Systemic Immune-Inflammation Index: A Promising, Non-Invasive Biomarker for Crohn's Disease Activity and Severity Assessment.

Abstract

Purpose: Crohn's disease (CD) is a chronic inflammatory disorder with periods of exacerbation and remission. We aim to evaluate the systemic immune-inflammation index (SII) as a prognostic biomarker in CD and its utility in predicting disease activity and severity.

Patients and methods: This retrospective study analyzed CD patients using the Harvey-Bradshaw index (HBI) for disease stratification and the Simple Endoscopic Score for Crohn's Disease (SES-CD) for post-treatment evaluation. Data analysis was conducted using R software. Serological indices underwent predictive analysis through the receiver operating characteristic (ROC) curve. The least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression identified independent prognostic factors to construct nomograms. Model validation was performed using the Concordance index (C-index), calibration analysis and decision curve analysis (DCA).

Results: In this study, 254 patients with Crohn's disease (CD) were enrolled, including 171 males and 83 females, with ages ranging from 13 to 74. SII was significantly elevated in active CD (p<0.001), correlating with disease severity (p<0.001). Although SII decreased in patients with mucosal healing (p<0.001), its prognostic accuracy (AUC=0.719) was lower than other biomarkers. However, SII emerged as an independent predictor for CD activity and severity with higher efficacy (AUC=0.774 and 0.807). The CD activity and severity prediction nomograms showed high C-indices (0.8038 and 0.8208), indicating strong predictive performance.

Conclusion: SII is a valuable biomarker for assessing CD severity and monitoring mucosal healing post-treatment. The SII-based nomograms offer a reliable model for evaluating CD progression, aiding in personalized treatment approaches and enhancing clinical decision-making. We recommend randomized controlled trials (RCTs) or studies with larger sample sizes to improve the model.