- 作者：George A. Antoniou ; George S. Georgiadis ; Stavros A. Antoniou ; Polyvios Pavlidis ; Dimitrios Maras ; George S. Sfyroeras ; Efstratios I. Georgakarakos ; Miltos K. Lazarides
- 刊名：Journal of Vascular Surgery
- 出版年：2013
- 出版时间：October, 2013
- 年：2013
- 卷：58
- 期：4
- 页码：1091-1105
- 全文大小：3110 K
Background
Despite the intuitive advantages of endovascular repair (EVAR) of ruptured abdominal aortic aneurysms (AAAs), uncertainty remains about the optimal management in the absence of convincing high-quality evidence. Our objective was to undertake a comprehensive literature review and perform a meta-analysis of outcome data of treatment modalities for ruptured AAAs.Methods
Systematic searches were conducted of electronic information sources to identify studies comparing perioperative outcomes of EVAR and open repair for AAA rupture. Summary estimates of odds ratios (ORs) or standardized mean difference and 95 % confidence intervals (CIs) were obtained with a random-effects model. Meta-regression models were formed to explore potential heterogeneity as a result of changes in practice over time.
Results
We selected 41 studies for analysis. The entire meta-analysis population comprised 59,941 patients (8201 EVAR patients and 51,740 open repair patients). EVAR was associated with a significantly lower incidence of in-hospital mortality (OR, 0.56; 95 % CI, 0.50-0.64; P?< .01; meta-analysis of risk-adjusted observational studies and randomized controlled trials: OR, 0.58; 95 % CI, 0.46-0.73; P?< .01). EVAR patients had a significantly decreased risk of developing respiratory complications (OR, 0.59; 95 % CI, 0.49-0.69; P?< .01) and acute renal failure (OR, 0.65; 95 % CI, 0.55-0.78; P?< .01) and a trend toward a reduced incidence of cardiac complications (OR,??0.02; 95 % CI,??0.03 to 0.00; P?= .05) and mesenteric ischemia (OR, 0.66; 95 % CI, 0.44-1.00; P?= .05). Patients treated with EVAR had significantly less?requirements of intraoperative blood transfusion (standardized mean difference,??0.88; 95 % CI,??1.06 to??0.70; P?< .01). Random-effects meta-regression revealed no statistical evidence for an association between death and year of?publication (P?= .19).
Conclusions
Our analysis provides evidence to motivate the adoption of an EVAR-first policy in a nonelective setting and the establishment of standardized protocols for the management ruptured AAAs.