| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2007;84:880-887
© 2007 The Society of Thoracic Surgeons
a Department of Cardiology, Children’s Hospital, Boston, Massachusetts
b Department of Cardiac Surgery, Children’s Hospital, Boston, Massachusetts
a Department of Pediatrics, Harvard Medical School, Boston, Massachusetts
b Department of Surgery, Harvard Medical School, Boston, Massachusetts
Accepted for publication February 28, 2007.
* Address correspondence to Dr McElhinney, Department of Cardiology, Children’s Hospital, 300 Longwood Ave, Boston, MA 02115 (Email: doff.mcelhinney{at}cardio.chboston.org).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
Methods: Patients followed up at Children’s Hospital Boston who underwent takedown of a Fontan circulation to an intermediate palliative circulation within 1 year of Fontan completion were reviewed.
Results: Between 1979 and 2006, 53 patients underwent Fontan takedown at a median age of 2.3 years (range, 0.3 to 36.5 years). Takedown was performed during the Fontan procedure itself in 12 patients (22%), within the first postoperative month in 31(58%), and between 1 month and 1 year in 10 (18%). Overall, 29 patients (55%) survived the early period after takedown, and 19 ultimately underwent successful Fontan completion a median of 4.6 years after takedown; all but one was alive a median of 6.4 years later. Thirteen (68%) of the 19 had treatable abnormalities contributing to Fontan failure.
Conclusions: Fontan takedown can provide effective stabilization of the acutely or subacutely failing Fontan circulation, although a substantial number of patients die early despite Fontan takedown. Subjects surviving the perioperative period can often undergo uneventful redo Fontan. A thorough evaluation for treatable abnormalities should be performed in all patients with a failing Fontan circulation and in patients who undergo Fontan takedown.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Takedown of the Fontan circulation—performed either acutely or subacutely—can provide temporary stabilization of the failing single-ventricle circulation. Little is known about the long-term outcome of these patients, however. Moreover, little is known about the best approach to subsequent management, including which patients may be candidates for a second attempt at Fontan completion despite a previous failure.
Other options include heart transplantation or extended intermediate palliation with a cavopulmonary anastomosis or an additional systemic-to-pulmonary arterial shunt, or both [10–16]. Yet, Fontan completion may be preferable if the underlying cause(s) of the prior Fontan failure can be treated or prevented such that a second attempt at Fontan can be performed with acceptable risk. The purpose of this study is to better understand the outcome of patients undergoing Fontan takedown and to determine which patients may be suitable candidates for second attempt at Fontan completion.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Data Analysis
Data were reviewed from patient records in accordance with a protocol approved by the Committee for Clinical Investigation, Children’s Hospital Boston, which waived the need for parental consent for patients reviewed in this study. Continuous and categoric variables between groups were compared using the appropriate parametric or nonparametric tests, as indicated. Multivariable logistic regression analysis was performed using stepwise forward entry. Data are presented as mean ± standard deviation, median (range), or number (frequency). Early death was defined as any death occurring within 30 days of Fontan takedown or before discharge.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Among 43 patients in whom the Fontan circulation was taken down acutely, a median of 1 day had elapsed since the Fontan operation (0–27 days), including 12 patients who underwent takedown during the same operation as the Fontan procedure itself. Among 10 patients who underwent subacute Fontan takedown, the median duration between the Fontan operation and takedown was 3.1 months (range, 1 to 12 months).
|
|
|
Sources of pulmonary blood flow immediately after Fontan takedown are summarized in Table 2. Additional procedures at the time of Fontan takedown were performed in 4 patients, including pulmonary artery and Fontan thrombectomy in 3, pulmonary vein augmentation in 1, and branch pulmonary artery augmentation in 1.
