Embolization As Cause of Bowel Ischemia After Endovascular Abdominal Aortic Aneurysm Repair

CORE Metadata, citation and similar papers at core.ac.uk

Provided by Elsevier - Publisher Connector From the Eastern Vascular Society

Embolization as cause of bowel ischemia after endovascular abdominal aortic aneurysm repair

Wayne W. Zhang, MD,a Mahmoud N. Kulaylat, MD,b Paul M. Anain, MD,c Hasan H. Dosluoglu, MD,a Linda M. Harris, MD,a Gregory S. Cherr, MD,a Merril T. Dayton, MD,b and Maciej L. Dryjski, MD, PhD,a Buffalo, NY

Objective: We investigated the incidence, cause, and outcome of large bowel and small bowel ischemia after endovascular abdominal aortic aneurysm (AAA) repair. Methods: Medical records for all patients undergoing endovascular AAA repair from December 1999 to December 2003 were reviewed. The incidence, cause, and outcome of clinically detected postoperative bowel ischemia were analyzed. Results: Seven hundred two endovascular AAA repairs were performed. In 10 patients (1.4%) acute bowel ischemia developed. Six of these patients sustained concurrent small bowel necrosis, and the remaining 4 had isolated colon ischemia. Seven patients underwent exploratory laparotomy. In 6 of these bowel resection was performed, and in 1 patient the ischemic bowel was unsalvageable. Of the 6 patients with small and large bowel ischemia, 4 had segmental or patchy necrosis, which was separated by normal-appearing intestine, and 1 had extensive ischemia that involved most of the small bowel and the entire colon, with pathologic evidence of microembolization. Three patients had preoperative occlusion of the inferior mesenteric artery. One had unilateral and 1 had bilateral hypogastric artery interruption. Five of the 6 patients with small bowel ischemia had thrombus or atheroma in the proximal aneurysmal necks. All patients with isolated colon ischemia survived. All 6 patients with concurrent small bowel ischemia died. Conclusion: The total incidence of clinically evident bowel ischemia after endovascular AAA repair is similar to that after open surgery. However, small bowel ischemia occurs more commonly in patients with endovascular repair, and is associated with extremely high mortality. The direct pathologic evidence and the patterns of segmental, skipped, or patchy ischemia in most patients imply that microembolization has an important role. (J Vasc Surg 2004;40:867-72.)

Bowel ischemia is a devastating complication after open Dadian et al6 reported that colon ischemia after endo- or endovascular abdominal aortic aneurysm (AAA) repair. vascular aneurysm repair (EVAR) occurred in 1 of 122 Retrospective studies have documented the incidence of patients (0.8%) who required unilateral or bilateral hypo- colonic ischemia at 1.2% to 2.0% after open procedures1-5 gastric artery interruption and in 7 of 156 patients (4.5%) and 2.9% after endovascular repair,6 and prospective without hypogastric artery interruption. Microemboliza- colonoscopic investigations have shown that bowel isch- tion as a cause of postoperative colon ischemia may be more emia may occur as frequently as 6% to 30% in patients after important in patients who undergo EVAR than in those 6 open surgery.7-10 Severity varies from mucosal to transmu- who undergo open repair. Other possible causes include ral ischemia, and the mortality rate ranges from 0% to intraoperative hypotension, reperfusion injury, anatomic 100%.4-7,10Interruption of the inferior mesenteric artery features, bowel manipulation, and iatrogenic injury, but and hypogastric artery have been presumed to be the major these mechanisms are not likely to have a role in EVAR. Although EVAR has been widely used for treatment of cause. However, in a multicenter study the incidence of AAA in the past several years, the incidence and cause of postoperative colonic ischemia was 1.3% in 7 centers in postoperative bowel ischemia, especially small bowel isch- which inferior mesenteric artery revascularization was rou- emia, remain unclear. The purpose of this study was to tinely performed, and 1.1% in 28 centers in which inferior investigate the incidence, cause, and outcome of large and mesenteric artery revascularization was not performed small bowel ischemia after endovascular AAA repair. routinely.1This suggests that other causes may be more important. METHODS

