Endoscopic ultrasound features in predicting noncurative resection of endoscopic submucosal dissection in patients with rectal neuroendocrine neoplasms.

Abstract

Purpose: To explore the factors influencing noncurative resection (NCR) in patients with rectal neuroendocrine neoplasms (R-NENs) on the basis of endoscopic ultrasonography (EUS) findings and to construct and validate a nomogram prediction model based on these factors.

Methods: This retrospective cohort study included 244 patients with pathologically confirmed R-NEN who underwent endoscopic submucosal dissection (ESD) at the Affiliated Hospital of Qingdao University between March 2016 and June 2023. The patients' EUS and clinical features were collected to identify independent factors influencing NCR following ESD. A nomogram prediction model was constructed, and its performance was evaluated with receiver operating characteristic (ROC) curve analysis.

Results: A larger tumor diameter, irregular borders, and submucosal involvement on EUS were identified as independent risk factors for NCR following ESD in patients with R-NENs. A nomogram model integrating these three predictors effectively predicted the occurrence of NCR. ROC curve analysis was used to compare the clinical predictive efficacy of the independent influencing factors and their combination. The results revealed that the area under the curve for the combination of factors was 0.791, with a sensitivity of 68.6% and specificity of 86.0%, indicating good clinical diagnostic value. The Hosmer-Lemeshow goodness-of-fit test ( P  = 0.178) indicated satisfactory model calibration.

Conclusion: The nomogram model achieved good predictive performance. This model can assist endoscopists in dynamically assessing the risk of NCR in real time, but its clinical applicability requires verification.