Journal of Vascular Surgery
Volume 45, Issue 2 , Pages 227-235.e1, February 2007

Performance of endovascular aortic aneurysm repair in high-risk patients: Results from the Veterans Affairs National Surgical Quality Improvement Program

Presented at the Sixtieth Annual Meeting of The Society for Vascular Surgery, Philadelphia, Pa, June 2006.

  • Ruth L. Bush, MD, MPH

      Affiliations

    • Michael E. DeBakey Department of Surgery, Baylor College of Medicine, the University of Houston, College of Pharmacy, Houston, Tex.
    • Corresponding Author InformationReprint requests: Ruth L. Bush, MD, MPH, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, 1709 Dryden, Suite 1568, Houston, TX 77030.
    • ,
  • Michael L. Johnson, PhD

      Affiliations

    • Michael E. DeBakey Veterans Affairs Medical Center, Baylor College of Medicine, Houston, Tex.
    • ,
  • Nasim Hedayati, MD

      Affiliations

    • Michael E. DeBakey Department of Surgery, Baylor College of Medicine, the University of Houston, College of Pharmacy, Houston, Tex.
    • ,
  • William G. Henderson, PhD

      Affiliations

    • University of Colorado Health Outcomes Program, Denver, Colo.
    • ,
  • Peter H. Lin, MD

      Affiliations

    • Michael E. DeBakey Department of Surgery, Baylor College of Medicine, the University of Houston, College of Pharmacy, Houston, Tex.
    • ,
  • Alan B. Lumsden, MD

      Affiliations

    • Michael E. DeBakey Department of Surgery, Baylor College of Medicine, the University of Houston, College of Pharmacy, Houston, Tex.

Received 7 June 2006; accepted 4 October 2006.

Article Outline

Objective

Recent results after endovascular abdominal aortic aneurysm repair (EVAR) have brought into question its value in patients deemed at high-risk for surgical intervention. The Department of Veteran Affairs (VA) National Surgical Quality Improvement Program (NSQIP) is the largest prospectively collected and validated United States surgical database representing current clinical practice. The purpose of our study was to evaluate outcomes after elective EVAR performed in high-risk veterans.

Methods

Using NSQIP data from 123 participating VA hospitals, we retrospectively evaluated patients who underwent elective aneurysm repair from May 2001 to December 2004. High-risk criteria were used to identify a cohort for analysis (EVAR, n = 788; open, n = 1580). High-risk criteria analyzed included age ≥60 years, American Society of Anesthesiology (ASA) classification 3 or 4, and the comorbidity variables of history of cardiac, respiratory, or hepatic disease, cardiac revascularization, renal insufficiency, and low serum albumin level. Our primary end points were 30-day and 1-year all-cause mortality, and we evaluated a secondary end point of perioperative complications. Statistical analysis included univariate analysis and multivariate modeling.

Results

Veterans who were classified as high-risk underwent elective EVAR with significantly lower 30-day (3.4% vs 5.2%, P = .047) and 1-year all-cause mortality (9.5% vs 12.4%, P = .038) than patients having open repair. EVAR was associated with a decrease in 30-day postoperative mortality (adjusted odds ratio [OR], 0.65; 95% confidence interval [CI], 0.42 to 1.03; P = .067) as well as 1-year mortality (adjusted OR, 0.68; 95% CI, 0.51 to 0.91; P = .0094) despite the presence of severe comorbid conditions. The risk of perioperative complications was significantly lower after EVAR (16.2% vs 31.0%; P < .0001; adjusted OR, 0.41; 95% CI, 0.33 to 0.52; P < .0001). A subset analysis of higher-risk patients (ASA 4 and the above comorbidity variables) still demonstrated an acceptable 30-day mortality rate.

Conclusion

In veterans deemed high-risk for surgical therapy, outcomes after elective EVAR are excellent, and the procedure is relatively safe in this special patient population. Our retrospective data demonstrate that patients with considerable medical comorbidities and infrarenal abdominal aortic aneurysms benefit from and should be considered for primary EVAR.

 

Endovascular abdominal aortic aneurysm (AAA) repair (EVAR) is a major advance in minimally invasive surgical technology because of the potential for reducing perioperative morbidity and mortality as well as recovery time, especially in patients deemed at high surgical risk by conventional open standards. The rapid evolution of endoluminal devices and surgical and anesthetic techniques during the past decade has allowed higher-risk patients, including octogenarians and others considered medically unfit for conventional open repair, to be offered EVAR as an alternative to observation alone.

