Original Article

Introduction

Endovascular aortic aneurysm repair (EVAR) is a method for repairing abdominal aortic aneurysms (AAAs) that is increasingly used for patients with suitable anatomy.1,2) EVAR requires continued surveil-lance because up to 11% of patients need reintervention

for major adverse events, in particular endoleak.3) Persistent type II endoleak has been associated with a higher incidence of adverse outcomes, including aneurysmal sac growth, need for reintervention and for conversion to open repair, and rupture.4,5) The most commonly involved branches are the inferior mesen-teric artery (IMA) and lumbar arteries (LA), which are usually covered with endografts. Various strategies have been advocated for managing persistent type II endoleak with some investigators recommending pre-emptive adjunctive procedures such as IMA coil embo-lization to preclude type II endoleak.6–9)

However, some studies have questioned the effec-tiveness of preoperative embolization of aortic side branches for the prevention of type II endoleak.9–13)

CT Findings of Risk Factors for Persistent Type II Endoleak from Inferior Mesenteric Artery to Determine Indicators of Preoperative IMA Embolization

Tetsuya Fukuda, MD, PhD,1 Hitoshi Matsuda, MD, PhD,2 Yoshihiro Sanda, MD, PhD,1 Yoshiaki Morita, MD, PhD,1 Kenji Minatoya, MD, PhD,2 Junjiro Kobayashi, MD, PhD,2 and Hiroaki Naito, MD, PhD1

Purpose: To identify the computed tomography (CT) findings of persistent type II endoleak from the inferior mesenteric artery (IMA) which indicate the need for preoperative IMA embolization.Materials and Methods: Included were 120 patients (96 males, 49–93 years old, mean: 77.7) who underwent endovascular aortic aneurysm repair (EVAR) between June 2007 and October 2010. The relationship between persistent type II endoleak and CT findings of IMA orifice was examined.Results: CT showed no type II endoleak from IMA in 106 patients (89%; Group N), and transient type II endoleak from IMA in 10 patients (8.3%; Group T). CT showed persistent type II endoleak from IMA in 4 patients (3.3%; Group P) and three of them underwent reintervention. Univariate Cox-Mantel test analysis indicated that stenosis (p = 0.0003) and thrombus (p = 0.043) in IMA orifice were significant factors for per-sistent type II endoleak. The ratios of patients with proximal IMA more than 2.5 mm diameter in Groups N, Y, and P were 26/106 (24%), 5/10 (50%) and 4/4 (100%), respectively.Conclusion: Indicators for embolization of IMA prior to EVAR for the prevention of type II endoleak appear to be: (1) more than 2.5 mm in diameter and (2) no stenosis due to calcification or mural thrombus in IMA orifice.

Keywords: CT, inferior mesenteric artery, type II endoleak, endovascular aneurysm repair

1Department of Radiology, National Cerebral and Cardio-vascular Center, Suita, Osaka, Japan2Department of Cardiovascular Surgery, National Cerebral and Cardiovascular Center, Suita, Osaka, Japan

Received: January 14, 2014; Accepted: June 22, 2014Corresponding author: Tetsuya Fukuda, MD. Department of Radiology, National Cerebral and Cardiovascular Center, 7-5-1 Fujishirodai, Suita, Osaka 565-8565, JapanTel: +81-6-6833-5012, Fax: +81-6-6872-7486E-mail: tetsuyaf@ncvc.go.jp

274 Annals of Vascular Diseases Vol. 7, No. 3 (2014)

Ann Vasc Dis Vol. 7, No. 3; 2014; pp 274–279 Online Month August 30, 2014©2014 Annals of Vascular Diseases doi:10.3400/avd.oa.14-00008

05_08_OA_Fukuda_00008.indd 274 2014/09/09 13:27:36

CT Findings of High Risk IMA

Considering the indications as well as the complica-tions of catheterization of the IMA, computed tomog-raphy (CT) findings of the IMA orifice such as stenosis due to mural thrombus or calcification may well be significant. To the best of our knowledge, however, there have been no reports concerning the relationship between persistent type II endoleak from the IMA and CT findings of the IMA orifice. The purpose of the present study was therefore to identify the CT findings of which reveal persistent type II endoleak and consti-tute indications for preoperative IMA embolization.

