|
|
||||||||
Eur J Cardiothorac Surg 2007;31:36-41. doi:10.1016/j.ejcts.2006.09.018
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved
a Department of Cardiovascular Surgery, Université Claude Bernard, Inserm E0226, Louis Pradel Hospital, Bron, France
b Department of Interventional Radiology, CREATIS, CNRS UMR 551, INSERM U 630, Louis Pradel Hospital, Bron, France
c Department of Cardiology and Intensive Care, Louis Pradel Hospital, Bron, France
Received 30 August 2006; received in revised form 19 September 2006; accepted 19 September 2006.
* Corresponding author. Address: Department of Cardiovascular Surgery B (Pr Jegaden), Hôpital Cardiovasculaire et Pneumologique Louis Pradel, Université Claude Bernard, INSERM E0226, 28, avenue du doyen Lépine, 69677 Bron, Cedex, France. Tel.: +33 4 72 35 75 29; fax: +33 4 72 35 75 32. (Email: fadi.farhat{at}chu-lyon.fr).
| Abstract |
|---|
|
|
|---|
Key Words: Aortic dissection David Reimplantation technique Arch replacement Direct aortic cannulation
| 1. Introduction |
|---|
|
|
|---|
The reimplantation aortic valve sparing technique (David I) has been proposed as an alternative for repair of aortic dissections based upon the principle of totally resecting the aortic root, but leaving in place a macroscopically intact aortic valve [9]. Since the sinuses are very frequently involved in the dissection, this technique seems to be very appealing. The present study was carried on prospectively to evaluate if a reimplantation valve sparing technique done systematically is associated with an increased perioperative morbidity or mortality.
| 2. Patients |
|---|
|
|
|---|
Diagnosis of type A aortic dissection, defined as involvement of the ascending aorta regardless to the distal extension of the lesions, was assessed by echography, and confirmed in 11 cases by CT-scan. All of the diagnoses were done in external hospitals before transfer to our institution. All the patients had tricuspid aortic valve. Six presented with a preoperative pericardial effusion, eight had and aortic insufficiency > 2/4 and seven had an aortic aneurysm > 50 mm. All the patients were transferred in emergency into the operation room for surgical management. In two cases, the patients presented with a preoperative collapse needing intubation, inotropic support and fluid venous filling. Two other patients had an aortic rupture at the start of the surgery. Four patients had preoperative malperfusion (limbs, three patients and cerebral, one patient). A 38-year-old woman had coronary artery dissection in association with the type A aortic dissection, without extension to the subclavian arteries.
2.2 Surgical technique
All the procedures were performed under general anesthesia, standard tracheal intubation and transesophageal echography (TEE) monitoring. Two catheters were inserted into the right radial and the left femoral arteries for continuous blood pressure monitoring. Bispectral index was used for cerebral monitoring. In four cases, arterial cannulation was performed into the right femoral artery before sternotomy because of extreme haemodynamic instability. After exposition of the heart and the great vessels, venous cannulation was done into the right atrium using a double stage cannula. For 11 patients, the arterial cannulation was made directly in the concavity of the aorta, at the junction between the ascending segment and the arch. After the cardiopulmonary bypass (CPB) was started, we began general cooling to reach a rectal temperature of 30 °C. The aortic root was dissected free from the pulmonary artery, the right ventricle and the left atrium. After cross clamping, the aorta was opened transversally and cold crystalloid cardioplegia (Celsior®, Imtix Sangstat, Wien, Austria) was delivered into the coronary ostia and repeated when necessary. The coronary ostia were dissected, leaving two coronary buttons of approximately 1 cm diameter, as well as the sinuses, up to a remnant of 34 mm. The tops of the aortic commissures were suspended with one U-stitch of 4-0 polypropylene each. The diameter of the aortic annulus was assessed using a Hegar dilator to determine the size of the DacronTM tube. Six threads of 2-0 coated polyester fibre were passed in a U-fashion underneath the aortic annulus from inside to outside in a horizontal way. These threads secured a Valsalva shaped DacronTM Tube (Gelweave Valsalva graft, Vascutek, Ann Harbor, Mi) to the aortic annulus. The tops of the aortic commissures were then attached inside the tube. The valve remnant layers forming three cuffs were reimplanted into the tube using three 4-0 polypropylene running sutures, without glue. The coaptation of the three cusps was then examined carefully to determine any residual leakage.
