Safety of coil embolization of the internal iliac artery in endovascular grafting of abdominal aortic aneurysms Frank J. Criado, MD, a Eric P. Wilson, MD, a Omaida C. Velazquez, MD, b Jeffrey P. Carpenter, MD, b Clyde Barker, MD, b Eric Wellons, MD, a Omran Abul-Khoudoud, MD, a and Ronald M. Fairman, MD, b Baltimore, Md, and Philadelphia, Pa Purpose: During endovascular grafting of an abdominal aortic aneurysm (AAA), iliac limb extension to the external iliac artery may be indicated when the common iliac artery is ectatic or aneurysmal. Preliminary or concomitant coil embolization of the internal iliac artery (IIA) is thus necessary to prevent potential reflux and endoleak. We sought to determine the safety of hypogastric flow interruption in this setting. Methods: We retrospectively reviewed 156 patients who underwent stent-graft AAA repair at two institutions between February 1, 1998, and January 31, 1999. Coil embolization of one or both IIAs was undertaken when the diameter of the common iliac artery was more than 20 mm to enable limb endograft extension to the external iliac artery. Bilateral procedures were staged. Results: Thirty-nine (25%) of 156 patients were selected for coil embolization of one (n = 28) or both (n = 11) IIAs. The interventions were performed before (n = 31) or during (n = 8) the stent-graft procedure. Complications included groin hematomas in 3 patients, iliac artery dissection in 1, failure to catheterize the IIA in 2, and transient rise in the serum creatinine level in 3. One patient had erectile dysfunction, and five patients (13%) had buttock claudication after unilateral occlusion. Serious ischemic complications were not observed. Conclusion: Coil embolization of one or both IIAs appears to be safe in the setting of endovascular grafting of AAA. Buttock claudication is a relatively significant problem and may limit applicability of this strategy to patients who are unfit for standard open repair. (J Vasc Surg 2000;32:684-8.) From the Center for Vascular Intervention and Division of Vascular Surgery, Union Memorial Hospital/MedStar Health, a and the Department of Surgery, Hospital of the University of Pennsylvania. b Competition of interest: nil. Presented during the Twenty-eighth Annual Symposium on Vascular Surgery of the Society for Clinical Vascular Surgery, Rancho Mirage, Calif, Mar 15-19, 2000. Reprint requests: Frank J. Criado, MD, 3333 North Calvert Street, Suite 570, Baltimore, MD 21218 (e-mail: frankc@helix.org). Copyright 2000 by The Society for Vascular Surgery and The American Association for Vascular Surgery, a Chapter of the International Society of Cardiovascular Surgery. 0741-5214/2000/$12.00 + 0 24/6/110052 doi:10.1067/mva.2000.110052 Endovascular stent-graft repair of abdominal aortic aneurysm (AAA) is rapidly gaining acceptance as clinical studies begin to show proof of clinical efficacy 1 and devices receive approval by the Food and Drug Administration. Aortoiliac morphology and aneurysm extent are crucially important at the time of choosing a technical strategy for endograft placement. Aneurysmal involvement (or ectasia) of the common iliac artery (CIA) may constitute an impediment to endoluminal exclusion or necessitate limb extension to the external iliac artery (EIA) to bypass the CIA. In such a case, coil embolization of the ipsilateral internal iliac artery (IIA) is indicated to prevent reflux (endoleak) into the sac. Occasionally, bilateral IIA interruption may be necessary for proper endovascular treatment of complex, extensive aortoiliac aneurysms (Fig 1). Both coil embolization and endograft coverage of the IIA origin were first described by Parodi 2 and others in some of the earliest publications on stent-graft repair. With few exceptions, the body of published information on IIA interruption relates to the consequences of surgical ligation during standard operation. 3-11 The risks of bilateral hypogastric artery flow interruption are well documented and are known to be particularly high when the inferior mesenteric artery is occluded. The principle of preserving at least 684
Volume 32, Number 4 Criado et al 685 A B C Fig 1. A, Aortoiliac aneurysm with thrombus-lined ectatic CIAs. B, Bilateral coil embolization of IIAs was performed in two separate interventions. Ipsilateral iliac limb extension to the EIA has been deployed. C, Completed aortobilateral EIA stent-graft. one IIA continues to stand as near-doctrine in vascular surgery. Nonetheless, the implications of hypogastric artery occlusion may not be quite so serious in the setting of endovascular repair of AAA. In this retrospective study, we sought to determine the safety of coil embolization of the IIA during endovascular exclusion of AAA. METHODS Between February 1, 1998, and January 31, 1999, we performed stent-graft exclusion of AAA on 156 patients (140 men, 16 women; age range, 59-89 years; mean, 71.6 years). They were evaluated preoperatively with contrast angiography and spiral computed tomography. Transcatheter coil embolization of one or both IIAs was undertaken to eliminate a potential source of backflow endoleak when aneurysmal involvement of the CIA, as defined by a diameter of more than 20 mm at the landing zone, required endograft limb extension to the EIA. The intervention was undertaken before (2 days or more) or during the stent-graft procedure and involved deployment of
