-pair testing was carried out: The areas under the ROC curves were statistically significant. (P < 0.001) The area under the ROC curve for the 3-h post-ERCP lipase was 0.778, the 95% confidence interval was 0.673–0.862, and optimal cut-off value was 1834 U/L. The area under the ROC curve for the 3-h post-ERCP serum amylase was 0.780, and the 95% confidence interval was 0.676–0.864. The optimal cut-off is 380 U/L, and there was no statistically significant difference between the two for diagnostic accuracy. According to gender, 206 patients were divided into 2 groups, and the ROC curves were drawn respectively. Based on statistical analysis, there was no statistically significant difference in the diagnostic accuracy of the two groups. In the male group, 436 U/L serum amylase provided the greatest diagnostic accuracy with sensitivity (SE) of 70.5%, specificity (SP) of 89.2%, positive predictive value (PPV) 87.5%, and negative predictive value (NPV) 78.1%. Whereas, in the female group, 357 U/L serum amylase provided the greatest diagnostic accuracy with sensitivity of 76.9%, specificity of 81.2%, positive predictive value of 80.4%, negative predictive value of 77.9%.

Conclusions

1. The 3-h post-ERCP serum amylase level is a useful measurement for predicting post-ERCP pancreatitis. 2. There was no significant difference between serum amylase and lipase 3-h post-ERCP for predicting PEP. 3. There was no statistically significant difference between male and female using the 3-h post-ERCP serum amylase level to predict PEP. For female, the optimal cut-off value was 357 U/L, whereas male 436 U/L .

Measurement error and timing of predictor values for multivariable risk prediction models are poorly reported

OBJECTIVE

Measurement error in predictor variables may threaten the validity of clinical prediction models. We sought to evaluate the possible extent of the problem. A secondary objective was to examine whether predictors are measured at the intended moment of model use.

METHODS

A systematic search of Medline was used to identify a sample of articles reporting the development of a clinical prediction model published in 2015. After screening according to a predefined inclusion criteria, information on predictors, strategies to control for measurement error, and intended moment of model use were extracted. Susceptibility to measurement error for each predictor was classified into low and high risks.

RESULTS

Thirty-three studies were reviewed, including 151 different predictors in the final prediction models. Fifty-one (33.7%) predictors were categorized as high risk of error; however, this was not accounted for in the model development. Only 8 (24.2%) studies explicitly stated the intended moment of model use and when the predictors were measured.

CONCLUSION

Reporting of measurement error and intended moment of model use is poor in prediction model studies. There is a need to identify circumstances where ignoring measurement error in prediction models is consequential and whether accounting for the error will improve the predictions.

Patient perception and preference of EUS-guided drainage over percutaneous drainage when endoscopic transpapillary biliary drainage fails: An international multicenter survey

Background and Objectives:

EUS-guided biliary drainage (EUS-BD) is a feasible procedure when ERCP fails, as is percutaneous transhepatic BD (PTBD). However, little is known about patient perception and preference of EUS-BD and PTBD.

Patients and Methods:

An international multicenter survey was conducted in seven tertiary referral centers. In total, 327 patients, scheduled to undergo ERCP for suspected malignant biliary obstruction, were enrolled in the study. Patients received decision aids with visual representation regarding the techniques, benefits, and adverse events (AEs) of EUS-BD and PTBD. Patients were then asked the choice between the two simulated scenarios (EUS-BD or PTBD) after failed ERCP, the reasons for their preference, and whether altering AE rates would influence their prior choice.

Results:

In total, 313 patients (95.7%) responded to the questionnaire and 251 patients (80.2%) preferred EUS-BD. The preference of EUS-BD was 85.7% (186/217) with EUS-BD expertise, compared to 67.7% (65/96) without EUS-BD expertise (P < 0.001). The main reason for choosing EUS-BD was the possibility of internal drainage (78.1%). In multivariate analysis, the availability of EUS-BD expertise was the single independent factor that influenced patient preference (odds ratio: 3.168; 95% of confidence interval, 1.714–5.856; P < 0.001). The preference of EUS-BD increased as AE rates decreased (P < 0.001).

Conclusions:

In this simulated scenario, approximately 80% of patients preferred EUS-BD over PTBD after failed ERCP. However, preference of EUS-BD declined as its AE rates increased. Further technical innovations and improved proficiency in EUS-BD for reducing AEs may encourage the use of this procedure as a routine clinical practice when ERCP fails.

Anesthetist-Directed Sedation Favors Success of Advanced Endoscopic Procedures

OBJECTIVES

Sedation is required to perform endoscopic retrograde cholangiopancreatography (ERCP) and endoscopic ultrasound (EUS) given the duration and complexity of these advanced procedures. Sedation options include anesthetist-directed sedation (ADS) vs. gastroenterologist-directed sedation (GDS). Although ADS has been shown to shorten induction and recovery times, it is not established whether it impacts likelihood of procedure completion. Our aim was to assess whether ADS impacts the success of advanced endoscopy procedures.

METHODS

We prospectively assessed the sedation strategy for patients undergoing ERCP and EUS between October 2010 and October 2013. Although assignment to ADS vs. GDS was not randomized, it was determined by day of the week. A sensitivity analysis using propensity score matching was used to model a randomized trial. The main outcome, procedure failure, was defined as an inability to satisfactorily complete the ERCP or EUS such that an additional endoscopic, radiographic, or surgical procedure was required. Failure was further categorized as failure due to inadequate sedation vs. technical problems.

RESULTS

During the 3-year study period, 60% of the 1,171 procedures were carried out with GDS and 40% were carried out with ADS. Failed procedures occurred in 13.0% of GDS cases compared with 8.9% of ADS procedures (multivariate odds ratio (OR): 2.4 (95% confidence interval (CI): 1.5-3.6)).This was driven by a higher rate of sedation failures in the GDS group, 7.0%, than in the ADS group, 1.3% (multivariate OR: 7.8 (95% CI: 3.3-18.8)). There was no difference in technical success between the GDS and ADS groups (multivariate OR: 1.2 (95% CI: 0.7-1.9)). We were able to match 417 GDS cases to 417 ADS cases based on procedure type, indication, and propensity score. Analysis of the propensity score-matched patients confirmed our findings of increased sedation failure (multivariate OR: 8.9 (95% CI: 2.5-32.1)) but not technical failure (multivariate OR: 1.2 (0.7-2.2)) in GDS compared with ADS procedures. Adverse events of sedation were rare in both groups. Failed ERCP in the GDS group resulted in a total of 93 additional days of hospitalization. We estimate that $67,891 would have been saved if ADS had been used for all ERCP procedures. No statistically significant difference in EUS success was identified, although this sub-analysis was limited by sample size.

CONCLUSION

ADS improves the success of advanced endoscopic procedures. Its routine use may increase the quality and efficiency of these services.