Download this article's Factoid (PDF & PPT for Gold Subscribers)

In low-to-intermediate-risk patients with asymptomatic abdominal aortic aneurysm (AAA), open surgery and endovascular repair (EVAR) provide similar long-term survival with no differences in major or minor complications, according to findings from a French randomized trial published online January 26, 2011, ahead of print in the Journal of Vascular Surgery. However, EVAR is more likely to require reintervention.

For the multicenter ACE (Anévrysme de l’aorte abdominale: Chirurgie versus Endoprothèse) trial, investigators led by Jean-Pierre Becquemin, MD, of Hôpital Henri Mondor (Creteil, France), randomized 299 low-to-intermediate-risk patients with asymptomatic AAA that was anatomically suitable for EVAR to receive either open repair (n = 149) or EVAR (n = 150) at 25 French centers between March 2003 and March 2008.

EVAR was associated with less blood transfusion and shorter duration of intervention and ventilatory support. However, EVAR involved longer X-ray exposure time and more contrast use.

Both Procedures Relatively Safe

In the postoperative period, rates of complications (0.6% for open surgery vs. 2.0% for EVAR) and reintervention (1.3% vs. 5.3%) did not significantly differ between the procedures. Three deaths occurred: 1 due to an MI in each arm and 1 due to cardiac arrest following conversion to open surgery. Unsurprisingly, hospital stay was longer after surgery than after EVAR (10.4 ± 8.3 days vs. 5.8 ± 5.5 days; P < 0.0001).

After median follow-up of 3 years (range 0-4.8 years), there was no difference between the treatment arms for the coprimary endpoints of all-cause death and major adverse events (composite of MI, permanent stroke, permanent hemodialysis, major amputation, paraplegia, and bowel infarction). Rates of several individual endpoints also were similar (table 1).

Table 1. Outcomes at 3 Years

In addition, Kaplan-Meier estimates of cumulative survival at 3 years were equivalent between open repair and EVAR at 86.7 ± 4.4% and 86.3 ± 3.4%, respectively (P = 0.24). There was 1 death related to aneurysm or treatment in the surgical arm (0.6%) vs. 6 in the EVAR arm (4%; P = 0.14). Among EVAR patients, 2 deaths were perioperative, 2 were due to aortic rupture, and 2 occurred after reintervention.

Rates of survival free of major adverse events were also similar for the 2 groups at 3 years: 85.1 ± 4.5% for surgery and 82.4 ± 3.7% for EVAR (P = 0.09). Three ruptures occurred in the EVAR group, all more than 2 years after intervention.

On the other hand, cumulative survival free of vascular reintervention was higher in the open repair arm than in the EVAR arm: 85.8 ± 4.5% vs. 76.1 ± 4.6% (P = 0.01). Reintervention occurred in 16% of EVAR patients compared with 2.7% of surgical patients (P < 0.0001), with a trend toward higher aneurysm-related mortality after EVAR (4% vs. 0.7%; P = 0.12).

There was no overall difference in rates of minor complications between the 2 groups. However, surgical patients experienced more minor cardiac and incisional complications, while EVAR patients more frequently suffered buttock claudication (table 2).

Table 2. Minor Complications at 3 Years

CT scans identified endoleaks in 41 EVAR patients (27%); 10 were type I and 31 were type II. In addition, assessment of sexual function at 1 year showed no difference between the 2 arms (7.4% for open surgery and 4.7% for EVAR), although there was a trend toward more sexual dysfunction in the surgical group.

The majority of patients (92.6%) were treated according to randomization, but 1 patient did not undergo intervention and 21 patients crossed over from their assigned treatment (11.4% in the surgical arm vs. 2.7% in the EVAR arm; P < 0.01). In light of the fact that these patients did not receive the assigned treatment, a per-protocol analysis was performed. However, the findings did not change.

Ruptures Still an EVAR ‘Achilles’ Heel’

“AAA ruptures remain the Achilles’ heel of EVAR,” Dr. Becquemin and colleagues write. However, they suggest that 2 of the 3 ruptures seen in ACE may have been prevented by more expedient treatment of identified endoleaks and the third by a longer limb overlap in the common iliac artery.

The trial results, in particular the high rate of reintervention, “reflect current stent graft technology,” the investigators say. “With continuous advances in stent graft design, it may be possible . . . that the durability of EVAR will improve in the future.”