Materials and Methods

Of the 173 patients who underwent EVAR using commercially available devices between June 2007 and October 2010, 120 (96 males, 49–93 years old; mean: 77.7) were included in the present study. The reasons for exclusion of the remaining 53 patients were: occlusion of the IMA at its orifice in 36 patients, allergy to contrast media or renal dysfunction in 9 patients, which made evaluation of the IMA with enhanced CT impossible, and the IMA originating from the aorta above the aneurysm in 8 patients.

All clinical data including radiographic data were obtained retrospectively from each of the patient charts.

All CT scans were performed with intravenous con-trast material and thin collimation. After an unen-hanced scan, a bolus injection of contrast medium (1.5–2.0 ml/s) was administrated after a delay of 20–25 s for preparation. The images were recon-structed from 2 mm-thick slices. Images of the delayed phase were obtained for all patients 3 min after the early phase scan. CT scan with 2-mm slice thickness was used to analyze the anatomical factors of the IMA such as diameter, stenosis, thrombus, and calci-fication at its orifice. Stenosis of the IMA orifice was defined as 75% stenosis due to mural thrombus and/or calcification seen on the axial and curved mul-tiplaner image of thin-slice CT. Mural thrombus was defined as a dense (more than 5 mm) thrombus around the IMA orifice. Calcification was defined as protru-sion of calcified plaque around the IMA orifice. These CT findings for the IMA were determined by two radiological specialists.

All patients included in this study underwent an enhanced CT scan before and one week after the ini-tial EVAR procedure. Subsequently, patients were clinically evaluated with follow-up CT scans 6 and

12 months after EVAR, and at least yearly thereafter. The presence of type II endoleak was determined only from the CT scan for this study.

When the increase in size of the aneurysm sac was 5 mm more than the preoperative maximal sac diameter, selective angiography of the IMA via the superior mesenteric artery and meandering artery or the lum-bar arteries was indicated to determine the source of type II endoleak and the indicators of embolization.

Patients were divided into three follow-up groups based on the CT findings. Group N (106 patients; 89%): no type II endoleak from the IMA; Group T (10 patients; 8.3%): transient type II endoleak from the IMA; Group P (4 patients; 3.3%): persistent type II endoleak from the IMA with or without secondary intervention. When the type II endoleak spontaneously disappeared during the follow-up period, it was defined as transient (Fig. 1).

The preoperative variables and anatomical factors potentially associated with persistent type II endoleak were assessed using univariate analysis to identify the risk factors for persistent type II endoleak from the IMA.

JMP software (SAS Inc., Cary, North Carolina, USA) was used for all statistical analyses. Univariate com-parisons of patient demographic and preoperative risk factors were performed with the Cox-Mantel test.

This study was approved by the institutional review board of the National Cerebral and Cardiovascular Center.

Results

All EVAR procedures were performed successfully with an Excluder (W.L. Gore & Associates, Flagstaff, Arizona, USA) for 55 patients, a Zenith (Cook Medical, Bloomington, Indiana, USA) for 47, and a Powerlink (Endologix, Irvine, California, USA) for 18.

Type II endoleak from the IMA was detected in 14 patients (11.6%) during the follow-up period, but disappeared spontaneously in 10 patients (8.3%) from Group T 2.2–23.9 (mean: 10.6) months after EVAR (Fig. 1). In addition, in 8 of these 10 patients (80%) type II endoleak from the IMA disappeared more than 6 months after EVAR. No significant dif-ference was recognized between three devices in the detection of type II endoleak from IMA (p = 0.47). In addition, no significant difference was recognized in

Annals of Vascular Diseases Vol. 7, No. 3 (2014) 275

05_08_OA_Fukuda_00008.indd 275 2014/09/09 13:27:36

Fukuda T, et al.

superior mesenteric artery was performed for 3 of the 4 patients of Group P (Fig. 2).

