Twenty-one cases of aortoenteric fistula: lessons for the general surgeon.

ABSTRACT

We retrospectively reviewed our experience from 1984 to 2001 with 21 cases of aortoenteric fistula (AEF) in 19 patients. The majority of cases were in men (13 of 19, 68%). One AEF was spontaneous, the other 20 developed after prior vascular reconstruction (95%). The majority of AEF were duodenal (48%) followed by small bowel (38%), colon (10%), and esophageal AEF (5%). The proximal anastomosis of the prior vascular repair was the site of AEF origin in 62 per cent of cases, the distal anastomosis accounted for 19 per cent, and the body of the graft for 14 per cent. The intestinal repair was chosen on a case-by-case basis by the general surgeon and consisted of a simple primary repair in 48 per cent, resection with primary anastomosis in 38 per cent, and patching with pleura or omentum in individual cases. Colostomies were created in the two cases with colonic AEF. The duodenum was excluded in one of 10 duodenal AEF. Six patients (32%) died in the 90 days following surgery. The biggest risk of postoperative death was presentation with sepsis (P = 0.069); interestingly, women were more likely to present with sepsis (P = 0.019) and experienced a disproportionate rate of postoperative death (male 23%, female 50%, P = 0.24). The method used to repair the bowel was linked to a higher rate of postoperative death, and patients that required bowel resection died more frequently (66%) than those who had a simple repair (10%, P = 0.07). Overall mortality with AEF remains high despite routine SICU care. The biggest risk for death is preoperative sepsis. Women presented with sepsis more frequently than men. The method of bowel repair appears to be related to overall survival and along with sepsis is, perhaps, a surrogate for the degree of erosion present at the site of the AEF. Simple bowel repairs were sufficient when technically possible. Duodenal exclusion is not an obligatory adjunct to duodenal repairs.