Early Serum Creatinine Change and Multidimensional Clinical Determinants of Kidney Replacement Therapy in Acute Kidney Injury

Background: Acute kidney injury (AKI) is a dynamic clinical syndrome, and a single serum creatinine measurement may not adequately reflect the severity or progression of renal dysfunction. The aim of this study was to assess whether early changes in serum creatinine provide better discrimination for renal replacement therapy (RRT) requirement than admission creatinine alone and to identify clinical factors associated with RRT and short-term renal outcomes. Methods: This retrospective observational cohort study included 115 adults with AKI admitted to the Nephrology Department or Intensive Care Unit of Sibiu County Clinical Emergency Hospital between January 2018 and December 2025. Clinical, biochemical, hemodynamic, fluid-balance and nephrotoxic exposure data were analyzed. Serum creatinine was assessed at admission or transfer to the Nephrology Department or Intensive Care Unit (time zero) and at 24 and 48 h thereafter. Multivariable logistic regression and receiver operating characteristic analyses were used to evaluate factors associated with RRT requirement and an eGFR below 60 mL/min/1.73 m2 at 30 days. Creatinine change was defined as the absolute difference from the value measured at study entry. Results: Fifty-two patients required RRT. The 48 h change in serum creatinine showed better discrimination for RRT requirement than admission creatinine (AUC 0.758 vs. 0.634; DeLong p = 0.012). An exploratory, cohort-derived 48 h creatinine-change threshold of 0.79 mg/dL showed a sensitivity of 90.4% and a specificity of 52.4% for distinguishing RRT status. In multivariable analysis, vasopressor requirement, maximum serum potassium and positive 24 h fluid balance were independently associated with RRT, while the overall model showed high apparent discrimination (AUC 0.925). RRT requirement was also associated with an eGFR below 60 mL/min/1.73 m2 at 30 days (adjusted OR 2.84, 95% CI 1.02–7.91). Conclusions: The 48 h creatinine change showed modest apparent discrimination of RRT status, but these findings are exploratory and do not establish prospective prediction. Patients requiring RRT also had greater hemodynamic instability, hyperkalemia and positive fluid balance and were more likely to have an eGFR below 60 mL/min/1.73 m2 at 30 days. Serial assessment of renal function, together with the patient’s clinical and metabolic status, may therefore be more informative than a single creatinine measurement.