Preoperative anatomical variables related to the IMA are summarized in Table 1. The univariate

the disappearance of type II endoleak from IMA by the device selection (p = 0.38). Because of the aneurysmal sac enlargement more than 5 mm during the follow up period, transcatheter arterial embolization via the

Fig. 1 A 74-year-old woman underwent EVAR with Powerlink for treatment of AAA, 46 mm in diameter (A). Stenosis of the orifice of the IMA was caused by mural thrombus and calcification (arrow). Computed tomography (CT) 1 week after EVAR showed type II endoleak from the IMA (B, arrow). However, CT 24 months after EVAR showed spontaneous disappearance of type II endoleak (C). AAA: abdominal aortic aneurysm; IMA: inferior mesenteric artery; EVAR: endo-vascular aortic aneurysm repair.

(A) (B) (C)

Fig. 2 A 76-year-old man with AAA, 57 mm in diameter. Preoperative computed tomography (CT) showed the IMA without stenosis or mural thrombus (A, arrow). The size of the aneurysm sac increased from 57 mm (B, 1 week after EVAR) to 65 mm (C, 2 years after EVAR), and selective angiography of the IMA via SMA and meandering artery (D) showed type II endoleak. Coil embolization of the aneurysm sac and IMA was performed (E). AAA: abdominal aortic aneu-rysm; IMA: inferior mesenteric artery; EVAR: endovascular aortic aneurysm repair; SMA: supe-rior mesenteric artery.

(A)

(D) (E)

(B) (C)

276 Annals of Vascular Diseases Vol. 7, No. 3 (2014)

05_08_OA_Fukuda_00008.indd 276 2014/09/09 13:27:37

CT Findings of High Risk IMA

that type II endoleak from the IMA can be managed conservatively more than 6 months, but three of four patients (75%) with persistent type II endoleak had to undergo secondary intervention. Our data also showed that the natural history of persistent type II endoleak is difficult to define because, even though a majority of such endoleak from the IMA disappeared, if it persists, it often leads to secondary intervention.

Transarterial retrograde embolization as well as translumbar embolization was performed for patients with sac enlargement of more than 5 mm, but the technical success rate was not high (17%–44%) because of the presence of a multitude of collateral networks of aortic side branches.18–20) Identification of patients at increased risk of persistent type II endoleak is still imprecise, and the occurrence of various asso-ciated adverse events has prompted some interven-tionists to come to accept that the best method for handling a type II endoleak is to prevent its develop-ment in the first place. Muthu, et al. reported the findings of their study of routine preoperative embo-lization of the IMA and thrombin injection into the aneurysm sac.21) This technique reduced the occur-rence of type II endoleak, but the difference failed to

Cox-Mantel test indicated that the absence of stenosis of the IMA at its orifice was more likely to be a sig-nificant factor of persistent type II endoleak (p = 0.0003), and thrombus at the orifice of the IMA was a negative predictor of persistent type II endoleak (p = 0.043).

The diameter of the proximal IMA, near its orifice, tended to be larger in Groups T and P but the differ-ences were not significant (p = 0.096). However, the ratios of patients whose proximal IMA was larger than 2.5 mm in diameter in Groups N, T, and P were 26/106 (24.5%), 5/10 (50%), and 4/4 (100%) respectively (Fig. 3). Moreover, in 8/10 (80%) patients of Group T, CT showed stenosis of the proximal IMA, but no stenosis in any patients of Group P.

The total number of patients with an IMA more than 2.5 mm diameter and without stenosis was 12 (11.2%) (Fig. 3). The sensitivity and specificity of persistent type II endoleak in patients with an IMA more than 2.5 mm diameter and without stenosis were 4/4 (100%) and 108/116 (93.1%) respectively. In addition, the respective positive and negative predictive values were 4/12 (33%) and 108/108 (100%).

