| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Section of Interventional Radiology, the Department of Radiology (R.A.B., C.M.T., M.C.S.); and the Department of Surgery (J.P.C., O.C.V., C.F.B., M.A.G., A.M.P., R.M.F.), University of Pennsylvania Medical Center, 3400 Spruce St, Philadelphia, PA 19104. Received March 12, 1999; revision requested May 3; final revision received October 12; accepted November 23. Address correspondence to R.A.B. (e-mail: baumr@rad.upenn.edu).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXReferences |
|---|
MATERIALS AND METHODS: Fifty patients who underwent endovascular repair of abdominal aortic aneurysms were examined. If an endoleak was identified at 30-day postoperative computed tomography, conventional arteriography was performed to identify and eliminate its source. After the exclusion of attachment site leaks, a catheter was placed selectively in the superior mesenteric artery (SMA). If retrograde filling of the IMA and aneurysm was identified, coil embolization was attempted through the SMA and middle colic artery. Intrasac pressures were measured at embolization.
RESULTS: Eight of 50 patients (16%) had type II endoleaks that were attributed to retrograde flow in the IMA. Intrasac measurements demonstrated systemic pressure in six patients and one-half systemic pressure in two patients. The IMA was embolized through the SMA and left colic artery in seven patients and through the translumbar aorta in one patient.
CONCLUSION: Retrograde flow in the IMA is responsible for many type II endoleaks. Systemic pressures are transmitted into the aneurysm sac from the IMA. The IMA can be embolized successfully with an SMA approach in most patients.
Index terms: Aneurysm, abdominal, 981.732 • Aneurysm, aortic, 981.732 • Aneurysm, therapy, 981.1264, 981.1268 • Aortography, 981.1213 • Arteries, therapeutic embolization, 956.1264 • Stents and prostheses, 981.1268
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXReferences |
|---|
Late failures of the devices due to persistent or recurrent flow into the aneurysm, or endoleaks, can be classified as one of four types (2–4). Type I endoleaks involve a mechanical separation of the device from the native vascular system. Blood flows around the stent-graft at an attachment site, which results in filling of the aneurysm sac. Because systemic arterial pressure is transmitted to the aortic wall, the aneurysm remains at risk of rupture (5). Creation of a tighter seal at these attachment sites by using balloon catheters, additional stents, and stent-graft extensions usually can be used to treat this leak.
Type II endoleaks are more complex and occur as flow enters the aneurysm sac through the branch vessels—the lumbar arteries, the inferior mesenteric artery (IMA), the hypogastric arteries, and so on—that have reversed their normal direction of flow (Fig 1). As is the case in type I endoleaks, in type II endoleaks, arterial pressure is transmitted into the aneurysm sac, and the patient remains at risk for aneurysm rupture. The treatment of type II endoleaks requires the elimination of flow through these branch vessels to protect the aneurysm from systemic pressure (6,7).
|
Endoleaks can also be classified according to when they first appear. Leaks identified at final postcompletion intraoperative angiography are called primary, while those identified later are called secondary. This classification system is problematic, as there is the potential for small leaks to be unrecognized at intraoperative digital subtraction arteriography and identified at postoperative computed tomography (CT).
This article describes our experience with the diagnosis and treatment of type II, IMA endoleaks that occurred after the endovascular repair of infrarenal abdominal aortic aneurysms. The purpose of our study was to review the incidence and repair of these endoleaks.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXReferences |
|---|
The procedures were performed as part of separate phase II and phase III clinical trials (Talent, World Medical, Sunrise, Fla; and Ancure EGS, Endovascular Technology, Menlo Park, Calif) (Appendix). Thirty-three bifurcated grafts, 15 aortouniiliac grafts, and two tube grafts were placed (Table 1). Forty-four high-risk and six low-risk patients were treated as defined by the clinical protocols. All patients signed an institutional review board–approved informed consent form.
|
Dynamic images and images obtained before and after the administration of contrast material were compared for endoleaks. Images were reviewed on film and in three-dimensional form at a workstation. If an endoleak was identified at the initial postoperative CT examination and the patient remained clinically stable, the patient was discharged without intervention.
