Early Laboratory Predictors of Sepsis-Associated Acute Kidney Injury: Development and Validation of a Clinical Prediction Model

Abstract

Objective: To identify early laboratory predictors of sepsis-associated acute kidney injury (SA-AKI) and develop a practical clinical prediction model using routinely available laboratory investigations among adult patients with sepsis admitted to tertiary care hospitals in Pakistan.

Methods: This prospective multicenter observational cohort study was conducted over 12 months in the Departments of Nephrology and General Medicine of tertiary care teaching hospitals in Pakistan. A total of 250 adult patients diagnosed with sepsis according to the Sepsis-3 criteria were enrolled consecutively. Demographic and clinical characteristics, along with routine laboratory investigations including complete blood count, serum creatinine, blood urea nitrogen (BUN), C-reactive protein (CRP), serum albumin, serum bicarbonate, and urine output, were recorded within 24 hours of admission. The primary outcome was the development of SA-AKI within seven days according to the Kidney Disease: Improving Global Outcomes (KDIGO) criteria. Independent predictors were identified using multivariable logistic regression. The predictive performance of the model was assessed using receiver operating characteristic (ROC) curve analysis, the Hosmer–Lemeshow goodness-of-fit test, and internal split-sample validation.

Results: Of the 250 enrolled patients, 92 (36.8%) developed SA-AKI. Higher serum creatinine, BUN, CRP, neutrophil-to-lymphocyte ratio, and lower serum albumin, serum bicarbonate, and urine output were independently associated with the development of SA-AKI (p<0.05). The final prediction model demonstrated good discrimination (AUC=0.85) and satisfactory calibration (Hosmer–Lemeshow p=0.58).

Conclusion: A prediction model based solely on inexpensive and routinely available laboratory investigations effectively identified septic patients at high risk of developing acute kidney injury. The model may facilitate early risk stratification and timely intervention in resource-limited tertiary care hospitals in Pakistan.

Objective: To determine the frequency of MDR bloodstream infections and identify factors associated with their development among adults with newly diagnosed AML.

Methods: This multicenter analytical cross-sectional study was conducted over one year at three tertiary care hospitals in Pakistan. Adults with newly diagnosed AML and culture-confirmed BSI before or during induction chemotherapy were included. Clinical, laboratory, and microbiological data were collected, and antimicrobial susceptibility testing was performed according to CLSI guidelines. Multivariable logistic regression identified independent predictors of MDR BSI.

Results: Among 186 patients, 72 (38.7%) developed MDR bloodstream infections. Gram-negative bacteria, mainly Klebsiella pneumoniae, Escherichia coli, and Acinetobacter baumannii, predominated. Previous broad-spectrum antibiotic exposure, prolonged neutropenia, central venous catheterization, ICU admission, and prolonged hospitalization were independent risk factors. MDR infections were associated with longer hospital stay and increased in-hospital mortality.

Conclusion: MDR bloodstream infections are common among newly diagnosed AML patients in Pakistan. Early identification of high-risk patients, effective antimicrobial stewardship, and strengthened infection prevention strategies are essential to reduce infection-related morbidity and mortality.