ASSOCIATION BETWEEN EARLY URINARY POTASSIUM EXCRETION AND ACUTE KIDNEY INJURY IN CRITICALLY ILL PATIENTS WITH SHOCK: A HOSPITAL-BASED OBSERVATIONAL STUDY

Abstract

Acute kidney injury (AKI) is a frequent and clinically significant problem in critically ill patients, particularly those presenting with shock, where hemodynamic instability, inflammatory injury, and multi-organ dysfunction contribute to rapid renal impairment. Early recognition of AKI is essential to improving outcomes; however, conventional diagnostic markers such as serum creatinine and urine output rise relatively late in the course of disease, limiting opportunities for timely intervention. The present hospital-based observational study aimed to evaluate whether early urinary potassium excretion specifically urinary potassium concentration and fractional excretion of potassium (FeK) can predict the subsequent development of AKI in critically ill patients with shock admitted to the intensive care unit (ICU). Eighty-five adult patients (18–85 years) admitted to M.S. Ramaiah Hospitals ICU with shock, and catheterized within 24 hours, were prospectively evaluated between May 2023 and December 2024. Patients with pre-existing renal failure, diuretic use within the preceding 24 hours, nephrotoxic exposure, baseline dyskalemia, or anuria at presentation were excluded. Baseline demographics, vital signs, serum and urinary biochemical parameters, daily urine output, creatinine clearance, and severity indices (SOFA and APACHE II) were recorded. AKI diagnosis and staging were performed according to KDIGO criteria. Diagnostic performance of urinary potassium and FeK was assessed using receiver operating characteristic (ROC) analysis. The incidence of AKI in the cohort was 76.5% (n=65), with Stage 1 constituting 56.9% of cases. Septic shock was the predominant etiology (76.4%), and respiratory and urinary infections were the leading septic sources. Patients who developed AKI exhibited significantly higher urinary potassium (49.36 ± 20.15 mEq/L) and FeK (37.13 ± 36.83%) compared to those without AKI (p < 0.05). FeK demonstrated excellent diagnostic accuracy with an area under the ROC curve (AUC) of 0.871, sensitivity of 73.85%, and specificity of 85% at an optimal cut-off >10.25%.

Urinary potassium alone had moderate predictive ability (AUC 0.691) but high specificity (95%) at a cut-off ≤32.4 mEq/L. AKI was associated with significantly prolonged hospitalization (9.25 vs. 7.30 days, p = 0.002), while mortality occurred only in AKI patients (7.7%), though not statistically significant. This study demonstrates that early urinary potassium measurements, particularly FeK, serve as sensitive and practical biomarkers for early identification of AKI in critically ill shock patients. Incorporating these urinary indices into early ICU assessment protocols may facilitate timely recognition of renal deterioration, enabling earlier intervention and improved clinical outcomes.