Discussion

Dr William Quinones-Baldrich (Los Angeles, Calif): This is a review of events occurring after 30 days of implantation of the Zenith aortic endograft for the treatment of infrarenal abdominal aortic aneurysms. The authors report an incidence of type II endoleak of 5.8%, which is quite low, and no type I endoleaks. Similarly, graft occlusions were rare, only one requiring surgical treatment with a femorofemoral bypass. The authors also find that hypertensive patients with type II endoleaks more often require intervention for increase in the size of the aneurysm. The latter is an important finding as it strongly suggests that in the presence of a type II endoleak, blood pressure control is of benefit.

The authors also conclude that routine follow-up imaging plays a limited role in the identification and prevention of graft failure. They now recommend CT scan without contrast, noting that the size of the aneurysm and detection of migration are the most important factors during follow-up. Several reports are noting a gradual decrease in renal function in patients treated with endovascular grafts regardless of whether or not suprarenal fixation is present. Therefore, avoiding contrast during follow-up in these patients is important.

The occurrence of type III endoleaks leads the authors to increase the degree of intercomponent overlap to a minimum of 1½ stents. Their experience with graft limb occlusion has led them to be more aggressive in placing stents in areas of concern at the time of implantation. These are excellent recommendations based on the authors’ extensive experience with this device.

Unfortunately, this report only presents part of the picture. By omitting events during the first 30 days of follow-up, the incidence of postimplantation events is underestimated. In the manuscript, it is clear that several type I endoleaks were treated with proximal extensions, some limb occlusions were treated with thrombolysis and stent placement, and some type III endoleaks were also detected. They have also observed stent fractures, barb separation, and top stent separation. The latter events are more likely to have occurred after 30 days. I have several questions for the authors.

During follow-up you found that women tend to have more events than men. My first question is why are there so few women in your cohort? Does gender influence your recommendations? Even though it is not the emphasis of your presentation today, could you give us information on events occurring in the first 30 days? What is the overall incidence of type I, II, and III endoleaks in your experience? When and how often were the stent fractures, bar and/or stent separation observed? Do you have information on long-term renal function in patients receiving the Zenith endograft for treatment of their abdominal aortic aneurysms? And finally, in my practice I usually obtain a CT scan as baseline at a month, and if everything looks well, I will then alternate duplex scan and CT during the follow-up. Have you had any experience using duplex scan for follow-up?

I congratulate the authors on an excellent experience and thank you for providing me a copy of the manuscript in advance of the meeting. I also wish to thank the society for the privilege of discussing this paper.

Dr Jade Hiramoto: Thank you, Dr Quinones. With regard to your first question, we did not alter our recommendations for treatment based on gender. We did have very few women in our study. Part of this may be explained by the fact that a large proportion of our operations were performed at the VA, where essentially all of the patients are men. Gender may initially have played a role in our recommendations when we were worried that because of access issues they might not be appropriate candidates; however, we use the same anatomic criteria for all patients. We do not take gender into consideration now. We look at the diameters of our access vessels and make appropriate recommendations based on those size measurements.

Women did tend to do worse in terms of late graft failure. They were also more likely to require treatment for their type II endoleaks, and if you actually look at all interventions—early complications, late complications, treatment of type II endoleaks—it is statistically significant. It did not reach statistical significance when you just looked at each of these separately, but if you pooled all interventions they (women) did worse. I think it is difficult to explain. The women tended to be older on presentation for their aneurysm repair, but there did not appear to be an interaction between age and gender. Gender itself was an independent predictor.

Again, this talk was really devoted towards looking at the long-term complications since this device has been in use since 1998 at UCSF, and we wanted to analyze and present our long-term data. However, with regards to complications from within the first 30 days, there were a total of 10, for an overall late and early complication rate of 19 in the entire cohort. The majority of the early complications, as you might imagine, were limb thromboses. It was probably 7 out of those 10. There were probably 1 or 2 type I endoleaks and no treated type II endoleaks in the first 30 days.

With regards to stent fracture, barb fractures, and stent separation, there has been a total of four in our series and they have not required treatment. They have been recognized on the follow-up abdominal x-rays and followed but not treated.

In terms of the long-term renal function of these patients, I do not have all of their data analyzed at this point to determine the number of patients who have had worsening renal function or those who have required dialysis, but we can certainly look back at our data. I do have their preoperative values, and there was quite a range, but the average preoperative creatinine was 1.2. With regards to duplex ultrasound, I think that it is an excellent imaging modality. I think in the face of going to noncontrast CT scans if the 1-month postoperative contrast-enhanced CT does not demonstrate enlargement of the aneurysm or an endoleak, I think a duplex would be a great adjunctive measure to include. This would be especially useful in a patient who maybe has some renal insufficiency, gets a noncontrast CT scan, and there is a question of aneurysm sac enlargement; perhaps a duplex would be useful in that patient to determine if they have evidence of an endoleak.

8. “Late Complications After Endovascular Aneurysm Repair Using the Zenith Stent Graft.” Discussion by Benjamin Starnes, Tacoma, WA.

Dr Benjamin Starnes (Tacoma, Wash): I congratulate you on a very impressive experience. One of the unique aspects, and in my opinion, one of the favorable aspects of the Zenith system is that the main body is sized all the way down to the aortic bifurcation and this theoretically assists in preventing distal migration. You had one case where the renal artery was occluded, you presumed, because of overlap of the orifice of the renal artery. I am wondering, because I have had one similar case, have you ever seen any instance of proximal migration of the graft after aneurysm remodeling, especially in very large aortic aneurysms?

Dr Hiramoto: Not to my knowledge, and certainly not in this series of patients. This one case of renal artery occlusion that we presented today was clearly an intraoperative event that was not recognized at the time of the procedure. Looking back at the intraoperative angiogram, you see that the top of the covered portion of the stent graft was almost completely covering that renal artery. This was the only patient in which that occurred, and again, I think this was an early mistake that was recognized late. This was not a case of proximal migration, nor have we observed proximal migration in this series of patients.