| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Cardiothoracic Surgery, Department of Surgery, Sections of Pediatric Cardiology and Critical Care, Departments of Pediatrics and Quantitative Health Sciences, Medical College of Wisconsin and Herma Heart Center, Children's Hospital of Wisconsin, Milwaukee, Wisconsin
Accepted for publication May 21, 2009.
* Address correspondence to Dr Tweddell, MS 715, Children's Hospital of Wisconsin, 9000 W Wisconsin Ave, Milwaukee, WI 53226 (Email: jtweddell{at}chw.org).
Presented at the Fifty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Austin, TX, Nov 5–8, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Methods: A cross-sectional analysis of 256 consecutive patients undergoing a total cavopulmonary connection Fontan after superior cavopulmonary shunt between January 1, 1994, and June 30, 2007 were studied. Fenestration was used selectively. Fontan failure was defined as death, transplant, or takedown. Event-free survival was defined as freedom from death, transplant, Fontan takedown, functional class III to IV, pacemaker, antiarrhythmic medication, protein-losing enteropathy, stroke, or thrombus.
Results: Survival was 97% ± 1%, 96% ± 1%, and 94% ± 2%, respectively, at 1, 5, and 10 years. Event-free survival was 96% ± 1%, 87% ± 3%, and 64% ± 6%, respectively, at 1, 5, and 10 years. Factors predicting worse event-free survival included longer cross-clamp time (p = 0.003), fenestration (p = 0.014), and longer hospital length of stay (p = 0.016). Ventricular morphology did not predict outcome. Left ventricle (n = 113, 44%) versus right ventricle (n = 142, 56%) failure-free survival (death, transplant, or Fontan takedown) at 10 years was 92% ± 4% versus 91% ± 3%, respectively (p = 0.19). Left ventricle versus right ventricle event-free survival at 10 years was 75% ± 7% versus 67% ± 9%, respectively (p > 0.1).
Conclusions: Survival for patients undergoing a completion Fontan in the current era is excellent, but patients remain at risk for morbid events. In the intermediate follow-up period, we could not identify a difference in outcome between dominant left and right ventricle morphology.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
We undertook this study to look at our experience with the Fontan procedure in the era of staging with a prior CP shunt, TCPC, and the common use of a fenestration. We looked at the impact of a variety of patient and operative factors on the outcome of the completion Fontan.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Data Collection
Patients were identified from the Herma Heart Center Cardiology and Cardiothoracic Surgery Database. Additional information was abstracted from medical records, echocardiography, cardiac catheterization, and operative reports. Follow-up information was obtained from clinic records and test results. Ventricular morphology of the functional single ventricle was determined from preoperative echocardiography reports as well as operative findings and defined as a dominant left ventricle (LV), right ventricle (RV), or a ventricle of indeterminate morphology. All patients had a prior CP shunt and completion Fontan type was LT, with or without fenestration, or EC with or without fenestration. There were no direct right atrium to pulmonary artery Fontan connections performed during this period.
Operative Technique
All Fontan procedures were performed using cardiopulmonary bypass with moderate hypothermia at 28° to 32°C, unless concomitant procedures required deep hypothermia and altered perfusion, such as low flow bypass or circulatory arrest. The intracardiac portion of the procedure was performed using one of three cardiac management strategies: aortic cross-clamping with cold cardioplegia, ventricular fibrillation, or an empty beating heart. For the empty beating heart technique, the cardiac action was slowed with the use of hypothermia (28°C), the patient was placed in the Trendelenburg position, the cross-clamp was briefly applied to prevent ejection of air while the heart was opened, and a pump sucker was then placed in the ventricle for decompression and to temporarily render the atrioventricular valve incompetent. The cross-clamp was then released. By rendering the atrioventricular valve incompetent and emptying the single ventricle, ejection of air was prevented. The LT was constructed using a patch of Gore-Tex (W. L. Gore & Associates, Flagstaff, AZ) to create the baffle within the atrium. The EC was constructed using a Gore-Tex tube graft between 16 and 22 mm in diameter. Fenestrations, usually between 3.5 mm and 4.0 mm, were placed routinely at the time of the Fontan. For the LT Fontan, a coronary punch was used to create a fenestration in the Gore-Tex baffle. For the EC Fontan, a fenestration was created in the Gore-Tex tube graft with the coronary punch. Then the atriotomy that results from detaching the inferior vena cava from the right atrium was sewn to the Gore-Tex tube graft as a circle of about 2.5 cm, with the fenestration in the center of the circle. That prevents the adjacent atrial wall from impacting the size of the fenestration. At the discretion of the surgeon, the fenestration was sometimes closed in the operating room. This decision was made after weaning from bypass and assessment of the hemodynamics and transesophageal echocardiogram. In the current era, fenestrations are routinely left open.
