Invited commentary

I enjoyed reading this article, which is important because of its size and length of follow-up (at least 3 years in 121 of 325 patients). The clinical results are excellent, with a low rate of reintervention and aneurysm-related death beyond 30 days. That being said, the follow-up is primarily from 30 days to 5 years, defined as “mid-term” by Society for Vascular Surgery reporting standards,1 and should be taken in that context.

I was initially taken aback when I read that “Large prospective studies and registries have failed to produce the kind of device specific data…needed for informed decisions on patient selection, device selection, and follow-up.” I suspect the authors of the DREAM, EVAR, EUROSTAR, and other studies would take issue with that comment.2, 3, 4, 5 Nonetheless, I think Hiramoto et al are pointing out that a large single-center study can bring a unique perspective when it combines consistency in evaluation and follow-up with access to the patient’s chart and raw imaging data. Of course, this unique perspective can lead to bias, but it can also provide unique insights.

Most of the insights in this paper have a sound basis in fact, as well as some controversy: (1) The Zenith device has very good mid-term results, consistent with EUROSTAR, clinical trials, and other studies. I will avoid discussing the “barbs” regarding other devices, as this is not a concurrent comparative study including other devices. (2) Opinions regarding intervention for type II endoleak are changing. I would like to have seen more information about how often the treatment of endoleak was successful in stopping the endoleak or resolving aneurysm enlargement, as this would have helped influence decision-making regarding follow-up and intervention. (3) When a device has a low rate of failures and reinterventions, any imaging modality will have a low yield. Even at an intervention rate of 8.5% over an average 2.3 years’ follow-up, however, postoperative imaging should clearly be performed. The authors make a good point that if there is no endoleak on initial computed tomography, if there is no migration of components, and if the aneurysm is shrinking, no contrast should be needed on future computed tomography for a device with this track record. Noncontrast computed tomography provides good information about migration, deformation, aneurysm sac size, and, in some cases, fracture (the only missing parameter is endoleak). This article should not be used as an excuse to avoid imaging studies, nor do the authors suggest that.