Late aortoenteric fistulae may occur in less than 1% of patients undergoing endovascular aneurysm repair (EVAR), with an increased risk in cases of emergency EVAR or those performed for pseudoaneurysm following previous aortic surgery. The data came from a study presented at the annual meeting of the European Society for Vascular Surgery (ESVS; 23–25 September, Porto, Portugal).
Andrea Kahlberg, working with the group of Roberto Chiesa at Vita-Salute University School of Medicine, San Raffaele Institute, Milan, Italy, explained that as the number of EVAR patients has grown in recent years, the number of related complications has also increased. Of these complications, an aortoenteric fistula is considered a “dramatic and lethal” problem. However, evidence regarding aortoenteric fistula is “scant” in the existing literature, largely based on single-centre case reports—to date 53 aortoenteric fistula patients have been documented in 20 case reports and seven small series.
As such, Kahlberg and colleagues initiated a retrospective multicentre study at eight Italian universities and hospital centres to investigate the incidence, clinical features, therapeutic options, and outcomes of aortoenteric fistulization following EVAR—the MAEFISTO Project. This is the first multicentre report of post-EVAR aortoenteric fistulae.
The research team collected electronic data from 1997–2013, identifying 3,932 EVAR patients, of whom 32 (0.8%) developed aortoenteric fistulae following surgery at a median time of 18.5 months post-EVAR.
The data indicated that there was an increased risk of aortoenteric fistulae if EVAR was initially indicated for pseudoaneurysm (35% vs. 5% in the global cohort, p<0.0001). Kahlberg suggested that this could be due to pre-existing subclinical infections or aortoenteric fistulae, the saccular shape of pseudoaneurysms or the mechanical stress of an EVAR procedure. Emergency EVARs were also shown to carry a higher risk of an eventual aortoenteric fistula (22% vs. 8% in the global cohort, p=0.01), possibly due to periaortic haematoma or compression, or an increased inflammatory response in emergency procedures.
Melena (observed in 66% of patients) and fever (observed in 63% of patients) were the two most frequent symptoms of post-EVAR aortoenteric fistula, and Kahlberg reported a relatively low rate of shock (19% of patients), suggesting that this may be a result of the protecting role of the stent graft against free bleeding.
In terms of treatment methods, Kahlberg et al found that conservative treatment of post-EVAR aortoenteric fistula initially resulted in a higher survival rate than conventional treatment, although there was a risk of recurrence and death at longer follow-up. Five patients were treated conservatively, of which two (40%) died at seven and 15 months respectively, while the remaining three were alive at a median follow-up of 12 months. The remaining 27 patients underwent aortoenteric fistula surgical treatment, with a perioperative mortality of 37% (n=10). No additional aortic-related death was recorded in operated patients at a median follow-up of 28 months.
In comparing different imaging modalities, the researchers found that computed tomography was the most reliable imaging technique, producing a 100% (27) positive predictive value after scanning 27 patients. Endoscopy was “not always useful”, with a positive predictive value in 76% (16) of the 21 patients scanned, while FDG positron emission tomography was “highly sensitive, but not specific” (positive predictive value of 100% in the 10 patients scanned).
Kahlberg closed by telling delegates that the data suggested that aortoenteric fistula after EVAR is a rare complication, the risk of which is increased in the case of EVAR for pseudoaneurysm or in an emergency. Conservative treatment of aortoenteric fistulae has a “possible initial role in selected cases”. Surgical treatment produces a high rate of perioperative mortality but survival may be maintained at long-term.