Prognostic value of the systemic immune-inflammation index in acute non-variceal upper gastrointestinal bleeding: a retrospective single-center cohort study.

Abstract

Background: Acute non-variceal upper gastrointestinal bleeding (NVUGIB) is a life-threatening emergency associated with significant morbidity and mortality. Current risk stratification tools have limitations in predicting short-term outcomes. The Systemic Immune-Inflammation Index (SII), a composite biomarker integrating platelet, neutrophil, and lymphocyte counts, has shown prognostic value across various acute conditions; however, its utility in NVUGIB remains unexplored.

Methods: In this retrospective single-center cohort study, 812 adults presenting with NVUGIB to the emergency department of Kartal Dr. Lütfi Kırdar City Hospital (Istanbul, Turkey) were analyzed. SII was calculated as (platelet count × neutrophil count) / lymphocyte count. Primary outcome was 28-day all-cause mortality; secondary outcome was 72-hour rebleeding. Discriminative performance was assessed using receiver operating characteristic (ROC) curve analysis and compared with established risk scores (Glasgow-Blatchford Score, pre-endoscopic Rockall Score, complete Rockall Score, and AIMS65). Independent predictors were identified using multivariable logistic regression and Cox proportional hazards analyses.

Results: The 28-day mortality rate was 6.5% (n = 53) and the 72-hour rebleeding rate was 11.1% (n = 90). Median SII was significantly higher in non-survivors compared to survivors [1,714.9 (IQR: 980.5-3,776.0) vs. 910.5 (IQR: 599.3-1,501.0); p < 0.001]. For 28-day mortality, SII achieved an AUC of 0.704 (95% CI: 0.622-0.785), comparable to established scoring systems (all p > 0.05 by DeLong test). Notably, SII demonstrated superior discriminative performance for 72-hour rebleeding (AUC = 0.767; 95% CI: 0.714-0.820), significantly outperforming GBS (AUC = 0.540), AIMS65 (AUC = 0.541), pre-endoscopic Rockall Score (AUC = 0.561), and complete Rockall Score (AUC = 0.554). In multivariable logistic regression, SII (per 100 units; OR = 1.02, 95% CI: 1.02-1.03; p < 0.001) was an independent predictor of mortality. Decision curve analysis demonstrated that adding SII to existing scores improved net clinical benefit.

Conclusions: SII is an independent prognostic marker in NVUGIB, offering discriminative performance for 28-day mortality comparable to established risk scores and superior performance for 72-hour rebleeding prediction. As a simple, routinely available laboratory index, SII may complement existing risk stratification tools in the emergency management of NVUGIB.