If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Queensland Paediatric Cardiac Services, Queensland Children's Hospital, Brisbane, AustraliaSchool of Clinical Medicine, Children's Health Queensland Clinical Unit, University of Queensland, Brisbane, Australia
Royal Children's Hospital, Melbourne, AustraliaGreen Lane Paediatric and Congenital Cardiac Service, Starship Hospital, Auckland, New ZealandMurdoch Children's Research Institute, Melbourne, AustraliaFaculty of Medicine, Department of Paediatrics, University of Melbourne, Melbourne, Australia
Murdoch Children's Research Institute, Melbourne, AustraliaFaculty of Medicine, Department of Paediatrics, University of Melbourne, Melbourne, Australia
Queensland Paediatric Cardiac Services, Queensland Children's Hospital, Brisbane, AustraliaSchool of Clinical Medicine, Children's Health Queensland Clinical Unit, University of Queensland, Brisbane, Australia
Queensland Paediatric Cardiac Services, Queensland Children's Hospital, Brisbane, AustraliaSchool of Clinical Medicine, Children's Health Queensland Clinical Unit, University of Queensland, Brisbane, Australia
Heart Centre for Children, The Children's Hospital at Westmead, Sydney, AustraliaFaculty of Health and Medicine, Sydney Medical School, Discipline of Child and Adolescent Health, University of Sydney, Sydney, AustraliaSydney Children's Hospital Network Cardiac Services, Sydney, Australia
Royal Children's Hospital, Melbourne, AustraliaMurdoch Children's Research Institute, Melbourne, AustraliaFaculty of Medicine, Department of Paediatrics, University of Melbourne, Melbourne, Australia
Address for reprints: Nelson Alphonso, FRACS, Cardiac Services, Clinical Directorate 7f, Queensland Children's Hospital, PO Box 3474, Stanley St, South Brisbane, 4101, Australia.
Queensland Paediatric Cardiac Services, Queensland Children's Hospital, Brisbane, AustraliaSchool of Clinical Medicine, Children's Health Queensland Clinical Unit, University of Queensland, Brisbane, Australia
Fontan takedown remains an option for the management of Fontan failure. We sought to evaluate early and late outcomes after Fontan takedown.
Methods
The Australia and New Zealand Fontan Registry was interrogated to identify all patients who had a Fontan takedown.
Results
Over a 43-year study period (1975-2018), 36 of 1540 (2.3%) had a Fontan takedown. The median age at takedown was 5.1 years (interquartile range [IQR], 3.7, 7.0). Nine (25%) patients had a takedown within 48 hours, 6 (16%) between 2 days and 3 weeks, 14 (39%) between 3 weeks and 6 months, whereas 7 (19%) had a late takedown (>6 months). Median interval to takedown was 26 days (IQR, 1.5, 127.5). Sixteen (44%) patients died at a median of 57.5 days (IQR, 21.8, 76.8). The greatest mortality occurred between 3 weeks and 6 months (<2 days: 1/9, 11%; 2 days to 3 weeks: 2/6, 33%; 3 weeks to 6 months: 11/14, 79%; >6 months: 2/7, 28%; P = .007). At median follow-up of 9.4 years (IQR, 4.5, 15.3), 11 (31%) patients were alive with an intermediate circulation (10 in New York Heart Association class I/II). Five (14%) patients underwent a successful second Fontan. Freedom from death/transplant after Fontan takedown was 59%, 56%, and 52% at 1, 5, and 10 years, respectively.
Conclusions
The incidence of Fontan takedown is low, but mortality is high. The majority of takedowns occurred within 6 months. Mortality was lowest when takedown occurred <2 days and highest between 3 weeks and 6 months. A second Fontan is possible in a small proportion of survivors.
The incidence of Fontan takedown is low. Factors influencing takedown remain unclear. The majority of takedowns occurred within 6 months (80%). Mortality was lowest when takedown occurred <2 days (11%) and greatest between 3 weeks and 6 months (79%). Midterm outcomes after takedown to an intermediate circulation are acceptable. A second Fontan is possible in a small proportion of survivors.
