- Email: [email protected]
Endovascular repair of ruptured abdominal aortic aneurysm does not confer survival benefits over open repair
From the Eastern Vascular Society
Endovascular repair of ruptured abdominal aortic aneurysm does not confer survival benefits over open repair Naveed Saqib, MD,a Sun Cheol Park, MD,a Taeyoung Park, PhD,b Robert Y. Rhee, MD,a Rabih A. Chaer, MD,a Michel S. Makaroun, MD,a and Jae-Sung Cho, MD,a Pittsburgh, Pa; and Seoul, Korea Objective: Endovascular repair of ruptured abdominal aortic aneurysm (rAAA) is being increasingly performed despite lack of good evidence for its superiority. Other reported studies suffer from patient selection and publication bias with limited follow-up. This study is a single-center propensity score comparing early and midterm outcomes between open surgical repair (OSR) and endovascular repair of rAAA (REVAR). Methods: A retrospective review from January 2001 to November 2010 identified 312 patients who underwent rAAA repairs. Thirty-one patients with antecedent AAA repair and three with incomplete records were excluded, leaving 37 REVARs and 241 OSRs. Propensity score-based matching for sex, age, preoperative hemodynamic status, surgeon’s annual AAA volume, and preoperative cardiopulmonary resuscitation was performed in a 1:3 ratio to compare outcomes. Thirty-seven REVARs were matched with 111 OSRs. Late survival was estimated by Kaplan-Meier methods. Results: Operative time and blood replacement were higher with OSR. Overall complication rates were similar (54% REVAR vs 66% OSR; P ⴝ .23), except for higher incidences of tracheostomies (21% vs 3%; P ⴝ .015), myocardial infarction (38% vs 18%; P ⴝ .036), and acute tubular necrosis (47% vs 21%; P ⴝ .009) with OSR. Operative mortality rates were similar (22% REVAR vs 32% OSR), with an odds ratio of 0.63 for REVAR (95% confidence interval ⴝ [0.24, 1.48]; P ⴝ .40). No differences in the incidences for secondary interventions for aneurysm- or graft-related complications were noted (22% REVAR vs 22% OSR; P ⴝ .99). Kaplan-Meier estimates of 1-, 2-, and 3-year survival rates were also similar (50%, 50%, 42% REVAR vs 54%, 52%, 47% OSR; P ⴝ .66). Conclusions: REVAR for rAAA does not seem to conclusively confer either acute or late survival advantages. Routine use of REVAR should be deferred until prospective, randomized trial data become available. ( J Vasc Surg 2012;56:614-20.)
While the majority of ruptured abdominal aortic aneurysms (rAAAs) are repaired by open surgical method (OSR), endovascular aneurysm repair of rAAA (REVAR) is being increasingly used to treat rAAA. This enthusiasm for REVAR was primarily borne from excellent results reported in case reports and single-center series that have adopted a standardized protocol for REVAR,1-7 metaanalyses,8-10 and population-based studies.11-15 However, these reports invariably suffer from patient selection bias with uncontrolled confounding factors and limited follow-up.16-18 In addition, the selection of patients is poorly reported in cohort studies, and some studies fail to discriminate urgent/ symptomatic from ruptured aneurysms. Also, poor results are rarely published in the literature. Undoubtedly, these factors may contribute to superior results observed with REVAR over From the Division of Vascular Surgery, Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgha; and the Department of Applied Statistics, College of Business and Economics, Yonsei University, Seoul.b Author conflict of interest: none. Presented at the Twenty-fifth Annual Meeting of the Eastern Vascular Society, Washington, DC, September 22-24, 2011. Reprint requests: Jae-Sung Cho, MD, Division of Vascular Surgery, Department of Surgery, University of Pittsburgh School of Medicine, UPMC Presbyterian A1010, 200 Lothrop Street, Pittsburgh, PA 15213 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214/$36.00 Copyright © 2012 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2012.01.081
614
OSR. Despite these limitations and the lack of good evidence for its superiority, REVAR has been espoused as the preferential mode of therapy, such that even the utility of a prospective randomized trial is debated.16,19 The aim of this study is to evaluate whether REVAR confers early and midterm survival benefits compared with OSR using propensity score matching for confounding variables. METHODS Patient population. A retrospective review of all patients with ruptured infrarenal abdominal aortic aneurysm (rAAA) treated with REVAR or OSR between January 2001 and November 2011 at the University of Pittsburgh Medical Center (UPMC) was performed using a prospectively maintained database. Ruptured abdominal aortic aneurysm was defined as either presence of free peritoneal blood or retroperitoneal hematoma on laparotomy, or presence of blood outside the aortic wall on computed tomography scan. Three hundred twelve patients were treated for rAAA during the study period. Patients with antecedent abdominal aortic aneurysm (AAA) repairs (four previous open repairs and 27 previous endovascular repairs) and three patients with incomplete medical records were excluded from the study. Symptomatic patients without rupture were also excluded. The remaining 288 patients form the basis for this study; 37 patients underwent EVAR and 251 patients received OSR.
JOURNAL OF VASCULAR SURGERY Volume 56, Number 3
Saqib et al 615
Table I. Baseline characteristics of patients undergoing repair of ruptured abdominal aortic aneurysm Variable Patient demographics Age (years) Male History of COPD History of CVA History of CAD History of diabetes mellitus History of PVD History of hypertension History of CRF History of hemodialysis Current smokers Preoperative characteristics Preoperative SBP ⬍80 mm Hg Preoperative CPR ER to OR time (minutes) Time of presentation 3 AM - 6 AM 6 AM - 1 PM 3 PM - 5 PM 8 PM - midnight Blood work Mean hemoglobin Mean platelet count Mean creatinine
REVAR (n ⫽ 37) 74.9 ⫾ 8.2 70.3 (26/37) 27.1 (10/37) 37.8 (14/37) 56.7 (21/37) 40.5 (15/37) 37.8 (14/37) 81.8 (30/37) 8.0 (3/37) 3.0 (1/37) 43.2 (16/37)
OSR (n ⫽ 111)
P value
75.6 ⫾ 8.4 62.2 (69/111) 31.1 (35/111) 36.8 (41/111) 63.7 (70/111) 33.2 (37/111) 21.6 (24/111) 86.7(96/111) 14.7 (16/111) 3.6 (4/111) 46.4 (41/111)
.76 .48 .53 .85 .42 .65 .50 .77 .20 .80 .66
24.3 (9/37) 8.1 (3/37) 94 ⫾ 218
44.1 (49/111) 11.7 (13/111) 76 ⫾ 212
.03 .35 .09 .1
22% (8/37) 38% (14/37) 22% (8/37) 19% (7/37)
16% (18/111) 24% (27/111) 20% (22/111) 40% (44/111)
11.3 ⫾ 2.3 (37) 213 ⫾ 40 (37) 1.3 ⫾ 0.8 (37)
10.8 ⫾ 2.1 (99) 207 ⫾ 38 (99) 1.7 ⫾ 1.0 (95)
.50 .50 .06
CAD, Coronary artery disease; COPD, chronic obstructive pulmonary disease; CPR, cardiopulmonary resuscitation; CRF, chronic renal failure; CVA, cerebrovascular accident; ER, emergency room; OR, operating room; OSR, Open surgical repair; PVD, peripheral vascular disease; rAAA, ruptured abdominal aortic aneurysm; REVAR, endovascular repair of rAAA; SBP, systolic blood pressure.
