|
|
||||||||
Ann Thorac Surg 2000;70:547-552
© 2000 The Society of Thoracic Surgeons
A Department of Cardiac Surgery, Oxford Heart Centre, The John Radcliffe Hospital, Oxford, England, GBR, UK
Address reprint requests to Dr Katsumata, Oxford Heart Centre, John Radcliffe Hospital, Oxford OX3 9DU, England
| Abstract |
|---|
|
|
|---|
Methods. Between April 1993 and February 1999, we treated 10 patients, aged 25 to 73 years, with anastomotic mediastinal false aneurysm originating from the proximal thoracic aorta. Nine had undergone prior operations on the ascending aorta (7, type A dissection repair; 1, aortitis; 1, root abscess) with a Dacron conduit (n = 5) or valved conduit (n = 4). The last patient had undergone valve replacement for excavating aortic root sepsis. False aneurysms were detected from 2 to 70 months after the most recent operation. Three patients had positive tissue cultures. The surgical procedure was direct suture repair of the disrupted anastomosis in 5, root or ascending aortic replacement with an aortic homograft in 4, and Dacron graft interposition in 1. Hypothermic low-flow perfusion with or without circulatory arrest was used in all patients.
Results. There was one hospital death caused by staphylococcal mediastinitis. A false aneurysm recurred after direct suture repair in 2 patients with underlying type A dissection or aortitis. This resulted in one late death. One patient experienced a neurologic event during removal of an infected vascular graft. All 8 surviving patients are alive and well after a mean follow-up of 2 years. Three patients with mycotic false aneurysms remain free from infection after aortic homograft replacement.
Conclusions. Mediastinal false aneurysms are surgically taxing. Low-flow hypothermic perfusion with or without circulatory arrest allows safe reentry. Radical surgery provides a satisfactory outcome in infected patients. Local repair of suture dehiscence in pathologic tissues may predispose to recurrence. We suspect that excessive use of formalin in gelatin-resorcin-formol glue may predispose to tissue necrosis.
| Introduction |
|---|
|
|
|---|
Without an operation, aortic false aneurysms progressively expand, compress and erode the surrounding structures, or are a source of persistent infection and systemic embolism. Reoperations for false aneurysms must provide secure aortic repair and eradicate infection. In practice, sternal reentry alone may precipitate fatal hemorrhage or cerebral air embolism if the oscillating saw enters the cavity.
Consequently, operations for retrosternal false aneurysm should begin with femoral arteriovenous cannulation and hypothermic profusion. Circulatory arrest can then be used in the event of major bleeding.
In this report we describe the techniques used for mediastinal false aneurysm repair in 10 consecutive patients operated on by one surgeon (S.W.) between 1993 and 1999. Well-defined surgical strategies can be distilled from this experience.
| Patients and methods |
|---|
|
|
|---|
|
Two patients had received a mechanical valved conduit for aortic root replacement in devastating root disease. The first of these (patient 3) was an insulin-dependent diabetic who had been operated on for viridans streptococcal endocarditis and required a DDD pacemaker after extensive debridement of the right fibrous trigone. The second, a female patient (patient 6), had previously had valve dehiscence after bileaflet aortic valve replacement in active inflammatory aortitis.
The remaining 7 patients had all undergone operation for type A dissection. Five had undergone ascending aortic replacement with conservation of the native aortic valve by resuspension. Gelatin-resorcin-formol glue had been used to reconstitute the dissected aortic layers in 4 of these patients, one of whom had the Ehlers-Danlos syndrome. Two patients had undergone aortic root replacement with a bileaflet valved conduit, one for chronic dissection associated with a regurgitant bicuspid aortic valve (patient 5), and the other for acute dissection, which occurred 3 months after mechanical valve replacement for aortic stenosis (patient 1).