When an endoleak was identified at 30-day CT, the patient was referred for conventional arteriography to identify the source of the leak and to determine if the leak could be occluded by using transcatheter embolization. Diagnostic conventional arteriography was performed through a common femoral or left brachial arterial approach by using a 5-F pigtail catheter. In type I endoleaks, contrast material was seen to enter the aneurysm sac between the stent-graft and the native aorta at the proximal, distal, or intragraft attachment site. In type II leaks, contrast material entered the sac from retrograde flow through an aortic branch vessel such as the IMA or lumbar artery. This typically occurred during the venous phase of the injection.
To exclude type I leaks, anteroposterior and lateral intraarterial digital subtraction aortography was performed, with the pigtail catheter positioned at the proximal stent-graft attachment site. Injection rates were 15–20 mL/sec for 2–3 seconds, with nonionic contrast material (iohexol). If the source of the endoleak could not be identified, additional injections were performed at the middle (graft-to-graft) and distal attachment sites.
After attachment site leaks were excluded, the aortic branch vessels were catheterized selectively to identify collateral filling and retrograde flow into the aneurysm sac, or a type II leak. Selective arteriography of the superior mesenteric artery (SMA) was performed to identify collateral and retrograde flow into the IMA. If still no leak was seen, selective internal iliac arteriography was performed to identify retrograde flow into the lumbar arteries.
If an endoleak was seen during an SMA injection, a subselective catheterization was performed. A 4- or 5-F catheter was placed in the proximal SMA, and a 3-F, 150-cm-long microcatheter (Fast Tracker 18; Boston Scientific, Natick, Mass) was used to cannulate the middle colic artery. The microcatheter was advanced through the middle colic artery and IMA into the aneurysm sac. After hand-injected digital subtraction angiography was performed to confirm the origin of the leak, intrasac pressures were measured and compared with systemic arterial pressures.
The microcatheter was then withdrawn to the proximal portion of the IMA, where microcoils (Boston Scientific/Target Vascular, Freemont, Calif) were deployed until stasis of blood flow was obtained. Arteriography of the SMA was performed subsequently to confirm the elimination of the leak. Postembolization CT was then performed to examine the aneurysm sac.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXReferences |
|---|
Ten patients had endoleaks at 30-day CT. Eight of these patients had leaks that were present at initial postoperative CT. One patient with a primary leak and with normal results of postoperative CT had repeat development of the leak at 30 days. Another patient with a leak at 30-day CT did not undergo immediately postoperative CT because of an elevated creatinine level. Nine of the 10 patients with persistent endoleaks were referred for conventional arteriography and repair (Table 2). The 10th patient died of unrelated causes after 30-day CT but before arteriography. Eight patients did not undergo 30-day CT because of an elevated serum creatinine level or because of other protocol issues.
|
Type II endoleaks from the IMA were noted in eight patients, all of whom were successfully treated with transcatheter embolization (Fig 2). Four of the eight patients had leaks that were not seen at aortic injection alone and that required selective arteriography of the SMA to be demonstrated.
|
|
|
|
One patient (patient 6) had a proximal attachment site leak that was treated with sac embolization and with repeat ballooning of the proximal attachment site. This patient then developed a type II leak 3 months later that was treated with IMA embolization. One patient (patient 10) with a recanalized IMA required two embolizations 6 weeks apart to eliminate the leak. One patient had a type II leak that was symptomatic, with severe abdominal pain that resolved after embolization. Six patients had systemic arterial pressures and waveforms, while two patients had mean sac pressures that were approximately one-half the systemic pressure, with dampened waveforms. There were no complications related to endoleak arteriography or embolization.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXReferences |
|---|
As with any graft, close follow-up and surveillance are necessary to maximize long-term success. Physical examination and imaging at regular intervals are needed to confirm durable stent-graft repair and to ensure that the aneurysm sac remains completely isolated from the systemic circulation (10–14). After stent-graft placement, aneurysms have decreased in diameter and have changed their geometric conformation (15). While this is taken as evidence of successful stent-graft repair, the conformational changes may cause the separation of graft attachment points, with resultant endoleak. Supported grafts can become dislodged, while unsupported grafts in which stents have been placed only at attachment points may kink and occlude (16).
