Reoperation for Giant False Aneurysm of the Thoracic Aorta: How to Reenter the Chest?

Reoperation for Giant False Aneurysm of the Thoracic Aorta: How to Reenter the Chest? Jean Bachet, MD, Manuel Pirotte, MD, François Laborde, MD, and Daniel Guilmet, MD Département de Pathologie Cardiaque, Institut Mutualiste Montsouris, Paris, Service de Chirurgie Cardio-vasculaire, Hôpital Foch, Suresnes, France Background. Giant false or pseudoaneurysm of the aorta is a rare but dreadful complication occurring several months or years after cardiac or aortic surgery. We describe a surgical approach that allowed safe reentry in the chest in five patients, with a mean follow-up of almost seven years. Methods. From December 1991 to October 1999, five patients aged 34 to 74 years (mean age, 55 11.6 years), who had previously undergone a total of nine operations in other institutions, required reoperation for giant false aneurysm of the ascending aorta in a mean delay of 22.6 20.3 months (3 months to 6 years) after the last surgical procedure. In order to avoid major mediastinal wound and patient s exsanguination during resternotomy, the following technique has been used: femoral artery cannulation; right atrial cannulation through the femoral vein; femoro-femoral full-flow cardiopulmonary bypass; rectal temperature lowered to 25 C; direct cannulation and cross-clamping of both carotid arteries through a direct cervical approach, and selective cerebral perfusion with cold blood (10 C to 12 C); circulatory arrest of the main circuit; chest opening; and mediastinal division. Results. Despite the fact that the false aneurysm was entered in all patients, reopening of the chest has been safe in all cases. In four cases, the aortic repair consisted of complete graft replacement (Dacron) of the compromised aortic segment (ascending aorta in two; both ascending aorta and aortic arch in two). In one case, reimplantation of the left coronary ostium and closure of a fistula with the left ventricle was carried out. One patient with ongoing mediastinitis died from intractable septicemia and multiorgan failure. Presently, two patients are in excellent condition; one suffers from light neurologic sequelae (oculomotor nerves palsy) and one patient had a nonrelated stroke one year postoperatively. Conclusions. The technique of separate carotid cannulation and selective antegrade brain perfusion with cold blood during circulatory arrest at moderate core hypothermia has, in our opinion, many advantages. In addition to allowing harmless opening of the chest in the presence of most dangerous mediastinal false aneurysms, it implies no general deep hypothermia, reduced duration of cardiopulmonary bypass, and circulatory arrest of the lower part of the body, and safe and permanent brain protection throughout chest opening and mediastinal division. It has allowed us to safely reoperate on patients who are generally considered as a major surgical risk. (Ann Thorac Surg 2007;83:1610 4) 2007 by The Society of Thoracic Surgeons Accepted for publication Dec 27, 2006. Address correspondence to Dr Bachet, Département de Pathologie Cardiaque, Institut Mutualiste Montsouris, 42, Boulevard Jourdan, 75014 Paris, France; e-mail: jean.bachet@imm.fr. After cardiac or aortic surgery, some patients may develop a false aneurysm requiring reoperation. In rare cases, the volume and the location of the lesion make direct resternotomy impossible because of the major risk of mediastinal wound and the patient s exsanguination (Figs 1A; 1B). For most authors [1 3], thorax reentry is one of the key factors of a successful reoperation, but even with a safe technique avoiding brutal exsanguination or life threatening mediastinal wound the hospital mortality may reach 17% to 20% [1, 3, 4]. The technique usually employed in these cases associates the following: femoro-femoral cardiopulmonary bypass (CPBP), core profound hypothermia (16 C to 18 C), and total circulatory arrest of the main circuit (CA) implemented systematically before reopening the chest [1, 2, 4]. The disadvantages of this method are well known: very long duration of CPBP, danger of prolonged heart fibrillation, possible air embolism and overall, limited time allowed for entering