Many retrospective, single institution studies documenting acceptable morbidity and mortality rates following EVAR in special high-risk patient populations are available in the literature.1, 2, 3, 4, 5, 6, 7 Furthermore, a recent analysis of high-risk patients using the Society of Vascular Surgery database for EVAR (registry established in 1998)8 concluded that endovascular repair is safe and effective in preventing aneurysm rupture.9

Conversely, in prospective, randomized trials comparing EVAR with open surgical repair, patients who were considered to be at excessive surgical risk and medically unfit for conventional open repair were excluded from participation.10, 11 In the recently published EVAR Trial 2 performed in the United Kingdom, this subset of excluded patients were further studied by being randomized to either EVAR or no intervention.12 The 30-day mortality rate of 9% and the 4-year mortality of 64% are both much higher than has been previously reported in high-risk patient cohorts. Furthermore, the investigators found no survival advantage after repair compared with those who had observation alone.

This prospective and randomized trial, with its level 1evidence, has thus challenged the promising outcomes considered to be associated with EVAR in high-risk individuals. These data have resulted in suggestions that perhaps EVAR should not be performed in select high-risk individuals owing to the possibility of poor outcomes and lack of improvement in survival, both at excessive financial expenditure.

In the United States, no prospective, randomized trial similar to EVAR Trial 2 has been completed to date. The availability of data that demonstrate positive outcomes with EVAR in high-risk patients, who represent those treated in routine surgical practices, is paramount to physician, patient, and policy-makers alike. The National Surgical Quality Improvement Program (NSQIP) represents a large, validated, prospective data set that is organized and implemented through the Department of Veteran Affairs (VA). The purpose of our study was to compare postoperative mortality and complications as well as survival in high-risk patients undergoing elective EVAR vs open repair within a national publicly funded health care system using a large prospective cohort.

Back to Article Outline

Patients and methods 

Data sources and sample 

All patients who underwent either EVAR or open abdominal aortic aneurysm repair between May 1, 2001, and December 31, 2004, were identified through the VA NSQIP database13 of surgical procedures performed at 123 participating VA hospitals. We chose this time frame because before May 2001, no suitable current procedural terminology (CPT) codes existed in the database for EVAR, although endovascular grafting systems were approved by the US Food and Drug Administration and became commercially available in September 1999.

Patients undergoing elective repair were defined by the primary International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) diagnostic code 441.4 (intact, nonruptured abdominal aortic aneurysm). Open AAA repairs were defined by CPT codes 35081 and 35102. EVAR was defined by codes 34800, 34802, and 34804. Patients with secondary diagnostic codes for ruptured AAA or thoracic or thoracoabdominal aortic aneurysm were excluded from the analysis. CPT codes representing open repair after EVAR (34830, 34831, 34832) were also excluded from primary analysis.

The NSQIP database contains very detailed, prospectively collected clinical data on all patients undergoing major surgical procedures within the VA.13 At the time of surgery, patients are enrolled in NSQIP, and baseline demographic, preoperative laboratory, and clinical information is collected by dedicated trained nurse reviewers. Additional perioperative data are prospectively collected by the nurses, including 30-day morbidity and mortality information.

To supplement the information in the NSQIP records with longer-term utilization and vital statistics data, we used encrypted Social Security numbers, date of admission, date of discharge, and primary operative date to link the NSQIP database with the VA Patient Treatment File (PTF), which contains abstracts of all patients discharged from all VA hospitals. To obtain additional information, we similarly linked the VA Outpatient Clinic File (OPC), which contains records for every outpatient visit to a VA facility. Linkage to the VA Beneficiary Identification Record Locator System (BIRLS) was performed to assess mortality data. The sensitivity of the BIRLS to estimate death rates of inpatients has been demonstrated in several studies to be approximately 87% to 95%.14, 15 Approval for the study was obtained from the Baylor College of Medicine Institutional Review Board and the Michael E. DeBakey VA Medical Center Research and Development Review Committee.

Definition of independent variables 

Patient demographic data are recorded prospectively in the NSQIP database, including age, gender, race, and hospital location. Comorbid conditions known to have an influence on the risk of cardiovascular morbidity and mortality were chosen for analysis.1, 16, 17 These independent variables were defined using NSQIP definitions and the ICD-9-CM codes. Within NSQIP, nurse reviewers recorded diabetes (persons using insulin or oral hypoglycemic agent), renal dysfunction or dialysis dependence, hepatic dysfunction, history of malignancy, history of congestive heart failure during the month preceding surgery, history of stoke or transient ischemic attacks, functional status, and tobacco use within the year preceding AAA repair.

We additionally identified 24 individual comorbid conditions and included a separate variable for elevated creatinine (≥2.0 mg/dL) that was based on the Revised Cardiac Risk Index.17 American Society of Anesthesiology (ASA) classification was included as a subjective estimation of a patient’s preoperative risk and baseline comorbid status. Both the inpatient (PTF) and outpatient (OPC) files were searched for matching ICD-9-CM codes 1 year before the operation date to ensure accuracy of comorbidity status.