The integrity of the seal at the proximal, mid, and distal stent-graft attachment sites must be maintained over time, or blood flow will violate the space between the stent-graft and the aneurysm wall and will place the patient at risk of continued aneurysm expansion and resultant rupture. Creation of a tighter seal between the stent-graft and the native arterial system with an extension can be used to repair these type I leaks. It is rare that the placement of a stent or simple angioplasty will suffice. Open surgical repair may be necessary for refractory type I endoleaks. It is interesting that only two patients in our series had this leak. This may be because we examined patients who had undergone the procedure 30 days previously. As aneurysm sacs continue to shrink over time, we expect to see more attachment site leaks.
A separate cause of continued sac expansion is the recruitment of aortic branch vessels that fill the aneurysm sac in a retrograde fashion, or the type II endoleak. Many different aortic branches, alone or in combination with others, may contribute to type II leaks. The lumbar artery and IMA are the most common contributors. In our series, we were surprised that this leak was far more common than the type I leak and that the IMA contributed to every type II leak seen.
To our knowledge, there is no consensus on how best to treat type II endoleaks (17). Some have advocated direct embolization of the feeding arterial supply to the sac (6). Others have suggested transcatheter embolization of the aneurysm sac itself by entering the sac directly and achieving thrombosis in the space between the stent-graft and the native arterial wall either proximally or distally (7,18). There are several problems inherent in this approach, however. The space between the stent-graft and the native arterial wall is a complex structure that presents a technical challenge to catheter manipulation. This is particularly true in patients with large mural thrombus burdens whose endoleaks result in channels of flow through a thrombus matrix and transmit systemic pressure to the aneurysm wall. In embolization with coils, the clot is shifted from one space to another, which may occlude some leaks but create others.
In addition, catheter manipulation at the proximal and distal ends of the stent-graft creates the potential for attachment site disruption and for the creation of type I leaks, particularly in grafts without fixation hooks. Sac embolization alone also does nothing to eliminate the source of the leak, and there is no guarantee that just because the aneurysm sac has achieved thrombosis, it is protected from systemic pressure.
Once a leak is identified at CT, the determination of its source can be difficult. Attachment site leaks can be identified at anteroposterior and lateral aortography, but type II leaks may not be identified unless catheters are placed selectively. In our series, four of eight IMA endoleaks were not seen with aortic injection alone. This is why we routinely place catheters in the SMA and internal iliac arteries when trying to find the cause of these leaks. We also suspect that this is why type II leaks were more common in our series than has been reported previously. Nonselective arteriography alone is not adequate in identifying the location and cause of these lesions.
Even with our high incidence of IMA endoleaks, we and others (19) think that there is little role for preoperative embolization. Although it is true that in all eight patients with IMA endoleaks the IMA was identified at preoperative arteriography, there are many more patients whose patent IMAs achieved thrombosis spontaneously after stent-graft deployment. If we were to perform embolization preoperatively in all patients with patent IMAs, many would undergo this potentially dangerous procedure unnecessarily.