the chest, dividing the mediastinal structures, and repairing the aortic lesion. In order to obviate those drawbacks, we have used a safer procedure based on selective cannulation of both carotid arteries through specific cervical approach and separate antegrade cerebral perfusion with cold blood and distal CA in moderate core hypothermia, as described as early as 1986 for surgery of the aortic arch [5 7]. Material and Methods Surgical Technique The concept of the procedure is to independently cannulate both carotid arteries through limited cervicotomies, after institution of full flow CPBP through the femoral vessels, and induction of moderate core hypothermia at 25 C. When this level of temperature is obtained, selec- 2007 by The Society of Thoracic Surgeons 0003-4975/07/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2006.12.044

Ann Thorac Surg BACHET ET AL 2007;83:1610 4 GIANT FA OF THE THORACIC AORTA 1611 tive cerebral cold (10 C to 12 C) perfusion is initiated, while the circulation is discontinued in the main circuit. This requires the addition, beyond the oxygenator, of a heat exchanger usually dedicated to cold blood cardioplegia, and a roller pump (Fig 2). The patient is prepared and anesthetized as usual (monitoring of central venous pressure, radial artery pressure, 5-lead electrocardiography, rectal and esophageal temperatures, intermittent blood gas analysis). A 12-lead electroencephalogram (EEG) is recorded permanently throughout the entire procedure. After access and division of the right common femoral artery and vein, the patient is anticoagulated with an appropriate dose of heparin and the femoral artery is cannulated with an 18 to 22F Bardyk cannula (Terumo, Ann Arbor, MI), whereas a long venous cannula (21F Medtronic/Bio- Medicus, Minneapolis, MN) is pushed into the right atrium through the femoral vein in order to establish full-flow CPBP. The rectal temperature is lowered to 25 C to 28 C. During cooling, a double cervicotomy is carried out allowing direct carotid cannulation (Fig 3), with specially designed cannulas available in several diameters (2.5 to 6 mm; Polystan A/S, Vaerløse, Denmark) or simply with coronary sinus balloon catheters (4, 6, 7, 8 mm, Polystan A/S). Cannulas are held by means of adventitial 5-0 polypropylene purse-string sutures. The common carotid arteries are then cross-clamped, and selective cold perfusion is initiated at a flow rate of 400 to 500 ml/minute. Because of the very low temperature of the perfusate, the EEG becomes silent in a mean of nine minutes. During selective cerebral perfusion (SCP) and CA, no adjuncts (such as barbiturates or steroids) are used to enhance the cerebral protection as we consider that such adjuncts would not bring any increased protection over the permanent perfusion with cold blood. Similarly, it seems to us useless to pack the patient s head in ice. Then CPBP is discontinued, while selective cold blood (10 C to 12 C) perfusion of the carotid arteries is maintained. Resternotomy and mediastinal division are then carried out. Even in case of direct entry of the false aneurysm there is no danger of exsanguination or cerebral air embolism. Depending on the location of the false aneurysm, the repair of the aortic lesion can be per- CARDIOVASCULAR 4 Fig 1. (A) Computed tomographic (CT) scan of a giant false aneurysm (FA) of the ascending aorta (Ao) after three surgical procedures in a Marfan patient. The arrow indicates the origin of the false aneurysm at the base of the valved conduit. (B) A CT scan showing a giant FA several year after surgery for type A dissection. It shows how the anterior chest wall and the sternum constitute the anterior limit of the FAs and demonstrates that it would have been quite impossible to enter the chest without exsanguinating the patient in the absence of circulatory arrest. (C) Preoperative angiogram of a giant FA of the Ao in a patient suffering from methicillin resistant Staphylococcus Aureus mediastinitis after two coronary artery bypass graft procedures. The long arrow points to the origin of the false aneurysm at the cannulation site on the ascending aorta. The short arrow indicates the deformity of the innominate artery induced by the voluminous FA.

