Development of an easily applicable risk score model for contrast-induced nephropathy prediction after percutaneous coronary intervention: a novel approach tailored to current practice

Abstract

Background: Several risk factors for contrast-induced nephropathy (CIN) after percutaneous coronary intervention (PCI) have been identified. The cumulative effect of these risk factors on renal function has been assessed with the development of risk score models in a number of studies. However, concerns were raised that estimates of the risk attributable to individual factors may be unreliable. We sought to develop a simple risk score for developing CIN after PCI irrespective of use of prophylactic measures and also capturing the effect of pre-intervention medication and presence of various co-morbidities.

Methods: Consecutive patients treated with elective or urgent PCI at our cardiac catheterization laboratory were enrolled (derivation cohort n = 488, validation cohort n = 200). CIN was defined as increase ≥ 25% and/or ≥ 0.5 mg/dl in serum creatinine at 48 h after PCI vs baseline. Multivariable logistic regression analysis was then performed to identify independent predictors of CIN (pre-existing renal disease, metformin use, history of previous PCI, peripheral arterial disease and ≥ 300 ml of contrast volume).

Results: The incidence of CIN in the development cohort was 10.2% with a significant trend across increasing score values (p < 0.001). The model demonstrated good discriminating power (c-statistic 0.759) and excellent calibration (calibration slope 0.91). The model was validated internally by bootstrapping in 1000 samples (c-statistic 0.753) and in a cohort of 200 patients (c-statistic 0.864) demonstrating stable performance.

Conclusions: The proposed risk score is easily applicable and allows for practically simple risk assessment compared to other published scores while at the same time overcomes drawbacks of previous model designs.