Acute Survival after Fontan Takedown
Twenty- four (45%) of the 53 patients died early after Fontan takedown, including 22 (51%) of 43 patients who underwent acute Fontan takedown and 2 (20%) of 10 in whom the takedown was subacute (Fig 1). Three of the 8 patients who underwent subacute takedown for intractable effusions had a nonfenestrated Fontan, and 7 survived. Fourteen of these 24 patients died on the same day as Fontan takedown; the other 10 died between 1 and 48 days after takedown. Patients who died early after Fontan takedown were more likely to have undergone Fontan completion before 1990, to have a circulation other than bidirectional cavopulmonary anastomosis before Fontan completion, to have an atriopulmonary Fontan connection, to have a fenestrated Fontan procedure, and to have a morphologic right ventricle as the dominant/single systemic ventricle (Table 2).
|
Follow-Up After Fontan Takedown
The 29 patients who survived the early posttakedown period were followed up for a median of 9.7 years (range, 0.4 to 20.4 years), during which time 3 patients died at 0.4, 11.8, and 18.5 years after takedown. One or more interventions after takedown but before redo Fontan completion or heart transplant (if applicable) were performed in 18 patients, and included coil embolization of arterial or venous collaterals, or both in 9, placement of a systemic-to-pulmonary arterial shunt in 7, balloon angioplasty/stenting of branch pulmonary arterial obstruction in 6, balloon dilation/stenting of recurrent aortic arch obstruction in 3, 2 each with reconstruction of discontinuous branch pulmonary arteries, superior cavopulmonary anastomosis, and placement of a permanent pacemaker; and 1 each with placement of a brachial arteriovenous fistula, stenting of a systemic-to-pulmonary arterial shunt, and stenting of an obstructed innominate vein.
Redo fontan completion
At the time of most-recent follow-up, 19 (66%) of 29 patients had undergone redo Fontan completion a median of 4.6 years (range, 0.6 to 13.1 years) after takedown. This included 3 of 7 patients with a previous atriopulmonary-type Fontan and 15 of 22 patients with a previous extracardiac or lateral tunnel-type Fontan. The redo Fontan procedure was a lateral tunnel in 16 patients and an extracardiac or intraatrial conduit Fontan connection in 3; all were fenestrated with a single 4-mm fenestration. Additional interventions at the time of redo Fontan were performed in 8 patients, including branch pulmonary arterioplasty in 4, augmentation of the pulmonary venous pathway in 2, and in 1 patient each, reconstruction of discontinuous branch pulmonary arteries, pulmonary valve excision, and tricuspid valvuloplasty.
On catheterization before redo Fontan, the median pulmonary artery pressure was 13 mm Hg (mean range, 7 to 17 mm Hg), pulmonary vascular resistance was 1.8 WU (range, 0.5 to 3.6 WU), ventricular end-diastolic pressure was 9 mm Hg (range, 4 to 12 mm Hg), and systemic arterial oxygen saturation was 82% (range, 76% to 88%).
The median hospital stay after redo Fontan completion was 9 days, and all but 2 patients were discharged after an uncomplicated perioperative course within 2 weeks of surgery. Two patients had protracted in hospital convalescence; in 1 this was related primarily to complications of stroke and infection and the other had persistent chylothorax.
At a median follow-up of 6.4 years (range, 0.5 to 14.3 years) after the redo Fontan operation, 18 of these 19 patients were alive with a Fontan circulation. One patient required takedown 6 months after the redo Fontan for persistent pleural effusions and ascites, which was also the reason for the first Fontan takedown; this patient died 4 months after takedown from complications of intractable protein-losing enteropathy. Interventions after successful Fontan completion occurred in 11of the remaining 18 patients, including device closure of the Fontan fenestration in 9, embolization of venous or arterial collaterals in 4, balloon dilation or stenting of the branch pulmonary arteries or Fontan pathway, or both, in 2; dilation of a Fontan fenestration in 2, closure of a baffle leak in 1, and placement of a permanent pacemaker in 1. The median arterial oxygen saturation at most recent follow-up was 92% (range, 86% to 96%).