Clinical presentation
The false aneurysms were identified between 2 and 70 months after the last operation (Table 2). Six of the 10 patients had suffered a persistent febrile illness since the last procedure (Fig 1 ), 1 of which presented with two recent episodes of stroke (patient 7). Two had blood cultures that were positive for Corynebacterium organisms.
|
|
|
|
Surgical methods
After general anesthesia, all patients were placed in the supine position on the operating table. Access to the right common femoral artery and vein was obtained, and the patients were anticoagulated with heparin and cannulated before sternotomy. A long 32F venous cannula was advanced until the tip was positioned in the right atrium. Cardiopulmonary bypass was then established with a flow of 2.2 L·min-1·m-2. Systolic blood pressure was pharmacologically adjusted to less than 60 mm Hg. A neuroprotective cocktail of the calcium-channel blocker, nimodipine (1 mg/h), thiopentone (6 mg/kg loading dose, then 6 mg·kg-1·h-1), and mannitol (500 mg/kg) was administered pending total circulatory arrest. In 9 patients, sternal reentry was undertaken at between 20°C and 22°C. In all but one case, the heart remained in sinus rhythm during the cooling period so that aortic regurgitation did not cause left ventricular distension. In 1 patient in whom the computed tomographic scan showed the aneurysm to be situated more than 2 cm from the midsternotomy, reentry was performed uneventfully at 28°C.
In 4 patients (including all 3 with active mediastinal infection), the sternal reentry opened the aneurysm causing profuse hemorrhage. Perfusion flow was then reduced or stopped with the head tilted downward to prevent embolism. Cardiotomy suction was continued as the sternal edges were separated carefully. This allowed the borders of the aneurysm to be dissected from the posterior sternal table and the native aorta to be identified distally.
The first objective was to circumscribe the distal aorta just before the innominate artery and apply a cross-clamp so that cardiopulmonary bypass could be recommenced.
Uneventful sternal reentry without entering the aneurysm (six cases) was followed by developing access to both the right atrium and distal ascending aorta. In the event of limited venous return, a second venous cannula was inserted directly into the right atrium to achieve full flow. If feasible, the left ventricle was vented through the apex or the superior pulmonary vein, depending on ease of access. When deep hypothermia (between 16°C and 18°C) was established, the left heart vent was switched off, the pump flow was reduced, and the aneurysm was opened. When direct suture closure of the fistula seemed appropriate, the pump was switched off and repair was performed during hypothermic circulatory arrest (4 patients). The method of repair was individually tailored in each patient. All infected tissue and prosthetic material were radically excised, the roof of the aneurysm cavity was removed, and the cavity was thoroughly debrided. In 3 patients with extensive aortic root and mediastinal infection (patients 2, 4, and 10), aortic root replacement was undertaken with a fresh antibiotic-soaked aortic homograft. Homograft aortic root replacement was also performed in the patient with bioprosthetic aortic valve detachment after previous endocarditis (patient 9). In each case, the coronary ostia were mobilized out of the surrounding aorta for direct reimplantation into the aortic homograft. The inflow suture line was undertaken with interrupted monofilament sutures of 3-0 polypropylene, and the coronary buttons were reimplanted with continuous 5-0 polypropylene. In another patient (patient 10) with an infected graft but normal native aortic valve, the ascending aorta was replaced using an aortic homograft in which the coronary arteries were ligated and the aortic cusps excised. The native aortic valve was scalloped and reimplanted within the homograft conduit. On two occasions, a right coronary ostium immediately adjacent to an infected Dacron graft had to be sacrificed to achieve a secure inflow suture line. A saphenous vein graft was interposed to the distal right coronary artery.
In 1 patient (patient 7), a partially dehisced proximal and distal aortic anastomosis (after previous type A dissection repair) was taken down and revised by adding a short length of Dacron conduit to the existing vascular graft. This method is only applicable in the absence of infection. The other 5 noninfected patients underwent direct suture closure of the entry site into the aneurysm cavity. One had a saphenous vein bypass graft to the right coronary artery for a concomitant coronary ostial lesion.
