Construction and Validation of Risk Prediction Model for Colorectal Polyps / 胃肠病学

ABSTRACT

Background: Colorectal polyp is a common lower gastrointestinal disease. Study of its risk factors is of great significance for prevention and treatment of colorectal polyps in clinical practice. Aims: To construct and verify a prediction model for risk of colorectal polyps. Methods: According to the inclusion and exclusion criteria, 254 subjects who were hospitalized for health examination in the Special Internal Medicine Ward of Shanghai Huadong Hospital from January 2019 to June 2021 were enrolled in the study. They were allocated into colorectal polyps group and non⁃polyp group based on the results of colonoscopy. The relevant risk factors of colorectal polyp were collected, including gender, age, cigarette smoking, alcohol drinking, hypertension, diabetes, hyperlipidemia, hyperuricemia, polyps/stones of gallbladder, fatty liver, etc. After screened by LASSO regression model, the selected factors were analyzed by multivariate Logistic regression to build the prediction model and nomogram. Furthermore, the prediction model was evaluated by ROC curve, C index, calibration curve and decision curve, and validated by internal samples. Results: Of the 254 subjects enrolled in the study, 116 cases were in colorectal polyps group and 138 in non⁃polyp group. The risk prediction model identified that gender (OR=2.11, 95% CI 1.06⁃4.27), age (OR=2.76, 95% CI 1.17⁃6.73), hypertension (OR=3.23, 95% CI 1.52⁃7.12), diabetes (OR=4.37, 95% CI 1.52⁃14.64), hyperlipidemia (OR=3.20, 95% CI 1.74⁃5.95) and fatty liver (OR= 2.21, 95% CI 1.13⁃4.35) were independent risk factors for colorectal polyps. The model showed good area under the ROC curve (0.807) and C index (0.807). The decision curve demonstrated that if the threshold probability of colorectal polyps was more than 12%, the model would be of clinical significance. Internal samples were randomly selected for validation, and the C index was 0.793. Conclusions: The prediction model and nomogram constructed by combination of risk factors including gender, age, hypertension, diabetes, hyperlipidemia and fatty liver have a substantial reference value for risk prediction of colorectal polyps.