| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2003;75:126-131
© 2003 The Society of Thoracic Surgeons
a Department of Cardiovascular Sciences, Ospedale "S. Maria Della Misericordia," Udine, Italy
Accepted for publication August 9, 2002.
* Address reprint requests to Dr Gelsomino, U. O. Cardiotoracica, Azienda Ospedale "S. Maria della Misericordia," Piazzale S. Maria Della Misericordia 11, 33100 Udine, Italy.
e-mail: sandrogelsomino{at}virgilio.it
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
METHODS: Between 1986 and 2002 forty-five patients (mean age 58.7 ± 13.8 years old, 84.4% male) underwent a Cabrol procedure. Aortic dissection was the most frequent cause of aortic disease in this series (n = 17), followed by annuloaortic ectasia (n = 10), atherosclerotic aneurysm (n = 5) and poststenotic dilatation (n = 5). Six patients (13.4%) had undergone a previous aortic operation, 8 (17.7%) had a Marfan syndrome and five (11.1%) underwent concomitant arch replacement. Mean clinical follow-up was 87.3 ± 24.3 months. Twenty-eight patients (93.3% of survivors) had a transesophageal echocardiography (TEE) performed at a mean of 64 ± 32 months postoperatively.
RESULTS: Early mortality was 20%. It was 9.1% for patients with an ascending aortic aneurysm and 41.2% for dissections (p = 0.026). Independent multivariate predictors of early mortality were: aortic dissection (p = 0.009), emergency operation (p < 0.001), operative year (p = 0.02), cross-clamp time (p = 0.001), and CPB duration (p < 0.001). Actuarial survival was 0.77 ± 0.06, 0.72 ± 0.06, 0.59 ± 0.04 and 0.59 ± 0.04 at 1, 5, 10, and 16 years, respectively. Multivariate analysis revealed age (p = 0.007), cross-clamp time (p = 0.0006) and CPB duration (p = 0.009) as strong predictors of poor late survival. A periprosthetic jet with significant valve regurgitation was detected by TEE in 3 patients. In one of them, an infected periprosthetic space-right ventricular fistula was demonstrated requiring reoperation. Altogether, freedom from reoperation and endocarditis at 16 years was 0.97 ± 0.02 and 0.94 ± 0.03, respectively.
CONCLUSIONS: The Cabrol technique demonstrated a nonnegligible incidence of early and long-term complications. It should be rarely used and only when a "button" technique is not feasible.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
The Cabrol procedure has been used in our institution since 1986 and the aim of the present study was to summarize our experience with this technique over a 16-year period.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
Twenty-nine percent of the procedures were performed by 1990 and 66.6% were carried out in the first decade (1986 to 1995). Altogether, between 1986 and 1996, the Cabrol operation was used in 20.6% of patients undergoing a composite ascending aorta and aortic valve replacement in our institution. In contrast, during the last 5 years it represented only 8% of the aorta and aortic valve replacement. In our practice, indications for this technique were: redo procedures, extensive calcification of the aneurysmal aorta and low coronary ostia (< 1.5-cm above the annulus). Preoperative patient characteristics are summarized in Table 1. A large number of patients (33/45, 73.3%) were in New York Heart Association (NYHA) functional class III or IV and 8 (17.7%) had a Marfan syndrome. Six patients (13.4%) had had prior cardiac surgery through a median sternotomy: aortic valve replacement (n = 3), separate aortic valve and ascending aortic valve replacement (n = 2), Bentall procedure (n = 1).
|
Operative data are listed in Table 2. Standard median sternotomy and extracorporeal circulation techniques were used in all of the patients. Femoral artery cannulation was employed in 11 patients (24.4%) and aortic arch cannulation in the rest. In patients with hemodynamic instability or those whose aneurysms were likely to be adherent to a prior sternal incision, the femoral artery and vein were cannulated before sternotomy. In the remaining cases, venous return was through single atrial cannulation. Myocardial protection was achieved by antegrade infusion of crystalloid cardioplegia. Systemic hypothermia (25°C to 28°C) was used when circulatory arrest was not required. Five patients (11.1%) required a concomitant arch replacement and one (2.2%) had a selective antegrade cerebral perfusion (ACP) using the technique first proposed by Kazui [7].
|
From the first postoperative day onward, all patients with mechanical prostheses started a regimen of lifelong treatment with warfarin sodium (Coumadin; Du Pont Pharmaceuticals, Wilmington, DE). The target International Normalized Ratio (INR) was 2.5 to 3.0. Antiplatelet therapy was continued in patients who had concomitant coronary artery bypass surgery.