The authors conclude, “The choice between [open repair] and EVAR should rely on the balance of different risks: more postoperative transfusions, a longer hospital stay, and incisional complications with [surgery] vs. the need of follow-up with repeat CT scans, a higher rate of vascular reinterventions, and a small but persistent risk of rupture with EVAR.”

ACE Anomalies?

In a telephone interview with TCTMD, Zvonimir Krajcer, MD, of the Texas Heart Institute (Houston, TX), pointed to several unusual features of the trial that may have affected the results.

The fact that this trial cohort, unlike previous randomized trials, consisted of low-to-intermediate-risk patients undoubtedly influenced early mortality and morbidity, Dr. Krajcer acknowledged. “But there were other significant differences that may explain [the results],” he added.

For example, the 30-day mortality in ACE is one of the lowest reported in the literature for both surgery and EVAR, he pointed out. And unique to France, all of the procedures were done by highly skilled vascular surgeons, making for very good surgery but perhaps only adequate EVAR.

In addition, in this randomized trial with relatively small numbers, 17 patients assigned to surgery switched to EVAR. “The authors mention patient preference, but you always wonder if that cross-over introduced a certain degree of bias,” Dr. Krajcer said. “For example, if patients with hostile abdomen were switched to EVAR because [the operators] thought [that condition] would increase the risk of surgery. In an intention-to-treat analysis, [the crossovers] certainly put the results in a different light.”

Another important factor is that almost all of the procedures were done under general anesthesia, which is known to increase early mortality and complications compared with local anesthesia, Dr. Krajcer observed. That may have helped level the playing field between the 2 procedures, he explained.

Dr. Krajcer also pointed to an issue that may have muddied the difference between the 2 procedures. A considerable number of the EVAR patients had aortouniiliac stent grafts. “That means that in reality you are performing surgery as well,” he said. “The only difference between open surgery and aortouniiliac grafting is that [with the latter] you’re not opening the abdomen, but you still have to do a femoro-femoral bypass, which is surgery. This technique does not offer as good results [as conventional EVAR] and is not the current standard of care.” Low-risk patients who have a complexity that would require an aortouniiliac approach should instead receive surgery, he said.

Dr. Krajcer also insisted that rates of certain complications associated with EVAR were out of the ordinary. For example, he has never seen lymphorrhea with an endovascular procedure, and buttock complications should be no more common than with surgery, he commented. Moreover, the rate of early reintervention after EVAR of 5.3% is an anomaly, he emphasized, saying it suggests either a lack of operator expertise or a considerable number of semisurgical procedures.

While there remains a fairly high rate of reintervention due to endoleak, Dr. Krajcer acknowledged, the great majority of endoleaks can be treated by an endovascular approach and should not translate into increased mortality.

With regard to long-term mortality, ACE confirms the results of earlier trials showing no advantage for EVAR. “A lot of people look at this negatively because they think EVAR should offer better results,” Dr. Krajcer said. “But in reality the benefit of EVAR, if it is done percutaneously and with local or regional anesthesia, is a significant improvement in the first 30 days after intervention.” Also, it is important to keep in mind that most AAA patients have serious comorbidities and in the long run die from causes unrelated to the aneurysm or its treatment, he added.

Study Details

The study population included 3 women and 296 men. Mean age was 69 ± 7 years. Inclusion criteria were based on anatomic and clinical assessment. The 2 treatment arms did not differ in baseline characteristics, except for a higher rate of category 2 coronary disease (16.8% vs. 8.0%; P < 0.05) and a more severe Society of Vascular Surgeons/American Association for Vascular Surgery score in the open surgical arm (grade 2: 69.1% vs. 54.7% for EVAR; P < 0.01).

In the EVAR group, patients received 1 of 4 stent grafts:

• Zenith (n = 81; Cook, Bloomington, IN)
• Talent (n = 52; Medtronic Vascular, Santa Rosa, CA)
• Excluder (n = 9; WL Gore, Flagstaff, AZ)
• Powerlink (n = 4; Endologix, Irvine, CA)

Related Stories:

• Endovascular Approach Takes Lead Over Surgery for AAA Repair
• Trials Show EVAR Provides No Long-term Mortality Benefit for AAA Patients
• Hospitals with Higher EVAR Volume Have Better Outcomes