| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2007;32:126-130. doi:10.1016/j.ejcts.2007.01.076
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved
Cardiothoracic Surgery, University Hospital, Coimbra, Portugal
Received 21 August 2006; received in revised form 11 January 2007; accepted 23 January 2007.
* Corresponding author. Address: Centro de Cirurgia Cardiotorácica, Hospitais da Universidade, 3000-075 Coimbra, Portugal. Tel.: +351 239 400418; fax: +351 239 829674. (Email: antunes.cct.huc{at}sapo.pt).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Patients and methods3. Results4. DiscussionAppendix AReferences |
|---|
Key Words: Prosthetic endocarditis • Prostheses • Reoperation • Allografts
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix AReferences |
|---|
Aortic PVE is more common than mitral PVE. When the disease is not controlled early, the infection extends beyond the prosthesis and involves the annulus and other neighbouring structures, such as the conduction system and the mitral valve. Abscess formation occurs and may become circumferential, resulting in partial or complete ventriculo-aortic endocardial discontinuity, which requires special technical steps besides simple replacement of the prosthesis.
The now generally accepted surgical strategy to treat aortic PVE accompanied by abscesses consists of radical debridement of infected tissue, to obtain a margin of healthy tissue in order to avoid recurrent or residual infection [5,6]. Several different techniques have been proposed, which include reconstruction of the annulus and replacement with another prosthetic valve, composite prosthetic valve conduit, Ross operation and aortic allograft implantation [7]. Aortic allografts are known to have excellent haemodynamic performance associated with a very low incidence of reinfection and late mortality [8].
In this retrospective study, we analysed the peri-operative and long-term outcome of complex PVE treated by aortic root replacement with a complete fresh valved aortic allograft.
|
2. Patients and methods |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix AReferences |
|---|
The allografts were collected sterile from multiorgan cadaveric donors or from the hearts removed from transplant patients. In cadavers, the cardiectomy and harvesting of the valves was performed within an average of 3 h after death and the dissection of the homografts was completed in the next 90 min. After the beginning of our transplantation program, the dissection of the homografts was performed in the same operating room and concluded in an median time of 60 min after the cardiectomy and preserved in a tissue culture media (TC Medium-199), with a small amount of Vancomycin (25 mg/l of TC) to prevent infection, at 4 °C, for a maximum of 8 weeks post harvesting. The mean time between harvesting of the allografts and the implantation was of 22.8 ± 17.3 days (1–55 days).
We have not used cryopreserved allografts because we do not have them in our institution, and our relatively peripheral geographic position in Europe makes it difficult to have easy access to Central European tissue banks. We reserve our aortic allografts for PVE, but have also used them in 35 cases of complex native endocarditis treated during this period. Some patients treated by direct implantation of biological or mechanical prosthesis, in the beginning of the period, were not included in this study.
The mean age of the group was 51.5 ± 13.7 (range 26–72) years and 37 (90.2%) patients were male. Four patients had a stented bioprosthesis in place and the remaining 37 (90.2%) had a mechanical aortic valve. Five (12.2%) had two previous aortic valve replacements.
Twenty-six patients (63.4%) had signs and symptoms of heart failure when presenting for surgery, six (14.6%) were referred with angina pectoris and nine (22%) were asymptomatic. We had one patient in whom the diagnosis of endocarditis was made because of a cerebral embolism. In 10 patients (24.4%), the surgery was performed on an emergency basis because of haemodynamic instability. Four patients had chronic renal failure and were on dialysis before surgery.
Although active infection was diagnosed in more than three-quarters of the cases, only 16 patients (39.0%) had either preoperative (blood) or postoperative (surgical specimen) positive cultures. The most common agent was the Staphylococcus (Table 1 ).
|
Twenty-eight patients (68.3%) had early endocarditis, defined as infection within 12 months after surgery. The mean interval between aortic valve replacement and reoperation for PVE was 26.8 ± 45.4 months (range: 1–241 months).
2.1 Operative technique
All operations were performed through a median sternotomy under moderate hypothermic (26–28 °C) cardiopulmonary bypass. After re-opening of the sternum, the pericardial adhesions over the atria and the ascending aorta were sharply dissected. The LV was vented through the right superior pulmonary vein and left atrium. The aortic root was opened and, under cold crystalloid cardioplegic cardiac arrest, the infected prosthesis was excised. All evidently infected tissues were debrided, but care was taken to preserve non-infected healed tissues, to avoid further damage of the neighbouring structures, and the final anatomical situation was evaluated. Fresh aortic allografts were used in all cases and implanted by a complete aortic root replacement technique (modified Bental-De Bono).
