| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;84:737-744
© 2007 The Society of Thoracic Surgeons
Division of Cardiovascular Surgery, Toronto General Hospital, and Department of Surgery, University of Toronto, Toronto, Ontario, Canada
Accepted for publication April 16, 2007.
* Address correspondence to Dr David, Division of Cardiovascular Surgery, Toronto General Hospital, Rm 4N-457, 200 Elizabeth St, Toronto, Ontario, M5G 2C4, Canada (Email: tirone.david{at}uhn.on.ca).
Presented at the Forty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 29–31, 2007.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: All patients with stentless valves undergoing redo aortic valve replacement (AVR) at our institution were examined (n = 57). Ten patients had a prior Freestyle valve (Medtronic, Minneapolis, MN), and 47 patients had a Toronto stentless porcine valve (SPV; St. Jude Medical St Paul, MN).
Results: Redo AVR was performed 8.4 ± 3.7 years (range, 0.1 to 16.5 years) after stentless valve implantation. Reoperations were elective in 27 patients (49%), and 30 (51%) underwent urgent or emergency procedures. The indication for redo AVR was structural valve dysfunction in 48 patients (84%), acute endocarditis in 7 (12%), and other in 2 (4%). Aortic insufficiency was present in 47 patients (82%). A total of 36 aortic root replacement operations (63% of patients) were required, of which 19 were secondary to severe adhesions between the stentless valve and the native aortic root. Operative mortality was 11% (n = 6) for the entire group. Mortality was higher in patients undergoing redo AVR less than 1 year after stentless valve implantation versus more than 1 year (67% versus 7%, p = 0.03). Long-term survival at 5 years postoperatively was 79% ± 7% in all patients, and 81% of survivors were in New York Heart Association class I or II.
Conclusions: Reoperation after stentless AVR is a challenging procedure that frequently requires aortic root replacement. Stentless valve reoperation is associated with an increased risk of death, particularly in patients operated on within 1 year of implantation.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The number of implanted stentless xenografts has increased during the past decade. The increased implantation rate is likely due to improved hemodynamic performance and favorable rates of morbidity and mortality compared with stented bioprostheses [5, 6]. The stentless porcine valves that were developed earlier, in particular the Toronto SPV (St. Jude, St. Paul, MN) and Freestyle (Medtronic Inc, Minneapolis, MN) bioprostheses, have been implanted for 10 to 15 years and are starting to reach the limits of their durability. It is therefore very likely that we will observe an increase in the number of stentless valve reoperations within the next decade. Because relatively little has been published on stentless valve reoperation in the literature, we undertook the current study to examine our results with this challenging set of patients.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Follow-up was obtained in 100% of patients, by telephone or mail questionnaire, or both, followed by contact with the patient’s cardiologist or family physician. Length of follow-up after the reoperation was 2.7 ± 2.5 years (range, 0 to 11.6 years).
Operative Considerations
Access to the mediastinum was through a median sternotomy in all patients, using an oscillating saw. The femoral vessels were not exposed before sternotomy, except in 1 patient with previous mediastinitis. The distal ascending aorta or aortic arch was used for arterial cannulation, and the right atrium was cannulated for venous return. The left heart was vented through the right superior pulmonary vein.
The hematocrit was maintained between 20% and 25% during cardiopulmonary bypass (CPB), pump flow rates between 2.0 and 2.5 L/(min · m2), and mean arterial pressure between 60 and 80 mm Hg. Systemic body temperature was allowed to drift to 34°C, with active rewarming at the end of CPB. Antegrade cold blood cardioplegia was delivered directly into the coronary ostia in all patients, except those with diseased coronary bypass grafts, in whom we used retrograde cardioplegia [7].
A transverse aortotomy was performed 1 to 2 cm above the right coronary ostium. The stentless valve was carefully inspected and then removed. Every attempt was made to preserve the native aortic root wall and avoid total root replacement. In some patients, removal of the stentless valve resulted in a defect in the native aortic root requiring patch repair with autologous or bovine pericardium.
Aortic root replacement was necessary in patients with stentless valves that were severely calcified or severely adherent to the native aortic wall, resulting in large defects or completely absent native aortic root tissue after valve resection. In some patients, resection of the stentless prosthesis also resulted in a lack of suitable annular tissue for root replacement. We therefore anastomosed a suitably sized Dacron (DuPont, Wilmington, DE) graft to the left ventricular outflow tract (muscular septum and anterior leaflet of mitral valve), and then implanted a prosthetic valve within the Dacron graft 5 to 8 mm above the proximal suture line and secured it with running or interrupted sutures with the knots on the outside of the graft. The coronary buttons were anastomosed to the Dacron graft in such patients before or after prosthetic valve implantation using standard techniques.