686 Criado et al October 2000 A B C Fig 2. Illustration of bell-bottoming technique on patient with AAA not included in the study group. A, 16-mm limb of AneuRx device (Medtronic AVE) does not appose to wall of ectatic left CIA. B, Bell-bottoming achieved with a 20-mm aortic cuff extender. C, Right iliac limb extended to EIA after preliminary coil occlusion of IIA on same patient. multiple Tornado 8- to 10-mm coils (Cook, Inc, Bloomington, Ind) as proximal as possible in the main trunk of the IIA after selective catheterization with a 5F system. Bilateral IIA interruption was always staged to avoid simultaneous hypogastric flow interruption. Embolization technique was designed to try to promote gradual vessel thrombosis by avoiding complete luminal obstruction during initial coil delivery. Patients were followed up carefully during their hospitalization and then examined every 3 months to assess the possible development of buttock claudication, sexual dysfunction, and other potential complications. Colonoscopy was not part of the surveillance protocol. The review focused on the frequency of perceived need for IIA interruption and the occurrence of ischemic complications that could be attributed to coil embolization of the IIA. RESULTS Endovascular AAA repair involved placement of a Talent stent-graft (Medtronic AVE, Santa Rosa, Calif); 125 patients underwent bifurcated repair and 31 patients, aorto-uni-iliac repair. Extension to the EIA (and need for IIA interruption) was considered necessary when the CIA diameter (one or both) exceeded 20 mm. Thus, coil embolization of the IIA was per-
Volume 32, Number 4 Criado et al 687 formed in 39 patients (36 men, 3 women; age range, 64-89 years; mean, 78.3 years) undergoing bifurcated (n = 27) or aorto-uni-iliac (n = 12) stent-graft repair. The interventions were undertaken 2 days or more before the AAA procedure in 31 patients and as a concomitant intervention at the time of AAA endografting in eight patients. Patient selection for early or concomitant intervention reflected personal preferences in some cases and scheduling/practical issues in others. Bilateral IIA embolization (n = 11) was always staged, which separated the interventions by at least 1 week. Complications included puncture-site hematoma in 3 patients, catheterization-related dissection of the EIA (treated with stent placement) in 1, failure to achieve cannulation of the IIA in 2, and transient rise of the serum creatinine level in 3. Bell-bottoming of the iliac limb (to achieve seal at the CIA level) was used in the two patients who could not undergo catheterization of the IIA (Fig 2). None of the patients undergoing coil embolization had a serious ischemic complication (ie, colorectal, pelvic/buttock, or spinal cord dysfunction). Possible occurrence of subclinical mucosal colonic ischemia was not assessed with routine colonoscopy. One patient had erectile dysfunction after unilateral IIA interruption in the presence of documented patency of the contralateral IIA. Five patients (13%) had buttock claudication, which was described as severe in two of them. All five instances of buttock claudication occurred after unilateral IIA embolization; none of the patients with bilateral embolizations had clinically manifest claudication during the follow-up period. The follow-up length ranged from 12 to 24 months; the mean was 11.5 months. One patient was lost to follow-up after 6 months, and another died of lung cancer 14 months after AAA repair. Neither patient had a complication attributable to IIA embolization. Buttock claudication improved in one patient and remained the same in the other four during the period of observation herein reported. DISCUSSION The concept of preserving patency of at least one IIA is a time-honored surgical principle. Abrupt, simultaneous interruption of IIA flow carries significant risk of serious morbidity, especially when the inferior mesenteric artery is occluded or requires concomitant ligation. Predictably, such powerful teachings have influenced investigators pursuing endoluminal treatment. 12 However, it is appropriate to question conventional surgical wisdom and reexamine the potential consequences of IIA interruption when undertaken in the setting of endovascular stent grafting of aortoiliac aneurysms. We speculate that avoiding sudden flow interruption and coil embolization of the vessel in a manner that may promote gradual thrombosis allows development of collateral flow compensation. 13 This may help explain the remarkable absence of serious ischemic complications in our series. However, other investigators have achieved similar results without paying any attention to such details of timing and technique. The isolated occurrence of sexual dysfunction was quite surprising, especially when considering that the contralateral IIA was patent. None of the other patients described any sexual problems when asked directly; objective measurements of penile blood flow were not used. On the other hand, buttock claudication is a well-recognized complication of IIA occlusion, whether bilateral or unilateral. 