Invited commentary

Article Outline

• Copyright

The issue of the absolute necessity of life-long surveillance in patients who have undergone endovascular abdominal aortic aneurysm repair (EVAR) is certainly an important and timely topic from a number of perspectives. Clearly, there are appropriate medical concerns regarding the repetitive use of iodinated contrast materials and their possible cumulative deleterious effects on renal function as well as apprehension regarding potential carcinogenic consequences of recurring radiation exposure. In addition, the issue of the societal cost of these studies cannot be completely overlooked from a global health care standpoint.

To address these concerns, a variety of alternative forms of surveillance in EVAR patients have been proposed, including noncontrast computed tomography (CT) studies, traditional and contrast-enhanced duplex ultrasound, gadolinium-enhanced magnetic resonance imaging, implantable wireless sac pressure sensors, and intravascular ultrasound. However, currently the precise role and accuracy of these alternatives remains ill defined.

The current study presents notable information. Patients with an absence of early endoleak on a 30-day CT scan had a significant reduction in aneurysm-related morbidity (ARM) at 1 to 5 years. It should be noted, however, that although the definition of ARM included obvious events such as aneurysm-related death, rupture, endograft migration, and open conversion, it also included “secondary interventions,” which intuitively would be expected to be more prevalent in patients with known endoleaks. Patients with a cumulative absence of endoleak at 6 months (79.4% of patients) and at 1 year (77.6% of patients), however, had an identical 5-year freedom from any type of ARM. The authors state, therefore, that routine surveillance for continued absence of endoleak at 6 months and 1 year did not greatly alter the predicted 5-year freedom from ARM.

They conclude that the absence of an early endoleak therefore identifies a cohort of patients at substantially reduced risk for ARM in whom more relaxed surveillance regimens may be appropriate. However, there was a 10.5% risk of ARM at 5 years in patients who had never experienced an endoleak. In addition, it is correctly pointed out that not all ARM (for example, graft limb thrombosis) will necessarily be predicted by an abnormality seen on a routinely performed imaging study; nor can all ARM be prevented by even very aggressive surveillance and monitoring programs.

Finally, in the Zenith trials there was a population of patients in whom a delayed endoleak occurred: A 12% to 15% risk of late endoleak, most often of type II, was noted even when no leak was identified on the initial 30-day CT scan. The clinical significance of these delayed leaks remains unclear. At our institution we have anecdotally seen a number of patients in whom 3 or 4 years of initially negative CT follow-up was subsequently followed by the novel appearance of an endoleak on the next annual CT scan. Several of these late leaks eventually required secondary intervention and treatment.

The recommendations of the current study, although attractive, should be interpreted with caution. As the authors correctly point out, these excellent results were obtained only using one type of commercially available aortic stent graft system. In addition, and perhaps even more importantly, these results were obtained in patients who underwent intervention as part of a strictly controlled clinical trial. As such, these patients represent a cohort of cases with ideal anatomic parameters, and the long-term results are likely not applicable to post-trial EVAR cases, in which preoperative anatomic selection criteria is likely to have been relaxed by the individual practitioner.

If the algorithm adopted in the article had been strictly followed in this cohort of patients, this would have resulted in 23.4% of patients having no routinely prescribed follow-up except for a yearly ultrasound study after their negative result on the 1-year CT scan. Even this excellent and compelling data and analysis do not support such a radical change in follow-up protocols at the present time. Although there undoubtedly exists a cohort of patients who may not require lifelong intensive surveillance imaging after EVAR, they remain imprecisely characterized at the present time.