From the Division of Vascular Surgery, Department of Surgery, State This study was performed as a retrospective chart re- University of New York at Buffalo,a the Department of Surgery, State view. No attempt was made to identify subclinical cases of University of New York at Buffalo,b and the Division of Vascular Surgery, bowel ischemia. Medical records for all patients who under- c Buffalo Catholic Health System. went endovascular repair from December 1999 to Decem- Competition of interest: none. Presented at the Eighteenth Annual Meeting of the Eastern Vascular Soci- ber 2003 in Kaleida Health System and Buffalo Catholic ety, Philadelphia, Pa, Apr 29–May 2, 2004. Health System were reviewed. Endografts including Aneu- Reprint requests: Maciej L. Dryjski, MD, PhD, Department of Surgery, Rx (Medtronic/AVE), Ancure (Guidant), Zenith (Cook), State University of New York at Buffalo, Millard Fillmore Hospital, 3 and Excluder (W. L. Gore & Associates) were used. Various Gates Circle, Buffalo, NY 14209 (e-mail: [email protected]). sizes of standard grafts and sheaths were placed, based on 0741-5214/$30.00 Copyright © 2004 by The Society for Vascular Surgery. the different sizes of aorta and access vessels in different doi:10.1016/j.jvs.2004.08.054 patients. 867 JOURNAL OF VASCULAR SURGERY 868 Zhang et al November 2004

Table I. Demographic data for patients with colon and small bowel ischemia after endovascular AAA repair

Operating Aneurysm time for Small diameter AAA Site of Colon bowel Previous bowel Other related Patient (cm) (min) aneurysm ischemia ischemia resection Management complications Graft Outcomes

1 5.7 420 Infrarenal Left Distal No Left Left “blue Ancure Death colon jejunum hemicolectomy; toes” (aspiration segmental pneumonia resection of 4 months distal jejunum later) 2 6.8 360 Juxtarenal Sigmoid Distal Right Left No* Ancure Death ileum hemicolectomy hemicolectomy; segmental resection of distal ileum 3 8 160 Infrarenal Left Distal No Subtotal Acute renal AneuRx Death colon ileum colectomy; failure segmental resection of small bowel 4 5.3 280 Infrarenal Sigmoid Distal No Extended left Acute renal AneuRx Death and ileum hemicolectomy failure rectum 5 7 115 Infrarenal Sigmoid No No Sigmoidectomy Left trash AneuRx Survival buttock 6 6.3 235 Infrarenal Entire Jejunum No Subtotal Acute renal AneuRx Death colon and colectomy; failure ileum segmental resection of small bowel; second exploration 7 5.7 50 Infrarenal Left No No Conservative No Ancure Survival colon 8 7.3 68 Infrarenal Left No No Conservative No Ancure Survival colon 9 6 165 Infrarenal Entire Jejunum Sigmoidectomy Exploratory Acute renal AneuRx Death colon and laparotomy failure ileum (open and close) 10 5.4 90 Infrarenal Sigmoid No No Conservative No AneuRx Survival

AAA, Abdominal aortic aneurysm. *Patient had chronic renal failure and had been receiving hemodialysis for years.

