Factors Associated with Early Sepsis-Associated Liver Injury and Predictors of 30-Day Mortality: A Retrospective Cohort Study

İlkce Akgun Kurtulmus,1 Sukriye Miray Kilincer Bozgul,1 Caner Acar,2 Gunes Ak,3 Rabia Ozmen,4 Hande Dik Avcı,5 Merve Dogru Efgan,6 Cagrı Uslu,7 Zehra Tuba Karaman,8 Devrim Bozkurt11Division of Intensive Care Medicine, Department of Internal Medicine, Faculty of Medicine, Ege University, Izmir, Türkiye; 2Division of Medical Oncology, Department of Internal Medicine, Faculty of Medicine, Ege University, Izmir, Türkiye; 3Department of Clinical Biochemistry, Faculty of Medicine, Ege University, Izmir, Türkiye; 4Division of Rheumatology, Department of Internal Medicine, Faculty of Medicine, Ege University, Izmir, Türkiye; 5Division of Endocrinology and Metabolism, Department of Internal Medicine, Bursa City Hospital, Bursa, Türkiye; 6Department of Internal Medicine, Izmir Atatürk Training and Research Hospital, Izmir, Türkiye; 7Faculty of Medicine, Ege University, Izmir, Türkiye; 8Department of Internal Medicine, Faculty of Medicine, Bandırma Onyedi Eylül University, Balıkesir, TürkiyeCorrespondence: Sukriye Miray Kilincer Bozgul, Division of Intensive Care Medicine, Department of Internal Medicine, Faculty of Medicine, Ege University, Izmir, Türkiye, Tel +90 232 390 35 01, Email miraybozgul@gmail.comBackground: Sepsis-associated liver injury (SALI) is associated with higher mortality, but the factors associated with early SALI and the determinants of mortality in patients with SALI remain incompletely understood.Methods: This retrospective cohort study included 524 consecutive patients with sepsis (Sepsis-3) in a tertiary intensive care unit (ICU). SALI was defined as total bilirubin ≥ 2 mg/dL with international normalized ratio ≥ 1.5, or alanine aminotransferase ≥ 68 U/L (twice the local upper limit of normal), within 24 hours of ICU admission. Factors independently associated with SALI and with 30-day mortality in SALI were identified by logistic regression after multiple imputation and elastic-net selection, with bootstrap internal validation.Results: Of 523 evaluable patients, 135 (25.8%) met the SALI definition, and their 30-day mortality was higher (63.4% vs 50.0%, p=0.010). Factors independently associated with SALI were macrophage activation-like syndrome (MALS) (aOR: 5.33), an abdominal focus (aOR: 4.03), disseminated intravascular coagulation (DIC) score ≥ 5 (aOR: 3.07), acute kidney injury on admission (aOR: 2.29) and log-ferritin (aOR: 1.31; 1.48 without MALS and DIC) (all p< 0.05; optimism-corrected area under the curve [AUC]: 0.770). Among patients with SALI, immunosuppressive therapy (OR: 8.13), a thoracic focus (OR: 8.62), log-ferritin (OR: 2.04) and activated partial thromboplastin time (aPTT) (OR: 1.08 per second) were independently associated with 30-day mortality (all p< 0.05); appropriate antibiotic therapy showed a non-significant protective trend (p=0.076). The mortality model showed good discrimination (optimism-corrected AUC: 0.857) but limited calibration (slope: 0.562).Conclusion: SALI was associated with higher 30-day mortality in sepsis. Because the factors associated with SALI were assessed concurrently with liver injury, they describe its clinical profile rather than predict its development. Immunosuppressive therapy, a thoracic focus, higher ferritin levels and prolonged aPTT were associated with 30-day mortality in SALI. These findings should be interpreted with caution and require external validation.Keywords: sepsis, sepsis-associated liver injury, ferritin, mortality, intensive care unit