A predictive model identifies patients at high risk for progression of acute kidney injury (AKI) to chronic kidney disease (CKD), according to a new study published in the Journal of the American Medical Association.
To develop the model, Matthew T. James, MD, PhD, of Foothills Medical Center, in Calgary, and his colleagues gathered data from 9973 Canadian patients (mean age 66 years) with a pre-hospitalization estimated glomerular filtration rate (eGFR) of more than 45 mL/min/1.73m2 and AKI (defined as a serum creatinine increase during hospitalization of 0.3 mg/dL or more or above 50% from baseline). The researchers then validated the risk models using data from a second cohort of 2761 hospitalized patients (mean age 69 years). Average baseline serum creatinine was 1.0 mg/dL, and more than one in 5 patients had stage 2 or 3 AKI.
Within a year of discharge, stage 4-5 CKD developed in 2.7% of the original cohort and 2.2% of the validation cohort. Dr James and the team found that certain factors independently related with CKD: older age, female sex, higher serum creatinine at baseline or discharge, albuminuria, and AKI severity. A model including these 6 variables demonstrated a C statistic of 0.81. It outperformed simpler models that included age, sex, and discharge serum creatinine alone. Discrimination improved by 2.6% and reclassification index by 13.5%. It also bested models including age, sex, and AKI stage alone. Discrimination improved by 8% and reclassification index by 79.9%.
“This model was able to predict advanced chronic kidney disease following hospitalization with acute kidney injury but requires evaluation of its utility in a clinical setting,” Dr James and colleagues stated. If confirmed, the model could improve the low rates of follow-up care in the outpatient setting.
James MT, Pannu N, Hemmelgarn BR, et al. Derivation and external Validation of prediction models for advanced chronic kidney disease following acute kidney injury. JAMA. 2017;318:1787-1797. doi:10.1001/jama.2017.16326