Outcome Variables
Fontan failure was defined as death, transplant, or takedown of the Fontan circuit to a CP shunt or other palliative procedure. Morbid events after Fontan palliation were defined as New York Heart Association (NYHA) functional classification of III or IV, need for an antiarrhythmic medication, pacemaker placement, protein-losing enteropathy, stroke, or thrombus. The duration of time a patient was free from death, transplant, Fontan takedown, NYHA functional class III or IV, pacemaker, antiarrhythmic medication, protein-losing enteropathy, stroke, or thrombus was deemed the event-free survival time.
Statistical Analysis
Descriptive characteristics of the sample are summarized by median and range because variables were skewed. Categorical variables are summarized with frequencies and percentages. Nonparametric Mann-Whitney tests was used to compare continuous variables, and the Fisher exact test or 
2 test was used to compare patients with a dominant LV versus RV for baseline characteristics. Kaplan-Meier survival analysis was used to assess survival, failure-free survival, and event-free survival for the whole group, with log-rank comparisons between the LV and RV groups. Six variables were considered for inclusion as covariates in a Cox proportional hazards model. Variable selection was done using backward and forward stepwise regression. The software used was SPSS version 16 (SPSS Inc, Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
|
|
|
|
|
|
|
Figure 5
shows the Kaplan-Meier curve for failure-free survival comparing patients with LV and RV morphology, and demonstrates a similar failure-free survival. Figure 6
shows the Kaplan-Meier curve for event-free survival comparing patients with LV and RV morphology and shows that ventricular morphology did not impact event-free survival.
|
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Our experience shows that in its current version the Fontan procedure achieves excellent survival. In our group of 256 patients, the hospital survival was 98% (251 of 256). There was little mortality during the follow-up period, with a 10-year actuarial failure-free survival of 94% ± 2%. These outcomes were achieved in a broad spectrum of single-ventricle anatomies, including hypoplastic left heart syndrome in 40% of the patients. Despite excellent survival, patients were at risk for ongoing morbidity. Among the 251 survivors, there were 6 late deaths and 43 patients were identified as having morbid events including development of decreased functional status, protein-losing enteropathy, arrhythmias, conduction abnormalities, stroke, and thrombus formation. The event-free survival (death, transplant, Fontan takedown, NYHA class III–IV, pacemaker, antiarrhythmic medication, protein-losing enteropathy, stroke, or thrombus) was 64% ± 6% at 10 years. Among the factors analyzed, we identified longer cross-clamp time, presence of a fenestration, and hospital length of stay as factors associated with poorer event-free survival.
Other authors have identified longer cross-clamp time as an important factor in Fontan outcome [7–10]. Longer cross-clamp time may be a surrogate for more complex anatomy or may be associated with additional procedures such as atrioventricular valve repair for insufficiency that add to the overall risk. In addition, a longer cross-clamp time may result in more diastolic dysfunction that could adversely impact hemodynamics, resulting in poorer event-free survival. In our experience, nearly half of the patients had no cross-clamp. In these patients, the myocardium was perfused with mildly hypothermic blood during the intracardiac portion of the procedure as the heart was either allowed to beat while decompressed or ventricular fibrillation was induced. A strategy of empty beating heart or ventricular fibrillation may be associated with less myocardial edema than cardioplegia and may result in less diastolic dysfunction. While the intracardiac portion of these procedures was likely to be short and not involve additional complex procedures, it is noteworthy that event-free survival for these patients was similar to that of patients who had a short cross-clamp time.
Not surprisingly, hospital length of stay was identified as a factor predicting poorer event-free survival and, like longer cross-clamp time, reflects a higher risk patient and patients with important complications. Hospital length of stay is also collinear with other risk factors, particularly duration of pleural drainage. Prolonged pleural drainage after the Fontan procedure was found to predictive of the development of protein-losing enteropathy [10].