After Institutional Review Board approval, demographic data, comorbidities, physiologic data at presentation, time of presentation, time interval between presentation and treatment, operative data, and postoperative data were collected from medical records for all patients. Case selection. The UPMC is a regional tertiary referral hospital with high volume of open and emergent aortic surgery cases. Selection of treatment in each case was made by the vascular surgeon based on aortic morphology, time of presentation, hemodynamic stability, and the surgeon’s own experience. Propensity score-based matching. To reduce the confounding effects of covariate imbalance between OSR and REVAR and selection bias, propensity score-based matching was performed. The variables used for matching included sex, gender, surgeon, hemodynamic stability, and preoperative cardiopulmonary resuscitation. Patients who underwent REVAR were matched to patients who underwent OSR in a 1:3 ratio due to relative sample sizes of REVAR and OSR patients and their distributions of propensity scores. Hence, 37 patients with REVAR were matched with 111 patients with OSR, and their outcomes were analyzed. Outcomes. All intraoperative, in-hospital and 30-day mortality, postoperative morbidity, length of stay, and transfusion of blood products were recorded. The Social Security Death Index was used to evaluate the long-term mortality. Statistical analysis. Kaplan-Meier analysis was used to estimate the late survival rates of both patient groups. Treatment selection bias was controlled by using propen-
sity score-based matching for sex, age, preoperative hemodynamic status, and surgeon’s annual AAA volume. All baseline variables were included in logistic regression model to predict the probability that patient would receive REVAR vs OSR upon presentation with rAAA. The fit of the propensity score model to the data was assessed using concordance index. RESULTS The demographic data, comorbidities, and preoperative characteristics in each treatment group are illustrated in Table I. There were 26 males out of 37 patients (70.3%) treated with REVAR and 69 males (62.2%) in 111 patients who underwent OSR (P ⫽ .48). The mean age was 74.9 years in REVAR group and 75.6 years in OSR group (P ⫽ .76). Patients presenting with hemodynamic instability, defined as systolic blood pressure ⬍80 mm Hg, were more likely to undergo OSR (P ⫽ .03). Forty-nine patients (44%) who underwent OSR were hemodynamically unstable on presentation, compared with only 24% of patients who underwent REVAR (nine patients). This difference results from the fact that the propensity score-based matching balances the probability of being treated with REVAR as a function of multiple covariates, not each covariate directly. In addition, due to the small sample size, it was not feasible to find exact matching between the groups. The delay from presentation until treatment was not statistically significant, but there was a trend toward delay in the REVAR group; the time of presentation did not differ, either.
JOURNAL OF VASCULAR SURGERY September 2012
616 Saqib et al
Table II. Intraoperative variables Variable
REVAR
OR time Intraoperative fluid (mL) Intraoperative colloids (mL) Intraoperative pRBC (units) Intraoperative FFP (6 packs)
OSR
P value
138 198 2183 3594 566 1290 3.06 6.86 0.625 3.89
.0001 .0007 .02 .02 .0001
FFP, Fresh frozen plasma; OR, operating room; OSR, open surgical repair; pRBC, packed red blood cells; rAAA, ruptured abdominal aortic aneurysm; REVAR, endovascular repair of rAAA.
Table III. Comparison of morbidity and mortality after REVAR and OSR
Mortality Morbidity LOS (days) MI ARI Hemodialysis Pneumonia Tracheostomy ACS Bowel resection Postop hemorrhage
REVAR
OSR
P value
21.6 (8/37) 54 (20/37) 7.4 ⫾ 17.6 (7/37) 21 (8/37) 5.9 (2/37) 8.8 (3/37) 2.9 (1/37) 11.4 (4/37) 8.6 (3/37) 5.7 (2/37)
31.5 (35/111) 66 (73/111) 23.2 ⫾ 37.7 (42/111) 47 (52/111) 24.5 (27/111) 24.5 (27/111) 20.8 (23/111) 11.3 (12/111) 10.4 (11/111) 10.4 (11/111)
.35 .23 .13 .03 .008 .02 .05 .015 1 1 .51
ACS, Abdominal compartment syndrome; ARI, acute renal injury; LOS, length of stay; MI, myocardial infarction; OSR, open surgical repair; rAAA, ruptured abdominal aortic aneurysm; REVAR, endovascular repair of rAAA.
Intraoperative variables are detailed in Table II. The operative time was significantly longer in the OSR group with higher fluid and blood product transfusion requirements. The length of stay in the intensive care unit and hospital stay were lower following REVAR, but these were not statistically significant (P ⫽ .13). The REVAR group had a 7.3-day mean and 4.5-day median length of stay, compared with a 23-day mean and 10.4-day median length of stay in the OSR group. The operative mortality rates, defined as intra-operative, in-hospital, and 30-day mortality, were not statistically different between the two groups (21.6% REVAR vs 31.5% OSR; P ⫽ .34). To account for the differences in the proportion of patients with hemodynamic instability (HI), a logistic regression analysis was performed with and without adjusting for HI. The difference in mortality between REVAR and OSR adjusting for HI was not significant (P ⫽ .51 with the estimated difference in log odds of mortality being ⫺.31). Overall complication rates were similar in both groups (54% REVAR vs 66% OSR; P ⫽ .23) (Table III). However, pulmonary failure requiring tracheostomy, myocardial infarction, and acute tubular necrosis occurred more frequently after OSR. The incidences of abdominal compartment syndrome and bleeding requiring return to the operating room did not differ. At a mean of 3-year follow-up, the aneurysm-, graft-, or procedure-related secondary interventions were similar in
both groups: nine patients (25%) after REVAR vs 26 patients (23.4%) following OSR (P ⫽ .99) (Table IV). Kaplan-Meier analysis (Fig) revealed no differences in late survival at 1, 2, and 3 years (P ⫽ .66). The probability of survival after REVAR was 50%, 50%, and 42%, respectively. The corresponding figures for OSR were 54%, 52%, and 47%, respectively. Following OSR, aneurysmrelated deaths occurred in 35 of 55 deaths in the first year; 14 deaths in year 2 and year 3 each were nonaneurysmrelated deaths. Following REVAR, there were nine out of 12 aneurysm-related deaths in the first year, three deaths not related to aneurysm in the second post-REVAR year, and four deaths not related to aneurysm during the third post-REVAR year. DISCUSSION The perceived benefits of REVAR and improved results from selected studies1-5 have led some investigators to promote an “endovascular-first” approach in the treatment of rAAA2,20 since the first report of successful REVAR by Yusuf et al in 1994.21 In fact, the utilization of REVAR has steadily increased in the United States; the frequency of REVAR increased over the 6-year study period from 5.9% in 2001 to 18.9% in 200612; a similar trend was also observed in the present study. Such a trend was accompanied by a decrease in mortality rates from 43% to 29% in a review of a Medicare database.22 The literature is replete with varying results after REVAR.17 Some publications have shown no difference in early outcomes between REVAR and OSR,3,23,24 while some have shown superior results with REVAR. While these improved results can be found in propensity scorematched13 or large population-based studies,11,12 the most salient finding in review of the literature is that centers with an instituted protocol are likely to have superior outcomes with REVAR.2,4,5,7 In a recent review of collected world experience with REVAR, the 30-day mortality was 21.2% with REVAR compared with 36.3% with OSR; in a subgroup analysis of 13 centers with standardized protocols with preferential REVAR the rate was 19.7%.5 Not all centers with protocols are successful, however. A recent international multicenter study with preferential REVAR failed to show early survival benefit even with heavy selection bias toward OSR for patients with cardiovascular instability.24 A recent survey of the National Surgical Quality Improvement Program (NSQIP) database showed no statistically significant reduction in mortality rate with REVAR.25 The same findings were noted in the current study. While the lack of differences in the current study may be due to the low mortality rate with OSR for rAAA observed in our institution26 and type II errors due to the small sample size, the results resonate the conclusion of a large populationbased study and propensity score-matched studies that stated “institutional and surgeon experience were essential to fully benefit from the use of this procedure for the treatment of rAAA” and advocated for regionalization of care to high-volume centers.11-13 It should also be noted that administrative databases are limited by their inability to
JOURNAL OF VASCULAR SURGERY Volume 56, Number 3
Saqib et al 617
Table IV. Early and late secondary interventions after REVAR and OSR Complications Procedure-related Re-exploration for ACS/hemorrhagea Herniorrhaphy Colectomy for ischemia Lysis of adhesion Cholecysectomy Small bowel resection Groin hematoma Graft-related Thrombosis Graft infection Endoleak Vascular-related Iliac PTA/stenting Total
Post-OSR interventions (n ⫽ 26 patients)
Post-REVAR interventions (n ⫽ 9 patients)
12/26 (46%) 4/26 (15%) 8/26 (31%) 3/26 (12%) 1/26 (4%) 3/26 (12%) N/A
4/9 (44%) N/A 3/9 (33%) N/A 0 0 2/9 (22%)
2/26 (7%) 2/26 (7%) N/A
4/9 (44%) 0 3/9 (33%)
1/26 (4%) 26/111 (23%)
1/9 (11%) 9/37 (25%)
P
.99
ACS, Abdominal compartment syndrome; OSR, open surgical repair; PTA, percutaneous transluminal angioplasty; rAAA, ruptured abdominal aortic aneurysm; REVAR, endovascular repair of rAAA. Some patients may have more than one reintervention. a Includes splenectomy.
