Frequency and outcomes of acute renal failure following thoracic aortic stent-graft placement

Abstract

We assessed the incidence, correlates, and outcomes of acute renal failure (ARF) after thoracic aortic stent-graft placement. Postprocedural ARF is an inherent complication of catheter-based interventional procedures that use intra-arterial contrast agents and has adverse effects on short- and long-term outcomes. However, few data exist on the incidence, predictors, and outcomes of ARF after thoracic aortic stent-graft placement. We analyzed data of 97 patients (64.4 +/- 11.6 years of age; 73% men) who underwent thoracic aortic stent-graft placement from July 1999 to October 2005. Postprocedural ARF was defined as an increase > or = 25% and/or > or = 0.5 mg/dl in preprocedural serum creatinine at 48 hours after the procedure. Baseline estimated glomerular filtration rate was 65 +/- 24 ml/min/1.73 m2. Chronic kidney disease (glomerular filtration rate < or = 60 ml/min/1.73 m2) at baseline was present in 45% of patients. During the stent-graft procedure, patients received 307 +/- 188 ml of nonionic contrast medium. Postprocedural ARF occurred in 33 patients (34%), and 3 required dialysis. Multivariable analysis identified American Society of Anesthesiologists class > 3 (odds ratio 5.53, 95% confidence interval 1.71 to 17.85, p = 0.004) and duration of the stent-graft procedure (odds ratio 1.01, 95% confidence interval 1.001 to 1.014, p = 0.022) as independent predictors of postprocedural ARF. Compared with patients without ARF, those with ARF had markedly higher 30-day (18.2 +/- 6.7% vs 1.6 +/- 1.6%, p = 0.0022) and 1-year (35.2 +/- 8.6% vs 10.1 +/- 3.9%, p = 0.001) mortality. In conclusion, postprocedural ARF is a frequent complication of thoracic aortic stent-graft placement and has a significant adverse effect on 30-day and 1-year survival. Baseline factors identified in our study as associated with increased risk of ARF may facilitate a comprehensive informed consent process by way of patient education. In addition, identification of an at-risk subset may allow modification of reversible periprocedural factors that may help decrease postprocedural ARF.