|
|
||||||||
a Division of Cardiothoracic Surgery, Emory University School of Medicine, Atlanta, Georgia
b Sibley Heart Center, Children's Healthcare of Atlanta, Atlanta, Georgia
Accepted for publication February 20, 2009.
* Address correspondence to Dr Kirshbom, 1365 Clifton Rd, Suite A2100, Atlanta, GA 30322 (Email: paul.kirshbom{at}emoryhealthcare.org).
Presented at the Fifty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Austin, TX, Nov 5–8, 2008.
| Abstract |
|---|
|
|
|---|
Methods: Two hundred thirty-six consecutive Fontan procedures were retrospectively reviewed. Group 1 (n = 21, with concomitant AVV repair [n = 19] or replacement [n = 2]) was compared with group 2 (n = 215, no AVV surgery) with regard to preoperative characteristics and perioperative outcomes. Atrioventricular valve regurgitation was graded as 1 (none or trivial) to 4 (severe).
Results: Group 1 patients were older (4.3 ± 3.7 versus 3.0 ± 2.6 years; p = 0.04) and had longer cardiopulmonary bypass (118 ± 38 versus 85 ± 28 minutes; p < 0.001) and aortic cross-clamp times (33 ± 32 versus 14 ± 21 minutes; p < 0.001). There were no differences between groups regarding diagnosis, weight, hospital or intensive care unit length of stay, ventilator time, or 12-hour chest tube output. Postoperative complications were similar between groups, including bleeding (0 of 21 versus 8 of 215; p = 0.8), neurologic injury (1 of 21 versus 9 of 215; p = 0.7), arrhythmias (1 of 21 versus 24 of 215; p = 0.6), and operative mortality (0 of 21 versus 1 of 215; p = 0.1). Group 1 AVV regurgitation significantly decreased after surgery (3.0 ± 0.9 preoperatively versus 1.7 ± 0.9 postoperatively; p < 0.001).
Conclusions: Atrioventricular valve surgery has been shown to improve late outcomes for single-ventricle patients. This study demonstrates that AVV surgery performed with the Fontan procedure increased operative times, but did not significantly increase perioperative morbidity or mortality. This information supports appropriate utilization of AVV surgery for single-ventricle patients.
| Introduction |
|---|
|
|
|---|
However, certain risk factors continue to be associated with impaired outcomes after the modified Fontan procedure. Included among these is moderate to severe atrioventricular valve (AVV) regurgitation, which has been demonstrated in up to 20% of patients with single-ventricle physiology [3]. The presence of AVV regurgitation may be associated with increased operative mortality and a negative impact on long-term outcomes. Although AVV repair or replacement has been shown to improve late outcomes in some series, there are mixed data regarding the associated perioperative morbidity and mortality when performed concomitantly with the Fontan procedure. The goal of this study is to evaluate the additional perioperative risks potentially associated with single-ventricle AVV surgery. In addition, we sought to determine the durability of AVV repair or replacement in patients with single-ventricle physiology.
| Patients and Methods |
|---|
|
|
|---|
|
50% of atrial area) [3–6].
Operative Procedures
All procedures were performed by means of a standard median sternotomy. The majority of cases were conducted with cardiopulmonary bypass (CPB) using bicaval and single arterial cannulation with moderate systemic hypothermia. Five operations in 2002 were performed without the use of CPB. Eighty-six were performed with cardioplegic arrest with a mean cross-clamp time of 42 minutes, and 150 were conducted with induced fibrillatory arrest (ie, empty, perfused fibrillation without aortic cross-clamping) with a mean fibrillation time of 27 minutes. One hundred seventy-five patients underwent a lateral tunnel Fontan procedure (173 fenestrated, 2 nonfenestrated), and 61 had an extracardiac conduit (52 fenestrated, 9 nonfenestrated).
Among the 21 patients with AVV surgery, 12 had an additional concomitant procedure, including pulmonary artery reconstruction (n = 7), pacemaker implantation (n = 5), and semilunar valve closure (n = 2). Two patients in this group had AVV replacement, whereas 19 underwent AVV repair. Repair techniques consisted of one or a combination of the following: De Vega annuloplasty (n = 13), Kaye annuloplasty (n = 4), suture commissuroplasty (n = 5), cleft repair (n = 3), and edge-to-edge Alfieri repair (n = 1). Those with structural abnormalities underwent repair techniques relevant to the anatomic defect, whereas the De Vega technique was used if regurgitation was thought to result from ventricular dilatation and annular enlargement.
