New model predicts acute kidney injury risk after cardiac surgery

A bedside prediction model provides a simple approach to identifying patients at high risk of developing acute kidney injury (AKI) requiring renal replacement therapy after cardiac surgery, reports a study in the Canadian Medical Association Journal.

The model was developed using prospectively collected data on 6061 patients undergoing cardiac surgery (other than transplantation) in Alberta between 2004 and 2009. Of these, 2.5% developed AKI requiring renal replacement therapy within 14 days after cardiac surgery.

Multivariable logistic regression identified eight independent predictors of AKI: congestive heart failure (adjusted odds ratio [OR] of 3.03), Canadian Cardiovascular Society angina class 3 or higher (OR of 1.66), diabetes (OR of 1.61), baseline eGFR (OR of 0.96 per 1-mL/min per 1.73 m2 increase), preoperative hemoglobin level (OR of 0.85 per 10-g/L increase), proteinuria (OR of 1.65), coronary artery bypass graft (CABG) plus valve surgery (OR of 1.25 versus CABG only), cardiac procedures other than CABG (OR of 3.11), and emergent surgery (OR of 4.63).

A model comprising these eight variables had excellent performance, with c statistics of 0.87 in the derivation cohort and 0.83 in a validation cohort of 4467 patients. Net reclassification improvement was 13.9% compared with the best existing prediction model (Cleveland Clinic Score).

On the basis of readily available clinical and laboratory data, the new model provides a practical and accurate tool for predicting the risk of AKI requiring renal replacement therapy after cardiac surgery. Although additional validation is needed, this simple score could be a useful aid in talking to patients about AKI risk before heart surgery [Pannu N, et al. A new model to predict acute kidney injury requiring renal replacement therapy after cardiac surgery. CMAJ 2016, in press].