Definition of a high-risk cohort 

High-risk criteria were used to identify a cohort for analysis. Minimum criteria for entry into our study included age ≥60 years and ASA classification 3 or 4. We further classified patients according to the comorbidity variables of history of cardiac, respiratory, or hepatic disease, cardiac revascularization, renal insufficiency, and low serum albumin. Because hypoalbuminemia has been shown to be associated with adverse surgical outcomes,18, 19 preoperative serum albumin concentration was analyzed as an independent variable. We defined a low albumin level of <3.4 g/L, which represents the 10th percentile in this cohort.

Outcome measures 

The outcomes of interest included 30-day and 1-year mortality and any perioperative (30-day) complications. The 30-day mortality was obtained from the NSQIP database, and 1-year mortality was calculated using death dates obtained from BIRLS and the PTF.

Perioperative complications from the NSQIP (see Appendix, online only) were aggregated into categories, including adverse cardiac events, renal dysfunction, pulmonary complications, wound complications, neurologic complications, postoperative bleeding requiring transfusion, and graft failure.20 Unfortunately, the original NSQIP guideline does not have a strict definition for what constitutes a “graft failure” except for a return to the operating room. This term is not specific to endovascular devices; thus, this variable may be misleading and difficult to analyze in a meaningful way. Within the data sets used, specific analysis of secondary interventions was not possible.

Postoperative length of stay and intensive care unit length of stay were calculated by using the date of surgical repair as the index date. Complete vital status information (either via death or follow-up records) was available on all patients at 30 days and at 1 year.

Statistical analysis 

All clinical outcomes of interest were tested for association with type of AAA repair and with the presence of the six additional high-risk comorbidities as defined. The effect of type of operation performed was then tested for its unique association with the morbidity and mortality outcomes (30-day, 365-day, any complication), after adjusting for the number of high-risk comorbidities and 20 additional demographic and clinical covariates20, 21 in multivariable logistic regression models. A significance level of 0.10 was required to stay in the final model. This level was selected arbitrarily to capture as many possible confounding factors that might be strongly associated with both the selection of EVAR and postoperative outcomes. Models were assessed for goodness of fit by the Hosmer-Lemeshow statistic and for discrimination by the c-index.22, 23, 24 Additional assessment of time to death was made using Kaplan-Meier estimators and log-rank tests. All P values reported are two-sided.

Back to Article Outline

Results 

Patient population 

From the NSQIP database, we initially identified 2966 elective repairs of nonruptured aneurysms that occurred between the time period of interest and that met the minimum criteria of age ≥60 and ASA class 3 or 4. Further classification with our high-risk criteria narrowed the cohort for analysis to 788 patients who underwent EVAR and 1580 patients who underwent open repair.

Basic demographic characteristics of the sample and prevalence of high-risk comorbid conditions are summarized in Table I. EVAR patients were on average age about 1 year older (P < .001). Less than 1% of all patients were women. Racial/ethnicity distribution did not vary by type of surgery; more than three quarters of all patients were white, with at least 7% black and 2% Hispanic. Approximately 58% of the patients had respiratory disease, 20% had prior revascularization, about 75% had cardiac disease, 4% to 5% had hepatic disease, and 16% had a low serum albumin level. With the exception of renal disease, which was more frequent among open repair patients (P = .014), the prevalence of high-risk conditions was similar for patients receiving both types of surgery.

Table I. Demographic characteristics of sample and prevalence of high-risk comorbid conditions
Open n = 1580 (%)EVAR n = 788 (%)P
Age (years)⁎71.8 ± 6.472.9 ± 6.7<.0001
Gender
Male1568(99.2)798(99.4).73
Race/ethnicity 0.26
White1281(81.1)606(76.9)
Black104(6.6)58(7.4)
Hispanic34(2.2)14(1.8)
Other/unknown161(10.2)110(14.0)
High-risk conditions
Respiratory916(58.8)455(57.7).91
Hepatic79(5.0)36(4.6).64
Cardiac1198(75.8)596(75.6).92
Prev cardiac revascularization321(20.3)168(21.3).57
Renal106(6.7)33(4.2).014
Low serum albumin251(15.9)132(16.8).59

EVAR, Endovascular aneurysm repair.

⁎Age is presented as mean (standard deviation); all other values are given as number (%); percentages for high-risk conditions do not total 100% because patients may have more than one condition.

Outcome measures 

The crude, unadjusted association of type of surgery with clinical outcomes, and the risk associated with increasing number of high-risk comorbidities are presented in Table II. The 30-day mortality rates were 5.2% for patients undergoing open repair compared with 3.4% for patients receiving EVAR (P = .047). One-year mortality and perioperative complications were similarly reduced in EVAR patients (P = .038 and P < .0001, respectively). About 20% of patients undergoing each type of surgery had none of the additional high-risk comorbidities when the NSQIP data were cross-linked to the PTF and OPC data sets, indicating that four fifths had at least one of the high-risk conditions. Our final analysis included only patients identified from the NSQIP that had at least one of the high-risk conditions in addition to age ≥60 years and ASA class 3 or 4.

