| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Division of Cardiothoracic Surgery, Department of Surgery, Emory University School of Medicine, Atlanta, Georgia
b Pediatric Cardiac Surgery, Children's Healthcare of Atlanta at Egleston, Atlanta, Georgia
Accepted for publication June 8, 2011.
* Address correspondence to Dr Kanter, Pediatric Cardiac Surgery, Emory University School of Medicine, 1405 Clifton Rd, Atlanta, GA 30322 (Email: kkanter{at}emory.edu).
Presented at the Fifty-seventh Annual Meeting of the Southern Thoracic Surgical Association, Orlando, FL, Nov 3-6, 2010.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Methods: From 2003 to 2010, 15 children had 23 supra-annular MVRs. At first supra-annular MVR, median age was 6.5 months (28 days to 47 months); median weight was 5.4 kg (3.3-11.8 kg). Twelve (80%) had prior operations, 8 (53%) had previous mitral repair. Eight had congenital mitral anomalies (4 with Shone's), 5 had atrioventricular septal defects, 1 had endocarditis, and 1 had a repaired anomalous left coronary artery. All primary MVRs used mechanical valves (
17 mm in 9 patients).
Results: There was one early death (93% survival) in an 11-month-old with congenital pulmonary vein stenosis. One intraoperative conversion from annular to supra-annular MVR developed heart block. Three pacemakers were implanted for supraventricular rhythm disturbances. Three children had valve thrombosis early postoperatively treated medically. On follow-up of 4.3 ± 2.8 years, 8 had reoperation including redo MVR in 6 for pannus formation or thrombus (1 had three redo MVRs). At redo, a larger valve was used in 5 and a bioprosthetic valve in 4 patients. There was one late death after third redo MVR with pulmonary vein stenosis relief (overall survival 87%).
Conclusions: Supra-annular MVR is useful for children with a small annulus. Operative survival is good with infrequent heart block. Complications are common, including redo MVR and need for left ventricular outflow tract obstruction relief. Pulmonary vein stenosis is a marker for poor outcome; all patients without pulmonary vein stenosis survive long term.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Patient Population
From January 2003 to October 2010, we performed 23 supra-annular MVRs in 15 children. During this time, 15 children < 5 years of age (range 5.3 months to 4 years) had MVR in the annular position. Table 1
shows the patient characteristics at the time of the original supra-annular MVR. Patient 15 originally had an MVR in the annular position with a 17-mm St. Jude prosthesis (St. Jude Medical, St. Paul, MN) at 16 months of age at the time of repair of a hypoplastic aortic arch and coarctation. At 31 months of age, to increase the prosthetic valve size, a 19-mm St. Jude valve was placed in the supra-annular position.
|
Operative Technique
The surgical technique is depicted in Figure 1
. All patients had insertion of a mechanical valve at the initial supra-annular MVR (Table 1). For the 5 patients with AV septal defects, a trans-septal approach through a right atriotomy was used. For the others, a more standard left atrial incision was used. Often, an additional separate right atrial incision was made to close an atrial communication and to allow placement of right atrial-based valve sutures. The prosthetic valve was implanted with interrupted pledgeted polyester sutures with the pledgets on the atrial side of the prosthesis. Only one patient had the prosthetic valve sewn into a sleeve of Dacron Hemashield (Meadox Medicals, Inc, Oakland, NJ), which was then sewn to the native valve annulus, as we [3] and others [4] have previously described. Occasionally, the left atriotomy incision was closed with a patch of homograft tissue or pericardium to augment left atrial volume.