On completion of the repair, cardiopulmonary bypass was reestablished, and thorough air removal was undertaken. In each case, cardiopulmonary bypass was discontinued without difficulty. The sternum was closed directly with interrupted wire sutures. All excised material, including Dacron and aneurysm wall, were sent for microbiologic screening. Culture of the Dacron graft from 2 septic patients grew the same organism (Corynebacterium), identified by blood culture. One patient who had a negative blood culture preoperatively grew Staphylococcus epidermidis from the excised graft material. In each infected patient, a 6-week course of intravenous antibiotics appropriate for the organism was commenced.
| Results |
|---|
|
|
|---|
The 62-year-old insulin-dependent diabetic (patient 3), who had undergone direct suture repair of the entry to his false aneurysm (12 months after aortic root replacement), had enterococcal mediastinitis. Culture of the excised aneurysm wall had not provided any evidence for preexisting infection. He had sternal instability and discharge from the wound, after which computed tomographic scan showed a retrosternal mediastinal abscess and partial sternal dehiscence. On the 14th postoperative day, the sternotomy wound was reexplored for debridement and drainage. A continuous mediastinal irrigation system was established, and intravenous vancomycin (2 g/d) and gentamicin (150 mg/d) were prescribed. The infection abated with this regimen, with disappearance of the inflammatory markers. He left the hospital and remains well.
One patient experienced focal motor seizures after homograft aortic root replacement for an infected vascular prosthesis. The excised prosthesis was covered with mobile vegetations within the conduit and multiple small low-density defects were detected on brain computed tomographic scan. The neurologic signs resolved with time, and the patient survived.
The most remarkable recovery was in the young patient with Ehlers-Danlos who presented with a massive mediastinal false aneurysm and separate abscess discharging through the sternum. He made a completely uneventful recovery after aortic root replacement and right coronary artery bypass and is completely well 3 years later.
Late events
One patient with aortic dissection who underwent local repair of a proximal aortic graft dehiscence (patient 8) had a separate dehiscence and free rupture of the proximal suture line while at home 3 weeks postoperatively. Autopsy showed the false aneurysm repair to be intact, but there was aortic necrosis within the root after previous gelatin-resorcin-formol glue repair.
The female patient with aortitis (patient 6), who had first undergone isolated aortic valve replacement, then root replacement, and then false aneurysm repair as a third operation, represented again 3 months after hospital discharge with a subcutaneous pulsatile mass in the midline. A fourth operation was undertaken, during which the valved conduit was found to be detached again from the native aortic annulus. All prosthetic material was excised and replaced with an aortic homograft. Once again, the patient recovered without complications and is well 12 months later. Eight patients remain alive and well after a mean follow-up of 2 years (range, 0.6 to 5 years). All patients who received an aortic homograft for infected mediastinal false aneurysms remain well (patients 2, 4, and 10), active, and at work (New York Heart Association class 1), without recurrence of infection.
| Comment |
|---|
|
|
|---|
Reported surgical experience with mediastinal false aneurysm is limited [912]. The principles of extra anatomic bypass and excision of an infected graft currently applied in abdominal aortic sepsis are clearly not suitable for mediastinal graft infection. Instead, we use radical excision of infected material and aortic replacement with antibiotic-treated aortic homografts [5, 8, 13].
In 1 patient, we preserved the native aortic valve and resuspended this in an appropriately sized homograft aortic root [5]. This technique provided the advantage of an autogenous nonantigenic valve within a homograft tube that will eventually calcify. As in aortic root sepsis, the homograft proved resistant to active infection [14], and we had no secondary infection in these patients. Consequently, we believe that for mediastinal graft infection, the advantages of homograft repair outweigh the shorter structural durability as compared with prosthetic valve conduits. In contrast, Coselli and colleagues [15] have successfully treated graft infection with a combination of surgical drainage, intravenous antibiotics, and adjuncts such as viable omentum and muscle flaps. Their experience with extended follow-up to 6.5 years suggests that graft infection caused by organisms sensitive to selected antibiotics can be controlled by high-dose intravenous therapy, followed by indefinite oral treatment after mediastinitis has abated. In their experience, excision of the graft was necessary only if infection was resistant to antibiotics, if there was residual aortic insufficiency or aortocardiac fistula, or if the original graft had inadequately replaced the native aortic disease. Patients with simple false aneurysms in the series by Coselli and associates [15] were treated by local debridement and direct closure using simple suture or insertion of a patch graft rather than replacement of the whole prosthesis. Our limited experience suggests caution with local repair, particularly after aortic dissection or aortitis. We experienced recurrent false aneurysm formation and fatal aortic rupture in patients who underwent conservative local repair. In these patients, the suture-holding capacity of the native aorta was insufficient to prevent an adverse outcome. In retrospect, both would have benefited from the more daunting procedure of radical root replacement.
