Medicare data: Late outcomes in AAA stent-grafting age-dependent, rival surgery
Boston, MA - Long-term survival after endovascular abdominal aortic aneurysm (AAA) repair appeared to be better, compared with surgical repair, in a broad Medicare-based population than would be predicted by the less-favorable outcomes in several randomized trials, according to the authors of a retrospective case-matched study that also saw the survival advantage of endovascular repair rise with patient age [1].
Moreover, and also contrary to the prior studies, the new analysis suggests that the two treatment approaches pose about the same risk of late complications and reinterventions; earlier results had suggested that surgery had a significant advantage there.
Those earlier trials, EVAR-1 [2] and DREAM [3], may have suggested—misleadingly—that the short-term advantages of endovascular stent-grafting disappear within a year or so in part because they enrolled small numbers of patients treated at "selected high-volume centers," speculate Dr Marc L Schermerhorn (Beth Israel Deaconess Medical Center, Boston, MA) and colleagues in the January 31, 2008 issue of the New England Journal of Medicine.
But the current study, based on Medicare records from >45 000 patients undergoing AAA repair from 2001 to 2004, supports and extends the short-term results of both EVAR-1 and DREAM, suggesting superior perioperative morbidity and mortality with endovascular repair compared with open surgery. "Our study takes that and demonstrates that those results are applicable to the entire Medicare population," Schermerhorn told heartwire.
He and his colleagues retrospectively tracked 22 830 pairs of patients who were >67 years old when they underwent AAA repair by stent-grafting or conventional surgery after matching them according to propensity scores based on demographic and clinical data covering the previous two years.
Comparing endovascular and open-surgery groups, perioperative mortality was 1.2% and 4.8%, respectively (p<0.001). Short-term survival was better with stent-grafting regardless of age, the group observed. But in a novel finding, the gap between the two approaches widened with increasing patient age.
Perioperative mortality* by age and procedure in 22 830 matched pairs
Age at time of procedure (y)
|
Stent-graft (%)
|
Open surgery (%)
|
67-69
|
0.4
|
2.5
|
70-74
|
0.8
|
3.3
|
75-79
|
1.3
|
4.8
|
80-84
|
1.6
|
7.2
|
>85
|
2.7
|
11.2
|
>67 (all patients)
|
1.2
|
4.8
|
|
All differences between groups, p<0.001
*Rate of death during hospitalization or within 30 days of the procedure
Long-term survival after endovascular repair was highly dependent on perioperative survival, Schermerhorn observed, such that among older patients it favored stent-graft recipients compared with those who underwent surgery.
Whereas the two groups' overall survival curves—which initially favored stent-grafting—crossed after about one or 1.5 years in the earlier trials, he said, his group's analysis suggests they actually cross at about three years overall and beyond four years for the oldest patients.
Schermerhorn also observed that, in prior studies, the risk of important late complications had been consistently lower with surgical repair than with stent-grafting. Sure enough, after four years in the Medicare analysis, aortic rupture and AAA-related major and minor reinterventions were significantly more common in the endovascular repair cohort.
But the earlier studies were misleading, according to Schermerhorn, because the tracked complications were limited to those related to the aneurysm. "What no one has ever looked at before is [the rate of] complications related to the laparotomy," he said, which were about twice as common among the patients who had conventional surgery. "The reintervention rate for aneurysms is higher with endovascular repair, but it's balanced by reinterventions due to the laparotomy at open surgery."
Four-year complication rates by procedure in 22 830 matched pairs
Complication
|
Stent-graft (%)
|
Open surgery (%)
|
p
|
AAA-rupture-related
|
|
|
|
Aortic rupture
|
1.8
|
0.5
|
<0.001
|
Major reinterventiona
|
1.6
|
0.6
|
<0.001
|
Minor reintervention
|
7.8
|
1.3
|
<0.001
|
Laparotomy-related
|
|
|
|
Repair of abdominal wall hernia
|
1.1
|
5.8
|
<0.001
|
Bowel resection
|
3.0
|
3.4
|
0.02
|
Hospitalizationb
|
8.1
|
14.2
|
<0.001
|
|
a. Includes open surgical repair with aortic bypass, conversion to open surgical repair, or repair of infected graft
b. For diagnosis of either bowel obstruction or abdominal-wall hernia
"All we're saying is, as it's being practiced right now in the Medicare population, the outcomes with endovascular repair are quite good, and they're at least as good as with open-surgery repair," commented Schermerhorn. His personal approach, he said, is to recommend endovascular over surgical repair regardless of patient age, as long as the aortic anatomy is suitable for stent-grafting—ie, there is adequate normal proximal and distal aorta for anchoring the graft.
But, he said, "in very old patients I would be willing to push the envelope a little more in terms of [accepting] suboptimal anatomy and put a stent-graft in them. And I think that's the way it's being practiced and why the results are as good as they are."
No potential conflicts of interest appeared in the report.
|
|