|
Statistics
Continuous variables are presented as mean ± standard deviation. Life-table analysis was done by the Kaplan-Meier survival method.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Patient 6 (Table 1) initially had annular placement of a 21-mm St. Jude valve. On weaning from cardiopulmonary bypass, there was evidence of ischemia in the circumflex coronary artery distribution. At the same operation, the valve was explanted and a 23-mm St. Jude valve was placed in a supra-annular position. Although this resolved the coronary ischemia, the child required postoperative ECMO support, but gradually recovered. This is our only patient who developed postoperative complete heart block requiring permanent pacemaker placement. One other patient (Patient 8) exhibited signs of coronary artery ischemia in the circumflex distribution after supra-annular MVR with a 17-mm St. Jude valve. This was rectified by removing one-third of the valve sutures superiorly in the left atrium near the aortic valve and repositioning this portion of the prosthetic valve more proximally in the left atrium during the same operation. Although this child did well hemodynamically, she developed recalcitrant supraventricular arrhythmias uncontrolled by multiple antiarrhythmics. Seven weeks postoperatively, she had placement of a biventricular pacemaker with percutaneous ablation of her atrioventricular node, which effectively controlled her arrhythmias. Two other patients (Patients 11 and 15) had pacemaker placement postoperatively for persistent nodal rhythm or sick sinus syndrome.
There were three early episodes of thrombosis of one of the prosthetic valve leaflets (Patients 6, 10, and 14). Two episodes were related to inadequate coagulation. Two responded to thrombolytic therapy without surgical intervention. The third patient is now 6 years after supra-annular MVR and tolerates one chronically stuck prosthetic leaflet. The other 2 patients had valve fluoroscopy both before and after the thrombotic event, which showed normal prosthetic valve leaflet movement.
Reoperations
On average follow-up of 4.3 ± 2.8 years, 8 of the 14 survivors required reoperation, including 8 redo MVRs in 6 patients (Table 2). The interval from original supra-annular MVR to first redo supra-annular MVR ranged from 5 months to 10 years (mean 3.8 years). The primary indication for redo supra-annular MVR was chronic pannus formation in four and acute thrombus in two. One patient had 3 redo supra-annular MVRs (Table 2). In 3 patients, one leaflet of the prosthetic valve was noted to be stuck due to pannus ingrowth. Four of the six children who had redo supra-annular MVR were converted to a bioprosthetic valve, due to difficulties with anticoagulation control, valve thrombosis, or bleeding difficulties. All patients had the new mitral valve placed in the supra-annular position—none had the new valve positioned in the native valve annulus. Five of the 6 patients had a larger valve placed at reoperation (Table 2).
|
|
The overall patient survival in this series is 87% with an actuarial survival at 10 years of 84% (Fig 3 ). At last visit, all of the survivors are doing well clinically, although Patient 13 has developed prosthetic mitral stenosis 7 years after her original supra-annular MVR so she will undergo redo supra-annular MVR in the near future.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
Early results with supra-annular MVR in children were discouraging. Schaffer and colleagues [5] from the University of Colorado used this technique in eight children with seven deaths (two early and five late). Adatia and others [6], from Children's Hospital in Boston, reviewed 17 children who received a supra-annular MVR. There were seven deaths, five within 6 months of surgery. Similarly, Daou [7], from Paris, described two perioperative deaths in six children who received a supra-annular MVR. The group from Great Ormond Street [8] used this technique in eight children. They found it to be a predictor of 30-day mortality. A more recent publication [1] from Boston still identified supra-annular MVR as a risk factor for mortality (24% early mortality). In that study, 1-year actuarial survival after supra-annular MVR was 67%, even in their more recent experience.
Other series have shown more favorable results. Kadoba and colleagues reported uniform survival in four infants with a supra-annular MVR [9] as did Masuda from Japan in seven children [10]. Previously, we found no difference in survival in children who had a supra-annular MVR compared with those who had annular implantation [3]. Currently, we describe 23 supra-annular MVRs in fifteen children with one operative death at the time of the first supra-annular MVR (early survival 93%). On average follow-up of 4.3 years, there was one late death during a fourth supra-annular MVR giving an actuarial 5-year survival of 83% (Fig 3). Both deaths were associated with preoperative pulmonary vein stenosis and severe pulmonary artery hypertension. It is unlikely that the pulmonary vein stenosis was related to the supra-annular mitral prosthesis. In the patient who died early (Patient 12), the congenital pulmonary vein stenosis was present prior to MVR. The other patient developed left pulmonary vein stenosis after three supra-annular MVRs (Patient 10). Direct inspection of the stenotic pulmonary vein at reoperation revealed that it was not related to mechanical impingement on the pulmonary vein by the prosthesis. It is remarkable that the only two deaths in this series were related to pulmonary vein stenosis; this may be a marker for poor outcome.