14 It was observed in five (13%) of our 39 patients, but described as severe by only two of the patients, reflecting perhaps the special characteristics of the patient population selected for coil embolization: aged, high-risk individuals with rather inactive lifestyles. Considerations of embolization technique are relevant at the time of preventing or minimizing potential ischemia and buttock claudication; there is anecdotal evidence to support the notion that complications may occur more frequently when the coils are delivered distally into secondary and tertiary branches of the vessel. We always make an effort to effect coil occlusion of the main trunk of the IIA, carefully preserving patency of more distal branches. IIA catheterization can be technically difficult in the face of challenging anatomy and can involve administration of a significant radiocontrast load. It is preferable to undertake the intervention before and not during the stent-graft procedure, with the added theoretical advantage of allowing for collateral pelvic flow compensation. The issue of bilateral IIA embolization is more controversial, with most surgeons (and interventionists) continuing to view it as highly risky or contraindicated. The recently proposed relocation operation was designed precisely to address such situations and, at the same time, preserve flow into at least one IAA. 12 In our experience, we have been impressed with the uneventful outcome of bilateral coil embolization. Staging and proper sequencing may be critically important; we feel strongly about the dangers of simultaneous bilateral occlusion, but cannot offer evidence to substantiate such fears. Published and unpublished accounts would appear to suggest that both staged and simultaneous bilateral embolizations are quite safe. Finally, it is pertinent to discuss the appropriateness of preserving antegrade hypogastric artery flow by placement ( landing ) of the distal endograft limb
688 Criado et al October 2000 into an ectatic CIA. It has been our policy, which is shared with other groups, to deploy proximal to the iliac bifurcation when the CIA is less than 20 mm in diameter. At least two investigational devices (Talent, Medtronic AVE; Zenith, Cook, Inc,) provide iliac limbs large enough to achieve seal in CIAs 20 to 24 mm in diameter. Bell-bottom configurations with an aortic cuff modular extension can also achieve a good seal in a large-diameter CIA (Fig 2). 15 Recently published evidence suggests that CIAs up to 30 mm in diameter have an extremely low expansion rate 16 and may be acceptable as distal attachment sites in the absence of circumferential thrombus. It is our view that in the future, endograft limb extensions to the EIA (with consequent need for IIA interruption) are likely to be considered less frequently as endovascular experience grows and larger-diameter endograft limbs become available. In addition, future technologic evolutions with development of hypogastricbranched stent grafts are all but certain. We conclude that transcatheter coil embolization of the IIA is a safe intervention when undertaken within the context of endovascular stent-graft repair of AAA. Both unilateral and bilateral IIA interruption appears to be well tolerated. Bilateral interruption should probably be reserved for patients who have large aneurysms and are medically (or otherwise) unfit for standard open repair. Buttock claudication is a significant problem, occurring in more than 10% of instances. The long-term impact of this complication remains to be elucidated. REFERENCES 1. Zarins CK, White RA, Schwarten D, et al. AneuRx stent graft versus open surgical repair of abdominal aortic aneurysms: multicenter prospective clinical trial. J Vasc Surg 1999;29:292-309. 2. Parodi JC. Endovascular repair of abdominal aortic aneurysms and other arterial lesions. J Vasc Surg 1995;21:549-57. 3. Johnson WC, Nasbeth DC. Visceral infarction following aortic surgery. Ann Surg 1963;86:65-73. 4. Picone AL, Green RM, Ricotta JR, et al. Spinal cord ischemia following operations on the abdominal aorta. J Vasc Surg 1986;3:94-103. 5. Iliopoulos JI, Howanitz P, Pierce GE, et al. The critical internal iliac circulation. Am J Surg 1987;154:671-5. 6. Fry PD. Colonic ischemia after aortic reconstruction. Can J Surg 1988;31:162-4. 7. Iliopoulos JI, Hermreck AS, Thomas JH, et al. Hemodynamics of the internal iliac arterial circulation. J Vasc Surg 1989;9:637-42. 8. Gloviczki P, Cross SA, Stanson AW, et al. Ischemic injury to the spinal cord or lumbosacral plexus after aortoiliac reconstruction. Am J Surg 1991;162:131-6. 9. Brewster DC, Franklin DP, Cambria RP, et al. Intestinal ischemia complicating abdominal aortic surgery. Surgery 1991;109:447-54. 10. Szilagyi DE. A second look at the etiology of spinal cord damage in surgery of the abdominal aorta. J Vasc Surg 1993;17:1111-3. 11. Paty PKS, Shah DM, Chang BB, et al. Pelvic ischemia following aortoiliac reconstruction. Ann Vasc Surg 1994;8:204-6. 12. Parodi JC, Ferreira M. Relocation of the iliac artery bifurcation to facilitate endoluminal treatment of abdominal aortic aneurysms. J Endovasc Surg 1999;6:342-7. 13. Criado FJ. Iliac bifurcation relocation: more complex and controversial [commentary]. J Endovasc Surg 1999;6:348-9. 14. Gough MJ, MacMahon MJ. A minimally invasive technique allowing ligation of the internal iliac artery during endovascular repair of aortic aneurysms with an aorto-uni-iliac device. Eur J Vasc Endovasc Surg 1998;16:535-6. 15. Henretta JP, Karch LA, Hodgson KJ, et al. Special iliac artery considerations during aneurysm endografting. Am J Surg 1999;178:212-8. 16. Santilli SM, Wernsing SE, Lee ES. Expansion rates and outcomes for iliac artery aneurysms. J Vasc Surg 2000;31:114-21. Submitted Mar 23, 2000; accepted Jun 28, 2000.