Charts were reviewed to identify all cases of clinically patients were men, and 6 were women; their age range was documented bowel ischemia. Bowel ischemia was diag- 68 to 85 years. Aneurysms were 5.3 to 8.0 cm in diameter. nosed on the basis of clinical ﬁndings at presentation, Nine aneurysms were infrarenal, and 1 was juxtarenal. The endoscopy, or exploratory laparotomy. In patients in whom patient with the juxtarenal aneurysm had end-stage renal acute ischemia developed, additional data, including initial disease, and had been receiving hemodialysis for several history and physical examination, laboratory analyses (com- years before EVAR. In this patient the juxtarenal aneurysm plete blood cell count; electrolytes, arterial blood gas, was repaired by deploying an Ancure graft above the renal lactate acid levels), anesthesia records, colonoscopy de- arteries. The percentage of bowel ischemia was 1.4% (6 of scriptions, operative reports, and pathologic results were 429) in patients who received AneuRx grafts, and 1.6% (4 collected. Preoperative and postoperative images, includ- of 255) in patients who received Ancure grafts. Excluder ing computed tomography (CT) scans and aortograms, in and Zenith grafts were placed in only 9 patients each with patients with bowel ischemia were reviewed. no bowel ischemia. Three patients had preexisting occlusion of the inferior RESULTS mesenteric artery. One patient underwent unilateral hypo- Seven hundred two patients underwent EVAR during gastric artery interruption with microcoils before EVAR, the study period. In 10 patients (1.4%) acute bowel isch- and 1 patient had bilateral hypogastric artery interruption emia developed. All 10 patients had colon ischemia, and 6 from endovascular grafts without revascularization. Of the sustained concurrent small bowel infarction (Table I). Four 10 patients with bowel ischemia, only 1 had mild (Ͻ50%) JOURNAL OF VASCULAR SURGERY Volume 40, Number 5 Zhang et al 869

Fig 1. Computed tomography scans show thrombus or atheroma in proximal neck of abdominal aortic aneurysm (arrows; A), and thrombus or atheroma in bilateral iliac arteries (arrows; B). These are possible sources of microembolization during endovascular aneurysm repair. superior mesenteric artery stenosis. The celiac arteries in all At presentation the 6 patients with both small and large 10 patients were patent. None of the 10 patients had bowel ischemia had postoperative abdominal distention, preexisting symptoms of mesenteric ischemia. One patient shock, and acidosis. All 6 patients, except 1 who received a had intraoperative hypotension, with systolic blood pres- nasogastric tube and sedation postoperatively, complained sure of 70 to 80 mm Hg for 15 minutes. One patient had of abdominal pain. Two patients had bloody diarrhea. The intraoperative oliguria, and 1 patient with end-stage renal 4 patients with isolated colon ischemia had abdominal pain, disease had been receiving hemodialysis for several years. and 3 had early postoperative diarrhea.

Table II. Anatomic features of AAA and ischemic bowel

Thrombus or Patency of hypogastric artery atheroma in Thrombus or Bowel Preoperative aneurysm atheroma in Type of Direct evidence of Patient ischemia Preoperative Postoperative patency of IMA neck access vessels ischemia microembolization

1 Colon/ Patent Patent Occluded Yes Bilateral iliac Transmural Skipped, segmental small arteries ischemia bowel 2 Colon/ Patent Patent Patent Yes No Transmural Skipped, segmental small ischemia bowel 3 Colon/ Patent Patent Patent Yes No Transmural Skipped, segmental small ischemia bowel 4 Colon/ Patent Bilaterally Occluded No No Transmural Skipped, segmental small interrupted ischemia bowel 5 Sigmoid Patent Right Occluded Yes Bilateral iliac Transmural None interrupted arteries 6 Colon/ Patent Patent Patent Yes No Transmural Microscopic small embolization bowel 7 Colon Patent Patent Patent No No Mucosal Patchy ischemia 8 Colon Patent Patent Patent No No Mucosal Patchy ischemia 9 Colon/ Patent Patent Patent Yes No Transmural None small bowel 10 Sigmoid Patent Patent Patent No No Mucosal Patchy ischemia

AAA, Abdominal aortic aneurysm; IMA, inferior mesenteric artery. JOURNAL OF VASCULAR SURGERY 870 Zhang et al November 2004