Fig. Kaplan-Meier curves of survival rates after endovascular repair of ruptured abdominal aortic aneurysm (rAAA) (REVAR) and open surgical repair (OSR) of rAAAs.
control for factors such as hemodynamic stability, surgeon’s experience and preference, suitability for REVAR, referral pattern, and delay between presentation and treatment. It is likely that the OSR mortality rates reported in the literature are adversely affected by unfavorable hemodynamic and anatomic characteristics, a phenomenon that has been well documented. That there was a significantly higher proportion of patients with hemodynamic instability in the OSR group may explain, at least in part, the higher (although statistical significance was not reached) mortality rate following OSR observed in the current study. Had the current study been matched for hemodynamic stability, the mortality rate with OSR would be lower and the gap narrower; the estimated difference in mortality rates after adjusting for hemodynamic instability became even smaller.
Multiple meta-analyses, systematic reviews, and populationbased studies have shown survival advantages with REVAR. However, all these studies are flawed with heavy patient selection and publication biases, poorly controlled confounding factors, and the lack of standardized reporting.27 Recognizing these limitations, the conclusions of these studies state that there is insufficient evidence to recommend widespread adoption of REVAR and acknowledge the need for a large prospective, multicenter randomized trial,9 in light of the fact that two earlier prospective randomized trials (the Nottingham and the Amsterdam Acute Aneurysm [AJAX] trials) have failed to show survival benefits with REVAR.27-29 However, some refute the validity of such a trial.16
JOURNAL OF VASCULAR SURGERY September 2012
618 Saqib et al
There have been four prospective, randomized trials that have either been conducted or are currently underway. The Nottingham trial failed to show any reduction in mortality with REVAR and was terminated due to poor recruitment.17 The AJAX and the Endovasculaire vs Chirurgie dans les Anévrysmes Rompus (ECAR) trials have been designed to recruit hemodynamically stable REVARsuitable patients. These trials, while avoiding the ethical dilemmas of randomizing hemodynamically unstable patients, will not be able to assess the role of an endovascularfirst strategy vs open repair on an intention-to-treat basis since randomization occurs only after EVAR eligibility is determined by a computed tomography scan. The AJAX trial also failed to demonstrate any survival benefits with REVAR and has been extended to recruit 120 patients, increased from its original 80-patient recruitment goal.27 The Immediate Management of the Patient with Rupture: Open Versus Endovascular repair (IMPROVE) trial is a large trial, planned to recruit 600 patients, that randomizes patients at the time of clinical diagnosis of rAAA regardless of cardiovascular stability and before REVAR eligibility is determined. As such, it will address the question of whether or not the endovascular-first strategy provides survival benefits over OSR.30 It is powered to show a difference of a 14% 30-day mortality rate and will identify factors that are predictive of survival benefits.27,30 Widespread adoption of the endovascular-first approach for the treatment of rAAA at first glance is attractive. It is minimally invasive and associated with lower complication rates, as shown in the current study. However, implementation of such a protocol involves massive systemic changes, such as stocking of endovascular stent grafts and auxiliaries and around-the-clock availability of skilled endovascular, radiologic, and nursing teams. It is not pragmatic to undertake such systemic and systematic changes in the absence of clear evidence of REVAR’s superiority. As such, an institutional protocol for REVAR has not been established at our institutions; given the size of the UPMC system, this would represent a monumental task in a variety of ways. Furthermore, REVAR is not generalizable at present. At best, about 50% of patients with rAAA may be treated with REVAR.25 Cardiovascular stability plays a major role in selection bias and outcome. Mehta et al31 revealed that the mortality rate following REVAR for hemodynamically unstable patients was 41%, similar to their open repair mortality rate of 40%.32 They also noted that mortality rates after REVAR for women (33%) and octogenarians (38%) were significantly higher than those for men (17%) and those ⬍80 years of age (15%). Furthermore, REVAR can cause an increased risk of complications such as abdominal compartment syndrome, which has been shown to be a harbinger of poor outcome with mortality in excess of 60%.31,33,34 A recent study also showed an increased risk of spinal cord ischemia with occlusion (temporary or permanent) of the hypogastric artery.35 The benefit of REVAR was not observed all across the hospitals. While a reduction in mortality was observed in teaching hospitals with
REVAR, a higher risk of in-hospital death was noted in community facilities.22,24 Identification of patient characteristics, aneurysm morphology, and intraoperative maneuvers that may influence the outcomes of REVAR is paramount. In regard to midterm survival rates, the early survival benefit does not seem to provide lasting effects. Survival probabilities up to 3 years were similar between the two groups in the present study. This is consistent with the findings of earlier reports.24,36,37 It has been well established that a sustained risk of mortality exists even after successful open repair of ruptured abdominal and thoracoabdominal aortic aneurysm,26,38,39 and patients appear to be susceptible to it after REVAR. There is still equipoise on what the best treatment for rAAA is. The data are conflicting, and many studies showing improved outcomes with REVAR employed it preferentially for patients with favorable anatomy and hemodynamic stability.18,37 Preferential REVAR should be deferred until results of prospective, randomized trials are available and predictive factors for its success are identified. Some patients may be harmed by indiscriminate insistence on REVAR. There are some limitations to this study inherent to its retrospective nature with relatively small sample size. Despite propensity score-based matching to reduce selection bias, due to the small number of patients, complete matching was not feasible. As such, there may be a selection bias on the choice of treatment rendered. The propensity scorebased matching can only remove overt bias, but unlike randomization, it cannot be expected to remove hidden biases. Additionally, the propensity score-based method is based on the hypothesis that it balances only for the covariates that were used to construct the score. Including irrelevant covariates in propensity score-based matching model may reduce efficiency. CONCLUSIONS Endovascular repair for rAAA does not seem to conclusively confer either acute or late survival advantages over open repair. Wide adoption and routine use of REVAR should be deferred until there is good evidence for its superiority. AUTHOR CONTRIBUTIONS Conception and design: JC Analysis and interpretation: JC, TP, NS Data collection: NS, JC Writing the article: NS, TP, JC Critical revision of the article: NS, TP, JC Final approval of the article: JC, NS, TP Statistical analysis: TP Obtained funding: Not applicable Overall responsibility: JC
JOURNAL OF VASCULAR SURGERY Volume 56, Number 3