In patients with prosthetic valve endocarditis, aggressive removal of all infected material and affected surrounding tissue was performed to minimize the chance of recurrence [8]. Aortic root replacement was therefore required in all such patients, and reconstruction of the base of the heart with a Dacron patch or bovine pericardium was also frequently required.
A short interposition saphenous vein graft to the coronary buttons was performed if the coronary arteries could not reach the new aortic root without tension or if the proximal coronary artery was involved in a perivalvular abscess.
Statistical Analysis
Categoric variables are expressed as percentages, and continuous variables are expressed as mean ± standard deviation. Standard definitions were used for patient variables and outcomes. In particular, operative mortality was defined as death during the same hospitalization. All statistical analyses were performed with the SAS 8.1 software (SAS Institute, Cary, NC). Comparison of categoric variables was performed with Fisher exact tests, and continuous variables were analyzed with unpaired t tests or the Wilcoxon rank sum test where appropriate. Multivariable logistic regression analysis with backwards, stepwise elimination was used to determine the independent predictors of operative mortality. Kaplan-Meier methods were used to analyze long-term survival. Statistical significance was defined as p lt; 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
|
Postoperative outcomes are listed in Table 3. Six patients (11%) died perioperatively:
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The current generation of stentless valves have been implanted since the early 1990s and are therefore starting to reach the limits of their durability. The Freestyle valve is showing very good durability, with 96% freedom from structural valve deterioration (SVD) 10 years postoperatively [11, 12]. Some investigators have demonstrated higher rates of SVD for the Toronto SPV, particularly Desai and colleagues [13], who found a 78% freedom from SVD 10 years postoperatively. In contrast, our group found a 99% 10-year freedom from SVD in patients older than 65 years and 77% in patients younger than 65 [14].
The improved hemodynamic performance of stentless aortic valves during exercise makes them a particularly attractive option in young, active patients [15]. However, implantation of stentless valves in younger patients will increase the chance that a reoperation will be required some time in the future. It is without question that more and more stentless valve reoperations will be required in the near future as these prostheses reach the limits of their durability. Because our center has a long and large clinical experience with stentless valves (more than 700 implants during a 15-year period), we undertook the current study to examine our results with reoperation after stentless valve implantation.
The operative mortality rate in the current study was 11%. Mortality was considerably higher in those patients who received a previous Freestyle valve. However, 2 of the 4 Freestyle patients who died had acute endocarditis with a large perivalvular abscess less than 1 year after implantation. Perivalvular abscess formation is a well-recognized risk factor for operative mortality because of the complexity of the required curative procedure [8]. Reoperation within 1 year of valve implantation is also a risk factor because of extensive pericardial adhesions that are frequently present. Indeed, operation within 1 year of implantation was the only independent predictor of perioperative mortality in the current series.
Although the mortality rate in the current study is acceptable given the complexity of the operations, it is notably higher than previous studies of redo AVR published from our institution. A recent study of the 216 patients who underwent redo AVR during a 12- year period (88% of which were stented or mechanical prostheses) revealed an operative mortality rate of 4.6%, which was not significantly different from first-time AVR procedure [16]. Potter and colleagues [17] also demonstrated a low 4.9% rate of operative mortality for redo AVR that was not significantly different from 3.1% for primary valve replacement [17]. We found a very low mortality rate of 2.4% in 82 patients undergoing reoperative aortic root replacement in the above-mentioned study [16]. In another study, we examined 31 patients undergoing aortic root replacement after a previous root replacement procedure and found a very low mortality rate of 3% [18]. It would therefore appear that the observed 11% mortality rate for the current study is significantly higher than our experience with redo AVR after implantation of other valve prostheses.
The reasons for increased mortality after stentless valve implantation are difficult to precisely identify given our limited sample size; however, it is our clinical impression that the redo stentless valve operation is a complex procedure. Trauma to the coronary ostia, aortic wall, aortic annulus, anterior mitral valve, and membranous septum can all occur when severe adhesions or calcification are present between the stentless valve and the native aortic root. In contrast, removal of a stented tissue or mechanical valve can usually be accomplished without extensive damage to surrounding tissues.
A recent publication from our institution looked at the pathologic findings in 30 explanted Toronto SPVs [19]. Although calcification of the valve leaflets was relatively uncommon, a significant proportion (40%) showed calcification of the porcine aortic wall or porcine muscle shelf. In addition, pannus and fibrosis was present on the exterior portion of the valve in 70%. Such calcification and pannus formation make dissection and removal of the prosthesis more complicated, thereby substantiating our clinical impressions.