1612 BACHET ET AL Ann Thorac Surg GIANT FA OF THE THORACIC AORTA 2007;83:1610 4 bypass and occurrence of a giant false aneurysm at the aortic cannulation site (Fig 1C). Results In all surgical procedures reopening of the chest has been safe as there was no catastrophic hemorrhage, despite the fact that the false aneurysm was directly entered in all patients. In four patients, the aortic repair consisted in performing a total graft replacement with a Dacron prosthesis (Hemashield; Boston Scientific, Natick, MA; Intervascular, La Ciotat, France) of the compromised aortic segment (ascending aorta in two; ascending aorta and aortic arch in two). In the fifth patient, reimplantation of the left coronary ostium (direct suture) and closure of a fistula between the left ventricle and the composite graft of a Bentall procedure was carried out (Table 1). Perfusion and Circulatory Arrest The mean durations of CPBP, SCP, and CA were 159 52 minutes (90 to 240 minutes), 79 23 minutes (50 to 120 minutes), and 34 9 minutes (14 to 45 minutes), respectively. Fig 2. The perfusion circuit. (P roller pump; HE heat exchanger; Ox oxygenator. formed under CA and SCP (especially if arch replacement is required), or after cross-clamping of the distal ascending aorta and resuming of CPBP. As soon as full-flow CPBP is resumed, SCP is discontinued, the carotid arteries are decannulated, and the patient is rewarmed. Patients All patients data have been reviewed retrospectively. From December 1991 to October 1999, five patients (four male and one female) were reoperated on for a false aneurysm (or ruptured chronic dissection self-contained by the mediastinal closed space) of the ascending aorta, in two departments of cardiac surgery (Hôpital Foch and Institut Mutualiste Montsouris, Paris, France). Patients surgical histories are summarized in Table 1. The mean age was 55 11.6 years (34 to 74 years) and the mean delay between redo surgery and prior operation was 22.6 20.3 months (3 months to 6 years). All patients had preoperative thoracic computed tomography (CT) or thoracic aortography. Doppler assessment of the carotid arteries was carried out in all patients. In addition, the carotid arteries were visualized on the preoperative thoracic aortic CT scan or angiograms in three patients. One patient had been treated for eight months for chronic mediastinitis, after two procedures of coronary artery Mortality and Morbidity The EEG had returned to normal and the end of the procedure in all patients but one (who returned to normal after six hours). One patient, already treated for chronic mediastinitis before reoperation (blood and mediastinal cultures with methicillin resistant Staphylococcus aureus [MRSA]), died from intractable septicemia (with polymicrobial cultures: MRSA and Pseudomonas aeruginosa) and multiorgan failure. Follow-up extends from 6 to 14 years with a mean of almost 7 years (83.5 46.5 months). Two patients are in excellent clinical condition 14 and 6 years after the aneurysm repair. One patient had a nonrelated stroke Fig 3. Cannulation of the carotid arteries through direct cervical approach. The carotid cannulation does not interfere with the sternal operative field.

Ann Thorac Surg BACHET ET AL 2007;83:1610 4 GIANT FA OF THE THORACIC AORTA Table 1. Patient History Patient Age Op No. 1 Op No. 2 Op No. 3 Op No. 4 First 34 Marfan Syndrome, acute aortic dissection Ruptured chronic dissection of the aortic arch Surgical procedure Bentall Arch replacement Second 47 Aortic regurgitation Aneurysm of the ascending aorta False aneurysm of the distal anastomosis Surgical procedure AVR Bentall Aortic arch replacement 1613 CARDIOVASCULAR Third 58 Marfan Syndrome, aortic and mitral regurgitation Aneurysm of the ascending aorta Surgical procedure DVR Ascending aorta replacement Postoperative posterior myocardial infarction Bentall and MVR Giant false aneursym Reimplantation of coronary ostia and closure of a fistula with the left ventricle Fourth 62 Ischemic cardiomyopathy Recurrence of ischemia chronic mediastinitis False aneurysm at the cannulation site Surgical procedure MIDCAB Double CABG Ascending aorta and arch replacement Fifth 74 Aortic regurgitation Ruptured chronic dissection of the aortic