The use of fenestration as part of the Fontan procedure has been associated with shorter duration of pleural effusions and shorter hospital length of stay [11, 12]. Fenestration use in this study was applied selectively, and the vast majority of patients left the operating room with a patent fenestration. Interestingly, event-free survival was superior among the 38 patients who underwent a Fontan with immediate intraoperative fenestration closure. There did not appear to be any specific event that predominated among patients who had a fenestration. We also entered ventricular morphology into the multivariable model and found that this did not change the impact of fenestration.
The potential benefits of a fenestration include a lower central venous pressure and better single ventricle preload, albeit at the expense of a right to left shunt and some increased cyanosis. These benefits would be expected to be greatest in the immediate postoperative period when, as a consequence of cardiopulmonary bypass and myocardial ischemia, myocardial function would be impaired and elevated pulmonary vascular resistance can be anticipated. Outside of the immediate postoperative period, these benefits may not outweigh the risks of persistent cyanosis and the potential for paradoxic embolism. Several studies have demonstrated that, in patients undergoing catheter-based device closure of a fenestration in the catheterization laboratory, late fenestration closure acutely results in increased arterial saturation but decreased cardiac output. The increase in arterial saturation is not enough to overcome the decrease in cardiac output, and as a result, there is a decrease in systemic oxygen delivery [13–16]. In addition, studies suggest a patent fenestration is of long-term benefit to the Fontan patient [17, 18]. Additional support for the benefit of persistent fenestration comes from small series in which a fenestration was created to improve patients with failing Fontan physiology [19, 20]. Alternatively, other studies indicate that after fenestration closure, in addition to a predictable increase in arterial saturation, patients have improved exercise tolerance, better somatic growth, and reduction in the use of cardiac medications [21, 22]. These studies suggest that a patent fenestration and persistent cyanosis can impair functional status. Taken together, these studies imply that a patent fenestration is not universally beneficial after the Fontan procedure.
Among our patients, fenestration closure in the operating room was performed in only 15% of the patients after weaning from cardiopulmonary bypass and only after observing the hemodynamics and postoperative transesophageal echocardiography. Fenestration closure testing was not attempted in the majority of patients, and no precise criteria were used to identify which patients were candidates for immediate fenestration closure. The survival curves comparing patients with and without a fenestration did not diverge until approximately 4 years after the Fontan procedure. There is substantial somatic growth during this time, and that might result in important reduction in the relative size of the fenestration. In addition, the fenestration was intentionally closed in some patients and spontaneously in some patients. Perhaps the lack of an adequate-size or patent fenestration is the factor that is responsible for the divergence of the survival curves. There are inadequate data at this point to determine the impact of persistent fenestration patency on outcome; all we can say is that for a select group of patients who left the operating room without a patent fenestration, event-free survival was superior compared with that of patients who had a patent fenestration.
This experience with the Fontan procedure includes a majority of patients with a dominant RV. Experience with corrected transposition and the atrial switch for D-transposition has shown that the RV has a substantial failure rate when placed in the systemic circulation [23, 24]. The crescent-shaped RV appears less suited to pressure work than the prolate spheroid-shaped LV, and the tricuspid valve with attachments to both the septum and free wall appears more prone to insufficiency as a consequence of ventricular dilation. Several groups have reported that patients with a dominant RV may be at increased risk for post-Fontan complications [25–27]. To further evaluate the right ventricle as the power source for the Fontan circulation, we undertook a separate analysis. Compared with patients with a dominant LV, patients with a dominant RV were younger and smaller at Fontan, more likely to have had a Norwood procedure, and had a longer cross-clamp time. Despite the higher incidence of these potential risk factors in the dominant RV group, these patients behaved similarly to patients with a dominant LV. Event-free survival was not different between groups. These findings are in agreement with two recent studies indicating that early and intermediate results were similar between patients with a dominant RV and a dominant LV [14, 28].
There are several explanations for the generally equivalent event-free survival of the dominant LV and RV. First, the etiology of factors impacting Fontan outcome such as rhythm disturbances, the development of protein-losing enteropathy, stroke, and thrombus are unknown, and these unknown factors may overwhelm the impact of ventricular morphology. Another explanation may be that the staged approach to single-ventricle palliation ameliorates the impact of ventricular morphology. Specifically, hypoplastic left heart syndrome is the most common single-ventricle diagnosis, and these patients are subject to a treatment algorithm that benefits from the increased experience with this diagnosis. Patients with hypoplastic left heart syndrome undergo predictable neonatal palliation, then undergo relatively early CP shunt that both relieves single-ventricle volume overload and improves cyanosis. These patients may be better prepared for the Fontan procedure than many patients with dominant LV, who may have variable anatomy, may present outside of the neonatal period, and as a consequence, may be subjected to greater periods of volume overload or cyanosis. Finally, the follow-up in this study and others identifying a similar behavior of the single RV and LV are relatively short, averaging just over 4 years, and the impact of ventricular morphology may occur outside of this limited window [10]. With time and accumulation of additional patients, a difference in failure-free survival and event-free survival may become apparent.