The allografts were implanted in the natural position to facilitate coronary reimplantation. The proximal anastomoses were constructed with interrupted sutures of 4/0 braided polyester in all but a few early cases where a continuous polypropylene suture was used. In some cases a strip of bovine pericardium was incorporated in the suture line for reinforcement and haemostasis. This suture line was usually positioned in the upper endocardial margin of the left ventricular outflow tract, in such a way that the root abscesses were excluded from the systemic circulation.
The coronary ostia had previously been mobilized as buttons of the aortic wall and the left coronary was first sewn to the allograft with a 6/0 polypropylene running suture. The distal aortic wall suture was then made with a continuous 4/0 polypropylene. If the anastomosis was in the distal ascending aorta, close to the arch, it was done in an open manner, during a very short period of circulatory arrest, after which the aorta was unclamped.
The right coronary ostium was last sutured to the graft while rewarming, during a new and very short period of aortic clamping. With this method, the authors believe to have minimized the myocardial ischaemic time and to be more precise in identifying the exact spot to anastomose the right coronary ostium, avoiding kinking of the artery.
Additional surgery was performed in three patients: mitral valve repair, coronary artery bypass grafting (CABG) and mitral repair with CABG, one case each.
Cardiopulmonary bypass time was of 107.8 ± 48.1 min and aortic cross clamping time was of 69.4 ± 17.4 min.
Antibiotic treatment was initiated as soon as the diagnosis of PVE was made (mean 28.6 ± 12.9 days) and continued for at least 4–6 weeks after surgery, depending on the duration of preoperative treatment and clinical response.
2.2 Follow-up and analysis
Follow-up was done through mail or telephone contact and was complete for all patients. Mean follow-up time was 54.3 ± 43.2 (range 2–166) months. Death or allograft failure requiring reoperation were the end points for calculation of survival. Continuous data was expressed as mean ± standard deviation.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix AReferences |
|---|
The majority of patients were extubated within 12 h of surgery. Inotropic support (>24 h or multi-drug) was required in six cases (14.6%) and one patient had intraaortic counterpulsation. Nine patients (21.9%) had acute renal failure (creatinine > 2 g/dl) due to antibiotic toxicity, but none necessitated haemodialysis. Complete heart block requiring a permanent pacemaker occurred in three patients (7.3%) (Table 2 ).
|
3.2 Late results
Two patients had late allograft failure due to degeneration and calcification of the valve and wall, respectively, at 61 and 82 months after implantation. Both were reoperated and subjected to aortic valve replacement only, without complications or mortality.
Eight patients (19.5%) died late. The cause of death could not be determined in two patients, two died of cardiac reasons (sudden death-one; AMI-one) and four died of non-cardiac causes (stroke-one; cholecystitis-one; traffic accident-one). Ten-year cumulative survival was 79% (Fig. 1 ).
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix AReferences |
|---|
But operation for PVE remains a surgical challenge, with some early series reporting mortality rates of 20–60% [1,10], although more recent reports indicate mortalities around 10%. In a more recent work, Sabik et al. [8] reported an in-hospital mortality of 3.9%. Additionally, there is a high rate of recurrence of infection when a new prosthesis is implanted. This has drawn attention to the use of allografts, which are considered to have the lowest incidence of postoperative recurrence of infection, when compared with other valve substitutes [11,12]. Besides this intrinsic resistance to infection, allografts have excellent haemodynamic properties and are associated to low rates of thromboembolism [13]. These characteristics may be related to a variety of factors, but the most important is probably the absence of synthetic material [14]. In the beginning of the period included in our study, we treated most patients with periannular abscesses by substitution of the infected prostheses by a new biological or mechanical prosthesis. The incidence of re-infection was high, many patients having required multiple re-interventions. Although these eventually resulted in cure of the infection in almost every patient, this experience led to our adoption of the allograft for treatment of this complex pathology, which is now our routine practice.
Aortic allograft implantation allows for an early surgical intervention, since the graft moulds well to the distorted infected annulus, even when the aortic root is destroyed. The early surgical approach is important to prevent secondary complications and contributed to the low rate of perioperative morbidity and mortality obtained in this study. Calderwood et al. [15] demonstrated that patients whose surgical treatment was postponed in their initial hospital admission had a high risk of progressive prosthetic dysfunction and worse prognosis. Nonetheless, provided that haemodynamic stability can be achieved, emergency or urgent surgery is usually not required and a period of preoperative antibiotic therapy may help reduce the rate of re-infection. Whenever possible, we attempt to do exactly this.