Intraoperatively, two of our patients developed signs of coronary ischemia in the circumflex coronary artery distribution which was recognized immediately and corrected. In the first instance (Patient 6), repositioning the prosthetic valve from the annular to the supra-annular position was effective. The coronary ischemia was corrected in the second patient (Patient 8) by repositioning the superior third of the prosthetic valve sewing ring further away from the native annulus. Compression of the coronary artery by the prosthetic valve is a potential pitfall of this surgical technique but can be corrected successfully if recognized early.
Only one patient in this series developed complete heart block requiring a permanent pacemaker after supra-annular MVR (Patient 6); this was in the child who initially had annular positioning of the mitral prosthesis but developed coronary ischemia which was relieved by repositioning the valve in the supra-annular position. Three patients had pacemaker implantation for supraventricular arrhythmias: one for persistent nodal rhythm, one for sick sinus syndrome, and one for medically refractory supraventricular arrhythmias eventually requiring atrioventricular node ablation and pacemaker implantation. This non-trivial incidence of postoperative supraventricular rhythm disturbances may be related to the generous atrial incisions necessary to achieve adequate exposure of the mitral valve in these small children or to the ventricularized portion of the left atrium between the mitral prosthesis and the native mitral annulus.
Early valve thrombosis was a problem in three patients in this series despite meticulous attention in the operating room to insure unimpeded movement of the prosthetic valve leaflets. This was successfully managed with thrombolytic therapy in two patients [11]; the third tolerated chronic immobility of one prosthetic valve leaflet long-term. Valve thrombosis was the indication for repeat supra-annular MVR in two children and emphasizes the difficulty of achieving optimal anticoagulation in these small children. Four of the six children who required redo supra-annular MVR had conversion from a mechanical valve to a bioprosthetic valve due to issues with chronically inadequate anticoagulation, recurrent valve thrombosis, or bleeding difficulties despite the recognition that the bioprosthetic valve will require fairly early re-replacement due to structural deterioration in these growing children. Hopefully, newer anticoagulation regimens being developed will be safer and more efficacious thus ameliorating the difficulties with optimal anticoagulation in these small children with mechanical prosthetic valves.
On average follow-up of 4.3 years, redo MVR was performed 5 months to ten years after the original MVR (mean, 3.8 years) in six children in this series (Table 2). All patients had the replacement valve placed in the supra-annular position recognizing that others recommend implantation of the new valve in the native annulus [1, 9, 10, 12]. Supra-annular positioning of the new valve at repeat operation has two advantages: First, it allows one to implant a larger valve more easily as was achieved in five of the 6 patients undergoing redo MVR (Table 2). Second, the risk of complete heart block is significantly decreased. There were no instances of heart block with redo supra-annular MVR in our series. This contrasts to a 33% incidence of heart block in fifteen patients from Boston who initially had a supra-annular MVR whose redo valve was positioned in the native annulus [1]. Although some investigators have expressed concern about the potential for persistent left atrial obstruction with consequent pulmonary hypertension with supra-annular MVR [1, 2], we have not recognized these problems although we have not routinely performed cardiac catheterization in these patients. Therefore, we feel comfortable employing the supra-annular technique at the time of valve re-replacement.