separated by normal-appearing intestine, and 2 had extensive ischemia involving most of the small bowel and the entire colon. One of these patients demonstrated direct pathologic evidence of massive microembolization. Four patients with combined ischemia also had acute renal failure postoperatively. Microembolization resulting in trash buttock occurred in 1 of these patients, and distal embolization with “blue toes” occurred in another patient. Review of records for patients with isolated colonic ischemia revealed that 3 of 4 had mucosal ischemia only, which was managed conservatively. In the 7 patients who underwent exploratory laparotomy, 6 had bowel resection, and in 1 patient the ischemic bowel was unsalvageable. Five of the 6 patients with concurrent small bowel ischemia died early, of multiple organ failure related to acute bowel ischemia, and 1 patient died of aspiration pneumonia 4 months postoperatively. All 4 patients with isolated colon ischemia survived. They have been followed up for 18 to 41 months, with no evidence of graft infection. DISCUSSION Colon ischemia after open AAA repair is a critical complication, and has been well addressed.11-13 However, the incidence, cause, and outcome of postoperative bowel, especially small bowel, ischemia after EVAR are poorly documented.6,14-15 Our data show that the incidence of clinically signifi- cant bowel ischemia after EVAR is 1.4%, which is approximately that after open surgery1-5; and small bowel ischemia occurred in 0.8% of patients. Sixty percent of patients with postoperative colon ischemia had concurrent small bowel ischemic infarction, which differs significantly from the ischemia noted after open repair. Prognosis with both large and small bowel ischemia is dismal, with mortality rate as high as 100% in our study. Most patients died of sepsis, shock, and multiple organ system failure. Isolated colon Fig 2. Thrombus or atheroma in aneurysm neck is dislodged while deploying the graft, and debris is flushed upstream into ischemia has a favorable outcome with conservative or superior mesenteric artery by blood flow turbulence caused by surgical treatment. guide wire manipulation, graft deployment, and balloon dilation. Interruption of the inferior mesenteric artery and hypogastric artery has been suggested as the major cause of Five of the 6 patients with both large and small bowel bowel ischemia after open AAA repair.11,12,16-19 However, ischemia had thrombus or atheroma in the proximal neck of our data indicate that inferior mesenteric artery interrup- the aneurysm. One of the 4 patients with isolated colon tion may not be so important as was suggested previously. ischemia had thrombus or atheroma in the proximal aneu- Although the inferior mesenteric artery is inevitably inter- rysmal neck and bilateral common iliac arteries (Fig 1). The rupted in all patients at EVAR, the incidence of postoper- presence of thrombus or atheroma in proximal neck and ative bowel ischemia in our group was not increased versus access vessels was more frequent in patients with concur- that in open procedures. This could be due to the preserved rent small bowel ischemia (5 of 6) than in the patients with collateral arteries in EVAR, which are frequently jeopar- isolated colon ischemia (1 of 4; Table II). CT scans were dized by dissection during open surgery. reviewed for 100 of 692 patients in whom bowel ischemia Another possible cause of bowel ischemia after open did not develop. Twenty-one patients (21%) had thrombus and endovascular AAA repair is hypogastric artery interrup- or atheroma in the aneurysm neck. The incidence of aneu- tion. However, previous studies have debated the necessity rysm neck thrombus or atheroma in patients with small of hypogastric artery revascularization in the patients with bowel ischemia (5 of 6, 90%) and without bowel ischemia bilateral hypogastric artery interruption.1,17,19-22 Eight of (21 of 100, 21%) was compared statistically with the Fisher 10 patients with bowel ischemia in our group had patent exact test, and was significantly different (P ϭ .003). hypogastric arteries bilaterally. Only 2 patients had inter- Of the 6 patients with combined small and large bowel ruption of the hypogastric artery: 1 patient had bilateral ischemia, 4 had segmental or patchy necrosis, which was hypogastric artery occlusion and 1 patient had unilateral JOURNAL OF VASCULAR SURGERY Volume 40, Number 5 Zhang et al 871 hypogastric artery occlusion, from endovascular grafting. inferior to the renal arteries, rather than dragging them Other recent studies have also documented that unilateral, down from above; and (4) open repair if a large amount of and even bilateral, hypogastric artery interruption is accept- thrombus is present in the proximal aneurysm neck. able during EVAR.6,19,20 Patients with thrombus or atheroma in the neck of Incidence, cause, and outcomes of small bowel isch- aneurysm should be closely followed up postoperatively. emia after EVAR are poorly documented. In our group, Colonoscopy should be performed if there is any suspicion small bowel ischemia occurred in 0.8% of all patients and in of microembolization, such as blue toes. Small bowel isch- 60% of patients in whom postoperative colon ischemia emia should be suspected in patients with clinical symptoms developed after EVAR. Why is small bowel ischemia more that suggest more severe ischemia than is found at colonos- common after EVAR versus open procedures? Interruption copy. Because the prognosis is extremely poor, early explor- of the inferior mesenteric artery or hypogastric artery has atory laparotomy should be considered if small bowel isch- not been demonstrated