Among the 18 patients who underwent successful redo Fontan completion, 13 were identified to have correctable/treatable abnormalities that may have contributed to the initial Fontan failure, including branch pulmonary artery stenosis or discontinuity in 8, ventricular inflow obstruction by the intraatrial Fontan pathway in 3, arrhythmia (junctional ectopic tachycardia or intraatrial reentrant tachycardia) in 3, hemidiaphragm paralysis/paresis in 2, and in 1 patient each, pulmonary venous obstruction, unilateral pulmonary arteriovenous malformations, large cavopulmonary thrombus, and recurrent arch obstruction. All of these either resolved or improved before redo Fontan (eg, arrhythmia, hemidiaphragm paresis), were treated successfully before or at the time of redo Fontan completion (eg, branch pulmonary artery dilation/stenting or reconstruction, dilation and stenting of arch obstruction, repair of pulmonary venous obstruction, brachial arteriovenous fistula for arteriovenous malformations, thrombectomy), or were avoided by modifying the Fontan connection at the redo procedure (eg, extracardiac conduit Fontan or atrial augmentation to avoid pulmonary venous pathway obstruction).
Heart transplantation
During follow-up, 3 patients underwent cardiac transplantation 0.5 to 14 years after takedown. All patients had persistent ventricular dysfunction after the Fontan takedown. Two patients were alive 8 and 10 months after transplantation, and the other died 4.5 years after transplantation.
Intermediate palliative circulation
At the most recent follow-up at a median of 3.2 years (range, 2 to 11.8 years), 7 patients remained with an intermediate palliative circulation and had not undergone a repeat attempt at Fontan completion or transplantation. One died 5 months posttakedown after placement of an additional systemic-to-pulmonary arterial shunt. The circulation at follow-up in the remaining 6 patients was a bidirectional cavopulmonary anastomosis in 3, and a classic Glenn shunt in the other 3, 1 of whom has antegrade flow the left lung; whereas the other 2 had no source of left lung blood flow other than systemic-to-pulmonary arterial collaterals. Five of these 6 patients were unfavorable candidates for a redo Fontan procedure in light of the prior Fontan failure and pulmonary vascular anomalies that would essentially leave them with a one-lung Fontan circulation: either discontinuous branch pulmonary arteries with inability to recruit the left pulmonary artery in 3, left pulmonary vein atresia and left pulmonary artery stenosis in 1, and scimitar syndrome with severe right lung and right pulmonary artery/vein hypoplasia in 1. The other patient lived abroad, and remained with a bidirectional cavopulmonary anastomosis at follow-up limited to 1.5 years post-takedown.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
As reported in a number of surgical series [3, 4, 6, 8, 9], a substantial proportion of patients (46%) who underwent acute or subacute Fontan takedown in our series did not survive the perioperative period. Similar to reported improvements in Fontan outcome during the past 10 to 15 years, acute survival among patients after Fontan takedown in our series was significantly better after 1990 than before. This transition also coincided with the routine staging to Fontan completion with a bidirectional cavopulmonary anastomosis and the switch from atriopulmonary modifications of the Fontan procedure to the fenestrated lateral tunnel Fontan connection. Patients with a dominant single right ventricle also had worse acute survival after takedown, and although this study was not designed primarily to evaluate factors associated with acute posttakedown survival, no differences existed in pre-Fontan hemodynamics between acute survivors and nonsurvivors.
The subset of patients in whom the Fontan takedown was performed in a subacute timeframe differed clinically from patients who underwent acute Fontan takedown and were generally not acutely ill. Most of these patients presented with intractable effusions, and Fontan takedown was elected after medical and sometimes other surgical management had failed. Overall, posttakedown survival was better among patients who underwent subacute takedown, although this difference did not reach statistical significance, partly owing to the small number of patients. Only 1 of the 8 patients who underwent subacute takedown for intractable effusions did not survive (Table 3).