REFERENCES 1. Hechelhammer L, Lachat ML, Wildermuth S, Bettex D, Mayer D, Pfammatter T. Midterm outcome of endovascular repair of ruptured abdominal aortic aneurysms. J Vasc Surg 2005;41:752-7. 2. Mehta M, Taggert J, Darling RC III, Chang BB, Kreienberg PB, Paty PS, et al. Establishing a protocol for endovascular treatment of ruptured abdominal aortic aneurysms: outcomes of a prospective analysis. J Vasc Surg 2006;44:1-8. 3. Coppi G, Silingardi R, Gennai S, Saitta G, Ciardullo AV. A single-center experience in open and endovascular treatment of hemodynamically unstable and stable patients with ruptured abdominal aortic aneurysms. J Vasc Surg 2006;44:1140-7. 4. Moore R, Nutley M, Cina CS, Motamedi M, Faris P, Abuznadah W. Improved survival after introduction of an emergency endovascular therapy protocol for ruptured abdominal aortic aneurysms. J Vasc Surg 2007;45:443-50. 5. Veith FJ, Lachat M, Mayer D, Malina M, Holst J, Mehta M, et al. Collected world and single center experience with endovascular treatment of ruptured abdominal aortic aneurysms. Ann Surg 2009;250: 818-24. 6. Mayer D, Pfammatter T, Rancic Z, Hechelhammer L, Wilhelm M, Veith FJ, et al. Ten years of emergency endovascular aneurysm repair for ruptured abdominal aortoiliac aneurysms: lessons learned. Ann Surg 2009;249:510-5. 7. Starnes BW, Quiroga E, Hutter C, Tran NT, Hatsukami T, Meissner M, et al. Management of ruptured abdominal aortic aneurysm in the endovascular era. J Vasc Surg 2010;51:9-17. 8. Sadat U, Boyle JR, Walsh SR, Tang T, Varty K, Hayes PD. Endovascular vs open repair of acute abdominal aortic aneurysms–a systematic review and meta-analysis. J Vasc Surg 2008;48:227-36. 9. Mastracci TM, Garrido-Olivares L, Cina CS, Clase CM. Endovascular repair of ruptured abdominal aortic aneurysms: a systematic review and meta-analysis. J Vasc Surg 2008;47:214-21. 10. Rayt HS, Sutton AJ, London NJ, Sayers RD, Bown MJ. A systematic review and meta-analysis of endovascular repair (EVAR) for ruptured abdominal aortic aneurysm. Eur J Vasc Endovasc Surg 2008;36: 536-44. 11. Egorova N, Giacovelli J, Greco G, Gelijns A, Kent CK, McKinsey JF. National outcomes for the treatment of ruptured abdominal aortic aneurysm: comparison of open versus endovascular repairs. J Vasc Surg 2008;48:1092-100. 12. McPhee J, Eslami MH, Arous EJ, Messina LM, Schanzer A. Endovascular treatment of ruptured abdominal aortic aneurysms in the United States (2001-2006): a significant survival benefit over open repair is independently associated with increased institutional volume. J Vasc Surg 2009;49:817-26. 13. Holt PJ, Karthikesalingam A, Poloniecki JD, Hinchliffe RJ, Loftus IM, Thompson MM. Propensity scored analysis of outcomes after ruptured abdominal aortic aneurysm. Br J Surg 2010;97:496-503. 14. Greco G, Egorova N, Anderson PL, Gelijns A, Moskowitz A, Nowygrod R, et al. Outcomes of endovascular treatment of ruptured abdominal aortic aneurysms. J Vasc Surg 2006;43:453-9. 15. Giles KA, Pomposelli F, Hamdan A, Wyers M, Jhaveri A, Schermerhorn ML. Decrease in total aneurysm-related deaths in the era of endovascular aneurysm repair. J Vasc Surg 2009;49:543-50. 16. Veith FJ, Powell JT, Hinchliffe RJ. Is a randomized trial necessary to determine whether endovascular repair is the preferred management strategy in patients with ruptured abdominal aortic aneurysms? J Vasc Surg 2010;52:1087-93. 17. Hinchliffe RJ, Powell JT, Cheshire NJ, Thompson MM. Endovascular repair of ruptured abdominal aortic aneurysm: a strategy in need of definitive evidence. J Vasc Surg 2009;49:1077-80. 18. Lee RW, Rhodes JM, Singh MJ, Davies MG, Wolford HY, Diachun C, et al. Is there a selection bias in applying endovascular aneurysm repair for rupture? Ann Vasc Surg 2008;22:215-20. 19. Veith FJ, Gargiulo NJ. Endovascular aortic repair should be the gold standard for ruptured AAAs, and all vascular surgeons should be prepared to perform them. Perspect Vasc Surg Endovasc Ther 2007;19: 275-82.
Saqib et al 619
20. Veith FJ, Ohki T. Endovascular approaches to ruptured infrarenal aorto-iliac aneurysms. J Cardiovasc Surg (Torino) 2002;43:369-78. 21. Yusuf SW, Whitaker SC, Chuter TA, Wenham PW, Hopkinson BR. Emergency endovascular repair of leaking aortic aneurysm. Lancet 1994;344:1645. 22. Lesperance K, Andersen C, Singh N, Starnes B, Martin MJ. Expanding use of emergency endovascular repair for ruptured abdominal aortic aneurysms: disparities in outcomes from a nationwide perspective. J Vasc Surg 2008;47:1165-70. 23. Anain PM, Anain JM Sr, Tiso M, Nader ND, Dosluoglu HH. Early and midterm results of ruptured abdominal aortic aneurysms in the endovascular era in a community hospital. J Vasc Surg 2007;46:898-905. 24. Peppelenbosch N, Geelkerken RH, Soong C, Cao P, Steinmetz OK, Teijink JA, et al. Endograft treatment of ruptured abdominal aortic aneurysms using the Talent aortouniiliac system: an international multicenter study. J Vasc Surg 2006;43:1111-23. 25. Davenport DL, O’Keeffe SD, Minion DJ, Sorial EE, Endean ED, Xenos ES. Thirty-day NSQIP database outcomes of open versus endoluminal repair of ruptured abdominal aortic aneurysms. J Vasc Surg 2010;51:305-9. 26. Cho JS, Kim JY, Rhee RY, Gupta N, Marone LK, Dillavou ED, et al. Contemporary results of open repair of ruptured abdominal aortoiliac aneurysms: effect of surgeon volume on mortality. J Vasc Surg 2008;48:10-7. 27. Powell JT. Time to IMPROVE the management of ruptured abdominal aortic aneurysm: IMPROVE trialists. Eur J Vasc Endovasc Surg 2009; 38:237-8. 28. Hinchliffe RJ, Bruijstens L, MacSweeney ST, Braithwaite BD. A randomised trial of endovascular and open surgery for ruptured abdominal aortic aneurysm–results of a pilot study and lessons learned for future studies. Eur J Vasc Endovasc Surg 2006;32:506-13. 29. Hoornweg LL, Wisselink W, Vahl A, Balm R. The Amsterdam Acute Aneurysm Trial: suitability and application rate for endovascular repair of ruptured abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 2007;33:679-83. 30. Powell JT, Thompson SG, Thompson MM, Grieve R, Nicholson AA, Ashleigh R, et al. The immediate management of the patient with rupture: open versus endovascular repair (IMPROVE) aneurysm trial–ISRCTN 48334791 IMPROVE trialists. Acta Chir Belg 2009;109:678-80. 31. Mehta M, Roddy SP, Sternbach Y, Taggert J, Kreienberg PB, Paty PS, et al. Outcomes of EVAR in hemodynamically stable and unstable patients with ruptured AAA: a prospective analysis. J Vasc Surg 2011;53:554-5. 32. Mehta M, Roddy SP, Paty PS, Kreienberg PB, Ozsvath K, Sternbach Y, et al. Factors impacting survival after endovascular repair of ruptured abdominal arotic aneurysms. J Vasc Surg 2011;54:920. 33. Mehta M, Darling RC III, Roddy SP, Fecteau S, Ozsvath KJ, Kreienberg PB, et al. Factors associated with abdominal compartment syndrome complicating endovascular repair of ruptured abdominal aortic aneurysms. J Vasc Surg 2005;42:1047-51. 34. Djavani GK, Wanhainen A, Bjorck M. Intra-abdominal hypertension and abdominal compartment syndrome after endovascular repair of ruptured abdominal aortic aneurysm. Eur J Vasc Endovasc Surg 2011; 41:742-7. 35. Peppelenbosch N, Cuypers PW, Vahl AC, Vermassen F, Buth J. Emergency endovascular treatment for ruptured abdominal aortic aneurysm and the risk of spinal cord ischemia. J Vasc Surg 2005;42:608-14. 36. Hinchliffe RJ, Braithwaite BD. Ruptured abdominal aortic aneurysm: endovascular repair does not confer any long-term survival advantage over open repair. Vascular 2007;15:191-6. 37. Verhoeven EL, Kapma MR, Groen H, Tielliu IF, Zeebregts CJ, Bekkema F, et al. Mortality of ruptured abdominal aortic aneurysm treated with open or endovascular repair. J Vasc Surg 2008;48:1396-400. 38. Cho JS, Gloviczki P, Martelli E, Harmsen WS, Landis ME, Cherry KJ Jr, et al. Long-term survival and late complications after repair of ruptured abdominal aortic aneurysms. J Vasc Surg 1998;27:813-9; discussion: 819-20. 39. Barbato JE, Kim JY, Zenati M, Abu-Hamad G, Rhee RY, Makaroun MS, et al. Contemporary results of open repair of ruptured descending thoracic and thoracoabdominal aortic aneurysms. J Vasc Surg 2007;45:667-76. Submitted Oct 14, 2011; accepted Jan 3, 2012.