Statistical Methods
Analyses were performed with SPSS 16.0 statistical analysis software (Chicago, IL). Data are presented as mean ± standard deviation or absolute numbers and percentages. Continuous variables were compared using Student's t test or analysis of variance, as appropriate. Categorical variables were compared using a
2 analysis. Probability values less than 0.05 were considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
|
|
| Comment |
|---|
|
|
|---|
These outstanding results have been attributed to several aspects of perioperative care, including better patient selection, enhanced postoperative management, and advances in surgical, anesthetic, and perfusion techniques [13, 17, 18]. Despite this, several risk factors continue to be predictors of adverse outcomes after the modified Fontan procedure, including elevated pulmonary artery pressures [10, 19–21], pulmonary artery distortion [20, 22], presence of a common AVV [13], impaired ventricular function [10], and predominant right ventricular morphology [2, 3]. Furthermore, right ventricular morphology and older age at Fontan have been associated with AVV regurgitation [3], which in turn has been identified as a significant independent predictor of morbidity and mortality after the Fontan procedure [19, 21, 23, 24].
Although the majority of studies have shown an improvement in long-term outcomes with AVV repair at the time of the Fontan procedure, the results have been somewhat mixed. Some studies have demonstrated an improvement in ventricular function and survival with AVV repair [25, 26]. However, despite 10-year survival rates comparable to other contemporary studies (84%), Imai and colleagues [27] showed an increase in operative mortality in patients with AVV valve regurgitation. Likewise, Sallehuddin and associates [28] had an almost twofold increase in operative mortality in 129 patients with AVV regurgitation who underwent repair (18.9%) versus no repair (10.9%), although these results were not statistically significant. In contrast, our results demonstrate equivalent operative mortality and postoperative morbidity despite having longer CPB and cross-clamp times in the AVV surgery group.
Although we do not have long-term follow-up in this series, we have shown AVV repair to be successful and durable. We found that the degree of AVV regurgitation decreased from a mean grade of 3.0 preoperatively to 1.7 postoperatively, and was preserved at 1.8 at a mean follow-up time of 2.6 years. Of the 21 patients in this series, 2 had valve replacement whereas 19 underwent repair using various techniques. Several successful repair techniques have been previously published. Kanter and coworkers [29] reported successful outcomes of the De Vega annuloplasty in 12 patients with single-ventricle physiology, 8 of whom underwent the Fontan procedure. Similarly, Imai and associates [27] have described a circumferential purse-string suture that they successfully used in 99 patients at the time of the modified Fontan procedure. Other techniques used include the edge-to-edge Alfieri repair, reported by Ando and colleagues [30] in 22 patients, and partial suture annuloplasty along the posterior leaflet of the tricuspid valve, isolated commissuroplasty, chordal shortening, and partial flexible ring annuloplasty reported by Ohye and coworkers [31].
Some previous studies have suggested that valve repair may not be necessary at the second-stage palliation for single-ventricle heart disease, as volume unloading at the second stage can improve ventricular distention and AVV regurgitation. For example, Mahle and colleagues [4] showed that moderate AVV regurgitation improves in approximately 20% of patients at the time of bidirectional cavopulmonary anastomosis without valve repair. Likewise, Sallehuddin and associates [28] demonstrated that among patients with moderate to severe AVV regurgitation, those who did not have valve repair derived the same degree of postoperative improvement in the grade of AVV regurgitation compared with those who underwent repair. In the current era in which the extracardiac Fontan modification can sometimes be performed without CPB, the decision to proceed with AVV surgery must be carefully considered. However, inasmuch as the Fontan procedure does not improve ventricular loading conditions and AVV surgery did not add to the perioperative risks in the current study, our results support repairing or replacing valves with moderate or severe regurgitation at the Fontan completion.
In summary, we propose that all patients with moderate or severe AVV regurgitation should undergo repair, if possible, or replacement at the time of the Fontan procedure. Despite the longer CPB and cross-clamp times, valve repair or replacement carries little added risk and can be accomplished safely with good mid-term durability.
| Discussion |
|---|
|
|
|---|
Could you elaborate on whether the same cardiologists who read the echos were blinded as to the time of the procedure and the name of the patient so as to remove any bias that that information may impart? Almost anecdotally, I was also intrigued to see that you briefly experimented doing the Fontan procedure off-pump early in your series with 5 patients and abandoned this. Doing a Fontan off-pump, at least for me, seems a triumph of technique over judgment, and I was wondering, did any of those 5 patients have an untoward or a more rocky recovery?
