Developing and validating a risk score model for prediction of acute kidney injury in non-ICU hospitalized patients

Background: Acute kidney injury (AKI) is a common complication in hospitalized patients. Several risk score models have been designed for intensive care unit (ICU) patients. We aim to establish a new risk prediction score for AKI patients in general wards. Materials and methods: All hospitalized ward patients who developed AKI were included in our study. To develop a new prediction score model we used the data collected from 107 patients who developed AKI. We used our prospective validation cohort (122 patients) to develop and validate this prediction score model of AKI. Results: Of 10,243 patients, 107 (1%) patients developed AKI 24 hours after admission to the general wards. Mortality rate was 26.2%. A score model of 15 points, based on clinical and laboratory data, was developed for prediction of AKI. We demonstrated a cutoff value >= 4 out of 15 as a predictor of AKI in non-ICU patients. The area under the receiver operating characteristic (AUC ROC) value of the score model was 0.950, 95% CI (confidence interval) and the p-value < 0.001 with sensitivity of 94.39 and specificity of 81.43. On applying this score model on a prospective group of patients (validation group n = 122), the AUC ROC value was 0.826. Conclusion: We developed and validated a new risk score model with a cutoff value >= 4 out of 15 for prediction of AKI in non-ICU patients. It will help in the early prediction of AKI in non-ICU patients.