| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2005;4:275-279. doi:10.1510/icvts.2004.094193 © 2005 European Association of Cardio-Thoracic Surgery
Systematic and exclusive use of intravascular ultrasound for endovascular aneurysm repair – the Lausanne experienceDepartment of Cardio-vascular Surgery and Angiology, University Hospital CHUV, Rue du Bugnon 46, 1011 Lausanne, Switzerland Received 5 August 2004; received in revised form 14 March 2005; accepted 14 March 2005
*Corresponding author. Tel.: +41-21-314-2316; fax: +41-21-314-2278.
Five years of experience with endovascular infrarenal aneurysm repair at our institution is reviewed. Implantation of endoprostheses in 88 patients has been performed by surgeons using exclusively intravascular ultrasound (IVUS) and fluoroscopy. IVUS identified the target site of deployment in all cases. In-hospital morbidity was 22% (19/88). Two percent mortality (2/88) and 5% early conversion (4/88) as a consequence of type I endoleaks were noted only in the first 53 patients with early devices (NS). Early endoleaks were present in 36% (32/88) including twenty-two type I, five type II and five type III endoleaks. Proximal endoleaks were associated with early devices (P<0.001), and technical difficulties with deployment. Tube grafts used in the beginning, performed poorly with 54% (7/13) type I endoleaks. Endoleaks diminished to 10% (9/88) by spontaneous closure and secondary endovascular procedures that were necessary in 24% (21/88) and consisted of coil embolization/cuff extension (9), late conversion (6), and limb recanalization or femoral cross-over bypass (6). Endovascular aneurysm repair using IVUS is a valid alternative technique. Improved devices and systematic use of bifurcated endoprostheses for infrarenal aneurysms reduce the occurrence of type I endoleaks.
Key Words: Endovascular; Intravascular ultrasound; IVUS; Aortic aneurysm; Endoleak
Intravascular ultrasound (IVUS) creates high-quality cross-sectional views of the vascular system. In aortic aneurysms IVUS allows for precise measurements and quality assessment of the aortic neck, and identification of the target site, namely, the lowermost renal artery [1]. Quality control by IVUS following endovascular repair includes expansion of the device, its apposition and patency of major branches. Therefore IVUS is considered an important adjunct for endovascular interventions [2,3]; however, its exclusive use is rare. In a previous study we demonstrated IVUS to be efficient for precise positioning of endoprostheses replacing completion arteriography [4]. The purpose of the present study was to analyze our experience with infrarenal aortic aneurysm treatment based on systematic and exclusive use of IVUS for the endoprosthetic repair.
2.1. Database and patient demographics From February 1998 to August 2002, a consecutive series of 88 patients with an infrarenal aortic aneurysm was treated by our institution. All relevant data were entered in a computerized vascular registry. Endovascular repair included both low and high-risk patients, many of them with comorbidities (Table 1).
2.2. Endoprostheses Early (first and second) and late generation devices of Talent (Medtronic World Medical, Sunrise, FL) and Excluder (ExcluderTM, WL Gore and Associates, Flagstaff, AZ) were used. In the beginning Zenith (Cook, Inc, Bloomington, IN), Vanguard (Boston Scientific, Oakland, NJ) and Stenway (Stenford Groupe Valendos S.A., Nanterre, France) endoprostheses were occasionally used, the latter taking both off the market. 2.3. Preoperative investigation Preoperative assessment consisted of standard calibrated arteriography and enhanced helical computed tomography. Coil embolization or balloon dilation was preferentially performed prior to aneurysm exclusion.2.4. IVUS-based aneurysm repair Endovascular repairs, including the operation of the IVUS machine, were performed by cardiovascular surgeons in the operating room. General anesthesia was used mostly, and epidural anesthesia in high-risk patients. Both common femoral arteries were exposed, and an 8 F introducer (Introducerkit, Boston Scientific, Meditech, Watertown, MA) inserted. A 6 F probe of an intravascular ultrasound catheter (Sonicath Ultra 6, 12.5 MHz, Boston Scientific) was advanced. The target site identification process was standardized. A manual pullback of the IVUS probe visualized the left renal vein (Fig. 1) and renal arteries that were marked under fluoroscopy on the patient's abdomen (Fig. 2). Pullback and fluoroscopy enabled positioning of a second marker at the distal end of the proximal neck, and a third marker at the aortic bifurcation, all within the fluoroscopic field. This ‘one fluoro position’ was used first. In 2001, we started to center the renal arteries in the fluoroscopic field of view to minimize parallax error. Then the endoprosthesis was deployed under fluoroscopic control (Fig. 2). Complete expansion of the device, and patency of the renal and internal iliac arteries were verified by IVUS.