Table II. Outcomes, adjusted odds ratios, and 95% confidence intervals for the association of type of surgery with outcomes⁎
EVAR n = 788 (%)Open n = 1580 (%)OR⁎ (EVAR)95% CIP
30-day mortality27(3.4)83(5.2)0.650.42,1.030.067
1-year mortality75(9.5)196(12.4)0.680.51,0.91.0094
Any complication128(16.2)490(31.0)0.410.33,0.52<.0001

EVAR, Endovascular aneurysm repair; OR, odds ratio; CI, confidence interval.

⁎Adjusted for count of high-risk conditions and 20 additional demographic and clinical covariables.

Table II also presents the adjusted odds ratios (OR) and 95% confidence intervals (CI) for the association of type of surgery with outcomes, after adjusting for the number of high-risk conditions and additional covariates. Risk of all clinical outcomes was reduced in patients undergoing EVAR compared with patients receiving open repair: the ORs for EVAR were 0.65 for 30-day mortality (95% CI, 0.42 to 1.03), 0.68 for 1-year mortality (95% CI, 0.51 to 0.91), and 0.41 for any complication (95% CI, 0.33 to 0.52). Among patients receiving open repair, this converts to a relative increased risk of 1.56 for 30-day mortality, 1.47 for 1-year mortality, and 2.33 for complications.

In stratified analyses (Fig 1, Fig 2, Fig 3), the risk of all clinical outcomes increased dramatically as the number of these conditions increased and was generally lower among patients receiving EVAR across strata. We were able to obtain 2-year data on patients in the cohort via linkage with the BIRLS database. Fig 4 shows the survival advantage in EVAR patients compared with open repair for 2-year follow-up (log-rank test χ2 = 5.23, P = .0222). In addition, postoperative length of stay averaged 5.6 days for patients receiving EVAR compared with 12.6 days for patients receiving open repair (data not shown, P < .0001).

  • View full-size image.
  • Fig 1. 

    Bar graph shows the unadjusted association between types of surgery, either endovascular aneurysm repair (EVAR) or open repair, and 30-day mortality rates stratified by number of surgical risk factors. The P value reflects comparison between Open (clear bars) and EVAR (shaded bars) within the level of risk factors.

  • View full-size image.
  • Fig 2. 

    Bar graph shows the unadjusted association between types of surgery, either endovascular aneurysm repair (EVAR) or open repair, and 1-year mortality rates stratified by number of surgical risk factors. The P value reflects comparison between Open (clear bars) and EVAR (shaded bars) within the level of risk factors.

  • View full-size image.
  • Fig 3. 

    Bar graph shows the unadjusted association between types of surgery, either endovascular aneurysm repair (EVAR) or open repair, and perioperative complication rates stratified by number of surgical risk factors. The P value reflects comparison between Open (clear bars) and EVAR (shaded bars) within the level of risk factors. There are fewer complications at all risk levels following EVAR.

  • View full-size image.
  • Fig 4. 

    Kaplan-Meier analysis shows a benefit to patients having endovascular aneurysm repair (EVAR; thick line) that remains greater than open repair (thin line) even at 2 years.

Highest-risk cohort 

A separate analysis of only the 721 patients with ASA class 4 showed a crude 30-day mortality rate of 5.6% (14/249) in EVAR patients and 7.0% (33/472) in open repair patients (P = .48). The 1-year crude mortality rates were 12.8% in EVAR patients and 16.7% in open patients (P = .17). Although neither the 30-day nor 1-year rates were statistically different, a beneficial trend was seen with EVAR. Furthermore, the proportion of patients with complications was significantly reduced in the EVAR patients (22.1%) compared with open repair patients (35.8%; P = .0002).

Back to Article Outline

Discussion 

In this unique cohort of US veterans, representing routine surgical practice, high-risk patients presenting for elective AAA repair who underwent EVAR had statistically significantly lower risk-adjusted perioperative mortality rates, lower complication rates, improved survival, and shorter lengths of stay compared to patients having open repair. Furthermore, even the highest risk patients, those with the ASA 4 classification, were found to benefit from EVAR with lower mortality and morbidity rates compared to open surgical counterparts. Our findings with regard to outcomes following EVAR and open AAA repair are consistent with several other published observational studies, with additional survival information via database linkage.2, 3, 4, 5, 9, 25, 26 The results herein are also similar to a previous study we published on a smaller, veteran cohort which included all elective aneurysm repairs. The similarities in findings are not surprising as others have suggested that patients presenting for treatment in the VA system may have more comorbidities and a greater illness burden than non-VA patients.27 Thus, a greater percentage of patients will match a “high-risk” definition. The fact that our study examined the outcomes following AAA repair in a specifically defined high-risk patient cohort is vital to determining the applicability, and potentially the reimbursement, of EVAR for this special patient population.

