Endovascular aneurysm repair (EVAR) has emerged as a routine and viable alternative in the treatment of abdominal aortic aneurysms (AAA). Its efficacy has been confirmed in many studies, and its role in the treatment of atherosclerotic aneurysms is becoming defined as long-term follow-up data is analyzed. Major complications such as migration, thrombosis, kinking of the graft, endoleaks and continued expansion with eventual rupture have been described and well studied. EVAR has also been shown to induce a systemic inflammatory response in patients leading to local perianeurysmal inflammation and fibrosis in some rare cases.^1,2 The etiology and natural history of periaortitis after aortic stent grafting, however, is unknown and unpredictable. Furthermore, the implications of this phenomenon as it relates to the use of EVAR for treatment of inflammatory AAAs are also unknown. We report 2 cases of periaortitis after EVAR for asymptomatic atherosclerotic AAA. The aortic inflammatory changes were incidentally discovered with computed tomography (CT) scan during routine post-operative endograft surveillance (Figure 1a).

Case 1. A 62-year-old man with a history of hypercholesterolemia was found to have a nontender pulsatile abdominal mass during routine physical exam. A computed tomography (CT) scan demonstrated a 5.4 cm infrarenal abdominal aortic aneurysm suitable for endograft repair. The patient underwent EVAR with an AneuRx modular device (Medtronic, Santa Rosa, CA) under general anesthesia. His post-operative course was routine and unremarkable. He was discharged home on post-operative day one. Two weeks post procedure, he presented with right leg claudication and right graft limb thrombosis was diagnosed. Successful treatment included thrombolytic therapy and endovascular stenting of the graft limb with 6 weeks of anticoagulation therapy. Routine follow-up CT scan at 1 month confirmed satisfactory position of the endograft, patency of the limbs, absence of endoleak, and no growth of the aneurysm. No other abnormalities were noted. The patient was doing well with no complaints. At his 6-month follow-up visit, however, he complained of mild, non-specific abdominal pain. Physical exam revealed a benign abdomen with no pulsatile masses. The patient was afebrile with no other symptoms. Serum leukocyte count and creatinine levels were normal. CT scan at this time demonstrated significant thickening of the aneurysm sac, resembling an inflammatory abdominal aortic aneurysm. No endoleak, growth of the aneurysm or other abnormality was noted. Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) level were slightly elevated at 30 mm/hr and 0.8 mm/L, respectively (reference range 0–25 mm/hr and < 0.5 mm/L, respectively). As the patient's exam and complaints were relatively benign and CT scan demonstrated no endoleak or growth of the aneurysm, the patient was managed with conservative observation. The abdominal pain resolved in less than two weeks.

Case 2. A 58-year-old man was incidentally found to have an asymptomatic abdominal aortic aneurysm during routine work-up for nephrolithiasis. Physical exam revealed a nontender pulsatile abdominal mass. A CT scan demonstrated an infrarenal atherosclerotic AAA measuring 5.1 cm at its greatest diameter, originating approximately 2 cm distal to the lowest renal artery. The patient underwent EVAR with an Ancure bifurcated graft (Guidant, Santa Clara, CA) without any complications. His post-operative course was uneventful, and he was discharged home on post-operative day #1. At one month and 6-month routine follow-up, the patient had no complaints and CT scans demonstrated a patent endograft with no migration, endoleak and or growth in the aneurysm sac. One year after endograft placement, the patient's surveillance CT scan now showed significant thickening of the AAA, again resembling an inflammatory abdominal aortic aneurysm without endoleak. The graft remained in satisfactory position and there was no growth in the aneurysm sac. No other abnormalities were noted. The patient was afebrile with no complaints and physical exam was benign. He denied any fatigue, malaise or abdominal discomfort. ESR and CRP levels were normal. The patient's 18-month CT scan demonstrated a patent graft with no endoleak and essentially no change in these aortic inflammatory changes. The patient remains asymptomatic with normal ESR and CRP levels (3 mm/hr and 0.39 mm/L, respectively).