to impair the blood supply to the emia cannot be ruled out. small bowel. The supraceliac aorta is not clamped during endovascular procedures, and the short period of aortic occlusion during balloon dilation does not result in isch- REFERENCES emia and reperfusion injury. Therefore the major cause of 1. Pittaluga P, Batt M, Hassen-Khodja R, Declemy S, Bas PL. Revascular- small bowel ischemia after EVAR is different from that after ization of internal iliac arteries during aortoiliac surgery: a multicenter open repair. In our 6 patients with concurrent small bowel study. Ann Vasc Surg 1998;12:537-43. ischemia, 5 patients had thrombus or atheroma in the 2. Ernst CB. Prevention of intestinal ischemia following abdominal aortic reconstruction. Surgery 1983;93:102-6. aneurysm neck or iliac arteries. We presume that the throm- 3. Fry PD. Colonic ischemia after aortic reconstruction. Can J Surg bus or atheroma in the neck was dislodged while the graft 1988;31:162-4. was deployed and that the debris was flushed upstream into 4. Brewster DC, Franklin DP, Cambria RP, Darling RC, Moncure AC, the superior mesenteric artery by blood flow turbulence Lamuraglia GM, et al. Intestinal ischemia complicating abdominal caused by graft deployment or balloon dilation (Fig 2). The aortic surgery. Surgery 1991;109:447-54. 5. Bjorck M, Bergquist D, Troeng T. Incidence and clinical presentation microemboli may be flushed into renal arteries, the inferior of bowel ischemia after aortoiliac surgery: 2930 operations from a mesenteric artery, the hypogastric artery, and distal vessels, population-based registry in Sweden. Eur J Vasc Endovasc Surg 1996; leading to acute renal insufficiency, colon ischemia, trash 12:139-44. buttock, and blue toes, as well. The direct pathologic 6. Dadian N, Ohki T, Veith FJ, Edelman M, Mehta M, Lipsitz EC, et al. evidence and the patterns of segmental, skipped, or patchy Overt colon ischemia after endovascular aneurysm repair: the importance of microembolization as an etiology. J Vasc Surg 2001;34:986-96. ischemia in most of our patients support the concept that 7. Ernst CB, Hagihara PF, Daughtery ME, Sachatello CR, Griffen WO Jr. microembolization has an important role in both small and Ischemic colitis frequency following abdominal aortic reconstruction: a large bowel ischemia after EVAR. Although Gitlitz et al23 prospective study. Surgery 1976;80:417-21. reported that no symptomatic embolic complications oc- 8. Hagihara PF, Ernst CB, Griffen WB Jr. Frequency of ischemic colitis curred in 19 patients with thrombus in the aneurysm neck, following abdominal aortic reconstruction. Surg Gynecol Obstet 1979; 149:571-3. subclinical thromboembolism developed in more than 10% 9. Ourie K, Fiore WM, Geary JE. Detection of occult colonic ischemia of their patients (2 of 19). during aortic procedures: use of an intraoperative photoplethysmo- The percentage of bowel ischemia was 1.4% in patients graphic technique. J Vasc Surg 1988;7:5-9. with AneuRx grafts (6 of 429) and 1.6% in patients with 10. Welch M, Baguneid MS, McMahon RF, Dodd PD, Fulford PE, Ancure grafts (4 of 255). Excluder and Zenith grafts were Griffiths GD, et al. Histological study of colonic ischemia after aortic placed in only 9 patients each, because of the timing of the surgery. Br J Surg 1998;85:1095-8. 11. Ernst CB. Colon ischemia following aortic reconstruction. In: study, and no bowel ischemia developed. Our data do not Rutherford RB, editor. Vascular surgery. 5th ed. Philadelphia (PA): suggest that the type of graft is related to the incidence of Saunders; 2000. p 1542-9. postoperative bowel ischemia. Seven of the 10 complica- 12. Valentine RJ. Intestinal ischemia complicating abdominal aortic recon- tions of bowel ischemia occurred within the first 2 years of struction. In: Ernst CB and Stanley JC, editors. Current therapy in endovascular AAA repair in our hospitals. The lower rate of vascular surgery. 4th ed. St Louis (MO): Mosby; 2001. p 294-8. 13. Elmarasy N, Soong CV, Walker SR, Macierewicz JA, Yusuf SW, bowel ischemia may be related in part to improved techni- Wenham PW, et al. Sigmoid ischemia and the inflammatory response cal skills, and possibly to careful patient selection. following endovascular abdominal aortic aneurysm repair. J Endovasc In summary, the total incidence of bowel ischemia after Ther 2000;7:21-30. EVAR is approximately that after open surgery. However, 14. Miahle C, Amicabile C, Becquemin JP. Endovascular treatment of small bowel ischemia occurs more commonly after EVAR, infrarenal abdominal aneurysms by the Stentor system: preliminary results of 79 cases. J Vasc Surg 1997;25:165-72. with extremely high mortality, especially in patients with 15. Moore WS, Kashyap VS, Vescera CL, Quinones-Baldrich WJ. Abdom- full-thickness ischemia. Microembolization is a major cause inal aortic aneurysm: a 6-year comparison of endovascular versus trans- of both small and large bowel ischemia after EVAR. To abdominal repair. Ann Surg 1999;230:298-308. prevent bowel ischemia caused by microembolization, we 16. Iliopoulos JI, Pierce GE, Hemreck AS, Haller CC, Thomas JH. Hemo- recommend the following: (1) careful review of preopera- dynamics of the inferior mesenteric arterial circulation. J Vasc Surg 1990;11:120-6. tive images to identify thrombus or atheroma in the neck of 17. Seeger JM, Coe DA, Kaelin LD, Flynn TC. Routine reimplantation of the aneurysm; (2) minimal manipulations during graft de- patent inferior mesenteric arteries limits colon infarction after aortic ployment; (3) direct deployment of the stent immediately reconstruction. J Vasc Surg 1992;15:635-41. JOURNAL OF VASCULAR SURGERY 872 Zhang et al November 2004