The NSQIP was prompted in 1986 by a congressional mandate and officially established in 1994 by the Veterans Health Administration.13 Though there are inherent deficiencies within retrospective data analyses, the NSQIP meticulous data collection methods and their effectiveness have been validated in multiple studies.28, 29, 30 We used the NSQIP database to identify our cohort based upon set criteria as outlined herein. With data from 123 institutions, the information contained within the NSQIP is robust with respect to variability in surgical techniques and skills as well as in cohort size, albeit procedure (and device) specific variables are not included. Moreover, data entry into NSQIP does not have the patient or aneurysm morphologic exclusion criteria that randomized trials contain, increasing the generalizability of our findings beyond VA patients. Furthermore, we cross-referenced the NSQIP data with both the inpatient and outpatient VA files to ensure accuracy of comorbidity and ICD-9 code reporting and thus, the actual presence of discrete co-morbidities of interest.

The criteria used for defining what constitutes a “high-risk” patient differ among clinical reports. Whether the patient is considered at “high-risk” or medically unfit for a specific operation, such as open AAA repair, may depend upon either specific criterion or upon guidelines established by the trial designers. In prospective, randomized trials comparing open to endovascular AAA repair, such as the Dutch and United Kingdom (UK) trials,11, 31 patients who were medically unfit for open surgical repair were essentially excluded from the randomization process. In the UK trial, the determination of fitness for surgery was left to the discretion of the local physicians participating in the trial.32 Guidelines for high-risk criteria, such as poor cardiac or respiratory status, were suggested but the presence or absence did not definitively determine trial eligibility. In many retrospective analyses, investigators establish inclusion/exclusion criteria based upon advanced age and coexisting medical conditions, for which the database of choice may be queried. There is now much information regarding successful operations in high-risk individuals due to improvements in anesthetic options, surgical techniques, and intensive care management.

Data from recent prospective, randomized trials evaluating elective AAA repair provides compelling Level One evidence for performing EVAR in patients “judged fit for open repair”.11, 31 Patients who were not deemed “medically fit for open repair” were then placed into EVAR Trial 2 in which patients were randomized to EVAR or no intervention. Within the high-risk cohort, twenty-five percent of the patients randomized to observation ultimately underwent aneurysm repair at the request of the patients or treating physician. These post-randomization crossovers may lead to bias if the crossover occurred in order to potentially avoid a worse outcome. Thus, the inability to evaluate the true effect of the “intervention”, EVAR or surveillance, may have resulted. However, the EVAR Trial 2 authors do state that an per protocol analysis demonstrated similar results. Many reasons may exist for the reported mortality rates in either group: time interval between randomization and repair leading to ruptures prior to repair, complex patient care secondary to coexisting medical conditions, or lack of preoperative medical treatments (statins and β-blocker medication use). At 4 years of follow-up, there was no difference in all-cause mortality between the EVAR and no intervention patients.This is not surprising as aneurysm-related events are not common beyond the perioperative period and patients generally succumb from causes unrelated to their AAA. Furthermore, aneurysm-related deaths can only be accurately determined by a post-mortem examination or rupture seen on an imaging study prior to death. However, obtaining an autopsy is generally not a common occurrence especially if a patient dies outside of a hospital setting.

Several studies have performed cost analyses and have determined that EVAR is a more expensive treatment than open aneurysm repair33, 34, 35, 36 and no intervention.12 The main difference in the two types of surgical repair is due to the increased costs of endovascular grafts and the lifelong surveillance and follow-up costs after EVAR none of which were available to us through the NSQIP. In an analysis of specific prospective studies, a recent Belgium document has concluded that EVAR is not cost-effective, and thus, there is not validation for widespread use.37 A full discussion on health care reimbursement, and whether or not certain costly procedures such as EVAR should be restricted, is beyond the scope of this manuscript.

Equal opportunities for good surgical outcomes do not exist among all persons. Circumstances transpire where a person’s illness or disability significantly reduce their opportunity to respond to a therapy, such as any of the criteria that we used in our definition of a high-risk surgical patient. Society is not obligated to spend whatever it must to provide them with such a therapy. However, society is obligated to provide persons with the chance to optimize their enjoyment of life, within the constraints of their disease or disability. In some, the presence of a large or expanding aneurysm, may contribute significantly to their psychological ability to enjoy life, as well as knowing that the outcomes following ruptured AAA repair remain dismal. Quality of life is not reported to the NSQIP but most recent studies have shown that though quality of life is improved in the perioperative period following EVAR compared to open repair, there is no difference when patients are surveyed at remote time points.12, 38, 39 It is clear from these published studies that EVAR is associated with early improvements in physical mobility and pain as well as a likelihood for a patient to be discharged to home rather than an institution.40, 41 Rationing parameters for health care resources are admittedly controversial and generate ethical dilemmas. It has been suggested that perhaps EVAR should be restricted to those patients who are “fit” for surgery, by whatever definition deemed appropriate, as the procedure, the endovascular devices, the continuing need for surveillance imaging, and possible secondary reinterventions are costly.

