The majority of data on rAAA outcomes is derived from a variety of retrospective studies (case series, registries, large databases), which is common for emergent diagnoses such as this. Several authors have made efforts to correct for bias and confounding by using multivariable regression, propensity score matching, and subgroup analyses. Despite the retrospective nature of these data, the large body of published work has allowed for the performance of carefully structured meta-analyses and systematic reviews.

In 2008, Hoornweg and colleagues published a systematic review and meta-analysis of 116 studies from 1991 to 2006 on open repair of rAAA [15]. They sought to evaluate trends in mortality over time, as well as the effect of hospital volume on outcomes. Among approximately 60,000 patients, the weighted mean overall perioperative mortality was 48.5 % (95 % CI, 48.1–48.9 %). The intraoperative mortality was reported in 37 studies, and the weighted mean was 13.3 % (95 % CI, 12.3–14.3 %). Over the 33 years of study, there was no significant change in overall mortality as a function of time (1.6 % reduction, p = NS). Analysis of the 58 studies that included hospital volume revealed a significant positive association between annual hospital volume of rAAA repairs and overall mortality (p = .04). Unfortunately, the authors were unable to determine a threshold of rAAA hospital volume that is associated with improved mortality. Despite the limitations inherent to this study, the inclusion of over 60,000 patients provides valuable data on outcomes following open repair of rAAA (Fig. 17.2).

In 2008, Mastracci and colleagues published a systematic review and meta-analysis of 18 studies between 1994 and 2006 on endovascular repair (EVAR) of rAAA [13]. The pooled mortality was 21 % (95 % CI, 13–29), with a broad range of reported outcomes across studies (0–45 % mortality). In fact, the authors conclude that in the presence of such substantial heterogeneity, the aggregated calculated point estimate cannot be considered reliable. Subgroup analyses that included only series where >30 cases were reported demonstrated a reduction in heterogeneity and a mortality of 19 % (95 % CI, 10–28). While EVAR for rAAA is widely practiced, rigorous analysis and interpretation of its results have been difficult.

Several advantages of EVAR for rAAA repair compared to open surgery have been published, but the retrospective and other non-randomized study designs have been criticized for the potential selection bias that may be inherent to these studies. The primary concern has been that patients selected for EVAR may be more stable at their time of presentation, and this increased stability may be responsible for improved outcomes. In 2009, Hinchliffe and colleagues published a systematic review of the literature on EVAR for rAAA, specifically noting those that contained a “control” open repair group [12]. Despite having adopted EVAR for treatment of many patients with rAAA at the authors’ center, they conclude that the evidence for its benefit is lacking. Conflicting results have been reported, and questions regarding the comparability of the two patient groups remain (Table 17.1).