The increased complexity of redo stentless valve surgery can be also confirmed by the following observations from the current series: (1) 5% of patients required a patch repair of the native aortic root after resection of the prosthesis; (2) 63% required a total aortic root replacement procedure, over half of which were for nontraditional indications, including severe calcification and adherence between the stentless valve and the native aortic root; (3) 18% required graft reconstruction of the left ventricular outflow tract because of lack of annular tissue after removal of the stentless valve or because of abscess formation; and (4) 16% of patients required interposition grafts between the new aortic root and the proximal coronary arteries. These latter two procedures, in particular, are technically challenging and are probably best suited for surgeons and centers that perform high volumes of valvular surgery.
Relatively few studies with small sample sizes have examined operative mortality poststentless valve implantation:
Although perioperative mortality for the current study was noticeably higher than previous studies from our institution, the intensive care unit and hospital lengths of stay were relatively short for such a complex cohort of patients (see Table 3). It is also interesting to note that the long-term survival compared favorably with our previously published studies, with a 5-year survival of 79% compared with 71% after redo aortic root replacement [18]. In addition, 81% of surviving patients in the current study had NYHA class I or II symptoms during follow-up. These findings would suggest that patients do quite well after redo stentless valve surgery, provided they survive the complex operation.
The primary limitation of our study is its retrospective nature, and although we had a large group of patients, it is still a relatively small number. In addition, our center has extensive experience with stentless valve implantation and complex valvular surgery. We therefore believe that our clinical experience is extensive enough to comment on the relative complexity of these reoperative procedures.
Another limitation is that we have not determined the valve-specific long-term failure rates in the current study; however, this was purposefully avoided for three reasons. First, we have previously published our experience with long-term failure rates for the SPV and did not want to repeat this information. Second, we do not yet have enough long-term experience with the Freestyle valve to comment on our institution-specific failure rates for this prosthesis. Third, several patients within the current study received their initial stentless valve implantation at outside institutions. We do not know the number of stentless valve implantations at these other institutions; therefore, we cannot accurately determine the valve failure rate for the current study population. However, several other publications already exist on the valve failure rates for the SPV and Freestyle prostheses, and we do not have any reason to believe that our experience is significantly different from the literature.
Reoperation on stentless aortic valves is associated with an increased rate of mortality and an increased need for total aortic root replacement. Although we continue to be supportive of stentless aortic valve prostheses in certain patient subgroups, we believe that reoperation after a stentless valve is more complex than after a stented tissue or mechanical valve. Reoperation after stentless valve implantation may therefore be an operation that is better suited for surgeons and centers that routinely perform complex valvular surgery, particularly when performed in an elective, planned fashion.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
DR BORGER: No, sorry, that’s not correct. Only 3 patients required reoperation within 1 year of implantation, but 2 of them died; thus, a 67% mortality for these patients.
DR DION: Do they die because they develop a pannus, or because of a structural failure?
DR BORGER: These patients that were operated on within one year died because of a complicated operation for prosthetic valve endocarditis with a big perivalvular abscess. By the time you end up removing the entire prosthesis, removing the abscess and infected tissue, reconstructing the intervalvular body, reconstructing the left ventricular outflow tract and creating interposition grafts to the coronary arteries, it is a large and risky operation.
DR DION: You seem to indicate that the most important technical explanation of the mortality is the presence of a pannus, is that correct?
DR BORGER: For those patients without prosthetic valve endocarditis, I would say that extensive pannus formation or extensive calcification is the biggest problem, yes.
DR DION: And you have the impression that it is more in one type of valve than in the other one?
DR BORGER: No, we don’t have that data. We have a lot more SPV reoperations at our center, and it appears that the failure rate for SPV may be higher than the Freestyle. According to the literature, the 10-year freedom from structural valve dysfunction is greater than 95% for the Freestyle valve. Some centers, however, have shown freedom from SVD rates of 80% or less for the SPV, as was seen in the earlier presentation from Dr Desai from Sunnybrook Hospital in Toronto.
We implanted a lot of SPVs at our center in the 1990s, and now we are performing a substantial number of reoperations for SVD. It’s our impression that the Freestyle SVD rates are probably lower, but we don’t have enough data from our center to prove that.
DR ROGER C. MILLAR (Salt Lake City, UT): I take a little bit of concern of saying that stentless valves in the subcoronary position are difficult to take out. I think that is true of the SPV valve, but that’s a first-generation valve that is failing quicker because of no anticalcification. It also is covered with Dacron on the outside, I think, which explains the pannus ingrowth.
I have been implanting the Freestyle valve, over 600 valves since 1993. I have only had the opportunity to take 2 of those particular valves out, and I found them much easier to remove than any stented prosthesis or mechanical prosthesis that we have had. You cut the suture line and you can get in behind that and there is no reaction between the porcine aorta and the native aorta. It looks pristine when you get done.