Redo MVR was performed with a low mortality in this series with only one death at the time of third redo supra-annular MVR with concomitant relief of pulmonary vein obstruction. It is likely that all of these infants eventually will require redo MVR. The safety of redo MVR in children, as demonstrated in this series, has been reported previously by our group [13] as well as others [14, 15].
Operations for relief of left ventricular outflow tract obstruction (LVOTO) were necessary in one-third of our patients. Although some have implicated left heart obstruction as an indicator for poor outcome after supra-annular MVR [6], this was not the case in our series. The LVOTO is more likely related to the underlying congenital defect in these patients (particularly Shone's syndrome) rather than related to the supra-annular MVR itself. Although encroachment on the left ventricular outflow tract by a prosthetic mitral valve is a recognized complication when the valve is implanted in the native mitral annulus, the LVOTO we encountered was felt not to be caused by the supra-annular mitral prosthesis. In fact, the reduction of the potential for development of LVOTO by the prosthetic valve with supra-annular MVR is an advantage of this technique.
In all, eight of the 14 operative survivors (57%) in this series required reoperation either for redo MVR or relief of LVOTO during a relatively modest follow-up. The actuarial freedom from reoperation is 69% at 5 years and 21% at 10 years (Fig 2). Undoubtedly, all of these children will eventually require reoperation. Despite this, so far, the patient survival is reasonable (Fig 3).
In summary, although annular placement of a prosthetic mitral valve is our preferred technique in children when MVR is necessary, supra-annular MVR is useful for those with a small annulus in whom a more standard annular implantation cannot be performed safely. Operative survival is good. Heart block is rare although supraventricular rhythm problems are not uncommon. Complications are common including the potential for coronary ischemia intraoperatively as well as early and late valve thrombosis. Precise anticoagulation and avoidance of thrombotic and hemorrhagic complications in these young patients is challenging. Reoperation appears inevitable including redo supra-annular MVR and LVOTO relief, both of which can be performed safely. A larger prosthetic valve usually can be implanted at reoperation. Pulmonary vein stenosis is a marker for poor outcome: both deaths in this series occurred with operative relief of pulmonary vein stenosis; all patients without pulmonary vein stenosis survive long term. Despite the acceptable early and mid-term mortality in this series, in view of the demonstrated high complication rate and the predictable need for reoperation, every reasonable effort should be made to preserve the native mitral valve by surgical repair techniques or interventions in the cardiac catheterization laboratory. Furthermore, if MVR is necessary with a prosthetic valve, supra-annular placement should be reserved only for those patients in whom the native annulus is too small to accept a commercially available valve.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
There is a contemporaneous group in your institution of 15 intra-annular mitral valve replacement patients that are close to age matched, and I was just wondering if you had any insight into who gets a mitral valve in what position, or has the intra-annular procedure fallen out of favor for this patient group at your program?
DR KANTER: No, if we can fit a prosthetic valve in the annulus, that is the procedure of choice. For all of these patients with supra-annular mitral valve replacements, the annulus was too small except for the one in whom an intra-annular one was placed and then was repositioned to a supra-annular position because of coronary ischemia. We would not recommend supra-annular mitral valve replacement as a first line choice, only for those we can't put in the annular position.
DR FORBESS: I will skip to what was question four. I noticed your reoperations were all in the supra-annular position. I think I have always moved a valve from the supra-annular position when the patient is larger to the intra-annular position. Did you re-replace in the supra-annular position because you felt that the annulus in those instances had not developed well enough, or fear of creation of heart block?
DR KANTER: The Boston group had a fairly large series of supra-annular mitral valve replacements. At the time of reoperation, they made an effort to place the redo valve in the annular position and had a 33% incidence of complete heart block. I think that if we place the redo valve in the supra-annular position, we can upsize it without the risk of new heart block. In five of the six reoperations, we were able to upsize and usually two full valve sizes, so from a 17 to a 21, a 19 to a 23.
