Discussion

Article Outline

• Copyright

Dr Richard Cambria (Boston, MA): Dr Buth, thanks for bringing to our Society yet another chapter from the EUROSTAR database.

Your report seems to echo a recent review assembled by Mark Morasch and his colleagues from Northwestern and published in the JVS within the last year, calling attention to the fact that perhaps a cavalier attitude towards left subclavian artery coverage is inappropriate. And in the Tag 9901 study, Dr Mike Makaroun pointed out that left subclavian artery coverage seemed to be an increased risk factor for stroke, not paraplegia, perhaps being a surrogate for manipulation of stent grafts in the arch.

I have a few questions, and thank you for giving me a copy of the manuscript, it certainly is exhaustively detailed. Did you analyze the risk of paraplegia as a function of pathology, dissection versus degenerative aneurysm, and the clinical circumstances of the procedure, urgent procedure versus elective procedure? Also, the overall rates of both paraplegia and stroke in this registry are admirably low. Just about 25 minutes ago I showed a series of stent grafts, all in urgent cases, but with a stroke rate of 10%. And indeed, in our own 250 cases, that figure is 9%; so I think the overall results in this study are in fact excellent.

I presume, and it is in your conclusion, you are prepared to recommend that cavalier left subclavian artery coverage is no longer appropriate?

Dr Jacob Buth: Starting with your last comment. Yes, I certainly recommend to revascularize in most instances the left subclavian artery unless the preoperative imaging is impeccable and indicates clearly collateral communications with the spinal cord. The imaging in our experience is often not conclusive.

With regard to the previous articles that you mentioned, Dr Morasch had an increased incidence of complications because of left subclavian artery, because there are many other complications that may occur by covering this artery. I remember he more often observed endoleakage or arm complications, not so frequently paraplegia.

Dr Makaroun found that stroke significantly correlated with LSA covering. In the current study, this comparison just did not make the level of statistical significance. However, if we combined stroke and spinal cord ischemia, this rate was statistically significantly higher in the group with a non-revascularized left subclavian artery covering.

Regarding your first comments, degenerative aneurysms had 2.5 times as frequent paraplegia than other thoracic disease, i.e. 3.7% compared to a 1.5% rate of paraplegia. This difference was not statistically significant.

Dr Roy Greenberg (Cleveland, Ohio): Dr Buth, I also want to thank you for a copy of the manuscript ahead of time, and I enjoyed the presentation. I had a couple of questions regarding your analysis, and I was a little bit surprised (or confused) as to how the proximity of the aneurysm to the subclavian was not significant, while the need for subclavian artery coverage was significant. Is there another reason to cover the subclavian besides aneurysmal involvement? I had believed the two entities to be inexorably linked.

The other question falls back on Dr Cambria’s question, which would be to clarify the message we should take from the paper. The EUROSTAR database involves a broad range of institutions, and nonconsecutive patient enrollment with a variety of diseases and other issues that come into play. Should we now universally recommend preoperative carotid subclavian bypass? Are there specific factors that you can tell us where we should be doing that and specific times when we don’t need to?

Dr Buth: Regarding the fact that we did not find a correlation between SCI and the localization of the disease and the LSA, this distance it was not measured as such. It was represented rather as a thoracic aorta segment. Thus, we had no accurate information of the localization of the lesion in relation to the left subclavian artery. This clearly is the reason that that was not a significant factor, while the covering itself was a significant factor for the development of neurologic complications.

With regard to the management recommendation based on our observations, I would think that we still have to learn more about vertebral artery imaging. In particular, how are the collaterals to the anterior spinal artery, which is the dominant side of supply, and how is the anatomy of the basilar artery of the circle of Willis. Sometimes one encounters an unexpected case of paraplegia. For example in a recent report, a patient with a traumatic tear of the thoracic aorta was described. This patient had a covering of the left subclavian artery and paraplegia developed quite unexpectedly. I know of other similar cases. I think that imaging of the collateral communication between the vertebral artery and the spinal cord often is imperfect.

Dr Girma Tefera (Madison, Wis): Enjoyed your presentation. I have two questions: One, when the strokes occurred, which hemispheres did it involve? Two, what is your philosophy regarding spinal protection. We always use CSF drainage and pharmacological protection. Did you analyze data pertaining to adjuncts for CSF protection? Thank you.

Dr Buth: The localization and the type of intercranial stroke (ie, whether anterior and posterior strokes were observed) could not accurately be retrieved from our data; similarly for left and right strokes. From what we could find and from previous reports, we know that there is not a straightforward correlation. So strokes are not always in the left anterior territory.

With regard to spinal cord protection, I can only quote from the literature because there was very little information on used methods of protection or on the late spinal cord ischemia in the current series in the registry’s case record form. Protection measures in TEVAR [thoracic endovascular aneurysm repair] usually are based on what is customary and what is known to be effective in open thoracic aorta repair. Frequently spinal fluid drainage is used. In reality, we don’t know exactly whether it is really necessary.