And we have not seen the pannus ingrowth described in the SPV valves. And some of that may be the amount of Dacron that is on that valve. But I think we have to be careful and not interpret the few Freestyle valves that were included in this series, 4 of which you have pointed out had implantation problems or infection with an abscess.
My question to you is, have you found that the Freestyle valve in the subcoronary position, you find the same amount of pannus in that valve as you have in the SPV valve?
DR BORGER: The good news is that we only have 10 Freestyle valve reoperations to present. Nobody has a large experience with redo Freestyle procedures in the world, and that must mean that the freedom from SVD rates are good. But of the 10 reoperations that we performed, I can tell you from personal experience that they were tricky operations. Particularly if endocarditis was present.
As for pannus formation, of the 10 Freestyles, 6 of them were removed for reasons other than endocarditis. And we have observed pannus formation for these valves, although probably not as extensively as with the SPV. It’s true that there is not as much Dacron on the outside of a Freestyle valve as there is on an SPV, but Dacron is present. From reviewing all of the OR reports for these patients, I can tell you that there were a couple of valves that were difficult to remove.
DR MILLAR: I just would raise a voice of caution that we need more data to condemn all stentless valves in the subcoronary position until we have more data.
DR BORGER: I am not condemning stentless valves. We wrote a paper last year with over 700 stentless valves, which is the largest series to date, showing that the long-term gradients are half of what they are for stented valves and LV mass regression is improved. I am not trying to tell you that stentless valves are bad. The Freestyle is still our prosthesis of choice, both in Toronto and in Leipzig, for patients with aortic root pathology who are over 65 years of age or over 60 years if they have coronary disease or other concomitant life-limiting comorbidities. I think that it is an excellent valve.
The message I’m trying to convey is that these are complex reoperations and you cannot predict what you’re going to find until it is too late. There is currently no preoperative test that will tell you how much pannus there is or how difficult the valve will be to remove.
Based on our observations, I will still implant a stentless valve in a young, active patient who wants to avoid Coumadin. However, I will warn that patient that for the second operation, he or she should go to a center that does a lot of complex valve surgery. I would not recommend having a redo stentless valve operation at a center that performs 10 aortic valve replacements per year.
DR JÜRGEN ENNKER (Lahr, Germany): Congratulations for this interesting presentation. However, I also have mixed feelings concerning the combination of Freestyle and SPV as a total stentless valve. What was the percentage of Freestyle valves that were reoperated? Ten reoperations out of how many Freestyle valves?
DR BORGER: Half of those came from other centers, so I don’t know what the denominator is.
DR ENNKER: And do you know this for the SPV, the percentage?
DR BORGER: Again, some of those came from other centers, so I don’t know what the denominator is. However, Dr David presented our institution-specific results for the SPV a couple of years ago at the AATS meeting. Our ten-year freedom from SVD in patients over 65 was 99%; under 65 it was 77%. As I stated before, however, other centers have reported worse freedom from SVD rates for the SPV. Although we cannot make definitive comparisons, a review of the current literature would suggest that the freedom from SVD rates are better for the Freestyle valve.
DR ENNKER: And you did 4 subcoronary reoperations and 5 each subcoronary full root. What was the first operation, also subcoronary in those patients?
DR BORGER: Those were the first operations, 5 subcoronary and 5 total root.
DR ENNKER: And what did you do for kind of operation, total root in all, or the same procedure?
DR BORGER: Approximately two-thirds of all patients required a total root at the time of stentless valve operation. Some of those patients we put another Freestyle in, for older patients. There was 1 patient, it was a young woman who wanted to still have children. It was her second reoperation, that is, her third valve replacement surgery. She had an SPV the first time; they put in a Freestyle the second time. There was no annular tissue to sew to after removing the Freestyle valve. There was a leak, bleeding, coming from the base of the heart at the end of the procedure. As you can imagine, the surgeon was not too happy. He had also put in interposition grafts into both coronary ostia. He actually scrubbed out, went upstairs and talked to the patient’s husband and said, "Look, I’m sorry, but I’m putting in a mechanical valve. I don’t want the risk of ever having to come back with this young woman again." And he went back and put in a mechanical valved conduit.
So although we are very big tissue valve supporters, we still consider patient-specific factors when recommending the optimal prosthesis.
DR THOMAS GLEASON (Pittsburgh, PA): Mike, excellent presentation. One last final question. I guess 40% of the patients had some AI, and I think you said, it was 6 or 10 patients with endocarditis. What was the indication for the rest, stenosis? And how bad was the stenosis?
DR BORGER: Eighty-two percent had aortic insufficiency, so the vast majority; and 84% had structural valve deterioration. So mostly aortic insufficiency and only a little bit of stenosis.
DR GLEASON: So the indication, the primary indication for the whole group, was AI?
DR BORGER: In 82% of patients, yes.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