Endovascular repair of ruptured abdominal aortic aneurysm does not confer survival benefits over open repair Naveed Saqib, MD,a Sun Cheol Park, MD,a Taeyoung Park, PhD,b Robert Y. Rhee, MD,a Rabih A. Chaer, MD,a Michel S. Makaroun, MD,a and Jae-Sung Cho, MD,a Pittsburgh, Pa; and Seoul, Korea Objective: Endovascular repair of ruptured abdominal aortic aneurysm (rAAA) is being increasingly performed despite lack of good evidence for its superiority. Other reported studies suffer from patient selection and publication bias with limited follow-up. This study is a single-center propensity score comparing early and midterm outcomes between open surgical repair (OSR) and endovascular repair of rAAA (REVAR). Methods: A retrospective review from January 2001 to November 2010 identified 312 patients who underwent rAAA repairs. Thirty-one patients with antecedent AAA repair and three with incomplete records were excluded, leaving 37 REVARs and 241 OSRs. Propensity score-based matching for sex, age, preoperative hemodynamic status, surgeon’s annual AAA volume, and preoperative cardiopulmonary resuscitation was performed in a 1:3 ratio to compare outcomes. Thirty-seven REVARs were matched with 111 OSRs. Late survival was estimated by Kaplan-Meier methods. Results: Operative time and blood replacement were higher with OSR. Overall complication rates were similar (54% REVAR vs 66% OSR; P ⴝ .23), except for higher incidences of tracheostomies (21% vs 3%; P ⴝ .015), myocardial infarction (38% vs 18%; P ⴝ .036), and acute tubular necrosis (47% vs 21%; P ⴝ .009) with OSR. Operative mortality rates were similar (22% REVAR vs 32% OSR), with an odds ratio of 0.63 for REVAR (95% confidence interval ⴝ [0.24, 1.48]; P ⴝ .40). No differences in the incidences for secondary interventions for aneurysm- or graft-related complications were noted (22% REVAR vs 22% OSR; P ⴝ .99). Kaplan-Meier estimates of 1-, 2-, and 3-year survival rates were also similar (50%, 50%, 42% REVAR vs 54%, 52%, 47% OSR; P ⴝ .66). Conclusions: REVAR for rAAA does not seem to conclusively confer either acute or late survival advantages. Routine use of REVAR should be deferred until prospective, randomized trial data become available. ( J Vasc Surg 2012;56:614-20.)
While the majority of ruptured abdominal aortic aneurysms (rAAAs) are repaired by open surgical method (OSR), endovascular aneurysm repair of rAAA (REVAR) is being increasingly used to treat rAAA. This enthusiasm for REVAR was primarily borne from excellent results reported in case reports and single-center series that have adopted a standardized protocol for REVAR,1-7 metaanalyses,8-10 and population-based studies.11-15 However, these reports invariably suffer from patient selection bias with uncontrolled confounding factors and limited follow-up.16-18 In addition, the selection of patients is poorly reported in cohort studies, and some studies fail to discriminate urgent/ symptomatic from ruptured aneurysms. Also, poor results are rarely published in the literature. Undoubtedly, these factors may contribute to superior results observed with REVAR over From the Division of Vascular Surgery, Department of Surgery, University of Pittsburgh School of Medicine, Pittsburgha; and the Department of Applied Statistics, College of Business and Economics, Yonsei University, Seoul.b Author conflict of interest: none. Presented at the Twenty-fifth Annual Meeting of the Eastern Vascular Society, Washington, DC, September 22-24, 2011. Reprint requests: Jae-Sung Cho, MD, Division of Vascular Surgery, Department of Surgery, University of Pittsburgh School of Medicine, UPMC Presbyterian A1010, 200 Lothrop Street, Pittsburgh, PA 15213 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214/$36.00 Copyright © 2012 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2012.01.081
614
OSR. Despite these limitations and the lack of good evidence for its superiority, REVAR has been espoused as the preferential mode of therapy, such that even the utility of a prospective randomized trial is debated.16,19 The aim of this study is to evaluate whether REVAR confers early and midterm survival benefits compared with OSR using propensity score matching for confounding variables. METHODS Patient population. A retrospective review of all patients with ruptured infrarenal abdominal aortic aneurysm (rAAA) treated with REVAR or OSR between January 2001 and November 2011 at the University of Pittsburgh Medical Center (UPMC) was performed using a prospectively maintained database. Ruptured abdominal aortic aneurysm was defined as either presence of free peritoneal blood or retroperitoneal hematoma on laparotomy, or presence of blood outside the aortic wall on computed tomography scan. Three hundred twelve patients were treated for rAAA during the study period. Patients with antecedent abdominal aortic aneurysm (AAA) repairs (four previous open repairs and 27 previous endovascular repairs) and three patients with incomplete medical records were excluded from the study. Symptomatic patients without rupture were also excluded. The remaining 288 patients form the basis for this study; 37 patients underwent EVAR and 251 patients received OSR.
JOURNAL OF VASCULAR SURGERY Volume 56, Number 3
Saqib et al 615
Table I. Baseline characteristics of patients undergoing repair of ruptured abdominal aortic aneurysm Variable Patient demographics Age (years) Male History of COPD History of CVA History of CAD History of diabetes mellitus History of PVD History of hypertension History of CRF History of hemodialysis Current smokers Preoperative characteristics Preoperative SBP ⬍80 mm Hg Preoperative CPR ER to OR time (minutes) Time of presentation 3 AM - 6 AM 6 AM - 1 PM 3 PM - 5 PM 8 PM - midnight Blood work Mean hemoglobin Mean platelet count Mean creatinine
REVAR (n ⫽ 37) 74.9 ⫾ 8.2 70.3 (26/37) 27.1 (10/37) 37.8 (14/37) 56.7 (21/37) 40.5 (15/37) 37.8 (14/37) 81.8 (30/37) 8.0 (3/37) 3.0 (1/37) 43.2 (16/37)
OSR (n ⫽ 111)
P value
75.6 ⫾ 8.4 62.2 (69/111) 31.1 (35/111) 36.8 (41/111) 63.7 (70/111) 33.2 (37/111) 21.6 (24/111) 86.7(96/111) 14.7 (16/111) 3.6 (4/111) 46.4 (41/111)
.76 .48 .53 .85 .42 .65 .50 .77 .20 .80 .66
24.3 (9/37) 8.1 (3/37) 94 ⫾ 218
44.1 (49/111) 11.7 (13/111) 76 ⫾ 212
.03 .35 .09 .1
22% (8/37) 38% (14/37) 22% (8/37) 19% (7/37)
16% (18/111) 24% (27/111) 20% (22/111) 40% (44/111)
11.3 ⫾ 2.3 (37) 213 ⫾ 40 (37) 1.3 ⫾ 0.8 (37)
10.8 ⫾ 2.1 (99) 207 ⫾ 38 (99) 1.7 ⫾ 1.0 (95)
.50 .50 .06
CAD, Coronary artery disease; COPD, chronic obstructive pulmonary disease; CPR, cardiopulmonary resuscitation; CRF, chronic renal failure; CVA, cerebrovascular accident; ER, emergency room; OR, operating room; OSR, Open surgical repair; PVD, peripheral vascular disease; rAAA, ruptured abdominal aortic aneurysm; REVAR, endovascular repair of rAAA; SBP, systolic blood pressure.
