Invited commentary

The main strength of the accompanying study is its size. Even though the study group included two very different diseases, aneurysm and dissection, the numbers were still large enough to demonstrate significant variations in the rates of relatively rare neurologic complications. It appears that repeatedly instrumenting a diseased ascending aorta or arch in an elderly patient raises the risk of stroke, and occluding the left subclavian artery, isolating a long segment of aneurysmal thoracic aorta, and simultaneously repairing an abdominal aortic aneurysm (AAA) raises the risk of paraplegia. Admittedly, these conclusions rest on slightly tenuous assumptions. For example, the duration of the procedure was a surrogate for the extent of aortic instrumentation, and the number of stent grafts was a surrogate for the length of the covered segment. Nevertheless, the findings make sense and they agree with the findings of other studies.

In general, there are two ways to respond to this kind of information: change the procedure, or change the selection criteria. The current study indicates several risk factors for stroke that affect patient selection but not the conduct of the operation. All the possible procedural modifications relate to the risk of paraplegia. Although, the overall protective effect of carotid–subclavian bypass was quite modest, collateral flow through branches of the left subclavian artery may be more critical in patients who have other reasons for spinal arterial compromise. The same may be said of cerebrospinal fluid drainage, which has been shown in studies of open repair to have a spinal protective effect. The risks of simultaneous AAA repair have been noted before, but whether staging the operation would prevent paraplegia depends on the relative importance of hemodynamic instability, lumbar artery occlusion, and collateral development between stages.

Despite many publications on this subject, the occurrence of paraplegia remains a largely random event owing to the effects of currently unidentified risk factors. Some of this uncertainty may yield to new methods of imaging the spinal blood supply. Preliminary findings suggest that the source of spinal perfusion may be a strong predictor of paraplegia risk after open repair. The current study suggests that collateral pathways may be equally important. Imaging studies and other more direct measurements may also strengthen the analysis by providing a continuous variable, such as spinal perfusion, oxygenation, or metabolism, as an alternative to the current dichotomous outcome.