The Journal of Practical Medicine ›› 2025, Vol. 41 ›› Issue (5): 716-723.doi: 10.3969/j.issn.1006-5725.2025.05.015

• Clinical Research • Previous Articles    

The predictive value of the systemic inflammation response index for non⁃curative resection after endoscopic submucosal dissection for early colorectal cancer

Jiyue ZHU1,Bo ZHANG2,Yaru LI2,Liuye. HUANG2()   

  1. *.Department of Clinical Medicine,Qingdao University,Qingdao 266071,Shandong,China
  • Received:2024-11-25 Online:2025-03-10 Published:2025-03-20
  • Contact: Liuye. HUANG E-mail:huangliuyeyhd@163.com

Abstract:

Objective To evaluate the predictive value of the Systemic Inflammation Response Index (SIRI) for non?curative resection during endoscopic submucosal dissection (ESD) treatment of early?stage colorectal cancer (CRC), and to develop a nomogram?based prediction model. Methods Retrospective data were collected from 235 patients diagnosed with early?stage CRC who underwent ESD at our hospital between January 2016 and August 2024. Receiver operating characteristic (ROC) curves were constructed to evaluate the predictive performance of inflammatory markers, such as the SIRI, for non?curative resection following ESD. Logistic regression analysis was conducted to identify independent risk factors associated with non?curative resection, and a prediction model was developed based on these factors. The model was internally validated. Results The incidence of non?curative resection in the study population was 26.38% (62 out of 235 cases). Inflammatory markers, specifically SIRI and SII, demonstrated predictive value for non?curative resection following ESD in patients with early?stage CRC, with SIRI exhibiting a higher predictive accuracy (AUC = 0.704). Logistic regression analysis identified age, family history, CEA, SIRI, and SII as independent risk factors for non?curative resection (all P < 0.05). Based on these findings, a nomogram prediction model incorporating age, family history, CEA, and SIRI was developed, achieving a C?index of 0.741 (95% CI: 0.675 ~ 0.806). The model's performance was validated using the Bootstrap method, and the decision curve analysis indicated satisfactory predictive accuracy. Conclusions SIRI demonstrates superior predictive value compared to SII for non?curative resection following ESD in patients with early?stage CRC. Independent risk factors for non?curative resection after ESD include age, family history, CEA levels, SIRI, and SII. A nomogram prediction model constructed using these risk factors?specifically age, family history, CEA levels, and SIRI?can effectively predict the likelihood of non?curative resection after ESD.

Key words: systemic inflammation response index, early colorectal cancer, endoscopic submucosal dissection, non-curative resection

CLC Number: