Discussion

Dr Michael J. Singh (Rochester, NY). Dr Cho and his colleagues have addressed several important issues in the management of ruptured thoracoabdominal aortic aneurysms. Your study has cleared the air and provided a benchmark for direct open repair of nondissected ruptured thoracoabdominal aneurysms. The majority of previously published reports failed to separate aortic dissections from degenerative aneurysms. The authors have analyzed the outcomes, risk factors, and natural history of a select group of patients over an 8-year period.

Contemporary series have reported an elective mortality rate of 6% to 19% for open repair and a repair-free survival of 52% at 2 years. I was pleasantly surprised to learn that your operative mortality rate for a rupture was 26% and the 2-year survival was nearly 50%. Even more interesting is your Kaplan-Meier survival analysis that shows there is no significant difference in survival between octogenarians and those who are younger.

In review of your manuscript, I noticed that four of your 40 patients were inpatients at the time of rupture, fully 10%. This seems unusual. Was the timing of their rupture coincidental, or were the patients being observed for their aneurysms in-house? Were their outcomes improved by a presumed earlier diagnosis and expeditious operative procedure? Could you provide some additional information of the instituted changes that your group has made to improve your outcomes in the latter part of your series and has this trend continued?

Endovascular stent grafts have reduced mortality and complication rates in traumatic aortic injuries and ruptured aortic aneurysms; however, I am hesitant to imagine that ruptured thoracoabdominal aneurysms will have an improved outcome with an emergent fenestrated endovascular repair. At your institution, have you had any experience with an endovascular repair for this catastrophic problem? As endovascular techniques evolve, I suspect that many endovascular experts will reference this article for future comparison.

Dr Joel Barbato. Thank you very much for your comments. As far as your first question referring to those patients who ruptured while in the hospital setting, what I can tell you is that all of them were evaluated for rupture. They all had preoperative or in-hospital CT scans, or were referred from other hospitals with CT scans that failed to demonstrate a rupture. As far as whether this is unusual, I can’t speak to that. And I don’t think we broke down the outcomes in those particular patients who ruptured within the hospital. As I mentioned, those patients with a shorter duration from symptom onset to OR actually had worse outcomes, and that’s likely a selection bias.

As far as changes in practice at our institution, obviously none of the factors that I showed up on the screen had a significant impact in terms of mortality. One of the things that has been instituted recently is an increased use of Narcan, one of the particular surgeons uses that, although that wasn’t associated with reduction in mortality.

The other institutional change is that whereas a total of four surgeons performed the operations in the first half of the series, only two surgeons have been performing this operation since 2003 at our institution. And I think what we do see here is that with increasing surgeon and institutional experience, the outcomes, for whatever reason, seemed to be improved.

As far as the endovascular repair, the short answer is we have not used either a hybrid or a fenestrated graft for the repair, particularly in this setting, although we have used them for isolated descending thoracic aortic aneurysms, both in the traumatic setting as well as in the degenerative ruptured setting.

There are only two studies that I am aware of that actually exist in the literature that describe these hybrid technologies in the setting of emergent or rupture. In one of these studies, all three of the patients that had utilization of this hybrid technology died in the postprocedure period. So I think your comment is a good one, that there certainly is a steep learning curve to these complex technologies. And I think the applications, at least initially here, are going to be somewhat limited.