Discussion

The clinical significance of type II endoleak after EVAR has not been established yet and remain con-troversial.14,15) When compared with type I and III endoleaks, type II is considered to be usually benign because as many as 80% of type II endoleak occur-rences are resolved spontaneously within 6 months after EVAR.16,17) However, type II endoleak that per-sists more than 6 months is less likely to be resolved and is associated with a higher risk of adverse events than is transient type II endoleak.16)

Our data showed spontaneous disappearance of type II endoleak from IMA more than 6 months after EVAR in 8 of 10 patients (80%). This may suggest

Table 1 Demographic and anatomic variables of patients in three groups

Variable Group N Group T Group P p value

Patients 106 10 4Orifice of IMA Stenosis 82 (77%) 8 (80%) 0 (0%) 0.0003* Thrombus 52 (49%) 4 (40%) 0 (0%) 0.043* Calcification 64 (60%) 4 (40%) 1 (25%) 0.428Diameter of IMA (mm) Proximal 2.13 ± 0.05 2.51 ± 0.21 2.67 ± 0.06 0.096

IMA: inferior mesenteric artery, *: p <0.05, significant factor for persistent type II endoleak

Fig. 3 Prevalence of diameter in inferior mesenteric artery (IMA) orifice. White dots indicate IMA without stenosis, and black dots IMA with stenosis due to thrombus or calcification.

Annals of Vascular Diseases Vol. 7, No. 3 (2014) 277

05_08_OA_Fukuda_00008.indd 277 2014/09/09 13:27:38

Fukuda T, et al.

Ann Vasc Surg 1991; 5: 491-9. 2) Greenhalgh RM, Powell JT. Endovascular repair of

abdominal aortic aneurysm. N Engl J Med 2008; 358: 494-501.

3) Conrad MF, Adams AB, Guest JM, et al. Secondary inter-vention after endovascular abdominal aortic aneurysm repair. Ann Surg 2009; 250: 383-9.

4) van Marrewijk CJ, Fransen G, Laheij RJ, et al. Is a type II endoleak after EVAR a harbinger of risk? Causes and outcome of open conversion and aneurysm rupture during follow-up. Eur J Vasc Endovasc Surg 2004; 27: 128-37.

5) Silverberg D, Baril DT, Ellozy SH, et al. An 8-year experi-ence with type II endoleaks: natural history suggests selec-tive intervention is a safe approach. J Vasc Surg 2006; 44: 453-9.

6) Baum RA, Carpenter JP, Stavropoulous SW, et al. Diag-nosis and management of type 2 endoleaks after endovas-cular aneurysm repair. Tech Vasc Interv Radiol 2001; 4: 222-6.

7) Chikazawa G, Yoshitaka H, Hiraoka A, et al. Preopera-tive coil embolization to aortic branched vessels for pre-vention of aneurysmal sac enlargement following EVAR: early clinical result. Ann Vasc Dis 2013; 6: 175-9.

8) Nevala T, Biancari F, Manninen H, et al. Inferior mesen-teric artery embolization before endovascular repair of an abdominal aortic aneurysm: effect on type II endoleak and aneurysm shrinkage. J Vasc Interv Radiol 2010; 21: 181-5.

9) Nevala T, Biancari F, Manninen H, et al. Type II endoleak after endovascular repair of abdominal aortic aneurysm: effectiveness of embolization. Cardiovasc Intervent Radiol 2010; 33: 278-84.

10) Axelrod DJ, Lookstein RA, Guller J, et al. Inferior mes-enteric artery embolization before endovascular aneu-rysm repair: technique and initial results. J Vasc Interv Radiol 2004; 15: 1263-7.

11) Bonvini R, Alerci M, Antonucci F, et al. Preoperative embolization of collateral side branches: a valid means to reduce type II endoleaks after endovascular AAA repair. J Endovasc Ther 2003; 10: 227-32.

12) Parry DJ, Kessel DO, Robertson I, et al. Type II endoleaks: predictable, preventable, and sometimes treatable? J Vasc Surg 2002; 36: 105-10.

13) Sheehan MK, Hagino RT, Canby E, et al. Type 2 endoleaks after abdominal aortic aneurysm stent grafting with sys-tematic mesenteric and lumbar coil embolization. Ann Vasc Surg 2006; 20: 458-63.

14) Ohki T, Veith FJ, Shaw P, et al. Increasing incidence of midterm and long-term complications after endovascu-lar graft repair of abdominal aortic aneurysms: a note of caution based on a 9-year experience. Ann Surg 2001; 234: 323-34; discussion 334-5.