Limitations
This was a retrospective analysis and has all of the limitations of a retrospective study, including those outlined above. In particular, we cannot state with certainty that a fenestration results in poorer event-free survival. This question could only be answered by a randomized prospective study.
In conclusion, the Fontan procedure after interval CP shunt is associated with excellent operative and intermediate survival. Ventricular morphology does not appear to impact outcome.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
The first question is, as I understand the conclusions here, right ventricles do just as well as left ventricles, long cross-clamp times are bad, and open fenestrations may be bad. So the right ventricle versus the left ventricle and cross-clamp times, talk a little bit about cross-clamp times. You have shown in your manuscript, and you didn't spend a lot of time on it here, that times over 37 and a half minutes seemed to be associated with worse outcomes. When I thought about that, it made me think that you were doing other procedures, or Dr Tweddell and his partners were doing other procedures when they were doing these Fontan operations, and that perhaps the risk factor here was not some bugaboo about being at 35 or 40 or 45 or 30 minutes but in fact that there was something wrong with this heart or this circulatory system that required more work at the time of the Fontan. So my first question for you is, have you looked at that specific variable, additional procedures along with cross-clamp time, to see if it is the additional procedures perhaps rather than length of time that the cross-clamp is on?
The other question I had for you has to do with the fenestration, the open versus closed fenestration, and I know that this was not a prospective randomized study, so conclusions are hard to draw on, but it gives us more opportunity to speculate. You have shown here that having an open fenestration seems to be worse than having a closed fenestration, but the two survival curves that you showed, open versus closed, were the same at 5 years and only diverged later. So I would suggest that maybe having an open fenestration is a good thing, and it neutralized the risk factor of being a bad Fontan candidate, and eventually when you outgrew your fenestration or it closed that that advantage was lost and then you saw the divergence. So could you comment on that as well? And I will walk away, and thank you very much for your wonderful work.
MR NERESIAN: Thank you, Dr Jaquiss. Your first question, are additional procedures being a risk factor—I believe that we did have those data, but I am not quite sure if we have analyzed them yet. Doctor Tweddell, do you know?
DR TWEDDELL: I am certain that a longer cross-clamp time was a surrogate for a more complex operation. We did not analyze a specific interaction of an additional procedure and cross-clamp time, and I think that is a subanalysis that we probably should do.
MR NERESIAN: And, I am sorry, what was the exact second question?
DR JAQUISS: I think that having an open fenestration seems to be working than having a closed fenestration, at least when you compare the two curves. That is not true for the first 5 years. So maybe you are just outgrowing your fenestration in 5 years.
DR TWEDDELL: Fifty-nine patients underwent intentional fenestration closure, and some others, a handful, closed spontaneously. The impact of fenestration closure on event-free survival is a subanalysis that we have yet to perform. It is an obvious question but it didn't occur to us until very late in the analysis, and exactly how we are going to account for that I am not certain at this point. I must admit I hadn't thought about the particular idea that patients are outgrowing their fenestration as the explanation for our findings.
DR EDWARD L. BOVE (Ann Arbor, MI): Matthew, I enjoyed the paper. It was very well done. At the last American Surgical Association meeting, our group from Ann Arbor presented our results on 636 Fontan procedures that we did over the last 15 years, and the results are remarkably similar to yours. We did not find a difference in outcome between a morphologically dominant right versus a morphologically dominant left ventricle. We also did not find an incremental risk for associated procedures, including valvular procedures, at the time of the Fontan.
Having said that, I rise to ask you one question. One outcome we did find and perhaps didn't expect is that among the patients that had prolonged chest tube drainage, they were statistically more likely to develop PLE as the years went by. Did you look at that? You have a large series of patients and I am curious to know if you found anything similar. Thank you.
MR NERESIAN: We did look at PLE, but I am not quite sure if we looked at the comparison of PLE with prolonged chest tube drainage.
