Influencing factors for endoscopic retrograde cholangiopancreatography-related adverse events in novice trainees and establishment of its prediction model: a prospective study / 中华消化外科杂志

ABSTRACT

Objective:To investigate the influencing factors for endoscopic retrograde cholangiopancreatography (ERCP)-related adverse events in novice trainees and establishment of its prediction model.Methods:The prospective study was conducted. The clinical data of 12 novice trainees of ERCP in the First Hospital of Lanzhou University from July 2016 to July 2019 were selected. The operation was performed by 12 novice trainees of ERCP under the guidance of the endoscopic experts. Observation indicators (1) ERCP-related adverse events in novice trainees; (2) analysis of influencing factors for ERCP-related adverse events in novice trainees; (3) establishment of a prediction model for ERCP-related adverse events in novice trainees. Measurement data with normal distribution were represented as Mean± SD, and comparison between groups was analyzed by the t test. Measurement data with skewed distribution were represented as M(range), and com-parison between groups was analyzed by the Mann-Whitney U test. Count data were expressed as absolute numbers or percentages, and comparison between groups was analyzed by the chi-square test. The Logistic regression model was used for univariate and multivariate analyses. The regression coefficients were used to construct a prediction model. The receiver operating characteristic curve was drawn, and the area under curve was used to evaluate the predictive ability. Results:(1) ERCP-related adverse events in novice trainees. Of the 300 patients with ERCP operated by 12 novice trainees, 52 cases had ERCP-related adverse events and 248 cases had no ERCP-related adverse events. Cases in grade 1?2 or grade 3?4 of ERCP difficulty classification, score for intubation time, score for cannulation time, cases with or without completion of the cannulation, cases with or with-out basket stone removal, cases with or without stenosis expansion, score for contrast-enhanced interpretation, score for implementation of reasonable treatment and score for expected purpose reached were 22, 30, 8(range, 5?10), 20(rang, 9?20), 24, 28, 11, 41, 0, 52, 39±17, 39±19 and 44±23 for novice trainees with ERCP-related adverse events, versus 146, 102, 6(range, 4?9), 12(range, 8?20), 163, 85, 94, 154, 20, 228, 52±22, 80±20, 52±23 for novice trainees without ERCP-related adverse events, showing significant differences in the above indicators between them ( χ2=4.79, Z=?2.46, ?2.72, χ2=7.01, 5.30, 4.49, t=?4.00, ?2.97, ?2.29, P<0.05). (2) Analysis of influencing factors for ERCP-related adverse events in novice trainees. Results of univariate analysis showed that the diffi-culty classification of ERCP, intubation time, cannulation time, completion of cannulation, basket stone extraction, contrast-enhanced interpretation, implementation of reasonable treatment and expected purpose reached were related factors for ERCP-related adverse events in novice trainees ( odds ratio=1.95, 1.11, 1.08, 0.45, 0.44, 0.97, 0.98, 0.98, 95% confidence intervals as 1.07?3.58, 1.02?1.22, 1.02?1.14, 0.24?0.82, 0.22?0.90, 0.96?0.99, 0.96?0.99, 0.97?1.00, P<0.05). Results of multi-variate analysis showed that difficulty classification of ERCP and contrast-enhanced interpretation were independent influencing factors for ERCP-related adverse events in novice trainees ( odds ratio=2.08, 0.95, 95% confidence intervals as 1.10?3.96, 0.92?0.99, P<0.05). (3) Establishment of a predic-tion model for ERCP-related adverse events in novice trainees. According to the important outcome indicators of clinical training and results of multivariate analysis, 4 indicators including difficulty classification of ERCP, intubation time, cannulation time and contrast-enhanced interpretation were included to establish a prediction model for ERCP-related adverse events in novice trainees, which indicated that trainees with the predicted score >0.4 were in high risk of ERCP-related adverse events. The area under receiver operating characteristic curve of the prediction model was 0.72(95% confidence interval as 0.65?0.79, P<0.05), with the best cut-off value as 0.40, the sensitivity as 76.9% and the specificity as 63.3%. Conclusion:The difficulty classification of ERCP and contrast-enhanced interpretation are independent influencing factors for ERCP-related adverse events in novice trainees. Novice trainees with a predicted score >0.4 are high-risk groups of ERCP-related adverse events.