The use of the pulmonary autograft (Ross operation) has been proposed as an alternative option for the surgical treatment of the complex PVE, especially in young patients. Joyce et al. [16], in 1994, were the first to report a small series of patients who had prosthetic valve endocarditis treated by pulmonary autograft replacement, with no mortality and no recurrence of infection. However, after an initial enthusiasm with the Ross technique, some authors found that this procedure increased operating time, exposed the patient to a higher surgical risk and is not totally free from late complications, such as pulmonary allograft calcification and autograft dilatation [17]. Currently, there is no evidence of superiority of the Ross procedure over the simpler aortic root replacement by allograft.
Recurrence of infection is a major problem in these cases. Other groups that use the same approach attribute the recurrence of infection, in the vast majority, to intravenous drug abuse [8]. In this series, we had no recurrence of endocarditis during a follow-up period of up to 10 years. We believe that these results are due not only to the radical surgical approach, with extended debridement and resection of all infected valvular and paravalvular tissues, but also to the pre and postoperative antibiotic therapy policy discussed above. Considering this radical approach, all cases required allograft implantation as a full root replacement, which also allows for the correction of significant anatomical distortions. Some authors have advocated subcoronary implantation but this technique is mostly unsuited for these cases, because of the significant root disruption. Another potential advantage of the root replacement technique is the possibility of using the attached anterior mitral leaflet as a patch to restore the continuity of the tissues and repair intra-cavitary fistulae. Dossche et al. [11] considered that even if concomitant mitral valve replacement becomes necessary, aortic allograft remains the technique of choice.
Despite these advantages of the use of allografts in PVE, it is well known that they have a limited durability, especially in young patients in whom degeneration and calcification occur much earlier [9]. However, Yacoub et al. [18] still considered that the allograft remains the first choice, even in the young. In the present study, two patients required reoperation because of allograft failure due to degeneration and/or calcification of the valve and wall. The short intervals to degeneration may be related to the use of fresh allografts but they have also been observed in some patients with cryopreserved homografts, and fresh allografts have been used by others with comparable results [18]. Both patients had a direct valve replacement with a mechanical prosthesis and recovered uneventfully.
We believe that the low hospital mortality and infection recurrence justify the use of allograft, even when a reoperation is to be expected due to graft degeneration. Yacoub et al. [18] also demonstrated that, despite being technically more complex, replacement of an allograft carried a low perioperative mortality risk and was associated with a good long-term outcome.
Several types of morbidity, such as permanent pacemaker implantation for complete atrio-ventricular block (AVB), acute renal failure, respiratory complications and multiple organ dysfunction are frequently observed in the early postoperative period [14,19,20]. AVB is a common complication following radical surgery of the aortic annulus, especially when the interventricular membranous septum is involved, and may have a negative effect on survival. In this series, we observed an incidence of AVB of 7.7%; however, all patients had a satisfactory outcome after pacemaker implantation.
In our series, the cumulative survival was 79% after a follow-up of up to 10 years, which compares very well with those reported by other authors, which present a 5-year and 10-year actuarial survival of 85% and 56%, respectively [6,8]. This relevant result may also be explained by the low hospital mortality.
In conclusion, although this is a small series, we only found a larger number of patients in the series of Sabik et al. [8]. Other series with more patients could be found in the literature, but include both native and prosthetic endocarditis, many not relating to the use of allografts. The results of this study suggest that complex prosthetic valve endocarditis is optimally treated by aortic root replacement with an aortic allograft, resulting in a low hospital and late mortality, and is associated with a low incidence of late complications, with special relevance for a zero incidence of re-infection. In our experience, these results were superior to those obtained with other valvular substitutes.
|
Appendix A |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix AReferences |
|---|
Dr M. Irarrazaval (Santiago, Chile): Could you tell us something about the bacteriology here and your long-term antibiotic treatment?
Dr Calvinho: Unfortunately, and it is a thing that is written many times in many papers, only 40% of the patients had positive blood cultures. In these 40% of the patients, we found that staphylococcus was the most frequent micro-organism that we could find. Naturally, it has no statistical value because we only could find that in 40 patients. There are many reasons why it happened, but those are our results.
Dr Irarrazaval: How long did you keep them on antibiotics?
Dr Calvinho: We keep them on antibiotics before surgery for four weeks to try to stabilize them, and after surgery we keep them on 4–6 weeks also.
Dr R. Dion (Leiden, The Netherlands): I am extremely impressed by the low mortality and also the short clamping time. So my question is, how many of these patients had extra-aortic infection, for instance requesting the resection of the roof of the left atrium or a part of the septum?