No specific therapy for patent lumbar arteries detected preoperatively has been adopted. We have not embolized these vessels preoperatively or postoperatively and at the time this article was published we have not noted endoleak due to retrograde lumbar filling alone. We have most often seen these vessels provide antegrade egress from a type I or II endoleak (Fig 3). Others (6,19), however, have described type II endoleaks from the lumbar arteries. Because any endoleak requires flow to enter the sac through one source and to exit through another, we presume that the lumbar arteries in the patients in our series achieved thrombosis spontaneously in response to coverage by the endograft or in response to the correction of endoleak inflow. We do, however, expect to identify lumbar endoleaks as our experience grows.
|
The fate of patients with untreated endoleaks is uncertain and unpredictable. We found that some immediately detected endoleaks sealed spontaneously in the 30-day postoperative interval. Intraoperative arteriography is performed prior to heparin reversal, and some of the endoleaks noted at the time of the procedure may spontaneously seal simply with the restoration of normal coagulation parameters. We also suspect that if left untreated for longer than 30 days, additional leaks would have achieved thrombosis. However, not all endoleaks seal spontaneously, and it is clear that as long as the leak is present and is transmitting arterial pressure to the aneurysm sac, the patient remains at risk of rupture (5,20,21).
In summary, we have found that retrograde flow from the IMA contributes to many type II endoleaks. Selective arteriography may be necessary to demonstrate these leaks. Access to the aneurysm sac is possible with subselective catheterization of the SMA and of the middle colic artery. Systemic pressures can be transmitted into the sac even with small leaks. The IMA can be embolized successfully via the SMA by using microcatheters in most cases.
|
APPENDIX |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXReferences |
|---|
1. Phase II Investigation of the TALENT Endoluminal Spring Stent-Graft System for the Treatment of Sub-Renal Abdominal Aortic Aneurysms in Patients Who Are Not Candidates for Standard Surgical Intervention
2. Phase II Investigation of the TALENT Endoluminal Spring Stent-Graft System for the Treatment of Sub-Renal Abdominal Aortic Aneurysms in Patients Who Are Candidates for Standard Surgical Intervention
3. Use of the TALENT Endoluminal Spring Stent-Graft System in High Surgical Risk Patients with Abdominal Aortic Aneurysm: Emergency/Compassionate Use Protocol
4. Phase II Clinical Study of the Aortoiliac EGS System as Compared to the Standard Surgical Procedure in the Treatment of Abdominal Aortic Aneurysms
5. Phase III Clinical Study of the Safety and Efficacy of EVT Ancure Tube and Bifurcated Systems
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, R.A.B.; study concepts and design, R.A.B., J.P.C., R.M.F.; definition of intellectual content, R.A.B., J.P.C., R.M.F.; literature research, R.A.B., O.C.V.; clinical studies, all authors; data acquisition, R.A.B., J.P.C., C.M.T., O.C.V., A.M.P., R.M.F.; data analysis, R.A.B., J.P.C., C.M.T., A.M.P., R.M.F.; manuscript preparation, R.A.B., J.P.C., C.M.T., M.C.S.; manuscript editing and review, all authors.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Se Hwan Kwon, Joo Hyeong Oh, Sun Jin Park, and Ho Chul Park Endovascular Repair of a Spontaneous Right Common Iliac Artery--Inferior Vena Cava Fistula Due to Infrarenal Aortoiliac Aneurysm Vascular and Endovascular Surgery, June 1, 2008; 42(3): 279 - 283. [Abstract] [PDF]
|
|
|
|
 |
|
 S. W. Stavropoulos and S. R. Charagundla Imaging Techniques for Detection and Management of Endoleaks after Endovascular Aortic Aneurysm Repair Radiology, June 1, 2007; 243(3): 641 - 655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Tolia, R. Landis, P. Lamparello, R. Rosen, and M. Macari Type II Endoleaks after Endovascular Repair of Abdominal Aortic Aneurysms: Natural History Radiology, May 1, 2005; 235(2): 683 - 686. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. S. Maldonado and P. J. Gagne Controversies in the Management of Type 11 "Branch" Endoleaks Following Endovascular Abdominal Aortic Aneurysm Repair Vascular and Endovascular Surgery, January 1, 2003; 37(1): 1 - 12. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
