POS-031 Developing and validating a Risk Score Model for prediction of Acute Kidney Injury (AKI) in Non-ICU Hospitalized Patients

Abstract

IntroductionAbstractBackground: AKI is a common complication in hospitalized patients. Several risk score models have been designed for ICU patients. Our aim was to establish risk prediction score for AKI in general wards.MethodsAll hospitalized patients from 1/7/ 2019 to 31/12/2019 who developed AKI were included. We used our prospective cohort to develop and validate this prediction score model of AKI in general wards patients.ResultsA score model of 15 points, based on clinical and laboratory data, was developed for prediction of AKI. We demonstrated that a cutoff value ≥ 4 was a predictor of AKI. The AUC ROC value of score model was 0.950, 95% and P-value <0.001 with sensitivity of 94.39 and specificity of 81.43. By applying this score on validation group (n=122) showed AUC ROC value of 0.826.ConclusionsWe developed and validated a new risk score model with cutoff ≥ 4 for prediction of AKI in non-ICU hospitalized patients with a good sensitivity and specificity.No conflict of interest IntroductionAbstractBackground: AKI is a common complication in hospitalized patients. Several risk score models have been designed for ICU patients. Our aim was to establish risk prediction score for AKI in general wards. Abstract Background: AKI is a common complication in hospitalized patients. Several risk score models have been designed for ICU patients. Our aim was to establish risk prediction score for AKI in general wards. MethodsAll hospitalized patients from 1/7/ 2019 to 31/12/2019 who developed AKI were included. We used our prospective cohort to develop and validate this prediction score model of AKI in general wards patients. All hospitalized patients from 1/7/ 2019 to 31/12/2019 who developed AKI were included. We used our prospective cohort to develop and validate this prediction score model of AKI in general wards patients. ResultsA score model of 15 points, based on clinical and laboratory data, was developed for prediction of AKI. We demonstrated that a cutoff value ≥ 4 was a predictor of AKI. The AUC ROC value of score model was 0.950, 95% and P-value <0.001 with sensitivity of 94.39 and specificity of 81.43. By applying this score on validation group (n=122) showed AUC ROC value of 0.826. A score model of 15 points, based on clinical and laboratory data, was developed for prediction of AKI. We demonstrated that a cutoff value ≥ 4 was a predictor of AKI. The AUC ROC value of score model was 0.950, 95% and P-value <0.001 with sensitivity of 94.39 and specificity of 81.43. By applying this score on validation group (n=122) showed AUC ROC value of 0.826. ConclusionsWe developed and validated a new risk score model with cutoff ≥ 4 for prediction of AKI in non-ICU hospitalized patients with a good sensitivity and specificity.No conflict of interest We developed and validated a new risk score model with cutoff ≥ 4 for prediction of AKI in non-ICU hospitalized patients with a good sensitivity and specificity.