Discussion
We report on 2 cases of asymptomatic periaortic inflammation following EVAR for atherosclerotic AAA. Review of the literature renders only 2 case reports of this unusual complication, both involving ureteral obstruction following EVAR for non-inflammatory AAA.^3,4 Simons et al.³ presented a 59-year-old man who underwent EVAR for an atherosclerotic AAA with a Talent (Medtronic) device. One-month and six-month follow-up CT scans showed no evidence of periaortic inflammation. CT scan surveillance at one-year post placement revealed an enhancing inflammatory ring around the aneurysm. The patient remained asymptomatic at that time. By 17 months, however, the patient developed increased retroperitoneal fibrosis with subsequent ureteral obstruction and hydronephrosis. The patient underwent eventual nephrectomy and was treated with tamoxifen to avoid future complications when subsequent CT scan revealed worsening inflammation. A second case reported by Jetty et al.⁴ presents a 76-year-old man who underwent EVAR for a non-inflammatory AAA with an AneuRx device (Medtronic). One-month follow-up CT was free of an inflammatory changes and the patient was doing well. At five months post-procedure, however, the patient developed acute renal failure with bilateral ureteral obstruction and hydronephrosis. CT scan revealed significant periaortic inflammation. The patient underwent ureteral stenting and was started on tamoxifen therapy with improvement of his renal function and substantial regression of the inflammation.

Few studies have evaluated this unusual complication of periaortitis following EVAR, although inflammatory reactions to endoluminal stents have been demonstrated in animal models.^1,5 Sapoval et al.⁶ also reported a case of a patient with stent related inflammatory changes of an iliac artery after placement of a polyester-covered stent for dissection, which resolved spontaneously after 11 months. Only one retrospective study has reported on the development of periaortitis following EVAR for atherosclerotic AAAs.⁷ In a series of 61 patients, Vallabhaneni et al.⁷ found that post-operative low-grade periaortitis (diagnosed by CT scan) developed in 10 out of 55 (18%) of their patients undergoing EVAR for non-inflammatory AAA. Based on these results, the calculated risk of developing de novo periaortitis was 16.4% (95% CI 6.6–27.1%). Of these patients, all remained without symptoms or complications and all perianeurysmal inflammation resolved spontaneously by 36 months. The remaining 6 patients in the series had inflammatory AAAs (IAAAs) pre-operatively, and all of these patients had either persistence or worsening of the inflammation post EVAR. Two patients even developed ureteral obstruction from the worsening periaortitis.

Endovascular repair is an attractive alternative for treatment IAAA where periaortic fibrosis encasing retroperitoneal structures can lead to greater morbidity and mortality during laparotomy and cross-clamping of the aorta. Whereas open surgical repair is associated with regression of this fibrosis,^8–12 the reports following EVAR have demonstrated variable success. While there have been case reports of successful regression of perianeurysmal inflammation after EVAR for IAAA, these results are not consistent.
Several anecdotal reports describe regression of perianeurysmal fibrosis following EVAR for symptomatic IAAAs with no pre-operative ureteral obstruction.^13–16 Two other groups have reported successful regression of periaortic fibrosis and ureteral obstruction following EVAR.^17,18 Additionally, there is one case report of periaortitis regression after EVAR with a bare metal Wallstent endoprosthesis (Boston Scientific, Maple Grove, MN). Other case reports, however, have shown that EVAR fails to treat periaortic fibrosis.^19,20 Only 3 small series addressing EVAR for IAAA are found in the literature. One small series by Hinchcliffe et al. demonstrated no increase in perianeurysmal fibrosis after EVAR in 11 patients.²¹ Deleersnijder et al. reports on a series of 7 patients with IAAA who underwent EVAR with various devices.²² Four of the seven were symptomatic and five had ureteral obstruction. Although no patients developed worsening of the periaortic fibrosis, none showed complete regression, and partial regression was noted in only three cases. They concluded that while EVAR for IAAA is technically feasible, the utility in regression of any perianeurysmal inflammation remains unknown. The series by Vallabhaneni et al.⁷ concluded that pre-operative perianeurysmal fibrosis was a relative contraindication to EVAR, as none of the patients demonstrated regression of the periaortitis.

While the option of endoluminal repair of an AAA is attractive compared to open repair, extension of this technique to the treatment of inflammatory abdominal aortic aneurysms remains controversial. It is not clear how endoluminal exclusion of an aneurysm may exacerbate or attenuate any perianeurysmal fibrosis. Additionally, our two cases demonstrate that de novo periaortitis after EVAR can also develop. Full implications of this unusual complication are unknown. If this process remains localized, as in our cases, expectant management seems reasonable. Little is known to identify those patients who might progress to retroperitoneal fibrosis and ureteral obstruction. Thus, the long-term effects of periaortic inflammation after EVAR warrants further study.

vcheanvechai@smail.umaryland.edu