Redefining expectations of long-term survival after the Fontan procedure: twenty-five years of follow-up from the entire population of Australia and New Zealand.
Early Fontan failure presents in a variety of ways, including low cardiac output state, high Fontan circuit pressures, large fluid requirements, and prolonged high-volume pleural effusions.
Late Fontan failure often presents as intractable arrhythmias, protein-losing enteropathy (PLE), plastic bronchitis, or poor functional capacity. Irrespective of the etiology or presentation, Fontan failure often leads to death and needs to be managed aggressively.
One of the options of managing Fontan failure is takedown of the Fontan circuit to an intermediate palliation stage in the single-ventricle pathway (bidirectional cavopulmonary shunt and/or systemic-pulmonary artery shunt).
The aim of our study was to identify and study all Fontan takedowns from the entire population of Australia and New Zealand over a period of 43 years.
Methods
Sample
The institutional review board approval number for this study is HREC/18/QRCH/105, dated March 28, 2018. Fontan takedowns were initially identified from the Australia and New Zealand Fontan Registry, which includes all patients undergoing the Fontan operation in either country.
However, Fontan patients are included in the Registry only after hospital discharge. Patients who underwent Fontan takedown in the same hospital admission were identified by interrogating individual hospital databases in each contributing pediatric cardiac center in Australia and New Zealand.
Takedowns were stratified into immediate, early, intermediate, and late groups depending on the interval since the Fontan operation (immediate <48 hours; early: 2 days to 3 weeks; intermediate: 3 weeks to 6 months; late: >6 months).
Statistical Analysis
Descriptive statistics are presented using percentages, median, and interquartile range (IQR). Statistical comparisons were performed with nonparametric tests including the Fisher exact binomial P values for categorical independent variables, and Wilcoxon rank-sum P values for continuous independent variables. A P value < .05 was considered statistically significant.
Results
Over a 43-year study period (1975-2018), 1540 patients underwent a Fontan procedure. Thirty-six (2.3%) patients had subsequent takedown of the Fontan.
Baseline Characteristics
Of the 36 patients who underwent Fontan takedowns, 25 (69.4%) were male (Table 1). Six patients (16.7%) had heterotaxy whereas 4 patients (11.1%) had dextrocardia. The dominant ventricular morphology was left ventricle in 17 patients (47.2%), right ventricle in 17 (47.2%), biventricular in 1 (2.8%), and indeterminate in 1 (2.8%). Underlying diagnoses were as follows: hypoplastic left heart syndrome (n = 10, 27.8%), double inlet left ventricle (n = 7, 19.4%), double outlet right ventricle (DORV) (n = 5, 13.9%), tricuspid atresia (n = 4, 11.1%), complete atrioventricular septal defect (n = 4, 11.1%), pulmonary atresia with intact ventricular septum (n = 3, 8.3%), complete atrioventricular septal defect/DORV (n = 2, 5.6%), congenitally corrected transposition of the great arteries (n = 1, 2.8%).
Table 1Baseline characteristics of patients undergoing Fontan takedown
Characteristic
n (%)
Male
25 (69.4%)
Heterotaxy
6 (16.7%)
Dextrocardia
4 (11.1%)
Dominant ventricular morphology
Left ventricle
17 (47.2%)
Right ventricle
17 (47.2%)
Biventricular
1 (2.8%)
Indeterminate
1 (2.8%)
Baseline diagnosis
HLHS
10 (27.8%)
DILV
7 (19.4%)
DORV
5 (13.9%)
Tricuspid atresia
4 (11.1%)
CAVSD
4, (11.1%)
PA-IVS
3 (8.3%)
CAVSD + DORV
2 (5.6%)
ccTGA
1 (2.8%)
Type of Fontan
AP
5 (14%)
LT
14 (39%)
ECC
17 (47%)
Era of Fontan takedowns and concomitant Fontans during the same period (%)
Nineteen (52.7%) patients had bidirectional cavopulmonary shunt (BCPS) whereas 2 (5.5%) had Hemi-Fontan as stage II procedure. Fifteen (41.6%) patients were single-stage Fontan procedures. The median age at stage II procedure was 5.4 months (IQR, 3, 16.8) whereas the median interval from stage II procedure to Fontan was 3.5 years (IQR, 2.7, 5.1). Five (14%) patients had an atriopulmonary Fontan, 14 (39%) had a lateral tunnel Fontan, and 17 (47%) had an extracardiac conduit Fontan (P = .039). Pre-Fontan hemodynamic data were available in 31 (86%) patients of which 27 (87%) had mean pulmonary artery pressures ≤15 mm Hg. No patient had moderate or severe ventricular dysfunction or moderate or greater atrio-ventricular valve regurgitation preoperatively. An examination of era distribution of Fontan takedowns as compared to total Fontan procedures during the same period revealed no era effect (P = .069) (Table 1).