I remain most impressed with your length of stay data, and from that comes a question that is pretty multifaceted. Could you define a little bit your postoperative management in terms of specifically addressing ACE (angiotensin-converting enzyme) inhibitors, steroids, antiinflammatory mediators, approach to extubation, and defining aggressive diuresis in terms of its effect on the patients?
Only one mild criticism, and that was why you didn't actually use the TEEs (transesophageal echocardiograms) intraoperatively for comparison as well. Do you have any data on how much the regurgitation increased out of the OR (operating room) as a result of the afterload changes? Your percentage of single-ventricle patients receiving valve repair might actually be a little low compared to other series. Could you tell us how many of your first- or second-stage single-ventricle palliations received valve repair as a part of that, and did any of those patients require further valve repair in the Fontan series?
Finally, you used the De Vega pretty frequently, 12 times in your manuscript, and in some series its use has been questioned in terms of durability in the tricuspid valve. Did you notice any differences between those patients who got a De Vega solely versus those patients that had a cleft repair or a suture commissuroplasty?
Once again, congratulations on a very nice contribution to our literature and this meeting. Thank you.
DR KERENDI: Thank you, Dr Calhoon, for your comments. Going back to your first question as far as the cardiologists who read the echos, one cardiologist read all the echocardiograms at one time. He was blinded and he read them all at the same time so as to not have any bias in terms of reading them at different periods in time.
Your second question was were there any untoward effects on the patient who had the off-pump Fontan procedure. As far as we know, there were not.
In terms of perioperative management, all patients receive IV (intravenous) diuretics within 6 to 8 hours of surgery, and we attempt to extubate them early on. It is usually by the evening of surgery; if not, then by the morning following surgery. They all receive steroids on cardiopulmonary bypass but no antiinflammatories postop, and they all receive ACE inhibitors when they start to take oral medications as soon as they are extubated.
As far as the intraoperative TEE, we felt that the preop and the postop echos were reasonably accurate and that, with the hemodynamic changes under anesthesia, the intraop TEEs may not be entirely accurate. So that is the primary reason we didn't look at that, but certainly we could go back and pull those data.
Your next question was on valve repair and how many patients had had previous repair. I don't know offhand what the total numbers were. I know there were a handful of patients in the AV valve repair group that had had previous repair. The 2 patients who had valve replacement had undergone a previous repair that failed, so that those were at least 2, and I know there were 2 others who had an additional repair. But offhand, I don't know the total number of how many patients there were in the entire cohort.
And then your last question regarding the De Vega durability, Dr Kanter published a series from our institution a couple of years ago showing the efficacy and durability. We did not look at these independently in this analysis to see if there was a difference in the isolated De Vegas from the patients who had more complex repairs.
DR ANDREW C. FIORE (St. Louis, MO): Faraz, I was going to ask you a question. Presumably the valve repair patients underwent aortic cross-clamping with cardioplegia. Were the non–valve repair patients protected in the same way?
DR KERENDI: Not all of them were. I would say the majority of them were but not all of them.
DR FIORE: So some patients were done with fibrillatory arrest as opposed to cross-clamping and cardioplegia?
DR KERENDI: Correct.
DR FIORE: The other question I had relates to the pleural drainage. You analyzed drainage at 12 hours, but was the long-term drainage similar for both groups or did the AV valve repair patients do better in terms of how quickly their chest tubes came out?
DR KERENDI: We don't have data on when the chest tubes came out. What we do have data on is on recurrent pleural effusions requiring drainage, and there was no difference in that regard.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Nakata, Y. Fujimoto, K. Hirose, Y. Tosaka, Y. Ide, M. Tachi, and K. Sakamoto Atrioventricular valve repair in patients with functional single ventricle J. Thorac. Cardiovasc. Surg., September 1, 2010; 140(3): 514 - 521. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. P. Graham Jr The Year in Congenital Heart Disease J. Am. Coll. Cardiol., January 12, 2010; 55(2): 147 - 155. [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 |