Authors:
SR Walsh, JR Boyle, AG Lynch, U Sadat, JP Carpenter, TY Tang, ME Gaunt
Journal name: 
J Vasc Surg
Citation info: 
47(6):1364-1370
Abstract: 
BACKGROUND: Suprarenal fixation is widely used in endovascular aneurysm repair. Numerous small, underpowered studies have concluded that it does not increase the risk of renal impairment compared with infrarenal fixation. A recent meta-analysis demonstrated that renal infarction is more common with suprarenal fixation, but the effect on renal function remains unclear. METHODS: Electronic abstract databases, article reference lists, and conference proceedings were searched for series reporting renal function data after suprarenal fixation. There was considerable study heterogeneity with respect to key factors such as pre-existing renal dysfunction and length of follow-up. Authors were contacted to obtain individual patient data for a pooled reanalysis using standardized criteria. RESULTS: Of 46 potentially relevant citations, only 11 were eligible for inclusion in the meta-analysis. Complete data sets were available for four studies (1065 patients), with a median follow-up of 33 months. Kaplan-Meier curves were constructed for postoperative renal impairment in the suprarenal fixation and infrarenal fixation groups and compared by the log-rank test. Median time free of renal impairment was 38.5 months in the infrarenal fixation group compared with 32.4 months in the suprarenal fixation group (P = .0038). However, to account for significant methodologic differences, further analysis was required using a Weibull regression model fitted in open Bayesian inference using Gibbs sampling (BUGS). The pooled hazard ratio for deterioration of renal function after suprarenal fixation was 0.6 (95% confidence interval, 0.3-10). CONCLUSION: Currently available data are insufficient to determine the precise effect of suprarenal fixation on medium-term renal function. Conventional Kaplan-Meier analysis of the pooled data set suggested that suprarenal fixation increased the risk of renal dysfunction; however, the effect disappeared when sophisticated statistical modelling was performed to account for study heterogeneity. A randomised controlled trial of suprarenal fixation may resolve this issue.
DOI: 
http://doi.org/10.1016/j.jvs.2007.11.029
E-pub date: 
01 Jun 2008