Background: Acute renal failure (ARF) following percutaneous coronary intervention (PCI) has been shown to be associated with a worse outcome. Whether this event should be considered as a marker of disease severity or an independent contributor to mortality is still unclear.
Methods: In a multicenter, prospective cohort study we investigated the predictive variables and the impact of postprocedural ARF on 2-year all-cause mortality in 2860 consecutive patients (50% with stable angina and 50% with non-ST-elevation acute coronary syndromes) undergoing PCI. Serum creatinine determinations were made immediately before and 24 h after PCI. ARF was defined as an increase in serum creatinine of > or =0.5 mg/dl over baseline.
Results: One hundred and six patients (3.7%) experienced ARF. At logistic regression analysis, ARF was associated with pre-existing low values of estimated glomerular filtration rate, reduced left ventricular ejection fraction, hypertension, and prior coronary bypass surgery. Mortality data at 2 years were available for all patients: 119 patients (4.16%) had died, 3.9% of those without and 11.3% of those with ARF (univariate hazard ratio 3.16; 95% confidence interval 1.68-5.94; P = 0.0004). At Cox regression analysis, the significant predictors of mortality were age, ejection fraction, preprocedural estimated glomerular filtration rate, PCI failure, atrial fibrillation, diabetes mellitus, and fluoroscopy time. In this comprehensive mortality model, ARF maintained a borderline statistical significance (hazard ratio 1.83, 95% confidence interval 0.98-3.44; P = 0.06).
Conclusions: ARF following PCI occurs almost exclusively in patients with chronic kidney disease or left ventricular dysfunction. These risk factors are also among the most powerful predictors of long-term mortality and are likely to explain most of the association between postprocedural ARF and long-term mortality. After correction for clinical determinants, however, postprocedural ARF maintains a clinically significant impact on mortality that must be taken into account for benefit vs. risk evaluation of PCI in individual patients.