Decidedly, there are several limitations to our analysis in addition to the deficiencies inherent in a retrospective design. Furthermore, direct comparisons to current published literature are impossible due to dissimilarities in study design and content of data collected. Our comparison was between EVAR and open repair as we do not have a surveillance arm for control comparison. As previously stated, the NSQIP does not contain aneurysm size which has been shown to correlate with outcomes. Additionally, we do not have cause of death and hence, use the term “all-cause mortality” in stating our results. The use of the BIRLS database, which contains death dates regardless whether or not the death occurs in a VA facility, has been demonstrated to miss approximately 5% to 10% of veteran decedents.42 Nonetheless, we have no reason to believe that more EVAR deaths than open deaths were overlooked in our study. Lastly, patients served at VA hospitals are predominantly men. One might suggest that our results could be positively biased as women may have worse results after elective AAA repair (either open or EVAR). However, the proportion of men is the same for both cohorts.

Back to Article Outline

Conclusions 

In the absence of a prospective, randomized trial, this large patient cohort, representing routine clinical practice, demonstrated the superiority of endovascular repair over open repair in veterans deemed high risk for surgical therapy. Our outcomes after elective EVAR are excellent, although we compared EVAR with open repair and did not have a surveillance arm10 for comparison. Our data demonstrate that patients with considerable medical comorbidities and abdominal aortic aneurysms benefit from and should be considered for primary EVAR.

Back to Article Outline

Author contributions 


Conception and design: RB, MJ, ABL

Analysis and interpretation: RB, MJ, WH

Data collection: RB, MJ

Writing the article: RB, MJ, NH, ABL, WH

Critical revision of the article: PHL

Final approval of the article: RB, MJ, WH

Statistical analysis: MJ

Obtained funding: RB

Overall responsibility: RB

Back to Article Outline

 

We are grateful for the advice and support given by Drs Robert B. Rutherford and Ralph G. DePalma.

Back to Article Outline

Appendix 

Additional material for this article may be found online at www.jvascsurg.org.

Appendix (online only). Categories of aggregated national surgical quality improvement program complication variables and definitions
1. Cardiac
Cardiac arrest requiring CPR
• Absence of cardiac rhythm or presence of chaotic cardiac rhythm (ie,. ventricular fibrillation) requiring CPR
Myocardial infarction
• New transmural acute myocardial infarction occurring within 30 days of surgery
2. Neurologic
Cerebrovascular accident/stroke
• New stroke
Coma
• Impaired level of consciousness for >24 hours
Peripheral nerve injury
• As stated (either sensory or motor)
3. Pulmonary
Failure to wean >48 hours
• On ventilator >48 hours postoperative
Pneumonia
• CDC definition
Pulmonary embolism
• Either high probability VQ scan, positive pulmonary angiogram, or a positive CT scan
Reintubation for respiratory/cardiac failure
• Unplanned intubation after surgery
4. Renal
Acute renal failure
• Worsening renal function requiring hemodialysis, ultrafiltration, or peritoneal dialysis
Progressive renal insufficiency
• Rise of creatinine of >2 mg/dL from preoperative value
Urinary tract infection
• CDC definition
5. Wound
Deep wound surgical site infection
• CDC definition
Superficial surgical site infection
• CDC definition
Wound disruption or dehiscence
• Disruption of the fascia
6. Bleeding requiring >4 units RBC
• Any transfusion of packed RBCs given from the time the patient leaves the operating room
7. Graft failure
• Mechanical failure of a vascular graft or prosthesis

CPR, cardiopulmonary resuscitation; CDC, Centers for Disease Control and Prevention; VQ, ventilation perfusion; CT, computed tomography; RBC, red blood cells.