Outcomes After Fontan Takedown
The four basic outcomes in our cohort were (1) death, (2) survival after a subsequent (redo) Fontan procedure, (3) survival after heart transplant, and (4) extended palliation with an intermediate circulation. Among the 28 patients (54%) who survived acutely after Fontan takedown and were followed up for a median of 9.7 years, there were three deaths: one each after heart transplant, after failure of a second attempt at Fontan completion, and after a subsequent palliative operation. Nearly two-thirds of these early posttakedown survivors (n = 19) underwent a second attempt at Fontan completion between 0.6 and 13.1 years after the initial Fontan failure and takedown. Among early survivors who had originally undergone a lateral tunnel or extracardiac type Fontan procedure, 14 of 22 underwent successful redo Fontan, with all but 1 patient alive at follow-up.
Most of the patients (74%) undergoing redo Fontan had correctable, treatable, or avoidable anatomic, medical, or technical factors that were thought to contribute to the initial Fontan failure—including pulmonary arterial obstruction, perioperative arrhythmias, obstruction to ventricular inflow by the intraatrial Fontan pathway, pulmonary arteriovenous malformations, hemidiaphragm paralysis, and thrombus formation in the Fontan pathway—but 5 patients did not. All but 1 of the 19 patients who underwent a second attempt at Fontan completion survived and are alive with a Fontan circulation. The perioperative course after redo Fontan completion was uneventful in all but 2 patients: 1 had prolonged chylothorax that ultimately improved (the initial Fontan failure in this patient was acute and not due to persistent effusions) and 1 died after failure and takedown of a second Fontan procedure.
Only 3 patients in our series underwent heart transplantation. All of these patients had persistent ventricular dysfunction after Fontan takedown, in addition to other contributing abnormalities. Some investigators have advocated a "rescue transplant" for patients with acute Fontan failure [15], but this may not be the best option in many cases. Even in patients with evidence of ventricular dysfunction, peribypass ventricular and pulmonary vascular dysfunction may resolve and subsequent Fontan completion may be an option. Moreover, in a multiinstitutional study of 97 patients with a failing Fontan circulation who were listed for heart transplant, the Pediatric Heart Transplant Study Group found that shorter duration after the Fontan procedure, younger age, status I listing, and ventilator dependence (most of which are likely to be the case in patients with acute Fontan failure) were all risk factors for worse survival after transplant [16].
At the most recent follow-up, 7 patients in this series had not undergone redo Fontan or transplantation. One died several months after Fontan takedown after placement of an additional systemic-to-pulmonary arterial shunt. In 5 of the 6 surviving patients with an intermediate palliative circulation at most recent follow-up, 3 had functional discontinuity of the branch pulmonary arteries and 2 had significant unilateral hypoplasia/stenosis of the pulmonary arteries and veins, with unsuccessful recruitment of the affected lung into a cavopulmonary circulation. Successful Fontan completion in patients with discontinuous pulmonary arteries or unilateral pulmonary artery atresia is possible [21, 22], and although we have performed successful single-lung Fontan procedures, these 5 patients all failed Fontan completion once and were considered poor candidates for another attempt.
Importance of Pulmonary Vascular Abnormalities
Among the correctable abnormalities in patients who ultimately underwent successful redo Fontan completion, pulmonary artery stenosis or discontinuity was the most common. Pulmonary artery size was one of the original criteria specified by Fontan and Choussat for Fontan candidacy [23], and a number of studies have found that pulmonary artery size or abnormalities are important determinants of outcome [8, 20, 24, 25]. Although other subsequent reports have challenged the significance of pulmonary artery size per se in affecting Fontan outcome [21, 26], pulmonary vascular anatomy and function are among the most important factors in considering the risk of a Fontan procedure in any given patient. Pulmonary artery stenosis can be more difficult to identify in patients with a shunted or low-flow, nonpulsatile circulation, and particular attention should be paid to identifying pulmonary vascular anomalies, including pulmonary artery stenosis or hypoplasia, pulmonary venous obstruction, or pulmonary arteriovenous malformations.
Pulmonary venous obstruction may be more subtle than pulmonary arterial anomalies and require more focused evaluation, and occasionally, it is treatable in a univentricular circulation. Often however, pulmonary venous obstruction is not easily correctable—it was a contributing factor in the Fontan failure in 5 of our patients, only 1 of whom underwent redo Fontan completion.