All clinical data were obtained by retrospective review of medical records, and postoperative follow-up information was obtained by written or telephone communication. Cumulative follow-up totaled 460 patient-years and was 100% complete. Clinical follow-up extended to a maximum of 192 months and the mean follow-up interval was 87.3 ± 24.3 months (median 94 months).
Infectious, thromboembolic, and bleeding complications were recorded as required by the AATS/STS/EACTS guidelines [9].
Twenty-eight patients (93.3% of survivors) had a transesophageal echocardiography (TEE) performed at a mean of 64 ± 32 months postoperatively. Informed written consent was obtained from all patients before performing the procedure. In 2 patients the TEE was not used because the procedure was refused; therefore, these patients were evaluated by TTE on a Hewlett Packard Sonos 1000/2500/5500 systems with 2.5 and 5.0 MHz ultrasound transducers (HP, Andover, MA) and recorded on VHS videotape for subsequent review.
The following features were examined:
All data were analyzed with the SPSS for Windows, release 8.0 (SPSS, Inc, Chicago, IL) statistical package. Continuous data were presented as mean ± SD; discrete variables were given as percentages. Death and event-free survival estimates were calculated by the product-limit method of Kaplan and Meier and reported with 95% of confidence limit; the Mantel-Cox (Log-rank) test was used to test the hypothesis that there was no difference in survival among subgroups.
In univariate analysis discrete variables were analyzed by the X2 or Fisher exact text; continuous variables were analyzed by one-way ANOVA (see Appendix). Significant and borderline (
0.1) variables were entered into a multivariate (Cox model) proportional hazard regression analysis to assess which risk factors were significant, independent predictors of early or late mortality. A p value of less than 0.05 was judged to be statistically signicant.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
1993 (p = 0.01), cross-clamp time (p < 0.001), and CPB duration (p < 0.001). Independent significant risk factors for early mortality at multivariate analysis were: aortic dissection (p = 0.009, relative risk [RR] 6.4 [95% CI 3.3 to 30.8]), emergency operation (p < 0.001, RR 9.2 [95% CI 3.4 to 41.6]), operative year 1993 (p = 0.02, RR 3.1 [95% CI 1.7 to 14.4]), cross-clamp time (p = 0.001, RR 8.4 [95% CI 2.1 to 34.3]), and CPB duration (p < 0.001, RR 10.1 [95% CI 25 to 40.1]).
|
During the follow-up period there were six late deaths with an actuarial survival of 0.77 ± 0.06, 0.72 ± 0.06, 0.59 ± 0.04 and 0.59 ± 0.04 at 1, 5, 10, and 16 years, respectively (Fig 1). The Hazard [± SE] were 0.25 ± 0.07, 0.32 ± 0.09, 0.52 ± 0.15 and 0.52 ± 0.15 at 1, 5, 10, and 16 years, respectively. The causes of late deaths are illustrated in Table 4. Sixteen-year survival did not differ between patients with aneurysm or dissection (0.54 ± 0.1 vs 0.55 ± 0.2, log-rank = 0.721).
|
|
Age (p = 0.007, RR 0.96 [0.93 to 1.00]), cross-clamp time (p = 0.0006, RR 8.3 [1.5 to 35.8]) and CPB duration (p = 0.009, RR 7.08 [2.24 to 22.4]) were found to be independent significant risk factors for late mortality at multivariate analysis.
Valve-Related complications
Two patients (6.6%) had episodes of endocarditis, at 1 and 14 months from the operation, respectively, and the actuarial freedom from endocarditis was 0.94 ± 0.03 at 16 years. One patient required reoperation: cumulative actuarial 16-year freedom from reoperation was 0.97% ± 0.02%. Freedom from anticoagulant-related hemorrhage, valve thrombosis, and prosthetic failure was 100%.
TEE assessments
Prosthetic aortic valve function
Patients revealed no valve thrombosis or valve vegetations on two-dimensional echocardiography. Furthermore, the exam failed to demonstrate any alteration in valve function. Mean diameter of the tubular graft at TEE was 30 ± 1.4 mm and it correlated well with the graft size (R = 0.423; p = 0.002). Doppler examination showed a peak gradient of 14.2 ± 3.8 mm Hg and a mean gradient of 8.4 ± 2.6 mm Hg. The valve area, obtained by the modified continuity equation, was 2.4 ± 0.3 cm2. Twenty patients (71.4%) exhibted a transprosthetic physiologic retrograde "washing jet," localized anteriorly in the majority of the cases (n = 17, 85%) and, less frequently, posteriorly (n = 3, 15%). A periprosthetic jet due to a valve detachment was demonstrated in 3 patients (10.7%). It was localized posteriorly in 2 patients and in the other case it was anterior leading to an insufficiency graded mild in 2 (7.1%) and moderate in 1 (3.5%) patients. Whereas in the first 2 patients a reoperation was unnecessary, in the third patient the valve detachment was caused by endocarditis with a creation of a right ventricular-aortic fistula and the color Doppler revealed a diastolic bidirectional flow from the periprosthetic space to the fistula and the left ventricle and back. This condition required a reintervention, which consisted in a closure of the fistula and a resuture of the prosthetic valve.