15) Zarins CK, White RA, Hodgson KJ, et al. Endoleak as a predictor of outcome after endovascular aneurysm repair:

attain statistical significance. Although these preop-erative embolization techniques appear attractive, most patients with patent side braches will not develop type II endoleak, so that routine emboliza-tion before EVAR exposes many patients to unneces-sary procedure-related risks.22)

In this study, only four of 120 patients (3.3%) showed persistent type II endoleak from the IMA. This means that, if all patent IMAs in our study had been subjected to preoperative embolization, 116 (96.7%) embolization procedures would have had to be considered unnecessary. This constitutes a strong argument for selective preoperative endovascular inter-vention in view of the substantial risk of persistent type II endoleak.23)

To determine the indication of selective interven-tion of the IMA prior to EVAR, preoperative CT con-stitutes the most reliable and easily available image source. Univariate analysis of the anatomical variables of preoperative CT revealed that stenosis and thrombus of the IMA orifice is associated with both transient and permanent type II endoleak. Although the prox-imal diameter of the IMA showed an insignificant p-value, setting the cut-off diameter at 2.5 mm resulted in a sensitivity of 100%, specificity of 93% and neg-ative predictive value of 100%. If embolization had been indicated only for IMA more than 2.5 mm diameter and without stenosis or thrombus, the num-ber of candidate would have been 12 (11.2%), which seems more acceptable than performing preoperative embolization on all patent IMAs.

Small number of cases with persistent type II endoleak from IMA and retrospective design were thought to be the limitation of this study.

In conclusion, the indicators for embolization of the IMA prior to EVAR for the prevention of type II endoleak appear to be: (1) diameter of more than 2.5 mm and (2) no stenosis due to calcification or mural thrombus in the IMA orifice.

Disclosure Statement

Fukuda and the other co-authors have no conflict of interest to declare.

References

1) Parodi JC, Palmaz JC, Barone HD. Transfemoral intralu-minal graft implantation for abdominal aortic aneurysms.

278 Annals of Vascular Diseases Vol. 7, No. 3 (2014)

05_08_OA_Fukuda_00008.indd 278 2014/09/09 13:27:38

CT Findings of High Risk IMA

outcomes after treatment of type II endoleaks associated with aneurysm sac expansion. J Endovasc Ther 2012; 19: 182-92.

21) Muthu C, Maani J, Plank LD, et al. Strategies to reduce the rate of type II endoleaks: routine intraoperative embo-lization of the inferior mesenteric artery and thrombin injection into the aneurysm sac. J Endovasc Ther 2007; 14: 661-8.

22) Rhee SJ, Ohki T, Veith FJ, et al. Current status of man-agement of type II endoleaks after endovascular repair of abdominal aortic aneurysms. Ann Vasc Surg 2003; 17: 335-44.

23) Jones JE, Atkins MD, Brewster DC, et al. Persistent type 2 endoleak after endovascular repair of abdominal aor-tic aneurysm is associated with adverse late outcomes. J Vasc Surg 2007; 46: 1-8.

AneuRx multicenter clinical trial. J Vasc Surg 2000; 32: 90-107.

16) Gelfand DV, White GH, Wilson SE. Clinical significance of type II endoleak after endovascular repair of abdomi-nal aortic aneurysm. Ann Vasc Surg 2006; 20: 69-74.

17) Rayt HS, Sandford RM, Salem M, et al. Conservative management of type 2 endoleaks is not associated with increased risk of aneurysm rupture. Eur J Vasc Endovasc Surg 2009; 38: 718-23.

18) Solis MM, Ayerdi J, Babcock GA, et al. Mechanism of failure in the treatment of type II endoleak with percuta-neous coil embolization. J Vasc Surg 2002; 36: 485-91.

19) Kasirajan K, Matteson B, Marek JM, et al. Technique and results of transfemoral superselective coil emboliza-tion of type II lumbar endoleak. J Vasc Surg 2003; 38: 61-6.

20) Gallagher KA, Ravin RA, Meltzer AJ, et al. Midterm

Annals of Vascular Diseases Vol. 7, No. 3 (2014) 279

05_08_OA_Fukuda_00008.indd 279 2014/09/09 13:27:38