Unilateral pulmonary arteriovenous malformations are an unusual pulmonary vascular complication after Fontan completion but may result from unilateral streaming of hepatic venous effluent or occasionally be unrecognized before Fontan completion [27]. In such cases, the addition of a brachiocephalic arteriovenous fistula or systemic-to-pulmonary artery shunt to a bidirectional cavopulmonary anastomosis may facilitate resolution of pulmonary arteriovenous malformations and possible subsequent successful Fontan completion [27].
Eight of the patients who underwent Fontan takedown in this series (5 of whom survived acutely) had discontinuous branch pulmonary arteries. In 7 of these 8 patients, pulmonary artery continuity was restored at the time of the original Fontan operation, 4 of whom survived acutely after Fontan takedown. Among the 4 acute survivors of takedown who had undergone reconstruction of discontinuous branch pulmonary arteries during the Fontan operation, the pulmonary arteries were discontinuous at the time of takedown. This finding is consistent with previous series that reported that reconstruction of discontinuous pulmonary arteries at the time of Fontan operation often does not result in a patent connection, although patients may survive with a one-lung Fontan circulation [21, 22].
Implications
These findings are encouraging and have several potential implications. First, takedown of an acutely or subacutely failing Fontan circulation may facilitate stabilization of a tenuous circulation and contribute to survival, and as such may be the best management option in certain patients with a failing Fontan. Second, many patients with acute or subacute failure of a Fontan circulation have treatable abnormalities that may affect their suitability for a Fontan operation. Third, failure of Fontan completion does not necessarily mean that a patient cannot tolerate a Fontan circulation, and such patients should be thoroughly evaluated for correctable or occult abnormalities. Fourth, even in patients with ventricular dysfunction who may not be good candidates for a second attempt at Fontan completion, takedown to an intermediate circulation may allow improvement in the patient’s overall circulatory status to enable listing for heart transplant in a more stable condition.
Limitations
This is a retrospective evaluation of outcomes after Fontan takedown, which is performed infrequently and has become even less common with improvement in Fontan outcomes. The relative infrequency of acute or subacute Fontan failure in the current era, however, does not obviate a better understanding of this patient population. This study was not designed to assess risk factors for acute or subacute Fontan failure. Fontan failure and takedown has been considered in prior reports as akin to failure and death, and although patients who survive after Fontan takedown may differ from those who experience Fontan failure and die after Fontan completion (with or without takedown), this study was not intended to address this issue. Because patients who underwent takedown, particularly acute takedown, were generally judged to be incapable of survival with a Fontan circulation, our assumption is that early posttakedown deaths were more in spite of than because of Fontan takedown, although we recognize that the takedown procedure carries morbidity in its own right.
Similarly, this study does not address the question of whether Fontan takedown, extracorporeal mechanical support, transplantation, or any other option is most likely to result in survival of the patient with acutely or subacutely failing Fontan circulation. Despite these limitations, the findings of this study may help contribute to our understanding and to better management and outcomes in the increasingly small subset of patients who do not tolerate a Fontan circulation in the early postoperative period.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. A. Harris, M. T. Cosulich, M. J. Gillespie, K. K. Whitehead, T. I. Liu, P. M. Weinberg, and M. A. Fogel Pre-Fontan cardiac magnetic resonance predicts post-Fontan length of stay and avoids ionizing radiation J. Thorac. Cardiovasc. Surg., October 1, 2009; 138(4): 941 - 947. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Bacha, P. Lang, J. E. Mayer Jr, and D. B. McElhinney Connection of Discontinuous Pulmonary Arteries in Patients With a Superior or Total Cavopulmonary Circulation Ann. Thorac. Surg., December 1, 2008; 86(6): 1948 - 1954. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Salvin, M. A. Scheurer, P. C. Laussen, J. E. Mayer Jr, P. J. del Nido, F. A. Pigula, E. A. Bacha, and R. R. Thiagarajan Factors Associated With Prolonged Recovery After the Fontan Operation Circulation, September 30, 2008; 118(14_suppl_1): S171 - S176. [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 |