DR TWEDDELL: Pleural drainage was one of the 21 factors that we looked at, and it did not correlate with the composite morbid outcome, and I think that would have picked up a correlation with PLE.
DR JOSEPH FORBESS (Dallas, TX): I enjoyed the presentation very much as well. I just have one additional question. Do you have any sense as to whether or not there was any disproportionate dropout, pre-Fontan, of the morphologic right ventricles? I note that this is an analysis of patients who met your criteria to have a Fontan done. Was there significant attrition before getting to the Fontan in either of the morphologic subgroups?
MR NERESIAN: Thank you for that question. I know that we recently published an article looking at hypoplastic left hearts, and out of 116, I believe 91 of those either had a Fontan procedure or was a candidate for a Fontan. As far as having a left ventricular morphology, I am not quite sure. Doctor Tweddell?
DR TWEDDELL: We have published a great deal on interstage management of hypoplastic left heart syndrome as Matthew said, our most recent analysis of hypoplasts undergoing a Norwood with a BT shunt, there were 116 patients and 90 have made it to a Fontan or are Fontan candidates still. That is the only subgroup we have analyzed to that degree, so I can't really tell you what the remainder of the group looks like. The purpose of this paper was really to look at Fontan outcome and a cross-sectional follow-up of all single ventricles.
DR ROBERTO CANESSA (Montevideo, Uruguay): Right ventricles don't fail in Fontans, and they fail in corrected transposition and in Senning. It is a matter of time. We haven't reached the age. Because I have seen that corrected transposition, sometimes in the early ages they fail.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Monagle, A. K. C. Chan, N. A. Goldenberg, R. N. Ichord, J. M. Journeycake, U. Nowak-Gottl, and S. K. Vesely Antithrombotic Therapy in Neonates and Children: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines Chest, February 1, 2012; 141(2_suppl): e737S - e801S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Brosig Commentary: The Use of Health and Behavior Codes in a Pediatric Cardiology Setting J. Pediatr. Psychol., January 19, 2012; (2012) jsr125v1. [Full Text] [PDF]
|
|
|
|
 |
|
 P. Monagle, A. Cochrane, R. Roberts, C. Manlhiot, R. Weintraub, B. Szechtman, M. Hughes, M. Andrew, B. W. McCrindle, and Fontan Anticoagulation Study Group A Multicenter, Randomized Trial Comparing Heparin/Warfarin and Acetylsalicylic Acid as Primary Thromboprophylaxis for 2 Years After the Fontan Procedure in Children J. Am. Coll. Cardiol., August 2, 2011; 58(6): 645 - 651. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Canter Preventing Thrombosis After the Fontan Procedure: Not There Yet J. Am. Coll. Cardiol., August 2, 2011; 58(6): 652 - 653. [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Wallace, J. Jaggers, J. S. Li, M. L. Jacobs, J. P. Jacobs, D. K. Benjamin, S. M. O'Brien, E. D. Peterson, P. B. Smith, and S. K. Pasquali Center Variation in Patient Age and Weight at Fontan Operation and Impact on Postoperative Outcomes Ann. Thorac. Surg., May 1, 2011; 91(5): 1445 - 1452. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kutty, M. A. Frommelt, D. A. Danford, and J. S. Tweddell Medium-Term Outcomes of Kawashima and Completion Fontan Palliation in Single-Ventricle Heart Disease With Heterotaxy and Interrupted Inferior Vena Cava Ann. Thorac. Surg., November 1, 2010; 90(5): 1609 - 1613. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sakurai, H. Kado, T. Nakano, K. Hinokiyama, S. Oda, J. Sugiura, T. Ushijima, and Y. Ueda The impact of extracardiac conduit-total cavopulmonary connection on apicocaval juxtaposition Eur J Cardiothorac Surg, October 1, 2010; 38(4): 439 - 444. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. D. Simon, J. Berry, C. Feudtner, B. L. Stone, X. Sheng, S. L. Bratton, J. M. Dean, and R. Srivastava Children With Complex Chronic Conditions in Inpatient Hospital Settings in the United States Pediatrics, October 1, 2010; 126(4): 647 - 655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sathanandam, A. C. Polimenakos, C. Blair, C. El Zein, and M. N. Ilbawi Hypoplastic Left Heart Syndrome: Feasibility Study for Patients Undergoing Completion Fontan at or Prior to Two Years of Age Ann. Thorac. Surg., September 1, 2010; 90(3): 821 - 829. [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 |