DR FORBESS: I actually think that is one of the more valuable contributions of this series, because moving to the intra-annular position is clearly not without problems at the reoperation. Now, as far as thrombosis, who gets TPA and who gets a reoperation? I noticed in the manuscript there were patients that were reoperated on for acute thrombosis, but TPA was also used for acute thrombosis.
DR KANTER: In the early postoperative period, typically a thrombosed leaflet is often asymptomatic because there is still a significant valve-patient mismatch in that the valve is too large; so the only way you discover it is on echo or routine valve fluoroscopy. We would use tPA or thrombolytics as the first line of therapy. We would only go to reoperation for those in whom that was not successful. In fact one patient, we left the leaflet chronically stuck.
DR FORBESS: I think it was reported in Brian Kogon's TPA paper but it re-stuck again—after the paper was published, of course.
DR KANTER: Right. So we won't go to surgery unless they are hemodynamically unstable and we are unable to unstick it. But two of the three were able to be opened up with thrombolytic. The others were more of a subacute nature in that they developed a stuck leaflet within a few weeks.
DR FORBESS: Everybody has relatively low numbers, thankfully, of patients that get these kinds of operations, but I was curious about your use of a bioprosthesis in these fairly young patients because of "difficulties with Coumadin," as you term it. I would note for the audience that our cardiology group in Dallas has fairly quickly moved to aspirin and Lovenox for patients that give them the first hint of any problems with Coumadin—again, small numbers of patients. It sure seems, however, like we have gotten away from having to convert to a bioprosthetic valve in very young children. I do note, again in your manuscript, one of your late deaths is from severe mitral stenosis in one of those bioprosthetic valves.
DR KANTER: Well, we routinely used aspirin and Coumadin in these patients. Of the four patients who were converted over to a bioprosthetic valve, two of them were done just for persistent clotting of their mechanical valve, and the other two, one was from chronic nosebleeds and the other was just a family preference. But this is a difficult group of patients to anticoagulate correctly.
I am intrigued by the thought of alternate things. I understand that there are other anticoagulants down the road which are about to be approved in the next year or so, which may make us able to use mechanical valves forever.
DR FORBESS: Thanks very much, Kirk.
DR CONSTANTINE MAVROUDIS (Cleveland, OH): That was a great presentation, Kirk. I have two unrelated questions. The first one was, those three patients with supraventricular tachycardia, did they occur in the operating room and were you able to ablate these arrhythmias during the operation?
DR KANTER: No. I remember that patient you talked about with the atrial appendage thing?
DR MAVROUDIS: Yes.
DR KANTER: No.
DR MAVROUDIS: You didn't see any topical myocardial irregularities that might have signaled a focus of automatic tachycardia?
DR KANTER: No. In fact, I talked to you about one of these patients and you told me about that.
DR MAVROUDIS: And then the other issue is, did you consider performing an electrophysiologic study and perhaps ablate in the operating room?
DR KANTER: That child actually did have an EP study and a successful ablation for a good 18 hours, and then it came back, which for the cardiologist was a success.
DR MAVROUDIS: But what I mean is, did they refer it to you for an ablation in the operating room?
DR KANTER: No. The ablation was done in the cath lab.
DR MAVROUDIS: I know. What I am saying is that it was a failed catheter ablation in the cath lab.
DR KANTER: Of the focus.
DR MAVROUDIS: Of the focus. And so the next stage in some places would have been to perform an operative ablation at surgery. The next stage for them was AV node ablation and a pacemaker.
DR KANTER: I see.
DR MAVROUDIS: Is that a good trade? I am asking your opinion. Would you have rather taken that patient to the operating room and tried to do an ablation of the arrhythmia without having AV block and the necessity for a pacemaker? That is the first question.
DR KANTER: Our electrophysiologist assured me that it was very unlikely that there was a direct focus that we could get, and they were very reluctant to even try it in the cath lab. It was with much arm twisting that they did go to the cath lab.
