Elective treatment of abdominal aortic aneurysm with endovascular or open repair: The first decade

Patients and methods 

All patients undergoing elective repair of an infrarenal AAA between June 1996 and May 2005 were reviewed. Patients were identified from the Jobst Vascular Registry, which prospectively records patients undergoing vascular procedures at The Toledo Hospital. The operating room database and individual surgeon case lists were also used to corroborate the findings. Patient-related and AAA-related information were recorded from these databases in addition to lab records, operative notes, physician notes, and radiology records. The study was approved by the hospital’s Institutional Review Board.

Patients with ruptured aneurysms or aneurysms involving the renal arteries were excluded from this analysis. Elective repair of infrarenal AAA was performed in 677 consecutive patients. The type of repair was decided by the primary vascular surgeon and patient preference. Devices used for the EVAR group included EVT (Endovascular Technologies, Menlo Park, Calif), Vanguard (Boston Scientific, Natick, Mass), Lifepath (Baxter, Morton Grove, Ill), Lifepath (Edwards, Irvine, Calif), Zenith (Cook, Bloomington, Ind), Talent and AneuRx (Medtronic, Santa Rosa, Calif), AnCure (Guidant, Indianapolis, Ind), and Excluder (W. L. Gore & Associates, Flagstaff, Ariz).

Computed tomography (CT) scans and arteriography were obtained in nearly every case of EVAR early in the experience and evolved to CT scans alone as the sole imaging method during the latter part of the study. CT scans were the predominant imaging method for patients having open repair.

Main outcome measures were operative death (defined as death ≤30 days of the procedure or during the same hospitalization, whichever was longer), blood loss, length of hospital stay, status at discharge, and long-term mortality.

The EVAR group was monitored with physical examination and serial CT scans and plain abdominal radiographs at 1, 6, 12, 18, and 24 months, and yearly thereafter. Patients in the open repair group were monitored at 1, 6, and 12 months with physical examination and generally every 6 to 12 months thereafter. Postoperative imaging was not routinely performed in patients having open repair. Follow-up data were retrieved from the hospital database, the inpatient and outpatient data systems, and the Social Security Registry.

Results 

Patient characteristics 

Of the 677 patients who underwent elective infrarenal AAA repair, 417 had conventional open repair and 260 had EVAR. Demographic characteristics and comorbidities of the patients are listed in Table I. Compared with the EVAR group, the patients undergoing open repair were 3 years younger (71 vs 74 years, P < .001) but had larger aneurysms (6.01 ± 1.38 vs 5.45 ± 0.99, P < .001), greater familial predisposition (P = .039), and more current smokers (P = .011), which likely contributed to more chronic obstructive pulmonary disease (P = .001).

Table I. Demographics and comorbidities of abdominal aortic aneurysm patients

		Open⁎	EVAR⁎	P
	Males	324(77.7)	207(79.6)	.555
	Females	93(22.3)	53(20.4)	.555
	Age (years)	71.6 ± 7.9	73.7 ± 7.6	<.001
	Aneurysm size (cm)	6.01 ± 1.38	5.45 ± 0.99	<.001
	Creatinine (mg/dL)	1.13 ± 0.41	1.13 ± 0.49	.642
	Family history	39(9.3)	13(5)	.039
	Past smoker	227(54.4)	165(63)	.027
	Current smoker	134(32.1)	60(23.1)	.011
	COPD	170(40.8)	50(19.2)	<.001
	Hypertension	293(70.3)	178(68.5)	.620
	CAD	203(48.7)	141(54.2)	.160
	CHF	38(9.1)	35(13.5)	.076

COPD, Chronic obstructive pulmonary disease; CAD, coronary artery disease; CHF, congestive heart failure.

⁎ Data are presented as n (%) for categoric variables and mean ± standard deviation for continuous variables, where appropriate.

There were no differences in renal function, hypertension, coronary artery disease, or heart failure between the two groups. The open repair group had 324 men (77.7%) and 93 women (22.3%), and the EVAR group had 207 men (79.6%) and 53 women (20.4%), with no differences in gender distribution. Most patients in both groups were in their eighth decade of life. The mean follow-up time was 1078 ± 52 days for the EVAR group and 1665 ± 41 days for the open repair group (P < .0001).

Operative mortality 

Overall operative mortality was 3.1%. Operative mortality was 3.5% for the open repair group and 2.7% for the EVAR group (P = .627; Table II). A perceptible difference was found in operative mortality by age (Table II). Only three patients <70 years died from elective AAA repair (1.3%), and all had open repair. For patients ≥80 years of age, operative mortality increased to 9.5% in the open repair group and 6.6% in the EVAR group (P = NS).

Table II. Operative mortality by decade of age

	Age	Open N (%)	Mortality (%)	EVAR N (%)	Mortality (%)
	40-49	3(.8)	0	2(.8)	0
	50-59	27(6.5)	0	3(1.2)	0
	60-69	121(29)	3(2.5)	71(28)	0
	70-79	203(48)	5(2.5)	124(48)	3(2.4)
	≥80	63(15)	6(9.5)	60(22)	4(6.6)
	Total	417	14(3.5)	260	7(2.7)

EVAR, Endovascular aneurysm repair.

Other operative outcomes 

Statistically significant differences, all in favor of EVAR (Table III), were seen in the amount of estimated blood loss (2532 ± 1982 mL vs 536 ± 708 mL, P < .001), length of stay (9 ± 9.7 days vs 3.4 ± 3.7 days, P < .001), and patients discharged to home (83% vs 95%, P < .001).

Table III. Operative outcomes

		Open, n = 417	EVAR, n = 260	P
	Blood loss (mL)	2532±1982	536±708	<.001
	Length of stay (days)	9±9.7	3.4±3.7	<.001
	Discharged home (%)	345 (83)	245 (95)	<.001

EVAR, Endovascular aneurysm repair.

