Admission Leuko-Glycemic Index and In-Hospital Mortality in Patients with Acute Surgical Abdomen: An Exploratory Retrospective Cohort Study

Background and Objectives: The leuko-glycemic index (LGI) is an emerging immunometabolic biomarker that integrates systemic inflammation and stress hyperglycemia and has shown prognostic value in various acute conditions. Its utility in patients with acute surgical abdomen remains unknown. We hypothesized that elevated admission LGI values would be independently associated with higher in-hospital mortality in patients with acute surgical abdomen and would show discriminative performance at least comparable to established inflammation-based indices (NLR, MLR, PLR, SII). Materials and Methods: We conducted a single-center retrospective cohort study including 1927 adults admitted with acute surgical abdomen who underwent emergency surgery between March 2019 and March 2021. Patients were stratified according to in-hospital outcome into survivors (Group 1, n = 1625) and non-survivors (Group 2, n = 302). Demographic, hematological, and biochemical parameters, including inflammation-based ratios (NLR, MLR, PLR, SII) and the LGI, were collected upon admission. Non-parametric tests were used for between-group comparisons, and receiver operating characteristic (ROC) curve analysis assessed the discriminative performance of each index. Multivariable logistic regression was used to assess the independent association of LGI with in-hospital mortality. Results: Non-survivors were older and exhibited more pronounced hematologic and biochemical derangements, including higher leukocyte, neutrophil, glucose, urea, creatinine, and AST levels, along with lower hemoglobin, hematocrit, red blood cell, and lymphocyte counts. All inflammation-based indices were significantly elevated in Group 2 (all p < 0.0001). LGI was markedly higher in non-survivors (median 1986 [IQR 1263–3113] vs. 1568 [980–2083]; p < 0.0001) and showed the highest AUC among the indices tested (0.646, 95% CI 0.607–0.685), significantly higher than MLR, PLR and SII (DeLong p ≤ 0.0002) and nominally higher than NLR (p = 0.042). An LGI cut-off of 1665 yielded a sensitivity of 60.17% and a specificity of 57.73% for in-hospital mortality. In multivariable analysis, log-transformed LGI remained independently associated with in-hospital mortality (OR 1.44 per 1 SD, 95% CI 1.20–1.73; p = 0.0001). Conclusions: Admission LGI was independently associated with in-hospital mortality and outperformed the inflammation-based ratios tested, but its standalone discrimination was modest and its threshold unstable; these findings do not support its use for triage or risk stratification, and external prospective validation would be required before any clinical role could be considered.