Permpikul C.Wongtirawit N.Tongyoo S.Khemmongkon M.Naorungroj T.Mahidol University2026-06-222026-06-222026-01-01Siriraj Medical Journal Vol.78 No.1 (2026) , 59-67https://repository.li.mahidol.ac.th/handle/123456789/117471Objective: To investigate the clinical predictors for hospital mortality in patients with sepsis-associated acute kidney injury who required renal replacement therapy. Materials and Methods: A retrospective cohort study enrolling adult patients hospitalized in medical wards at Siriraj Hospital between 2018-2021, who were concurrently diagnosed with sepsis and acute kidney injury. Patients who previously received long-term renal replacement therapy were excluded. We compared clinical characteristics and treatment strategies, then analyzed the predictors of mortality according to hospital mortality. Results: Among 189 patients with acute kidney injury requiring renal replacement therapy, 132 (69.8%) died during hospitalization. A receiver operating characteristic curve analysis for predictors of hospital mortality revealed cutoff values for age > 60 years, SOFA score > 10, APACHE II score > 20, total colloid in 72 hr > 1,500 mL, maximum vasopressor > 0.3 mcg/kg/min, white blood cell count < 12 cells/μL, serum creatinine < 4 mg/dL, and serum albumin < 2.5 g/dL. Multivariate analysis identified age > 60 (OR 2.8, 1.27-6.38), APACHE II score > 20 (OR 2.57, 1.23-5.42), and maximum vasopressor dose > 0.3 μg/kg/min (OR 4.26, 1.94-9.86) as independent mortality predictors, while creatinine > 4 mg/dL was protective (OR 0.36, 0.17-0.75). Conclusion: Patients with sepsis-associated acute kidney injury who underwent renal replacement therapy had high hospital mortality. Age > 60 years, APACHE II score > 20, maximum vasopressor dose > 0.3 mcg/kg/min, and serum creatinine ≤ 4 mg/dL were predictors for hospital mortality.MedicineArticleSCOPUS10.33192/smj.v78i11.2757882-s2.0-10504175208422288082