Mid-term and long-term survival 

Operative mortality was similar between the two treatment groups. Long-term mortality at 1, 3, and 5 years for each group is shown in Table IV. Only one aneurysm-related death in the current series occurred after the postoperative period. An overall Kaplan-Meier log-rank survival analysis showed no difference between the two treatment groups (P = .20); however, χ2 analysis of the long-term mortality results indicates that there was a mid-term (3-year) survival benefit favoring open repair (P < .002; Fig 1). Since many believe the benefit of EVAR may be greater in older patients, we analyzed operative and long-term survival for patients ≥70 years (Fig 2) and found no difference.

Table IV. Long-term mortality

		Open, n = 417 (%)	EVAR, n = 260 (%)
	1 year	28(7)	23(8.8)
	3 year	40(9.5)	41(16)
	5 year	65(15)	41(16)

EVAR, Endovascular abdominal aneurysm repair.

View Large Image Download to PowerPoint

Fig 1. Kaplan-Meier survival curves for patients undergoing elective open repair and endovascular aneurysm repair (EVAR) for infrarenal abdominal aortic aneurysms. Note that at 3 years, the curves separate, demonstrating survival benefit in patients having open repair. The curves later merge, showing equivalent survival at 5 years.

View Large Image Download to PowerPoint

Fig 2. Kaplan-Meier survival curves of patients ≥70 years undergoing elective open and endovascular aneurysm repair (EVAR) for infrarenal abdominal aortic aneurysms.

Discussion 

EVAR is rapidly developing as the technique of choice for elective repair of infrarenal AAA. Although distinct short-term advantages to this less-invasive approach have been demonstrated,1, 3 no definitive guidelines have been established to determine which patients should undergo EVAR vs open repair, assuming anatomy is appropriate for both.

Procedure-related and long-term mortality are appropriately considered the primary focus for comparing these two techniques. Although others have shown a significant short-term survival benefit of EVAR,1, 2, 4 reducing operative mortality by up to 66%, that was not observed in this clinical series. Interestingly, when EVAR demonstrated benefit compared with open repair, the mortality in patients undergoing open repair approached 5%. If an operative mortality of <4% can be achieved, it is likely that procedure-related mortality will not be different between the two techniques. This point is illustrated by the current series in which the mortality of open repair was substantially lower than that observed in patients enrolled in randomized trials. One cannot exclude clinical judgment and surgeon bias as influencing results in any nonrandomized trial. Intuitively, physician bias would seem to favorably affect open repair by selecting higher-risk patients for EVAR. That did not appear to be the case in this series because, with the exception of age, open repair patients had more comorbidities than EVAR patients.

Trials reporting early survival benefit of EVAR have yet to demonstrate a sustained survival advantage.1, 2, 4 Mortality results reported here are surprisingly similar to those reported in the Lifeline registry. EVAR patients were older than those undergoing open repair, women were older than men, and women had smaller aneurysms than men. Freedom from aneurysm-related rupture remained excellent in both groups throughout follow-up. The survival curves in our patients and in those of the Lifeline registry (Fig 3) have a striking similarity. Although there is no difference in procedure-related or long-term mortality, there is a significant separation of mid-term (2.5-year to 3-year) mortality favoring those who have open repair. The reasons for this observation are speculative, but may include more intense patient evaluation and better control of cardiovascular risk factors over the mid-term in open repair patients.

View Large Image Download to PowerPoint

Fig 3. Kaplan-Meier life-table from the Lifeline registry superimposed upon the survival curves of the current series from Jobst Vascular Center (JVC). Both show a mid-term (3-year) mortality benefit of open repair. EVAR, Endovascular aneurysm repair.

Operative mortality is usually associated with age and is no different in this series. We were gratified to find that no patient <60 years of age died as a result of elective operation for their AAA, and only three patients <70 years old died. Although a much higher procedure-related mortality occurred in patients ≥80 years (9.5% open, 6.5% EVAR), the differences were not significant. Sicard et al5 reported similar findings, with postoperative mortality no different for open and EVAR groups. They also noted that the subgroup of patients ≥80 years had an overall higher mortality, but less so in patients undergoing EVAR.

Reduced procedure-related length of stay in the intensive care unit and in the hospital is a consistent observation of EVAR-treated patients compared with open repair.1, 2, 6 This advantage is muted by the increasing number of repeat procedures over time in EVAR patients, which are associated with requisite hospitalization. Carpenter et al6 reported that the procedure-related length-of-stay advantage to EVAR was lost during the first year of follow-up owing to the readmissions required for subsequent aneurysm-related procedures. When total hospital days were compared at 12 months, no difference existed between EVAR and open repair patients. As follow-up continues over the long term, the additional procedures required may actually place EVAR patients at a disadvantage in a length-of-stay analysis.

Patients undergoing EVAR have a significant reduction in blood loss and the need for transfusions. Translating the short-term benefit into a long-term advantage relates to the risk of blood-borne diseases. This risk in the United States is miniscule compared with other comorbidities facing patients and is unlikely to translate into meaningful clinical observations.

Conclusion 

Open repair and EVAR can both be performed safely in patients treated for elective infrarenal AAA. Owing to the low operative mortality of open repair in patients <70 years old, open repair should be liberally applied in younger patients with AAA. EVAR clearly has perioperative advantages of reduced blood loss, decreased length of hospital stay, and an increased number of patients discharged to home. Early and late survival curves are similar; however, the long-term follow-up in EVAR patients is limited. The midterm survival advantage in open repair patients noted here and in other reports deserves greater study. If specific factors can be identified that correlate with this midterm survival advantage, perhaps improved survival over the long term can be enjoyed by both treatment groups.

Author contributions