Chronic kidney disease classification stratifies mortality risk after elective stent graft repair of the thoracic aorta

Objective: Risk factors for perioperative and late mortality after thoracic endovascular aortic repair (TEVAR) remain ill-defined. In this study, we examined the prognostic significance of chronic kidney disease (CKD), a well-known predictor of death after thoracic aorta open repair, employing a stratification based on CKD stages derived from glomerular filtration rate (GFR) values. Methods: A prospective database was evaluated for 179 consecutive patients electively submitted to TEVAR between 1999 and 2007. Preoperative GFR was estimated by using the Cockcroft-Gault equation. Patient groups were stratified into four quartiles by baseline serum creatinine (SC) and GFR values, with quartile I being the lowest, and quartile IV the highest, and into the five CKD stages in reverse order (I GFR >= 90 ml/min/1.73 m(2); 1160-89; 11130-59; IV 15-29; V < 15). Prognostic significance of preoperative GFR values and CKD stages were investigated by means of univariate and multivariate analyses, and the Kaplan-Meier log-rank method. Results. A primary technical success was achieved in 166 of 179 patients (92.7%), and an initial clinical success in 158 (88.3%). Thirty-day mortality was 5% (nine cases). Paraplegia or paraparesis were observed in 11 (6.1%) patients, and completely resolved in six cases after cerebrospinal fluid drainage. Preoperative GFR quartiles and CKD stages were significant predictors of 30-day mortality (P=.004 and P<.0001 respectively), whereas SC quartiles; did not affect the outcome (P=.12). In particular, GFR quartile 1 (<60 ml/min/1.73 m2) was associated with a ten-fold greater risk of perioperative death compared with the other three quartiles (Odds Ratio 11.4, 95% Confidence Interval 2.3-57.0, P=.003). Midterm survival was 88.8% (159 of 179) at a mean follow-up of 35.6 +/- 23.7 months. Actuarial survival at 60 months was 57.8%, 81.1%, 92.3%, and 100% for GFR quartiles I to]IV respectively (P<.0001), and 0.0%, 66.7%, 59.2%, 88.6%, and 100% (P<.0001) for CKD stage V to I respectively. At univariate analyses, age (P =.019), preoperative SC quartiles (P=.001), GFR quartiles (P=.0002), and CKD stages (P<.0001)were all predictive of mid-term mortality. At multivariate Cox proportional hazards regression analysis, only CKD stages remained independently associated with the outcome (P=.008). Conclusions. GFR is an accurate prognostic predictor in patients submitted to TEVAR. Also, perioperative and midterm mortality directly correlate with the severity of CKD stages, allowing a risk stratification model to be employed both for risk-adjusted preoperative evaluation, and to establish accurate matching criteria for comparative studies. (J Vasc Surg 2009;49:296-301.)