At that moment, rectal temperature usually reached 30 °C. In case of aortic arch replacement, a brief circulatory arrest was performed and the aortic arch was resected. Two manually inflatable retrograde cardioplegia cannulas (MedtronicTM, Minneapolis, Mn) mounted on a Y-shaped injection line were inserted into the lumen of the arterial brachiocephalic trunk and the left carotid artery to provide continuous cerebral antegrade perfusion at a temperature of 30 °C and a flow rate of 1.21.6 l/min (1015 ml/kg). The adequate perfusion was determined regarding to the right arterial radial pressure and to the lateral pressure line on the left carotid artery cannula. Another part of a standard DacronTM prosthesis was anastomosed to the aortic isthmus with a 4/0 polypropylene running suture, after instillation of fibrin glue (Tissucol®, BaxterTM, Maurepas, France) between the two layers of the descending aorta. An aortic cuff, including the three cerebral trunks, was anastomosed on the top side of the graft. At the end of the suture, the cerebral perfusion cannulas were removed, and a new cannulation was performed directly into the arch tube. The CPB was started slowly in order to de-air the tube before clamping and re-warming.
The two coronary buttons (left then right) were anastomosed on the lateral side of the Valsalva tube at the level of the skirt-like segment. Finally, the two DacronTM tubes (ascending aorta and arch) were anastomosed end-to-end with a 3-0 polypropylene running suture. For the patients without arch replacement, the ascending tube was directly anastomosed underneath the arterial brachiocephelic trunk. The aorta was declamped after de-airing, and the CPB was weaned in a standard manner. Two temporary epicardial pacing wires were placed on the right ventricle. Pericardial drainage was made using two Ch16 redon catheters with high depression (700 mmHg) [10].
2.3 Follow-up
Postoperative follow-up was done by the surgeon at the hospital, after 3, 6, 12 months and every year. A transthoracic echography (TTE) and a CT-scan were done before each outcome visit. Aortic regurgitation was assessed semi quantitatively between 0 and IV/IV. Thrombosis of the false lumen of the descending aorta was noted as well. Clinically, general dyspnea was calculated using the New York Heart Association classification (NYHA).
| 3. Results |
|---|
|
|
|---|
One patient presented with a right heart failure at the weaning of the CPB. Harvesting of the right coronary artery (RCA) revealed an atheromatous chronic occlusion. The patient had a venous-to-RCA bypass on beating heart and with the use of a stabilizer (Octopus IVTM, Medtronic, Minneapolis, Mn). The proximal end of the vein was attached to the arterial brachiocephalic trunk. Another patient had a triple arterial graft (RCA, left anterior descending, marginal branch) using both internal thoracic arteries (ITA) for a pre-existing coronary dissection. These bypasses were also performed on beating heart and with the use of a stabilizer and systematic coronary shunt insertion.
3.2 Early postoperative course
One patient died suddenly on postoperative day (POD) 8 during a tracheal aspiration. No diagnostic autopsy was performed. Two patients (13%) underwent early (POD 1) revision for bleeding. One was previously treated with clopidogrel for severe arterial disease. No surgical causes were noted on reintervention. One patient (7%) had a right transient ischemic attack (TIA) with right hemiparesia, completely regressive within 24 h. Cerebral CT-scan did not show any ischemic signs. One patient with preoperative leg malperfusion presented with postoperative mesenteric ischemia needing surgical abdominal aortic fenestration on POD 1. One patient had on POD 7 an implantation of a descending aortic covered stent graft (TalentTM, Medtronic) for a pre-existing penetrating aortic ulcer (PAU) located 10 cm above the celiac trunk. This procedure was done under epidural analgesia because of severe COPD.