arch Surgical procedure AVR Ascending aorta and arch replacement AVR aortic valve replacement; CABG coronary artery bypass grafting; DVR double valve replacement; MIDCAB minimal invasive direct coronary artery bypass; MVR mitral valve replacement; Op operation. one year after the false aneurysm repair. One patient suffers from light neurologic sequelae (oculomotor nerves palsy), probably related to a left subclavian steal syndrome during perfusion or an anomalous circle of Willis. We have not observed any recurrence of false aneurysm on late systematic thoracic CT scan performed on a yearly basis. Comment Giant false aneurysm, occurring after cardiac or aortic surgery (3 to 72 months in our review), is considered to be extremely rare [1]. However, some retrospective reviews attest of an incidence that may reach 12% to 35% [2, 8]. The following presupposed risk factors are well-defined in redo patients [4]: graft infection or aortitismediastinitis, Marfan syndrome, previous acute type A aortic dissection, and even tissue necrosis by abusive use of gelatin-resorcin-formol glue. Other rare nonaortic related causes of false aneurysm, such as direct thoracic trauma or sternal wires [9, 10], tuberculosis, or human immunodeficiency virus, have also been described [11]. Associated clinical symptoms may be misleading [1, 8] (fever, chest discomfort caused by retrosternal mass), but generally lead to the proper diagnosis through twodimensional echocardiography or CT scan [12]. Some authors [1, 4] have emphasized the infectious etiology (50% of their patients) and the necessity of radical surgical repair as they have experienced recurrence of false aneurysm formation after local or conservative repair. Only one patient in the present series demonstrated an obvious relation between chronic mediastinitis and occurrence of a false aneurysm at the site of aortic cannulation. Moreover, this patient is the only one who died postoperatively. In such a case one may discuss the use of an aortic homograft. However, it may not be available, in particular when the reoperation is required emergently or urgently (as in our case). In addition, the infectious process may have already impaired the main vital functions of the patient who cannot overcome the surgical procedure despite a proper aortic replacement. Mohammadi and colleagues [3] have used about the same surgical technique in some of their patients (CA in 21% and cold SCP in 57%) before reopening the chest. Others have also proposed some variants using Heartport port-access technology (CardioVations, Ethicon/J&J, Somerville, NJ) avoiding cervical cannulation, long CA, and profound hypothermia, by cannulating the right axillary artery and inserting and inflating an EndoClamp (CardioVations) balloon in the ascending aorta [13], or simply by positioning the balloon occlusion catheter at

1614 BACHET ET AL Ann Thorac Surg GIANT FA OF THE THORACIC AORTA 2007;83:1610 4 the level of the disrupted aortic anastomosis [14 16]. Those techniques can easily be criticized. On the one hand, cannulation of the right axillary artery without cross-clamping the origin of the innominate artery does not eliminate the risk of massive hemorrhage during chest opening. On the other hand, the use of an EndoClamp balloon may prove dangerous as it may tear or disrupt the fragile aorta, or be displaced in front of the origin of the epiaortic vessels in the transverse arch. In view of the satisfactory results obtained in the present very limited experience, but also in more than 200 patients operated on for aortic arch replacement with the aid of cold blood selective antegrade brain perfusion at moderate core hypothermia [7], it seems largely demonstrated that the technique of cold blood selective antegrade cerebral perfusion provides safe cerebral protection. One may question the temperature of the perfusion used in the present experience. Indeed, many groups [2, 17 19] have reported excellent results by perfusing the brain at 23 C to 25 C. The present experience was carried out between 1991 and 1999 when the perfusate temperature used for selective cerebral protection in all patients undergoing surgery of the aortic arch, was 10 C to 12 C. In 2002 we turned to the Kazui