Dr Calvinho: By definition, all these cases were complicated and had extra-aortic abscesses. We had two patients with aortic ventricular fistula that we put a part of the pericardium in.
Dr Dion: You conclude that the allograft is probably better than the other valve substitutes. We have in Leiden abandoned the use of allografts in this setup, and we use stentless grafts, which have the advantage of readily being at disposal in every size on the shelf. In 7 years’ time we didn’t have one recurrence of endocarditis out of 30 patients with periaortic involvement. And the group of Tirone David has shown that the type of valve substitute is not really important: what is important is the resection of all infected tissue. So do you really want to carry on with allografts knowing that you had already two failures?
Dr Calvinho: I think for now we will probably continue using the homografts. We don’t have the cryopreserved homografts, but the bank of fresh homografts is regulated by us, so we have free access to the homografts.
Dr O. Wendler (London, UK): I wanted to go into the same direction as the previous speaker and would like to make a comment. I mean, if you look at other groups from Zurich or from Berlin, the long-term result, not due to infection because their reinfection rate is always low, but due to failure of the homograft was always relatively high, particularly in the endocarditis group. The Berlin group has shown very nicely, comparing their results between homografts and stentless valves, that obviously the reoperation rate in the stentless group was significantly lower, around 3–4% for the stentless valves, compared to the homografts, which is 15%.
I think the stentless valves should be rated higher in this condition, and I just said that your results are somehow different from other groups, for example, groups from Zurich, groups from Berlin, with respect to reoperation on these homografts in the next 5, 10 years after surgery, not due to reinfection, reinfection is always low, but due to graft failure, graft failure if you think about aortic regurgitation in the postoperative course.
Dr Calvinho: Our graft failures occurred late and were due to calcification and those two patients had undergone valve replacement with a mechanical prosthesis, both of them. Homografts do fail, but there is no proof that heterografts (stentless) do better.
Dr P. Kappetein (Rotterdam, The Netherlands): The remark is that the homografts fail. After 10 years there is quite a significant amount of patients that have to be reoperated. That was also shown yesterday by Dr Takkenberg. That is a concern, I think, and stentless valves may perform better.
Dr A. Abdelghany (Cairo, Egypt): My question is about the preoperative status of your patients. You said that most of them were in Class III New York Heart Classification. Are your patients a random study? Is this study a random study or a selective study? Of course, most of the aortic patients with endocarditis have cardiogenic shock. A highly significant number of them present to our hospital with cardiogenic shock. And you said that most of your patients are in Class III New York Heart Classification.
Dr Calvinho: As I told you, we operated on 10 patients on an emergency basis. So that means that they were in hemodynamic collapse. Most of them were in New York Class IV and III, as you saw by the median. So they were pretty ill patients. We try to improve their status before surgery in a way that the surgery can be the best, and naturally if we have to intervene when the patient is in cardiogenic shock or haemodynamic collapse, we do it. So you saw the results also.
Dr Abdelgany: My second question, how long do you extend your postoperative antibiotic therapy?
Dr Calvinho: I already said that. Four to six weeks.
Dr P. Kappetein: We can agree that it is a complex patient group, but whether to implant a homograft, a stentless valve or even a stented valve is not solved yet. We have to wait for a randomized study, I think.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Patients and methods3. Results4. DiscussionAppendix AReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Endorsed by the European Society of Clinical Micro, Authors/Task Force Members, G. Habib, B. Hoen, P. Tornos, F. Thuny, B. Prendergast, I. Vilacosta, P. Moreillon, M. de Jesus Antunes, et al. Guidelines on the prevention, diagnosis, and treatment of infective endocarditis (new version 2009): The Task Force on the Prevention, Diagnosis, and Treatment of Infective Endocarditis of the European Society of Cardiology (ESC) Eur. Heart J., October 1, 2009; 30(19): 2369 - 2413. [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Mestres Invited Commentary Ann. Thorac. Surg., May 1, 2008; 85(5): 1569 - 1570. [Full Text] [PDF]
|
|
|
|
|
|
K. Okada, H. Tanaka, H. Takahashi, N. Morimoto, H. Munakata, M. Asano, M. Matsumori, Y. Kawanishi, K. Nakagiri, and Y. Okita Aortic Root Replacement for Destructive Aortic Valve Endocarditis with Left Ventricular-Aortic Discontinuity Ann. Thorac. Surg., March 1, 2008; 85(3): 940 - 945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Mestres and J. M. Miro Editorial comment: Allograft aortic root replacement in complex prosthetic valve endocarditis Eur. J. Cardiothorac. Surg., July 1, 2007; 32(1): 131 - 132. [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 |