18. Iliopoulos JI, Howanitz PE, Pierce GE, Kueshkerian SM, Thomas JH, 22. Karch LA, Hodgson KJ, Mattos MA, Bohannon WT, Ramsey DE, Hemreck AS. The critical hypogastric circulation. Am J Surg 1987;154: McLafferty RB. Adverse consequences of internal iliac artery occlusion 671-5. during endovascular repair of abdominal aortic aneurysms. J Vasc Surg 19. Mehta M, Veith FJ, Ohki T, Cynamon J. Unilateral and bilateral hypogas- 2000;32:676-83. tric artery interruption during aortoiliac aneurysm repair in 154 patients: a 23. Gitlitz DB, Ramaswami G, Kaplan D, Hollier LH, Marin ML. Endo- relatively innocuous procedure. J Vasc Surg 2001;33(2 suppl):S27-32. vascular stent grafting in the presence of aortic neck ﬁlling defects: early 20. Wolpert LM, Dittrich KP, Hallisey MJ, Allmendinger PP, Gallagher JJ, clinical experience. J Vasc Surg 2001;33:340-4. Heydt K, et al. Hypogastric artery embolization in endovascular abdominal aortic aneurysm repair. J Vasc Surg 2001;33:1193-8. 21. Parodi JC. Relocation of iliac artery bifurcation to facilitate endoluminal treatment of AAA. J Endovasc Surg 1999;6:342-7. Submitted May 17, 2004; accepted Aug 8, 2004.