technique, in which the perfusate temperature is set to 23 C to 25 C. The results have been equally satisfactory and we acknowledge that this level of perfusate temperature is safe and that the perfusion circuit is simpler. Therefore, presently we would certainly recommend perfusing the brain at 23 C to 25 C in such a particular occasion. In conclusion, when dealing with giant redo false aneurysm of the ascending aorta and (or) aortic arch, the use of systematic selective antegrade carotid perfusion through separate cervicotomies, moderate hypothermia, and CA at the moment of chest reopening has many advantages and, in particular, provides excellent brain protection with no limitation of brain exclusion, allowing for unhurried mediastinal dissection and aortic repair. Associated with radical aortic replacement rather than local repair (such as direct redo anastomosis) those techniques represent to date, in our opinion, the best possible method to avoid critical exsanguination, cerebral or myocardial complications, and tragic intraoperative or immediate postoperative outcome after surgery of giant mediastinal aortic false aneurysm. References 1. Dumont E, Carrier M, Cartier R, et al. Repair of aortic false aneurysm using deep hypothermia and circulatory arrest. Ann Thorac Surg 2004;78:117 21. 2. Schepens MAAM, Dossche KM, Morshuis WJ. Reoperations of the ascending aorta and aortic root: pitfalls and results in 134 patients. Ann Thorac Surg 1999;68:1676 80. 3. Mohammadi S, Bonnet N, Leprince P, et al. Reoperation for false aneurysm of the ascending aorta after its prosthetic replacement: surgical strategy. Ann Thorac Surg 2005;79: 147 52. 4. Katsumata T, Moorjani N, Vaccari G, Westaby S. Mediastinal false aneurysm after thoracic aortic surgery. Ann Thorac Surg 2000;70:547 52. 5. Guilmet D, Roux PM, Bachet J, et al. Nouvelle technique de protection cérébrale: chirurgie de la crosse aortique. Presse Med 1986;15:1096 8. 6. Bachet J, Guilmet D, Goudot B, et al. Cold cerebroplegia. A new technique of cerebral protection during operations on the transverse aortic arch. J Thorac Cardiovasc Surg 1991; 102:85 94. 7. Bachet J, Guilmet D, Goudot B, et al. Antegrade cerebral perfusion with cold blood: a 13-year experience. Ann Thorac Surg 1999;67:1874 8. 8. Kazui T, Yamashita K, Terada H, et al. Late reoperation for proximal aortic and arch complications after previous composite graft replacement in Marfan patients. Ann Thorac Surg 2003;76:1203 8. 9. Muller MF, Ferencz J. False aneurysm of the ascending aorta caused by a sternal wire. Circulation 2005;111:e14. 10. Chiung-Lun K, Jen-Ping C. Aortic graft pseudo-aneurysm secondary to fracture of sternal wires. Tex Heart Inst J 2003;30:240 2. 11. Chello M, Tamburrini S, Mastroroberto P, Covino E. Pseudoaneurysm of the thoracic aorta in patients with human immunodeficiency virus infection. Eur J Cardiothorac Surg 2002;22:454 6. 12. You-Bin D, Chun-Lei L, Qing C. Chronic traumatic pseudoaneurysm of the ascending aorta causing right ventricular inflow obstruction. Circ J 2003;67:359 361. 13. Shetty R, Voisine P, Mathieu P, Dagenais F. Recannulation of the right axillary artery for complex aortic surgeries. Tex Heart Inst J 2005;32:194 7. 14. Pettersson G, Nores M, Gillinov AM. Transfemoral control of ruptured aortic pseudoaneurysm at aortic root reoperation. Ann Thorac Surg 2004;77:311 2. 15. Warnecke H, Franz N, Roethemeyer S, Ritter F, et al. Preoperative balloon sealing of a false ascending aortic aneurysm for safe redo sternotomy. Ann Thorac Surg 2005; 79:2166 8. 16. D Attellis N, Diemont FF, Julia PL, Cardon C, Fabiani JN. Management of pseudoaneurysm of the ascending aorta performed under circulatory arrest by port-access. Ann Thorac Surg 2001;71:1010 1. 17. Kazui T, Inoue N, Komatsu S. Surgical treatment of aneurysms of the transverse aortic arch. J Cardiovasc Surg (Torino) 1989;30:402 6. 18. Kazui T, Inoue N, Yamada O, Komatsu S. Selective cerebral perfusion during operation for aneurysms of the aortic arch: a reassessment. Ann Thorac Surg 1992;53:109 14. 19. Di Bartolomeo R, Pacini D, Di Eusanio M, Pierangeli A. Antegrade selective cerebral perfusion during operations on the thoracic aorta: our experience. Ann Thorac Surg 2000;70: 10 6.