Back to Article Outline

References 

  1. Buth J, van Marrewijk CJ, Harris PL, Hop WC, Riambau V, Laheij RJ, et al. EUROSTAR Collaborators Outcome of endovascular abdominal aortic aneurysm repair in patients with conditions considered unfit for an open procedure: a report on the EUROSTAR experience. J Vasc Surg. 2002;35:211–221
  2. Biebl M, Lau LL, Hakaim AG, Oldenburg WA, Klocker J, Neuhauser B, et al. Midterm outcome of endovascular abdominal aortic aneurysm repair in octogenarians: a single institution’s experience. J Vasc Surg. 2004;40:435–442
  3. Sicard GA, Rubin BG, Sanchez LA, Keller CA, Flye MW, Picus D, et al. Endoluminal graft repair for abdominal aortic aneurysms in high-risk patients and octogenarians: is it better than open repair?. Ann Surg. 2001;234:427–435discussion 435-7
  4. Lange C, Leurs LJ, Buth J, Myhre HO. Endovascular repair of abdominal aortic aneurysm in octogenarians: an analysis based on EUROSTAR data. J Vasc Surg. 2005;42:624–630discussion 630
  5. Minor ME, Ellozy S, Carroccio A, Oak J, Chae K, Agarwal G, et al. Endovascular aortic aneurysm repair in the octogenarian: is it worthwhile?. Arch Surg. 2004;139:308–314
  6. Bush RL, Lumsden AB, Dodson TF, Salam AA, Weiss VJ, Smith RB, et al. Mid-term results after endovascular repair of the abdominal aortic aneurysm. J Vasc Surg. 2001;33:S70–S76
  7. Chaikof EL, Lin PH, Brinkman WT, Dodson TF, Weiss VJ, Lumsden AB, et al. Endovascular repair of abdominal aortic aneurysms: risk stratified outcomes. Ann Surg. 2002;235:833–841
  8. Lifeline Registry of Endovascular Aneurysm Repair Steering Committee. Lifeline Registry: collaborative evaluation of endovascular aneurysm repair. J Vasc Surg. 2001;34:1139–1146
  9. Sicard GA, Zwolak RM, Sidawy AN, White RA, Siami FS Society for Vascular Surgery Outcomes Committee. Endovascular abdominal aortic aneurysm repair: long-term outcome measures in patients at high-risk for open surgery. J Vasc Surg. 2006;44:229–236
  10. Greenhalgh RM, Brown LC, Kwong GP, Powell JT, Thompson SG. Comparison of endovascular aneurysm repair with open repair in patients with abdominal aortic aneurysm (EVAR trial 1), 30-day operative mortality results: randomised controlled trial. Lancet. 2004;364:843–848
  11. Prinssen M, Verhoeven EL, Buth J, Cuypers PW, van Sambeek MR, Balm R, et al. A randomized trial comparing conventional and endovascular repair of abdominal aortic aneurysms. N Engl J Med. 2004;351:1607–1618
  12. EVAR Trial Participants. Endovascular aneurysm repair and outcome in patients unfit for open repair of abdominal aortic aneurysm (EVAR trial 2): randomised controlled trial. Lancet. 2005;365:2187–2192
  13. Khuri SF, Daley J, Henderson W, Hur K, Demakis J, Aust JB, et al. National VA Surgical Quality Improvement Program The Department of Veterans Affairs’ NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care. Ann Surg. 1998;228:491–507
  14. Dominitz JA, Maynard C, Boyko EJ. Assessment of vital status in Department of Veterans Affairs national databases (comparison with state death certificates). Ann Epidemiol. 2001;11:286–291
  15. Fisher SG, Weber L, Goldberg J, Davis F. Mortality ascertainment in the veteran population: alternatives to the National Death Index. Am J Epidemiol. 1995;141:242–250
  16. Ligush J, Pearce JD, Edwards MS, Eskridge MR, Cherr GS, Plonk GW, et al. Analysis of medical risk factors and outcomes in patients undergoing open versus endovascular abdominal aortic aneurysm repair. J Vasc Surg. 2002;36:492–499
  17. Lee TH, Marcantonio ER, Mangione CM, Thomas EJ, Polanczyk CA, Cook EF, et al. Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery. Circulation. 1999;100:1043–1049
  18. Gibbs J, Cull W, Henderson W, Daley J, Hur K, Khuri SF. Preoperative serum albumin level as a predictor of operative mortality and morbidity: results from the National VA Surgical Risk Study. Arch Surg. 1999;134:36–42
  19. Rich MW, Keller AJ, Schechtman KB, Marshall WG, Kouchoukos NT. Increased complications and prolonged hospital stay in elderly cardiac surgical patients with low serum albumin. Am J Cardiol. 1989;63:714–718
  20. Bush RL, Johnson ML, Collins TC, Henderson WG, Khuri SF, Yu HJ, et al. Open versus endovascular abdominal aortic aneurysm repair in VA hospitals. J Am Coll Surg. 2006;202:577–587
  21. Johnson ML, Bush RL, Collins TC, Lin PH, Liles DR, Henderson WG, et al. Propensity score analysis in observational studies: outcomes after abdominal aortic aneurysm repair. Am J Surg. 2006;192:336–343
