In Medicare Population, Early Survival Benefit of EVAR Lost After 3 Years

Endovascular treatment of abdominal aortic aneurysms (AAAs) yields better survival compared with open surgery both perioperatively and through about 3 years, according to a large observational study published in the July 23, 2015, issue of the New England Journal of Medicine. Thereafter, mortality is similar between the therapies.

Take Home: In Medicare Population, Early Survival Benefit of EVAR Lost After 3 Years

Despite the early mortality benefit and declining rate of reintervention over time with endovascular aneurysm repair (EVAR), “a higher risk of late rupture remains a concern and warrants further study,” say Marc L. Schermerhorn, MD, of Beth Israel Deaconess Medical Center (Boston, MA), and colleagues.

Drawing on data from Medicare beneficiaries who underwent elective AAA repair between 2001 and 2008, investigators analyzed 39,996 propensity score–matched pairs of patients (mean age 75.6 years; 77.6% men) who had EVAR or surgery.

Perioperative mortality seen with EVAR was less than one-third of what was observed with open repair, with the benefit extending across all age groups. The endovascular cohort also had fewer medical and surgical complications, a shorter length of stay, and higher likelihood of being discharged home (table 1).

Table 1. Perioperative Outcomes by Type of Repair

Mortality Tipping Point at 4 Years

EVAR offered better survival for up to 3 months (HR 0.64; 95% CI 0.58-0.71), after which the advantage disappeared, reaching a tipping point at 4 years. After 4 years, mortality was higher with endovascular repair (HR 1.05; 95% CI 1.00-1.09), although the authors say the difference was “not meaningful in practical or clinical terms.”

Moreover, the substantial early mortality benefit of endovascular repair had a lasting impact: At 4 years, survival among patients in the EVAR cohort was calculated to be an average 12.4 days longer than that among the open-repair cohort (P < .001), with the benefit remaining significant through 7 years of follow-up.

At 8 years, Kaplan-Meier estimates showed that mortality was similar between EVAR and open repair, but the incidence of rupture was 3-fold higher after EVAR. The rate of aneurysm-related reintervention was higher after endovascular repair, but reinterventions for laparotomy-related complications (mainly hernia repair) were more common after open repair (table 2).

Table 2. Estimated Long-term Outcomes by Type of Repair

During the study period, perioperative mortality decreased for both EVAR (P = .001) and surgery (P = .01). In the endovascular group, rates of conversion to open repair, reoperation for bleeding, and readmission within 30 days of discharge also declined (all P < .001).

In the EVAR group, rates of mortality and total reinterventions were lower for patients treated in 2007 compared with those treated in 2001, driven mainly by a decrease in coil embolization (both P < .001); there was no change in 2-year mortality over the same period in the open repair group. In addition, the difference between the EVAR and open repair cohorts in long-term risk of either rupture or reintervention was smaller in the second vs the first half of the study period.

The authors acknowledge that the Medicare database lacks anatomical and clinical details that may have affected patient selection and outcomes. 

The results of the study are strikingly similar to those of the randomized OVER trial, its principal investigator, Frank A. Lederle, MD, of the Minneapolis VA Health Care System (Minneapolis, MN), told TCTMD in a telephone interview.

Barry T. Katzen, MD, of Baptist Health South Florida (Miami, FL), concurred, saying that the new study, with its “incredibly large” sample size, “adds to the body of work that supports EVAR.”

“A lot of people are now saying, ‘This is the era of endovascular repair—period,’” Dr. Lederle observed, but he added that there are still pros and cons to be weighed.

Rupture, Reintervention With EVAR Should Give Pause

“We are seeing the same message from multiple sources: Early mortality [with EVAR] is lower, but a few years out it is the same,” Dr. Lederle said. “Now, there is value in those few years, but with EVAR there is also the [burden] of more frequent imaging and the small but real possibility of rupture—which was the main concern in the first place. As long as you have that hanging over you, I think it’s still reasonable for people to opt for open repair.”

As for the reinterventions associated with the respective therapies, Dr. Lederle said most researchers consider them “roughly equal.” But Dr. Katzen demurred, suggesting that laparotomy, which is far more common after open repair, is “a significant surgical procedure,” whereas reinterventions linked to EVAR are mostly minor.

While calling the incidence of late rupture after EVAR “concerning,” Dr. Katzen made 2 points: The ruptures in the surgical arm should dispel the misconception that open repair eliminates the risk entirely. As for the excess ruptures in the endovascular arm, it is unclear what proportion occurred in patients lost to follow-up, he said, noting that other studies have shown that these catastrophic events often occur when patients stop undergoing routine surveillance.

The reason for the late catch-up in mortality in the EVAR group remains a puzzle, Dr. Lederle said. On the other hand, trial investigators have long “wondered whether the [survival] lines were going to cross so that open repair would turn out to be better long term, but we’ve never seen that,” he noted.

Given the long-term equipoise between EVAR and open repair, the main factors in the choice between the 2 approaches are anatomy—not everyone is a candidate for EVAR—and patient preference, Dr. Lederle said.

EVAR’s reduced invasiveness and lower early mortality are major advantages and account for its wide use, Dr. Katzen said. But, he added, its Achilles’ heel remains the secondary intervention rate, which necessitates long-term follow-up with imaging. Potential advances in technology that reduce the incidence of reintervention—primarily for endoleaks—“would have a dramatic effect on the whole field,” he said, noting that the relative benefit of EVAR over surgery is “a moving target.”

Dr. Lederle agreed but added: “Still, we do not have any reason to think that endovascular therapy is better.”


Schermerhorn ML, Buck DB, O’Malley AJ. Long-term outcomes of abdominal aortic aneurysm in the Medicare population. N Engl J Med. 2015;373:328-338.

Related Stories:


  • The study was supported by NIH grants.
  • Dr. Schermerhorn reports receiving grant support from Cook Medical, Medtronic, and WL Gore and personal fees from Endologix.
  • Dr. Lederle reports no relevant conflicts of interest.
  • Dr. Katzen reports serving on the advisory board for Boston Scientific, Medtronic Vascular, and WL Gore.

We Recommend