Troponin I level at 24 h was 21 ± 30 µg/l. Mean bleeding at 24 h was 1165 ± 846 ml. Mean intubation and ICU times were respectively 9.5 ± 16.3 and 16.2 ± 20.9 days. Yet, four patients with severe preoperative comorbidities (malperfusion, alcohol addiction, pulmonary Rendu-Ossler disease, preoperative shock) had prolonged ventilations (range from 15 to 47.5 days). Other postoperative data are represented in Table 1 .
|
|
|
| 4. Discussion |
|---|
|
|
|---|
Supracommissural tube graft replacement in conjunction with reconstruction of the sinuses with glue is the easiest way to treat type A aortic dissection [46]. This technique allows in almost 90% of the patients the restoring of the geometry of the aortic sinuses and subsequently the valve competence, which is failing in many cases. If this surgical treatment brings entire satisfaction in immediate postoperative course, the long-term outcome remains controversial. Supracommissural replacement has been proven by some authors inasmuch as the freedom for reoperation for aortic valve dysfunction was elevated after 10 years of follow-up (up to 91%) [11,12]. Yet, others underscore that the progression of the disease, regardless to a pre-existing aortic aneurysm, could lead to aortic insufficiency by the absence of coaptation between the three leaflets. Simon et al. found that 29% of the patients with supracommissural tube graft for type A aortic dissection developed a sinus of Valsalva aneurysm within 44 ± 22 months after surgery [13]. Three reasons are often evoked as an explanation: first, the supracommissural replacement of the aorta leads to increased shear stress in the sinuses because of the absence of elasticity of the DacronTM, especially during diastole, thereby transferring the diastolic pressure to the only sinuses [14]. Second, a long aortic root remnant has by itself a dilatation risk due to cystic medial necrosis that is diffusely present in the aortic structure [15]. Third, the use of gelatineresorcinolformaldehyde has been incriminated in a higher incidence of false aneurysms or re-dissections, particularly in the proximal part of the aorta [16]. For all these reasons, some authors have strongly recommended total aortic root replacement by a composite graft, subsequently to eliminate the entire diseased aortic root [7,8]. Yet, solving a problem was to create another one: biological tubes lead to valve failure, and mechanical ones to lifetime anticoagulation. Thus, the valve preserving techniques (remodeling/reimplantation) have gained more interest these past years in type A aortic dissection treatment along with their development in elective cases, such as aortic root aneurysms. Kallenbach et al. have reported that the aortic valve preservation techniques can be performed with favorable functional results regardless to the underlying aortic lesion (type A aortic dissection or aneurysm) [17]. Yet, Leyh et al., from the same group, noted a higher failure rate in aortic root remodeling, in patients with type A aortic dissection, comparatively to a reimplantation technique [18]. These results can be easily understandable: the remodeling is based upon the suture of a DacronTM tube, which is previously three folded, directly to the 34 mm remnants of the sinuses of Valsalva. This suture is made on a diseased tissue, without any protection against further dilatation of the aortic annulus.
Our prospective series was carried on to find out whether a systematic performance of a reimplantation technique is accompanied with a higher mortality or morbidity risk. We could only note one death, occurring on POD 8. The patient was already extubated and was scheduled to leave the intensive care unit (ICU) on the same day. Before discharge, a tracheal aspiration was made for important pulmonary secretions. The aspiration was followed by an extreme bradycardia and a cardiac arrest that could not be resuscitated. Two patients underwent revision for bleeding, without any surgical cause (one was undergoing treatment by clopidogrel). The prolonged ventilations that we had in four cases were not directly related to the surgical technique.