DR MAVROUDIS: Okay. And so then the next point is anticoagulation. Joe Dearani was alluding to new drug trials that I believe are being conducted at the Mayo Clinic. Do you know anything about these drugs which have been noted to be easier to dose and manage when compared to warfarin?
DR ROBERT D.B. JAQUISS (Little Rock, AR): There is a drug called rivaroxaban, which has been used extensively in Europe and in Canada in orthopedic surgery in head-to-head trials with Coumadin. So it is the same level and intensity that we will need when we get our hips and these replaced, respectively. It does not require any monitoring. It is associated with equivalent efficacy of prevention of deep venous thrombosis and less bleeding. So it is close to the super drug that we have been waiting for. There is no data in children whatsoever, but there isn't in most of the stuff that we use. So I think it and a whole class of drugs that are "me-too" type, you know, different by one methyl group or one amino group that are going to be very similar, I think we will have that stuff available to play with in this country fairly shortly in the next couple of years.
DR KANTER: I, too, have heard of these drugs and they are supposed to be in the next year or so, and they don't work the same way as warfarin.
DR MICHAEL HINES (Winston-Salem, NC): Remember now, Lovenox doesn't require monitoring in adults and wasn't supposed to be required in kids, but it does; if you actually check it, they are all over the place. So be careful about these drugs that are.
DR PIROOZ EGHTESADY (Cincinnati, OH): I truly enjoyed your presentation, as always. Just a quick question: When do you think enough is enough? And the reason I ask this is I have often thought about the idea of whether we should use cardiac transplantation for some of these kids as opposed to trying back again and again. In part, a typical scenario is a kid who one of our cardiologists had seen, a transitional canal had been repaired at a different institution, re-repaired and then re-repaired, then replaced and then re-replaced, and they came back and they said that they were concerned that there was an issue that that was causing some degree of stenosis, the transition we repaired, and I said, look at the literature, and this kid already was getting into a lot of morbidity from all these reoperations. And I said, why don't we just replace everything?
DR KANTER: I still maintain that cardiac transplantation should be reserved for those patients for whom there are no standard medical and surgical therapies. So therefore you can just keep on replacing these valves. Don't forget that transplant in an infant has a half-life of 12 or 14 years. So you have got to ask yourself, do you think that the fourth valve replacement will get that patient out longer than a transplant?
DR HINES: Kirk, one more quick question. I think if I saw it right, you had two patients that had circumflex ischemia when you put the valve in, and you moved it supra-annular.
DR KANTER: No. One was annular, one was supra-annular.
DR HINES: Do you think the ischemia was from a stitch or distension from the valve and how after you moved it did you confirm that you fixed the ischemia?
DR KANTER: The annular one was just from the valve being too big and compressing the circumflex under a fair length, and the one that was supra-annular, I don't think it was a specific stitch. I think it was just compression since it resolved with just moving the sewing ring back a little. We make an effort when we do these to put it as close to the native valve as possible so that you don't have this big ventricularized portion of the left atrium. We use as our landmark the mouth of the left atrial appendage. The left atrial appendage is always on the low pressure side, because another concern is pulmonary vein stenosis with placement of the supra-annular valve too proximally in the left atrium.
DR HINES: I assume you knew you had ischemia when you had trouble coming off bypass?
DR KANTER: Right.
DR HINES: After the repositioning, did you then just attempt to wean from bypass or did you do any specific diagnostic study to look for improvement in circumflex flow?
DR KANTER: No, we just came off pump again and looked if the heart pinked up; very scientific.
DR HINES: Very nice paper. Thank you again.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Murashita, T. Hoashi, K. Kagisaki, K. Kurosaki, I. Shiraishi, T. Yagihara, and H. Ichikawa Long-Term Results of Mitral Valve Repair for Severe Mitral Regurgitation in Infants: Fate of Artificial Chordae Ann. Thorac. Surg., August 1, 2012; 94(2): 581 - 586. [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 |