After Institutional Review Board approval, demographic data, comorbidities, physiologic data at presentation, time of presentation, time interval between presentation and treatment, operative data, and postoperative data were collected from medical records for all patients. Case selection. The UPMC is a regional tertiary referral hospital with high volume of open and emergent aortic surgery cases. Selection of treatment in each case was made by the vascular surgeon based on aortic morphology, time of presentation, hemodynamic stability, and the surgeon’s own experience. Propensity score-based matching. To reduce the confounding effects of covariate imbalance between OSR and REVAR and selection bias, propensity score-based matching was performed. The variables used for matching included sex, gender, surgeon, hemodynamic stability, and preoperative cardiopulmonary resuscitation. Patients who underwent REVAR were matched to patients who underwent OSR in a 1:3 ratio due to relative sample sizes of REVAR and OSR patients and their distributions of propensity scores. Hence, 37 patients with REVAR were matched with 111 patients with OSR, and their outcomes were analyzed. Outcomes. All intraoperative, in-hospital and 30-day mortality, postoperative morbidity, length of stay, and transfusion of blood products were recorded. The Social Security Death Index was used to evaluate the long-term mortality. Statistical analysis. Kaplan-Meier analysis was used to estimate the late survival rates of both patient groups. Treatment selection bias was controlled by using propen-
sity score-based matching for sex, age, preoperative hemodynamic status, and surgeon’s annual AAA volume. All baseline variables were included in logistic regression model to predict the probability that patient would receive REVAR vs OSR upon presentation with rAAA. The fit of the propensity score model to the data was assessed using concordance index. RESULTS The demographic data, comorbidities, and preoperative characteristics in each treatment group are illustrated in Table I. There were 26 males out of 37 patients (70.3%) treated with REVAR and 69 males (62.2%) in 111 patients who underwent OSR (P ⫽ .48). The mean age was 74.9 years in REVAR group and 75.6 years in OSR group (P ⫽ .76). Patients presenting with hemodynamic instability, defined as systolic blood pressure ⬍80 mm Hg, were more likely to undergo OSR (P ⫽ .03). Forty-nine patients (44%) who underwent OSR were hemodynamically unstable on presentation, compared with only 24% of patients who underwent REVAR (nine patients). This difference results from the fact that the propensity score-based matching balances the probability of being treated with REVAR as a function of multiple covariates, not each covariate directly. In addition, due to the small sample size, it was not feasible to find exact matching between the groups. The delay from presentation until treatment was not statistically significant, but there was a trend toward delay in the REVAR group; the time of presentation did not differ, either.
JOURNAL OF VASCULAR SURGERY September 2012
616 Saqib et al
Table II. Intraoperative variables Variable
REVAR
OR time Intraoperative fluid (mL) Intraoperative colloids (mL) Intraoperative pRBC (units) Intraoperative FFP (6 packs)
OSR
P value
138 198 2183 3594 566 1290 3.06 6.86 0.625 3.89
.0001 .0007 .02 .02 .0001
FFP, Fresh frozen plasma; OR, operating room; OSR, open surgical repair; pRBC, packed red blood cells; rAAA, ruptured abdominal aortic aneurysm; REVAR, endovascular repair of rAAA.
Table III. Comparison of morbidity and mortality after REVAR and OSR
Mortality Morbidity LOS (days) MI ARI Hemodialysis Pneumonia Tracheostomy ACS Bowel resection Postop hemorrhage
REVAR
OSR
P value
21.6 (8/37) 54 (20/37) 7.4 ⫾ 17.6 (7/37) 21 (8/37) 5.9 (2/37) 8.8 (3/37) 2.9 (1/37) 11.4 (4/37) 8.6 (3/37) 5.7 (2/37)
31.5 (35/111) 66 (73/111) 23.2 ⫾ 37.7 (42/111) 47 (52/111) 24.5 (27/111) 24.5 (27/111) 20.8 (23/111) 11.3 (12/111) 10.4 (11/111) 10.4 (11/111)
.35 .23 .13 .03 .008 .02 .05 .015 1 1 .51
ACS, Abdominal compartment syndrome; ARI, acute renal injury; LOS, length of stay; MI, myocardial infarction; OSR, open surgical repair; rAAA, ruptured abdominal aortic aneurysm; REVAR, endovascular repair of rAAA.
Intraoperative variables are detailed in Table II. The operative time was significantly longer in the OSR group with higher fluid and blood product transfusion requirements. The length of stay in the intensive care unit and hospital stay were lower following REVAR, but these were not statistically significant (P ⫽ .13). The REVAR group had a 7.3-day mean and 4.5-day median length of stay, compared with a 23-day mean and 10.4-day median length of stay in the OSR group. The operative mortality rates, defined as intra-operative, in-hospital, and 30-day mortality, were not statistically different between the two groups (21.6% REVAR vs 31.5% OSR; P ⫽ .34). To account for the differences in the proportion of patients with hemodynamic instability (HI), a logistic regression analysis was performed with and without adjusting for HI. The difference in mortality between REVAR and OSR adjusting for HI was not significant (P ⫽ .51 with the estimated difference in log odds of mortality being ⫺.31). Overall complication rates were similar in both groups (54% REVAR vs 66% OSR; P ⫽ .23) (Table III). However, pulmonary failure requiring tracheostomy, myocardial infarction, and acute tubular necrosis occurred more frequently after OSR. The incidences of abdominal compartment syndrome and bleeding requiring return to the operating room did not differ. At a mean of 3-year follow-up, the aneurysm-, graft-, or procedure-related secondary interventions were similar in
both groups: nine patients (25%) after REVAR vs 26 patients (23.4%) following OSR (P ⫽ .99) (Table IV). Kaplan-Meier analysis (Fig) revealed no differences in late survival at 1, 2, and 3 years (P ⫽ .66). The probability of survival after REVAR was 50%, 50%, and 42%, respectively. The corresponding figures for OSR were 54%, 52%, and 47%, respectively. Following OSR, aneurysmrelated deaths occurred in 35 of 55 deaths in the first year; 14 deaths in year 2 and year 3 each were nonaneurysmrelated deaths. Following REVAR, there were nine out of 12 aneurysm-related deaths in the first year, three deaths not related to aneurysm in the second post-REVAR year, and four deaths not related to aneurysm during the third post-REVAR year. DISCUSSION The perceived benefits of REVAR and improved results from selected studies1-5 have led some investigators to promote an “endovascular-first” approach in the treatment of rAAA2,20 since the first report of successful REVAR by Yusuf et al in 1994.21 In fact, the utilization of REVAR has steadily increased in the United States; the frequency of REVAR increased over the 6-year study period from 5.9% in 2001 to 18.9% in 200612; a similar trend was also observed in the present study. Such a trend was accompanied by a decrease in mortality rates from 43% to 29% in a review of a Medicare database.22 The literature is replete with varying results after REVAR.17 Some publications have shown no difference in early outcomes between REVAR and OSR,3,23,24 while some have shown superior results with REVAR. While these improved results can be found in propensity scorematched13 or large population-based studies,11,12 the most salient finding in review of the literature is that centers with an instituted protocol are likely to have superior outcomes with REVAR.2,4,5,7 In a recent review of collected world experience with REVAR, the 30-day mortality was 21.2% with REVAR compared with 36.3% with OSR; in a subgroup analysis of 13 centers with standardized protocols with preferential REVAR the rate was 19.7%.5 Not all centers with protocols are successful, however. A recent international multicenter study with preferential REVAR failed to show early survival benefit even with heavy selection bias toward OSR for patients with cardiovascular instability.24 A recent survey of the National Surgical Quality Improvement Program (NSQIP) database showed no statistically significant reduction in mortality rate with REVAR.25 The same findings were noted in the current study. While the lack of differences in the current study may be due to the low mortality rate with OSR for rAAA observed in our institution26 and type II errors due to the small sample size, the results resonate the conclusion of a large populationbased study and propensity score-matched studies that stated “institutional and surgeon experience were essential to fully benefit from the use of this procedure for the treatment of rAAA” and advocated for regionalization of care to high-volume centers.11-13 It should also be noted that administrative databases are limited by their inability to
JOURNAL OF VASCULAR SURGERY Volume 56, Number 3
Saqib et al 617
Table IV. Early and late secondary interventions after REVAR and OSR Complications Procedure-related Re-exploration for ACS/hemorrhagea Herniorrhaphy Colectomy for ischemia Lysis of adhesion Cholecysectomy Small bowel resection Groin hematoma Graft-related Thrombosis Graft infection Endoleak Vascular-related Iliac PTA/stenting Total
Post-OSR interventions (n ⫽ 26 patients)
Post-REVAR interventions (n ⫽ 9 patients)
12/26 (46%) 4/26 (15%) 8/26 (31%) 3/26 (12%) 1/26 (4%) 3/26 (12%) N/A
4/9 (44%) N/A 3/9 (33%) N/A 0 0 2/9 (22%)
2/26 (7%) 2/26 (7%) N/A
4/9 (44%) 0 3/9 (33%)
1/26 (4%) 26/111 (23%)
1/9 (11%) 9/37 (25%)
P
.99
ACS, Abdominal compartment syndrome; OSR, open surgical repair; PTA, percutaneous transluminal angioplasty; rAAA, ruptured abdominal aortic aneurysm; REVAR, endovascular repair of rAAA. Some patients may have more than one reintervention. a Includes splenectomy.