Different points are to be discussed in our global strategy. First, we did direct aortic cannulation into the dissected aorta for 12 patients. This technique was first described by Lijoi et al., who performed two cases of aortic replacement in type A aortic dissection, with direct aortic cannulation and deep hypothermia [19]. Minatoya et al. reported a series of 14 patients with direct aortic cannulation, without rupture or peroperative malperfusion [20]. The difference with the Lijoi description is that the arch replacement is performed under mild (28 °C) hypothermia and antegrade cerebral perfusion, like reported in our series. Beside the aortic rupture risk, the elective perfusion of the false lumen is the other pitfall of this technique. This problem can only be tracked down at cross clamping, when the proximal intimal aortic tear is excluded and does not anymore constitute an entry site for the perfusion of the true lumen. This complication can be suspected easily while the right radial pressure drops (along with a preservation of the pressure in the femoral artery), witnessing of a malperfusion of the arterial brachiocephalic trunk. Minatoya et al. [20,21] have also reported this complication. We propose a simple fenestration of the intima in the arch, that to recreate a downstream re-entry site and to perfuse adequately the true lumen. The second point of our technique is the arch replacement under mild hypothermia with selective antegrade cerebral perfusion. Minatoya et al. have recently reported the absence of difference upon neurological outcome with antegrade cerebral perfusion, comparatively between three groups with a body temperature of 20, 24, and 28 °C [21]. Karck et al. performed prolonged circulatory arrest times (up to 62 ± 14 min) in combined aortic arch and descending aortic replacement, using a 28 °C antegrade cerebral perfusion, without major cerebral complications [22]. This technique brings the advantage of shortening the CPB time, reducing the platelets dysfunction and subsequently the postoperative bleeding. The third point of our approach is the use of a Gelweave ValsalvaTM prosthesis for the root replacement. This tube is designed to reproduce the anatomic and physiologic features of the normal aortic root. The root of the tube, shaped as a skirt, has vertical crimps, allowing it to expand transversally along with the cardiac systole, contrarily to the body of the tube, which has horizontal crimps. In a normal aortic root, the sinuses of Valsalva create eddy currents, first described by DaVinci, with a major importance in the aortic leaflets motion: on valve opening, they prevent the cusps edges from impacting the aortic wall, and they participate to the initiation of the aortic valve closure [23,24]. One of the suggested reasons for mid term failure of the reimplantation technique is the use of a straight DacronTM tube without any stretching structure allowing the transversal expansion during systole. The consequence is repetitive impaction of the cusps free edges on the tube's walls, leading to a progressive fibrosis retraction of the cusps with subsequent coaptation default. We believe that the Valsalva shaped tube prevents from this evolution be recreating the anatomy and the haemodynamic of a normal aortic root. Along with the progression of the series, we noticed during echographic control that the Valsalva segment was far from the free edge of the cusps during systole.
Follow up was done for 11.1 ± 4.0 months and was completed in all cases. We didnt observe any evolution towards aortic insufficiency in any patient, and previous aortic aneurysm didnt constitute a risk factor for aortic repair failure [25]. None of the patients had anticoagulant treatment. Mean aortic gradients and effective permeability index were acceptable, without difference between the postoperative period and follow-up. CT-scan reconstructions allowed a perfect visualization of the aortic root, showing the shape of the Valsalva skirt. In six patients, we noted a thrombosis of the false lumen, without any evolution towards aneurismal dilatation of the descending aorta. The persistence of perfusion in the other patients is probably related to the re-entry sites situated distally on the descending and abdominal segments.
In summary, and based upon this preliminary series; the reimplantation technique could be performed systematically in type A aortic dissection without emphasizing perioperative risk. Results during follow-up, along with others published by many authors, seem to be encouraging, and push up to keep on performing this technique routinely in type A aortic dissection. We also think that the use of a Valsalva tube grandly contributes to these good results.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. K. von Segesser, E. Ferrari, D. Delay, O. Maunz, J. Horisberger, and P. Tozzi Routine use of self-expanding venous cannulas for cardiopulmonary bypass: benefits and pitfalls in 100 consecutive cases Eur. J. Cardiothorac. Surg., September 1, 2008; 34(3): 635 - 640. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Farhat, T. Sassard, and O. Jegaden Central cannulation in acute aortic dissection repair: What else? J. Thorac. Cardiovasc. Surg., August 1, 2007; 134(2): 547 - 548. [Full Text] [PDF] |
||||
![]() |
A. L. Estrera and H. J. Safi Editorial comment: Proximal reconstruction during Type A aortic dissection Eur. J. Cardiothorac. Surg., January 1, 2007; 31(1): 41 - 42. [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 |