Despite our policy of native aortic valve conservation and glue repair in acute type A dissection (with an overall hospital mortality of 7%) we have certain reservations about the effects of the polymerizing formalin on the tissues [16]. In 3 of our patients after type A dissection repair, there was fragmentation of the native aorta adjacent to the dehisced suture line. We suggest that excessive use of the formaldehyde solution proved toxic, causing tissue necrosis and the propensity for late anastomotic dehiscence [17, 18]. In contrast, judicious use of gelatin-resorcin-formol glue has greatly reduced the hospital mortality for dissection repair and has increased the scope for native valve preservation in this condition [17]. It remains to be seen whether biologic glues without formalin polymerization can provide competitive results.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Emaminia, A. A. Amirghofran, M. Shafa, A. Moaref, and R. Javan Ascending aortic pseudoaneurysm after aortic valve replacement: Watch the tip of the cardioplegia cannula! J. Thorac. Cardiovasc. Surg., May 1, 2009; 137(5): 1285 - 1286. [Full Text] [PDF] |
||||
![]() |
P. Salinas, T. Lopez, A. Gonzalez, L. Pena-Conde, R. Moreno, M. Moreno, and J. L. Lopez-Sendon Rupture of a thoracic aorta pseudoaneurysm: rare presentation and role of real-time 3D transoesophageal echocardiography Eur J Echocardiogr, May 1, 2009; 10(3): 473 - 475. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Rispoli, G. Varetto, F. M. Savia, and M. Rinaldi Large post-stenting innominate artery pseudoaneurysm Interactive CardioVascular and Thoracic Surgery, June 1, 2008; 7(3): 444 - 446. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Gabbieri, P. M Dohmen, J. Linneweber, A. Lembcke, C. von Heymann, and W. F Konertz Mycotic Pseudoaneurysm of the Ascending Aorta at Site of Aortic Cannulation Asian Cardiovasc Thorac Ann, April 1, 2008; 16(2): e15 - e17. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Watanuki, H. Ogino, H. Matsuda, K. Minatoya, H. Sasaki, T. Fukuda, and S. Kitamura Dilatation of the Aneurysmal Sac After Total Arch Replacement Ann. Thorac. Surg., February 1, 2008; 85(2): 639 - 641. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Coppola, R. Bonifazi, M. Gucciardo, and P. Pantaleo Ruptured aortic arch aneurysm: transposition of aortic arch branches after insertion of thoracic endovascular stent with extra-anatomic brain perfusion Interactive CardioVascular and Thoracic Surgery, June 1, 2007; 6(3): 376 - 378. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Bachet, M. Pirotte, F. Laborde, and D. Guilmet Reoperation for Giant False Aneurysm of the Thoracic Aorta: How to Reenter the Chest? Ann. Thorac. Surg., May 1, 2007; 83(5): 1610 - 1614. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. F. Coutinho and M. J. Antunes Pseudoaneurysm of an Aortic Homograft Ann. Thorac. Surg., December 1, 2006; 82(6): 2280 - 2282. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. A. Atik, J. L. Navia, L. G. Svensson, P. R. Vega, J. Feng, M. E. Brizzio, A. M. Gillinov, B. G. Pettersson, E. H. Blackstone, and B. W. Lytle Surgical treatment of pseudoaneurysm of the thoracic aorta. J. Thorac. Cardiovasc. Surg., August 1, 2006; 132(2): 379 - 385. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Villavicencio, T. A. Orszulak, T. M. Sundt III, R. C. Daly, J. A. Dearani, C. G.A. McGregor, C. J. Mullany, F. J. Puga, K. J. Zehr, and H. V. Schaff Thoracic aorta false aneurysm: what surgical strategy should be recommended? Ann. Thorac. Surg., July 1, 2006; 82(1): 81 - 89. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Maselli, G. Santise, A. Montalto, and F. Musumeci Endovascular Aortic Clamping for Pseudoaneurysms of the Aortic Root With Aortic Regurgitation Ann. Thorac. Surg., October 1, 2005; 80(4): 1303 - 1308. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Okubo, J. Isobe, J. Kitamura, and Y. Ueno Mediastinitis and pseudoaneurysm of brachiocephalic artery long after the resection of invasive thymoma and postoperative irradiation J. Thorac. Cardiovasc. Surg., September 1, 2005; 130(3): 918 - 919. [Full Text] [PDF] |
||||
![]() |
B. Parihar, L. S. D. Choudhary, A. P. Madhu, M. K. Alpha, R. Thankachen, and V. Shukla Pseudoaneurysm of Ascending Aorta After Aortic Valve Replacement Ann. Thorac. Surg., February 1, 2005; 79(2): 705 - 707. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Dumont, M. Carrier, R. Cartier, M. Pellerin, N. Poirier, D. Bouchard, and L. P. Perrault Repair of aortic false aneurysm using deep hypothermia and circulatory arrest Ann. Thorac. Surg., July 1, 2004; 78(1): 117 - 120. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. N. Messner, C. Vatcharasiritham, I. Gregoric, B. Radovancevic, P. Odegaard, S. D. Flamm, and O. H. Frazier Prosthetic graft remnant-related pseudoaneurysm after left ventricular assist device explantation: A case report J. Thorac. Cardiovasc. Surg., January 1, 2004; 127(1): 259 - 261. [Full Text] [PDF] |
||||
![]() |
A. Z. Apaydin, H. Posacioglu, F. Islamoglu, and A. Telli A practical tool to control bleeding during sternal reentry for pseudoaneurysm of the ascending aorta Ann. Thorac. Surg., March 1, 2003; 75(3): 1037 - 1038. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. W. Downing What are the risks of using biologic glues? Ann. Thorac. Surg., March 1, 2003; 75(3): 1063 - 1063. [Full Text] [PDF] |
||||
![]() |
K. Tanaka, H. Makuuchi, Y. Naruse, T. Kobayashi, I. Hayashi, T. Takayama, and Y. Namifusa False Aneurysm Due to Suture Loosening After Aortic Arch Replacement Asian Cardiovasc Thorac Ann, December 1, 2002; 10(4): 346 - 348. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Imanaka, S. Kyo, S. Takamoto, M. Kato, H. Tanabe, H. Ohuchi, H. Asano, and Y. Yokote Periannular abscess and aorta-left ventricular fistula after infection in the false lumen of an aortic dissection J. Thorac. Cardiovasc. Surg., October 1, 2002; 124(4): 841 - 843. [Full Text] [PDF] |
||||
![]() |
Y. Kawachi, A. Nakashima, T. Onzuka, and T. Yamauchi False aneurysm of the ascending aorta concomitant with chronic mediastinitis after tube graft replacement in octogenarian Eur. J. Cardiothorac. Surg., September 1, 2002; 22(3): 450 - 453. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Westaby, S. Saito, and T. Katsumata Acute type A dissection: conservative methods provide consistently low mortality Ann. Thorac. Surg., March 1, 2002; 73(3): 707 - 713. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Carrier, L. P. Perrault, M. Pellerin, R. Marchand, P. Auger, G. B. Pelletier, M. White, N. Racine, and D. Bouchard Sternal wound infection after heart transplantation: incidence and results with aggressive surgical treatment Ann. Thorac. Surg., September 1, 2001; 72(3): 719 - 723. [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 |