Fig. Kaplan-Meier curves of survival rates after endovascular repair of ruptured abdominal aortic aneurysm (rAAA) (REVAR) and open surgical repair (OSR) of rAAAs.
control for factors such as hemodynamic stability, surgeon’s experience and preference, suitability for REVAR, referral pattern, and delay between presentation and treatment. It is likely that the OSR mortality rates reported in the literature are adversely affected by unfavorable hemodynamic and anatomic characteristics, a phenomenon that has been well documented. That there was a significantly higher proportion of patients with hemodynamic instability in the OSR group may explain, at least in part, the higher (although statistical significance was not reached) mortality rate following OSR observed in the current study. Had the current study been matched for hemodynamic stability, the mortality rate with OSR would be lower and the gap narrower; the estimated difference in mortality rates after adjusting for hemodynamic instability became even smaller.
Multiple meta-analyses, systematic reviews, and populationbased studies have shown survival advantages with REVAR. However, all these studies are flawed with heavy patient selection and publication biases, poorly controlled confounding factors, and the lack of standardized reporting.27 Recognizing these limitations, the conclusions of these studies state that there is insufficient evidence to recommend widespread adoption of REVAR and acknowledge the need for a large prospective, multicenter randomized trial,9 in light of the fact that two earlier prospective randomized trials (the Nottingham and the Amsterdam Acute Aneurysm [AJAX] trials) have failed to show survival benefits with REVAR.27-29 However, some refute the validity of such a trial.16
JOURNAL OF VASCULAR SURGERY September 2012
618 Saqib et al
There have been four prospective, randomized trials that have either been conducted or are currently underway. The Nottingham trial failed to show any reduction in mortality with REVAR and was terminated due to poor recruitment.17 The AJAX and the Endovasculaire vs Chirurgie dans les Anévrysmes Rompus (ECAR) trials have been designed to recruit hemodynamically stable REVARsuitable patients. These trials, while avoiding the ethical dilemmas of randomizing hemodynamically unstable patients, will not be able to assess the role of an endovascularfirst strategy vs open repair on an intention-to-treat basis since randomization occurs only after EVAR eligibility is determined by a computed tomography scan. The AJAX trial also failed to demonstrate any survival benefits with REVAR and has been extended to recruit 120 patients, increased from its original 80-patient recruitment goal.27 The Immediate Management of the Patient with Rupture: Open Versus Endovascular repair (IMPROVE) trial is a large trial, planned to recruit 600 patients, that randomizes patients at the time of clinical diagnosis of rAAA regardless of cardiovascular stability and before REVAR eligibility is determined. As such, it will address the question of whether or not the endovascular-first strategy provides survival benefits over OSR.30 It is powered to show a difference of a 14% 30-day mortality rate and will identify factors that are predictive of survival benefits.27,30 Widespread adoption of the endovascular-first approach for the treatment of rAAA at first glance is attractive. It is minimally invasive and associated with lower complication rates, as shown in the current study. However, implementation of such a protocol involves massive systemic changes, such as stocking of endovascular stent grafts and auxiliaries and around-the-clock availability of skilled endovascular, radiologic, and nursing teams. It is not pragmatic to undertake such systemic and systematic changes in the absence of clear evidence of REVAR’s superiority. As such, an institutional protocol for REVAR has not been established at our institutions; given the size of the UPMC system, this would represent a monumental task in a variety of ways. Furthermore, REVAR is not generalizable at present. At best, about 50% of patients with rAAA may be treated with REVAR.25 Cardiovascular stability plays a major role in selection bias and outcome. Mehta et al31 revealed that the mortality rate following REVAR for hemodynamically unstable patients was 41%, similar to their open repair mortality rate of 40%.32 They also noted that mortality rates after REVAR for women (33%) and octogenarians (38%) were significantly higher than those for men (17%) and those ⬍80 years of age (15%). Furthermore, REVAR can cause an increased risk of complications such as abdominal compartment syndrome, which has been shown to be a harbinger of poor outcome with mortality in excess of 60%.31,33,34 A recent study also showed an increased risk of spinal cord ischemia with occlusion (temporary or permanent) of the hypogastric artery.35 The benefit of REVAR was not observed all across the hospitals. While a reduction in mortality was observed in teaching hospitals with
REVAR, a higher risk of in-hospital death was noted in community facilities.22,24 Identification of patient characteristics, aneurysm morphology, and intraoperative maneuvers that may influence the outcomes of REVAR is paramount. In regard to midterm survival rates, the early survival benefit does not seem to provide lasting effects. Survival probabilities up to 3 years were similar between the two groups in the present study. This is consistent with the findings of earlier reports.24,36,37 It has been well established that a sustained risk of mortality exists even after successful open repair of ruptured abdominal and thoracoabdominal aortic aneurysm,26,38,39 and patients appear to be susceptible to it after REVAR. There is still equipoise on what the best treatment for rAAA is. The data are conflicting, and many studies showing improved outcomes with REVAR employed it preferentially for patients with favorable anatomy and hemodynamic stability.18,37 Preferential REVAR should be deferred until results of prospective, randomized trials are available and predictive factors for its success are identified. Some patients may be harmed by indiscriminate insistence on REVAR. There are some limitations to this study inherent to its retrospective nature with relatively small sample size. Despite propensity score-based matching to reduce selection bias, due to the small number of patients, complete matching was not feasible. As such, there may be a selection bias on the choice of treatment rendered. The propensity scorebased matching can only remove overt bias, but unlike randomization, it cannot be expected to remove hidden biases. Additionally, the propensity score-based method is based on the hypothesis that it balances only for the covariates that were used to construct the score. Including irrelevant covariates in propensity score-based matching model may reduce efficiency. CONCLUSIONS Endovascular repair for rAAA does not seem to conclusively confer either acute or late survival advantages over open repair. Wide adoption and routine use of REVAR should be deferred until there is good evidence for its superiority. AUTHOR CONTRIBUTIONS Conception and design: JC Analysis and interpretation: JC, TP, NS Data collection: NS, JC Writing the article: NS, TP, JC Critical revision of the article: NS, TP, JC Final approval of the article: JC, NS, TP Statistical analysis: TP Obtained funding: Not applicable Overall responsibility: JC
JOURNAL OF VASCULAR SURGERY Volume 56, Number 3