2.5. Postoperative monitoring and follow-up examination Office visits within 1 month of surgery, and duplex sonography of the aneurysm were performed during hospital stay and at 6 and 12 months. Plane radiographs of the abdomen and an enhanced CT scan were obtained after twelve months and then yearly. Arteriography was performed selectively on the base of a persistent endoleak or aneurysm expansion. 2.6. Statistical analysis All data are presented as mean±SED. Differences between groups were evaluated with 2 test and reported as significant if the P value was less than 0.05.
3.1. Patient demographics There was an average of 2.4 comorbidities per patient (Table 1). Follow-up ranged from 7 to 43 months (34±16 months). 3.2. Procedure Mean time for anesthesia, procedure and radiation were 272±87 min, 200±79 min and 19±17 min, respectively. The operation of the IVUS machine was simple. IVUS identified renal arteries, neck and aortic bifurcation in all cases. Arteriography was not necessary. Device deployment was difficult in 13 cases due to high friction force during sheath retraction resulting in a lower than intended device position in 12 cases. A higher position with covering of one renal artery orifice was detected by IVUS, and patency was restored by distal displacement. Incomplete aneurysm exclusion due to a short iliac limb was detected once by IVUS and treated by an extension. In five cases IVUS detected incomplete alignment of a prosthetic limb requiring additional balloon dilation.3.3. Mortality and morbidity Thirty-day perioperative mortality was 2% (2/88) due to congestive heart failure and consumptive coagulopathy. Four perioperative conversions (5%) were necessary to resolve an important endoleak. Three tube grafts were involved. Nineteen patients had major complications including conversion with a total 30-day major morbidity of 22% (Table 2). Acute conversions and deaths all occurred in the first 53 patients who received early devices (NS, Table 3). Minor morbidity was 6% (one hematoma, two lymphoceles and two groin infections). Late death unrelated to the aneurysm or its repair occurred in 8 patients (9%) during follow-up.
3.4. Endoleaks Intraoperative and early endoleaks were present in 36% (32/88). Duplex sonography prior to discharge revealed 18 proximal or distal attachment site endoleaks (type I; 20%), 5 retrograde side-branches (type II; 6%), and 4 graft defects (type III; 5%).
3.4.1. Type I endoleaks
3.4.2. Type II endoleaks They showed spontaneous sealing in three of five patients at 12 months, in one of them with aneurysm shrinkage. Coil embolization was performed in one patient. In the other patient a small endoleak persisted with a stable aneurysm diameter.
3.4.3. Type III endoleaks After one year the total percentage of endoleaks decreased to 10% (9/88). 3.5. Secondary procedures Twenty-one of 88 patients (24%) underwent secondary procedures. Coil embolization or extension for endoleaks was performed in nine patients during follow-up. Late conversion was necessary in six patients (7%). Two patients with first-generation devices had claudication due to kinking and disintegration of both prosthetic limbs. All tolerated surgery well. Four patients (5%) had occlusion of one limb during the initial 30 days, and 2 patients (2%) during follow-up. Limb occlusion was treated by thrombolysis, stent placement or thrombectomy in 3 patients, by crossover bypass in 3 patients, and by late conversion in one patient.3.6. Aneurysm size Aneurysm size was calculated for 51 patients with a follow-up of 12 months or more. Maximum aortic diameter decreased by 5 mm or more in 15 patients (29%). It remained unchanged (change <5 mm) in 33 of 51 patients (65%), and increased by 5 mm or more in 3 of 51 patients (6%). Two patients with an aneurysm increase had an endoleak on CT scan. The third patient who was on dialysis showed no endoleak on CT scan, but the entire thoraco-abdominal aorta extensively calcified, and the patient was unfit for any intervention.
