Study Evaluates Novel Tool for Prediction of Renal Complications in Patients Undergoing PCI

June 4, 2013—Hitinder S. Gurm, MD, et al published a study in the Journal of the American College of Cardiology (JACC) on a computational tool for predicting the risk of contrast-induced nephropathy (CIN), a common complication for patients undergoing percutaneous coronary intervention (PCI) that is associated with adverse short- and long-term outcomes.

According to the authors, this new computational tool can reliably calculate the risk of CIN and new requirement for dialysis (NRD) among patients undergoing PCI, and it may prove useful for risk adjustment and clinical decision making.

As stated by the authors in JACC, the study cohort was made up of 68,573 PCI procedures performed between January 2010 and June 2012 in 46 hospitals in Michigan. Seventy percent (48,001) were randomly selected for training the models, and 30% (20,572) were used for validation. Random forest models were developed using 46 preprocedural clinical and laboratory variables to estimate the risk of CIN. Fifteen of the most influential variables were selected for inclusion in a reduced model. Model performance estimating risk of CIN and NRD was evaluated in an independent validation data set using area under the receiver-operating characteristic curve (AUC), with net reclassification improvement used to compare full and reduced model CIN prediction after grouping in low-, intermediate-, and high-risk categories.