Altogether, there was a very low incidence of significant prosthesis regurgitation over the follow-up period; it was judged to be trivial or mild in 64.3% of the patients (n = 18) and absent in 35.7% (n = 10).
Morphology and dimensions of the tubular graft and perigraft aneurysm
Two patients (8%) had a periprosthetic aneurysm at TEE examination (diameter 6.7 cm and 6.4 cm, respectively) localized anteriorly and one patient (4%) showed a posterior aneurysm (5.6 cm). Patients who had periprosthetic aneurysms were the same as those with flow measured in the periprosthetic space; in this situation, the flow from the left ventricle to the periprosthetic space was responsible for the nonclosure of the perigraft space along with the patency of the Cabrol fistula (see below).
In both patients showing an anterior aneurysm there was no evidence of tension or dehiscence at level of the anastomotic sites and they did not need reoperation. In contrast, the posterior aneurysm was associated with prosthetic valve endocarditis, valve detachment, creation of a right ventricular-aortic fistula and persistence of the Cabrol fistula and this patient was the only one undergoing a reintervention in our series.
In the remaining patients, the periprosthetic space seemed to be completely occluded by thromboses.
Persistence of the right atrial fistula
A patent perigraft-right atrial fistula was detected by color Doppler in three patients (12%) who were the same as those with valve detachment and perigraft aneurysms. The systolic unidirectional flow from the left ventricle to the perigraft space and to the right atrium was mapped by color Doppler. The presence of a patent Cabrol fistula did not cause any congestive heart failure and was not per se responsible for reoperation.
Coronary graft morphology and function
The TEE allowed the visualization of the coronary graft in all study patients. The transverse diameter of the graft was 7.6 ± 0.3 mm and was strongly correlated to the graft size (R = 0.86, p < 0.001).
A color Doppler examination was carried out in all the patients to study the graft patency: 1 patient (4%) revealed a complete occlusion of the graft limb of the left coronary artery, with echocardiographic signs of myocardial infarction. However, it was asymptomatic and did not require reoperation.
Persistence of an intimal flap in patients operated for aortic dissection
The presence of an intimal flap at level of the aortic arch and descending aorta was documented in 2 patients who were operated because of a Stanford type A aortic dissection. In one of them, a flow signal was noticed by color Doppler, which detected a unidirectional flow toward the false lumen, while in the second patient, the false lumen was completely obliterated.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
For these reasons many modifications to the original technique have been proposed, the most common being the coronary button (Carrel patch) [19, 20] and the Cabrol techniques [5, 9].
Theoretical advantages of the coronary button technique include a better exposure of the anatomic structures, an improved surgical access for the control of the hemostasis and a more anatomical reconstruction [21]. Potential disadvantages of the "open" technique include the time necessary to mobilize the coronary ostia, the risk of damaging these vessels and the possibility of occlusion caused by tension [18, 22]. With the Cabrol technique all bleeding sites can be easily visualized and it avoids the formation of pseudoaneurysms at the coronary ostia; however, it may carry the potential risk for various late complications such as kinking of the limbs of the coronary Dacron graft and its occlusion [23]. There is little information regarding the late results of this technique, and this study aimed at exploring the results, over a 16-year period, after a Cabrol repair, which has been used in our institution since 1986 and it was used in 20.6% of patients undergoing a composite ascending aorta and aortic valve replacement between 1986 and 1996. In our study, early mortality was not negligible (20%), and it was higher in patients with aortic dissection than in those with ascending aortic aneurysm (p = 0.026). In addition, the 30-day mortality was significantly higher during the first half of the observation (25% vs 11.7%, p = 0.002), and the year of operation ( 1993) was a significant predictor of early mortality. This might be explained by the late referral to surgery during the earlier years, by an improvement in perioperative management and operative techniques, and by the learning period necessary for this complex procedure. In contrast, long-term survival was satisfactory (0.59 ± 0.04 at 15 years), with a risk of death which became constant up to 10 years (0.52 ± 0.15 at 16 years). Age (p = 0.007), cross-clamp time (p = 0.0006), and CPB duration (p = 0.009) turned out to be independent significant risk factors for late mortality and no difference in long-term survival was detected by log-rank test between patients with aneurysm