The rationale for exclusive use of IVUS in endovascular aneurysm repair is the disposal of an easy and quick setup and the reduction of technical and personnel equipment. In a former study we demonstrated the equivalence of IVUS to arteriography [4], and the present study supports the feasibility of an exclusively IVUS-based repair without arteriography. Besides the pre- and post-deployment assessment, IVUS was also useful in verifying proper position of the guide within the contralateral limb, as experienced by others [5]. IVUS was effective to visualize incomplete expansion in some endoprostheses due to the 360° cross-sectional view. We appreciated the fact to renounce completely on contrast dye for the implantation procedure in patients with a known prevalence of 13 to 21% renal failure [6], IVUS seems to be advantageous over intraoperative arteriography in terms of radiation exposure. It probably reduces fluoroscopy time considerably. In this study mean fluoroscopy time was 19 min consisting of regular fluoro-scopy. Using an IVUS-based repair a cineaquisition mode associated with a high radiation dose is avoided. Lipsitz et al. reported a mean fluoroscopy time of 39 min per case in 47 endovascular aneurysm repairs using intraoperative arteriography [7]. They used high-level fluoroscopy due to ‘road mapping’ in up to 37% of the fluoroscopy time. A downside of this IVUS-based strategy is the risk of missing endoleaks escaping prompt treatment. Although endoleaks nowadays account for 2 to 3% in centers of excellence, a meta-analysis of clinical studies demonstrated an incidence of 24% [8,9]. We noted a significant relation between early devices and type I endoleaks. The high friction force during the deployment in some of these devices resulted in an accidentally low placement in twelve cases predisposing to insufficient sealing. Ease in device deployment is a prerequisite for precise positioning and nowadays standard. In fact, improved devices perform better [9–11]. Another reason for type I endoleaks in our study is the ‘one fluoro position’ technique. The divergence of the X-ray beams is highest at the outskirts of the fluoroscopic screen resulting in parallax error. Using this technique, the more angulated and rotated the aortic neck is, the lower the endoprosthesis is placed related to the renal arteries. The most accurate device position is achieved by the technique used by Broeders et al. [12]. They propose centering of the neck in the fluoroscopic field, and adapting the C-arm to the angulated neck by proximal tilting and left-right rotation. This technique is equally important for aortography and IVUS. The inclusion of challenging neck configurations in our study contributed also to endoleaks. As endovascular surgery evolved, hostile neck anatomy was identified as an important risk factor accounting for 36% proximal endoleaks or even higher in important neck angulation [13]. Finally, the use of tube grafts contributed considerably to endoleaks. Today they are no longer recommended regarding an inherent 26 to 57% risk of attachment site endoleaks [14]. A comment deserves the type III endoleaks in our study. They were mainly due to the impossibility to implant the contralateral leg. Although it is not specifically mentioned in the literature, we experienced canulation of the second limb often as the most time consuming step. Here, IVUS was of limited benefit. The mortality of 2% in our study was low comparable to others [6,10]. The total morbidity was high, secondary to various pre-existant comorbidities and technical complications. Acute lower limb ischemia due to prosthetic limb occlusion was present in 5% consistent with the literature [6]. Acute conversions were necessary only in patients receiving early devices. Nowadays the risk of acute conversion is less than 2% due to improved graft designs and endovascular experience [6,11]. Secondary interventions to maintain aneurysm exclusion or graft patency were necessary in one fourth of our patients. So far endovascular aneurysm treatment is considered not as durable as standard surgical repair with a need for secondary procedures in 10 to 27%, yet in most cases it is successfully resolved by endovascular means [10]. Late conversion is kept in reserve for complex graft failure and required only in 2 to 4% [15]. We noted 7% in our study mostly due to insufficient proximal fixation. Modern endografts design eliminated the structural faults that marred withdrawn devices [11]. They significantly reduced conversion and the need for secondary endovascular interventions. In conclusion, systematic and exclusive use of IVUS for implantation of endoprostheses is a valid alternative technique, and complications related to the use of IVUS are very rare. The identification of the target site by IVUS is easy and reliable. Centering and adjusting the fluoroscopic view when localizing the IVUS probe, and selection of straight aortic necks with few angulation enhance precise deployment. Early generation devices and tube grafts were mainly responsible for a high rate of attachment site endoleaks in this study.
Author: Ludwig K. von Segesser (CHUV, Switzerland) eComment: What is the outcome for first versus last generation covered stents? Author: Bettina Marty: (CHUV, Switzerland) eResponse: Thank you for this important comment emphasizing the ongoing evolution in stent technology and knowledge. Indeed, the results with new devices have markedly improved. Better performance of newer devices became obvious also in our study. In the last 35 patients in our series consisting of a total of 88 patients there were no deaths, conversions or early type I endoleaks. Regarding type I (attachment site endoleaks) there was a significant difference. In fact, in the literature, attachment site endoleaks was a relatively frequent complication in early series with 20% or more, whereas in current series it is reported in about 5%.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
