RESUMO
BACKGROUND: Preoperative diagnosis of the cause of adhesive small bowel obstruction (ASBO) is very challenging for surgeons. We aimed to develop a nomogram model for the identification of banded adhesions (BA) and matted adhesions (MA) of ASBO. METHODS: This retrospective study enrolled patients with ASBO between January 2012 and December 2020, classified into BA and MA groups according to the intraoperative findings. A nomogram model was developed by using multivariable logistic regression analysis. RESULTS: A total of 199 patients were included, with 117 (58.8%) cases of BA and 82 (41.2%) cases of MA. There were 150 patients designed for training the model, and the other 49 cases for validation. Multivariate logistic regression analysis showed that prior surgery for once (p = 0.008), white blood cells (WBC) (p = 0.001), beak sign (p < 0.001), fat notch sign (p = 0.013), and mesenteric haziness (p = 0.005) were independently associated with BA. The AREA under the receiver operating characteristic curve (AUC-ROC) of the nomogram model in the training and validation sets were 0.861 (95% CI 0.802-0.921) and 0.884 (95% CI 0.789-0.980), respectively. The calibration plot demonstrated a good agreement. A decision curve analysis demonstrated that the nomogram model was clinically useful. CONCLUSIONS: The multi-analysis of the nomogram model might have a favorable clinical applicability for the identification of BA and MA in patients with adhesive small bowel obstruction.