REFERENCES 1. Hechelhammer L, Lachat ML, Wildermuth S, Bettex D, Mayer D, Pfammatter T. Midterm outcome of endovascular repair of ruptured abdominal aortic aneurysms. J Vasc Surg 2005;41:752-7. 2. Mehta M, Taggert J, Darling RC III, Chang BB, Kreienberg PB, Paty PS, et al. Establishing a protocol for endovascular treatment of ruptured abdominal aortic aneurysms: outcomes of a prospective analysis. J Vasc Surg 2006;44:1-8. 3. Coppi G, Silingardi R, Gennai S, Saitta G, Ciardullo AV. A single-center experience in open and endovascular treatment of hemodynamically unstable and stable patients with ruptured abdominal aortic aneurysms. J Vasc Surg 2006;44:1140-7. 4. Moore R, Nutley M, Cina CS, Motamedi M, Faris P, Abuznadah W. Improved survival after introduction of an emergency endovascular therapy protocol for ruptured abdominal aortic aneurysms. J Vasc Surg 2007;45:443-50. 5. Veith FJ, Lachat M, Mayer D, Malina M, Holst J, Mehta M, et al. Collected world and single center experience with endovascular treatment of ruptured abdominal aortic aneurysms. Ann Surg 2009;250: 818-24. 6. Mayer D, Pfammatter T, Rancic Z, Hechelhammer L, Wilhelm M, Veith FJ, et al. Ten years of emergency endovascular aneurysm repair for ruptured abdominal aortoiliac aneurysms: lessons learned. Ann Surg 2009;249:510-5. 7. Starnes BW, Quiroga E, Hutter C, Tran NT, Hatsukami T, Meissner M, et al. Management of ruptured abdominal aortic aneurysm in the endovascular era. J Vasc Surg 2010;51:9-17. 8. Sadat U, Boyle JR, Walsh SR, Tang T, Varty K, Hayes PD. Endovascular vs open repair of acute abdominal aortic aneurysms–a systematic review and meta-analysis. J Vasc Surg 2008;48:227-36. 9. Mastracci TM, Garrido-Olivares L, Cina CS, Clase CM. Endovascular repair of ruptured abdominal aortic aneurysms: a systematic review and meta-analysis. J Vasc Surg 2008;47:214-21. 10. Rayt HS, Sutton AJ, London NJ, Sayers RD, Bown MJ. A systematic review and meta-analysis of endovascular repair (EVAR) for ruptured abdominal aortic aneurysm. Eur J Vasc Endovasc Surg 2008;36: 536-44. 11. Egorova N, Giacovelli J, Greco G, Gelijns A, Kent CK, McKinsey JF. National outcomes for the treatment of ruptured abdominal aortic aneurysm: comparison of open versus endovascular repairs. J Vasc Surg 2008;48:1092-100. 12. McPhee J, Eslami MH, Arous EJ, Messina LM, Schanzer A. Endovascular treatment of ruptured abdominal aortic aneurysms in the United States (2001-2006): a significant survival benefit over open repair is independently associated with increased institutional volume. J Vasc Surg 2009;49:817-26. 13. Holt PJ, Karthikesalingam A, Poloniecki JD, Hinchliffe RJ, Loftus IM, Thompson MM. Propensity scored analysis of outcomes after ruptured abdominal aortic aneurysm. Br J Surg 2010;97:496-503. 14. Greco G, Egorova N, Anderson PL, Gelijns A, Moskowitz A, Nowygrod R, et al. Outcomes of endovascular treatment of ruptured abdominal aortic aneurysms. J Vasc Surg 2006;43:453-9. 15. Giles KA, Pomposelli F, Hamdan A, Wyers M, Jhaveri A, Schermerhorn ML. Decrease in total aneurysm-related deaths in the era of endovascular aneurysm repair. J Vasc Surg 2009;49:543-50. 16. Veith FJ, Powell JT, Hinchliffe RJ. Is a randomized trial necessary to determine whether endovascular repair is the preferred management strategy in patients with ruptured abdominal aortic aneurysms? J Vasc Surg 2010;52:1087-93. 17. Hinchliffe RJ, Powell JT, Cheshire NJ, Thompson MM. Endovascular repair of ruptured abdominal aortic aneurysm: a strategy in need of definitive evidence. J Vasc Surg 2009;49:1077-80. 18. Lee RW, Rhodes JM, Singh MJ, Davies MG, Wolford HY, Diachun C, et al. Is there a selection bias in applying endovascular aneurysm repair for rupture? Ann Vasc Surg 2008;22:215-20. 19. Veith FJ, Gargiulo NJ. Endovascular aortic repair should be the gold standard for ruptured AAAs, and all vascular surgeons should be prepared to perform them. Perspect Vasc Surg Endovasc Ther 2007;19: 275-82.
Saqib et al 619
20. Veith FJ, Ohki T. Endovascular approaches to ruptured infrarenal aorto-iliac aneurysms. J Cardiovasc Surg (Torino) 2002;43:369-78. 21. Yusuf SW, Whitaker SC, Chuter TA, Wenham PW, Hopkinson BR. Emergency endovascular repair of leaking aortic aneurysm. Lancet 1994;344:1645. 22. Lesperance K, Andersen C, Singh N, Starnes B, Martin MJ. Expanding use of emergency endovascular repair for ruptured abdominal aortic aneurysms: disparities in outcomes from a nationwide perspective. J Vasc Surg 2008;47:1165-70. 23. Anain PM, Anain JM Sr, Tiso M, Nader ND, Dosluoglu HH. Early and midterm results of ruptured abdominal aortic aneurysms in the endovascular era in a community hospital. J Vasc Surg 2007;46:898-905. 24. Peppelenbosch N, Geelkerken RH, Soong C, Cao P, Steinmetz OK, Teijink JA, et al. Endograft treatment of ruptured abdominal aortic aneurysms using the Talent aortouniiliac system: an international multicenter study. J Vasc Surg 2006;43:1111-23. 25. Davenport DL, O’Keeffe SD, Minion DJ, Sorial EE, Endean ED, Xenos ES. Thirty-day NSQIP database outcomes of open versus endoluminal repair of ruptured abdominal aortic aneurysms. J Vasc Surg 2010;51:305-9. 26. Cho JS, Kim JY, Rhee RY, Gupta N, Marone LK, Dillavou ED, et al. Contemporary results of open repair of ruptured abdominal aortoiliac aneurysms: effect of surgeon volume on mortality. J Vasc Surg 2008;48:10-7. 27. Powell JT. Time to IMPROVE the management of ruptured abdominal aortic aneurysm: IMPROVE trialists. Eur J Vasc Endovasc Surg 2009; 38:237-8. 28. Hinchliffe RJ, Bruijstens L, MacSweeney ST, Braithwaite BD. A randomised trial of endovascular and open surgery for ruptured abdominal aortic aneurysm–results of a pilot study and lessons learned for future studies. Eur J Vasc Endovasc Surg 2006;32:506-13. 29. Hoornweg LL, Wisselink W, Vahl A, Balm R. The Amsterdam Acute Aneurysm Trial: suitability and application rate for endovascular repair of ruptured abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 2007;33:679-83. 30. Powell JT, Thompson SG, Thompson MM, Grieve R, Nicholson AA, Ashleigh R, et al. The immediate management of the patient with rupture: open versus endovascular repair (IMPROVE) aneurysm trial–ISRCTN 48334791 IMPROVE trialists. Acta Chir Belg 2009;109:678-80. 31. Mehta M, Roddy SP, Sternbach Y, Taggert J, Kreienberg PB, Paty PS, et al. Outcomes of EVAR in hemodynamically stable and unstable patients with ruptured AAA: a prospective analysis. J Vasc Surg 2011;53:554-5. 32. Mehta M, Roddy SP, Paty PS, Kreienberg PB, Ozsvath K, Sternbach Y, et al. Factors impacting survival after endovascular repair of ruptured abdominal arotic aneurysms. J Vasc Surg 2011;54:920. 33. Mehta M, Darling RC III, Roddy SP, Fecteau S, Ozsvath KJ, Kreienberg PB, et al. Factors associated with abdominal compartment syndrome complicating endovascular repair of ruptured abdominal aortic aneurysms. J Vasc Surg 2005;42:1047-51. 34. Djavani GK, Wanhainen A, Bjorck M. Intra-abdominal hypertension and abdominal compartment syndrome after endovascular repair of ruptured abdominal aortic aneurysm. Eur J Vasc Endovasc Surg 2011; 41:742-7. 35. Peppelenbosch N, Cuypers PW, Vahl AC, Vermassen F, Buth J. Emergency endovascular treatment for ruptured abdominal aortic aneurysm and the risk of spinal cord ischemia. J Vasc Surg 2005;42:608-14. 36. Hinchliffe RJ, Braithwaite BD. Ruptured abdominal aortic aneurysm: endovascular repair does not confer any long-term survival advantage over open repair. Vascular 2007;15:191-6. 37. Verhoeven EL, Kapma MR, Groen H, Tielliu IF, Zeebregts CJ, Bekkema F, et al. Mortality of ruptured abdominal aortic aneurysm treated with open or endovascular repair. J Vasc Surg 2008;48:1396-400. 38. Cho JS, Gloviczki P, Martelli E, Harmsen WS, Landis ME, Cherry KJ Jr, et al. Long-term survival and late complications after repair of ruptured abdominal aortic aneurysms. J Vasc Surg 1998;27:813-9; discussion: 819-20. 39. Barbato JE, Kim JY, Zenati M, Abu-Hamad G, Rhee RY, Makaroun MS, et al. Contemporary results of open repair of ruptured descending thoracic and thoracoabdominal aortic aneurysms. J Vasc Surg 2007;45:667-76. Submitted Oct 14, 2011; accepted Jan 3, 2012.