  22. Lemeshow S, Hosmer DW. A review of goodness of fit statistics for use in the development of logistic regression models. Am J Epidemiol. 1982;115:92–106
  23. Harrell FE, Califf RM, Pryor DB, Lee KL, Rosati RA. Evaluating the yield of medical tests. JAMA. 1982;247:2543–2546
  24. Hosmer DW, Taber S, Lemeshow S. The importance of assessing the fit of logistic regression models: a case study. Am J Public Health. 1991;81:1630–1635
  25. Hynes N, Kok N, Manning B, Mahendran B, Sultan S. Abdominal aortic aneurysm repair in octogenarians versus younger patients in a tertiary referral center. Vascular. 2005;13:275–285
  26. Jordan WD, Alcocer F, Wirthlin DJ, Westfall AO, Whitley D. Abdominal aortic aneurysms in “high-risk” surgical patients: comparison of open and endovascular repair. Ann Surg. 2003;237:623–629discussion 629-30
  27. Petersen LA, Normand SL, Daley J, McNeil BJ. Outcome of myocardial infarction in Veterans Health Administration patients as compared with medicare patients. N Engl J Med. 2000;343:1934–1941
  28. Daley J, Forbes MG, Young GJ, Charns MP, Gibbs JO, Hur K, et al. National VA Surgical Risk Study Validating risk-adjusted surgical outcomes: site visit assessment of process and structure. J Am Coll Surg. 1997;185:341–351
  29. Daley J, Khuri SF, Henderson W, Hur K, Gibbs JO, Barbour G, et al. Risk adjustment of the postoperative morbidity rate for the comparative assessment of the quality of surgical care: results of the National Veterans Affairs Surgical Risk Study. J Am Coll Surg. 1997;185:328–340
  30. Khuri SF, Daley J, Henderson W, Hur K, Gibbs JO, Barbour G, et al. Risk adjustment of the postoperative mortality rate for the comparative assessment of the quality of surgical care: results of the National Veterans Affairs Surgical Risk Study. J Am Coll Surg. 1997;185:315–327
  31. EVAR Trial Participants. Endovascular aneurysm repair versus open repair in patients with abdominal aortic aneurysm (EVAR trial 1): randomised controlled trial. Lancet. 2005;365:2179–2186
  32. Brown LC, Epstein D, Manca A, Beard JD, Powell JT, Greenhalgh RM. The UK Endovascular Aneurysm Repair (EVAR) trials: design, methodology and progress. Eur J Vasc Endovasc Surg. 2004;27:372–381
  33. Michaels JA, Drury D, Thomas SM. Cost-effectiveness of endovascular abdominal aortic aneurysm repair. Br J Surg. 2005;92:960–967
  34. Hayter CL, Bradshaw SR, Allen RJ, Guduguntla M, Hardman DT. Follow-up costs increase the cost disparity between endovascular and open abdominal aortic aneurysm repair. J Vasc Surg. 2005;42:912–918
  35. Angle N, Dorafshar AH, Moore WS, Quinones-Baldrich WJ, Gelabert HA, Ahn SS, et al. Open versus endovascular repair of abdominal aortic aneurysms: what does each really cost?. Ann Vasc Surg. 2004;18:612–618
  36. Forbes TL. Costs of open versus endovascular abdominal aortic aneurysm repair. J Am Coll Surg. 2004;198:329;author reply 329-30
  37. Bonneaux L, Cleemput I, Vrijens F, Vanoverloop J, Galloo P, Ramaekers D. HTA Le traitement electif endovasculaire de l’anevrysme de l’aorte abdominale (AAA). Brussels: KCE Reports; 2005;
  38. Lottman PE, Laheij RJ, Cuypers PW, Bender M, Buth J. Health-related quality of life outcomes following elective open or endovascular AAA repair: a randomized controlled trial. J Endovasc Ther. 2004;11:323–329
  39. Prinssen M, Buskens E, Blankensteijn JD. Quality of life endovascular and open AAA repair (Results of a randomised trial). Eur J Vasc Endovasc Surg. 2004;27:121–127
  40. Geraghty PJ, Sicard GA. Abdominal aortic aneurysm repair in high-risk and elderly patients. J Cardiovasc Surg (Torino). 2003;44:543–547
  41. Lee WA, Carter JW, Upchurch G, Seeger JM, Huber TS. Perioperative outcomes after open and endovascular repair of intact abdominal aortic aneurysms in the United States during 2001. J Vasc Surg. 2004;39:491–496
  42. Ashton CM, Souchek J, Petersen NJ, Menke TJ, Collins TC, Kizer KW, et al. Hospital use and survival among Veterans Affairs beneficiaries. N Engl J Med. 2003;349:1637–1646

 Additional material for this article may be found online at www.jvascsurg.org.Competition of interest: none.

PII: S0741-5214(06)01849-0

doi:10.1016/j.jvs.2006.10.005

Journal of Vascular Surgery
Volume 45, Issue 2 , Pages 227-235.e1, February 2007

Article Tools

* Image Usage & Resolution