or dissection (p = 0.721). Our long term results are comparable with those of the Cabrol group, who, in their last report [23], showed a 9-year actuarial survival of 0.66 ± 0.045 in 260 patients undergoing a Cabrol operation. In their study Bachet and coworkers [24] reported a 0.58 ± 0.035 8-year actuarial survival in a group of 26 patients undergoing a Cabrol operation. It has been demonstrated that TEE permits a prompt diagnosis of the potential complications of this procedure [25] allowing a complete visualization of both valved conduit and coronary conduit and of all the anastomotic sites. Valve function, identification of anastomotic leaks, assessment of grafts’ patency, detection of valve vegetations or thrombi, assessment of perigraft-right atrial fistula patency, can be all easily achieved by TEE [26]. Thus, in our series, 28 of 30 survivors underwent TEE examination. A periprosthetic jet, due to a valve detachment, occurred in three patients; in these subjects, the flow from the left ventricle to the periprosthetic space was responsible for the nonclosure of the perigraft space along with the patency of the Cabrol fistula. Two of these patients indicated a mild valve incompetence; in addition, the periprosthetic aneurysm did not lead to any tension or dehiscence of the anastomotic sites and the Cabrol fistula was not responsible for any congestive heart failure. Thus, these patients did not require reoperation. In contrast, in the third patient, the valve detachment was caused by endocarditis with a creation of a right ventricular-aortic fistula; this patient required a reoperation and was the only patient undergoing reintervention in our series. Altogether, in our study, the reoperation rate was low (2.2%) and 16-year actuarial freedom from reoperation was 0.97% ± 0.02%. Sixteen-year freedom from endocarditis was 0.94 ± 0.03. Moreover, follow-up was marked by a complete absence of anticoagulant-related hemorrhage, valve thrombosis, and prosthetic failure. Finally, one patient had a complete obliteration of the left part of the coronary graft with ECG evidence of an asymptomatic myocardial infarction that did not require reoperation.
The inclusion technique, as originally described by Bentall and De Bono, was in the past our method of choice for the composite replacement of ascending aorta and aortic valve. In our experience, this technique reduced the risk of recurrent proximal aortic aneurysm, removing the entire diseased aortic wall. A Cabrol operation was used only in specific conditions: patients undergoing redo procedures, in the presence of an extensive calcification of the aneurysmal aorta or when the coronary ostia were quite low (< 1.5 cm above the annulus) and the direct reimplantation was difficult or it might lead to an excessive tension on ostial anastomoses. In our current policy, the open technique including direct implantation of coronary buttons and a distal anastomosis to the completely divided ascending aorta is the preferred method of root replacement. This technique allows an increased length of artery with a reduced gap between the aneurysm wall and the graft with minimal tension on the anastomoses. The employment of the "button" technique further reduced, in our practice, the indications for the Cabrol operation, which is actually rarely used and only when the excision of the coronary ostial buttons is really difficult or not feasible (i.e. some redo procedures or complex repairs).
Our study is limited by its retrospective nature and by the small number of patients; however, as far as we know, it represents the longest published clinical report regarding this surgical procedure.
|
Conclusions |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
|
Appendix |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Leverich, C. Johnston, B. Stiles, L. Girardi, G. Hartman, and N. J. Skubas Cabrol Composite Graft for Aortic Root Replacement: Echocardiographic Imaging Anesth. Analg., April 1, 2009; 108(4): 1107 - 1109. [Full Text] [PDF]
|
|
|
|
 |
|
 P. D. Patel and R. R. Arora Pathophysiology, diagnosis, and management of aortic dissection Therapeutic Advances in Cardiovascular Disease, December 1, 2008; 2(6): 439 - 468. [Abstract] [PDF]
|
|
|
|
 |
|
 O. Cohen, J. Odim, D. De La Zerda, C. Ukatu, R. Vyas, N. Vyas, K. Palatnik, and H. Laks Long-Term Experience of Girdling the Ascending Aorta With Dacron Mesh as Definitive Treatment for Aneurysmal Dilation Ann. Thorac. Surg., February 1, 2007; 83(2): S780 - S784. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Cetin, A. Ozkara, E. Tireli, O. Koner, and K. Suzer Myocardial Ischemia after Cabrol Operation Asian Cardiovasc Thorac Ann, June 1, 2005; 13(2): 187 - 189. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sioris, T. E. David, J. Ivanov, S. Armstrong, and C. M. Feindel Clinical outcomes after separate and composite replacement of the aortic valve and ascending aorta J. Thorac. Cardiovasc. Surg., August 1, 2004; 128(2): 260 - 265. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
