Stenting or Open Repair? EVAR, DREAM Trials Inconclusive

NICE, FRANCE -- Advocates of endovascular aneurysm repair had high hopes that three large randomized, controlled trials conducted in Europe--EVAR I, EVAR II, and DREAM--would establish the procedure's superiority over open repair. Recently published results have been equivocal, however, and the closest an expert speakers' panel could come to consensus was that informed patient preference should be the deciding factor for the time being.

"I think today it's too close to call," Dr. Jim A. Reekers said after presentations by the principal investigators and a heated audience discussion at the annual meeting of the Cardiovascular and Interventional Radiological Society of Europe.

"The patient can make up his own mind because we don't have a definite answer yet," said Dr. Reekers, a radiology professor at the University of Amsterdam, who discussed implications at the special session.

The three multicenter studies focused on abdominal aortic aneurysm repair.

British investigators led by Dr. Roger Greenhalgh conducted the Endovascular Aneurysm Repair (EVAR) I trial at 34 centers in the United Kingdom. They enrolled patients who had aneurysms at least 5.5 cm in diameter and were fit for either endovascular or open repair. EVAR I randomized 543 patients to stenting and 539 to open repair. Early results favored stenting, as its 30-day mortality rate of 1.7% was two-thirds less than the 4.7% reported for patients who had open repair (Lancet 2004;364:843-8).

By 4 years, all-cause mortality had leveled off at about 28% for both groups. The stenting cohort had fewer aneurysm-related deaths (4% vs. 7%), but more postoperative complications (41% vs. 9%).

"These results are not surprising, as endovascular AAA repair worked exactly as intended and protected patients from AAA-related death, but it did not impact mortality caused by coronary artery disease or cancer," said Dr. Brian Rubin, director of the noninvasive vascular laboratory, Barnes-Jewish Hospital, St. Louis, when asked to comment on this presentation.

"Instead of prompting physicians to question the role of endovascular repair, the EVAR data should prompt physicians to more aggressively identify and treat other causes of mortality, so the benefits derived from successful endoluminal therapy can be maintained," Dr. Rubin added.

The Dutch Randomized Endovascular Aneurysm Management (DREAM) trial group led by Dr. Jan D. Blankensteijn enrolled patients with aneurysms at least 5 cm in length at 28 centers in the Netherlands and Belgium. It randomized 171 to stenting and 174 to open repair.

Again the early results favored stenting, which had an operative mortality rate of 1.2% vs. 4.6% in the open-repair group. Severe complications were fewer (N. Engl. J. Med. 2004;351:1607-18). Two years after randomization, both cohorts had cumulative survival rates approaching 90%. Though open repair had more aneurysm-related deaths (5.7% vs. 2.1% for stenting), the investigators attributed the difference to the perioperative period. Aneurysm-related mortality was similar after the first 30 days. About two-thirds of both groups were free of moderate to severe complications (N. Engl. J. Med. 2005;352:2398-405).

Only unfit patients who were not candidates for surgery entered the EVAR II trial at 31 hospitals in the United Kingdom. The investigators assigned 166 to stenting and 172 to no intervention.

All told, 197 patients (including 47 who had been assigned to no intervention) underwent some form of aneurysm repair. During the follow-up period, 142 patients died; in 42 cases the deaths were related to aneurysms. There were no significant differences in overall mortality or aneurysm-related survival (Lancet 2005;365:2187-92).

"The 9% mortality in the endovascularly treated patients in EVAR II was substantially higher than that reported in similar high-risk patients in other series. A soon to be published report of high-risk patients culled from multiple U.S. device trials identified a mortality rate of approximately 2%. Possible explanations for the inferior EVAR II outcomes might include inadequate preoperative patient optimization or lack of operator experience," Dr. Rubin commented.

All three studies reported higher costs with stenting. This was attributed, in part, to mandates for intensive follow-up in patients undergoing a new procedure.

The British team reported mean hospital costs per stenting patient for up to 4 years of follow-up as £13,257 in EVAR I and £13,632 in EVAR II. In comparison, open-repair patients in EVAR I added up to £9,946, with charges of £4,983 for the no-intervention group in EVAR II.

The Dutch investigators similarly found costs about 4,500 euros higher with stenting, according to Dr. Blankensteijn, a professor of vascular surgery at the Radboud University Nijmegen (the Netherlands) Medical Center. "You would expect EVAR to be cheaper because it has shorter procedure time, shorter hospital stay, lower morbidity, [and] lower mortality, but still EVAR is more expensive," he said.

Dr. Greenhalgh, head of the department of vascular surgery at the Imperial College School of Medicine and Charing Cross Hospital in London, rejected an audience suggestion that the results favored open repair. "If open repair were clearly superior I would say EVAR is dead. If EVAR and open repair are neck and neck, then it is possible that EVAR is ahead," he said. "At the 4-year point, there is a small but significant benefit of EVAR. Therefore EVAR is close to open repair or better."

From a patient's perspective, he added, improved short-term risk with EVAR could be the deciding factor. For a man who is going to become a grandfather in 6 months, the early advantage could be more important.

"But you have to say it comes at a cost, that you [the patient] will be chained to your institution for a period of time and might need repeat interventions." Dr. Greenhalgh added that increased complications with stenting did not increase mortality, with most complications minor.

"All three trials confirm that AAA patients have a relatively reduced life expectancy. Compared to open AAA repair, another advantage of endoluminal repair is that much less of the remaining lifespan is spent rehabilitating after a major operation," commented Dr. Rubin.

"Our recommendation is that we find a way to preserve the small but significant perioperative survival advantage," Dr. Blankensteijn said, proposing that the costs of endografts and stenting follow-up have to come down for endovascular repair to become cost effective. "And we need to find treatment for patients likely to die in the next 2 or 3 years," he said.

Pointing to poor results for both groups in EVAR II, Dr. Greenhalgh queried improving patient fitness rather than performing early EVAR. "We did not know when we designed the trial how important it was to annotate fitness."