RESUMO
BACKGROUND AND AIMS: Colonoscopy is a critical diagnostic tool for colorectal diseases; however, its effectiveness depends on adequate bowel preparation (BP). This study aimed to develop a machine learning predictive model based on Chinese adults for inadequate BP . METHODS: A multicenter, prospective study was conducted on adult outpatients undergoing colonoscopy from January 2021 to May 2023. Data on patient characteristics, comorbidities, medication use, and BP quality were collected. Logistic regression and four machine learning models (support vector machines, decision trees, extreme gradient boosting, and bidirectional projection network) were used to identify risk factors and predict inadequate BP. RESULTS: Of 3217 patients, 21.14% had inadequate BP. The decision trees model demonstrated the best predictive capacity with an area under the receiver operating characteristic curve of 0.80 in the validation cohort. The risk factors at the nodes included body mass index, education grade, use of simethicone, diabetes, age, history of inadequate BP, and longer interval. CONCLUSION: The decision trees model we created and the identified risk factors can be used to identify patients at higher risk for inadequate BP before colonoscopy, for whom more PEG or auxiliary medication should be used.
RESUMO
BACKGROUND AND AIMS: Despite its growing popularity, endoscopic submucosal dissection (ESD) for colorectal neoplasms is still technically challenging. The factors contributing to the failure of ESD are not yet comprehensively elucidated. Therefore, this systematic review was conducted to explore the potential risk factors associated with unsuccessful colorectal ESD. METHODS: A comprehensive search of Medline and Embase databases was conducted to identify relevant publications from inception until March 14, 2023. Unsuccessful ESD was defined as cases involving incomplete resection or the occurrence of adverse events, such as perforation and delayed bleeding. RESULTS: Among the 2067 citations initially identified, a total of 23 cohort studies and 16 case-control studies met the inclusion criteria. Following meta-analyses, several significant risk factors for incomplete resection were identified, including lesion diameter ≥40 or 50 mm, right-side colonic location, deeper submucosal invasion, and severe fibrosis. Similarly, lesion diameter ≥40 or 50 mm and severe fibrosis emerged as risk factors for perforation. However, no individual factor was found to be statistically associated with delayed bleeding. CONCLUSIONS: This meta-analysis identified risk factors correlated with incomplete resection and adverse events following ESD. The findings provide valuable insights that can guide clinical decision-making, aiding gastroenterologists in accurately identifying high-risk individuals.