Takedown Characteristics
The median age at takedown was 5.1 years (IQR, 3.7, 7.0). Nine (25%) patients had a Fontan takedown within 48 hours, 6 (16%) had an early takedown (2 days to 3 weeks), 14 (39%) had intermediate takedown between 3 weeks and 6 months, whereas 7 (19%) had a late takedown (>6 months). The median interval to Fontan takedown was 26 days (IQR, 1.5, 127.5). Indications for takedown were low cardiac output syndrome (19, 53%), intractable pleural effusions (12, 33%), PLE (3, 8%), thrombosis (1, 2.7%), and unknown (1, 2.7%).
Timing of Takedown
Immediate (<48 hours; n = 9; 25%)
All 9 patients developed intractable low cardiac output immediately after the Fontan. One patient could not be weaned off cardiopulmonary bypass and the Fontan was taken down intraoperatively.
Early (2 days to 3 weeks; n = 6, 16.7%)
All 6 patients presented with intractable low cardiac output. One patient also developed thrombosis of the Fontan circuit.
Intermediate (3 weeks to 6 months; n = 14; 38.9%)
Eleven of the 14 patients (79%) presented with intractable pleural effusions whereas 3 patients (21%) presented with low cardiac output. All 14 patients underwent multiple procedures in an attempt to ameliorate the pleural effusions including multiple chest tube insertions and pleurodesis.
Late (>6 months, n = 7, 19.4%)
These patients had a varied presentation (PLE, n = 3; low cardiac output, n = 1, pleural effusion, n = 1, thrombosis of the Fontan circuit, n = 1, unknown, n = 1). Overall, 80% of all Fontan takedowns occurred within 6 months of the index procedure.
Hemodynamics Post-Fontan Takedown
Post-Fontan takedown catheterization data were available in 11 (30%) patients. A catheter study was not performed in the remaining patients. Median pulmonary artery pressure was 13 mm Hg (IQR, 13, 18 mm Hg) whereas the median ventricular end diastolic pressure was 10 mm Hg (IQR, 8, 12 mm Hg). Median transpulmonary gradient was 4 mm Hg (IQR, 4, 5 mm Hg).
Immediate (n = 9). There was 1 death in a 5-year-old patient with a lateral tunnel Fontan despite mechanical support. A 4-year-old patient with DORV and hypoplastic left ventricle underwent an unsuccessful second attempt at the Fontan and ultimately underwent a heart transplant. Two patients underwent a successful second Fontan completion. Five patients are still alive with an intermediate circulation.
Early (n = 6)
Two patients died immediately following Fontan takedown (both procedures were performed in the 1980s). Two patients are alive with an intermediate circulation. Two patients underwent a successful Fontan at the second attempt 4.4 and 15.5 years after the initial Fontan takedown.
Intermediate (n = 14)
Before the decision to takedown the Fontan, 13 (27%) patients had 14 rescue procedures including Fontan revision (n = 5), enlargement of fenestration (n = 4), coiling of aortopulmonary collaterals (n = 2), pacemaker insertion (n = 1), left pulmonary artery stenting (n = 1), and removal of bronchial cast (n = 1). Other non-rescue procedures included thoracotomy and pleurodesis (n = 4), drainage of pleural effusion (n = 2), and drainage of pericardial effusion (n = 1). Eleven patients died, 9 of whom presented with pleural effusions. Of the 3 survivors, 2 are alive with an intermediate circulation and 1 patient with pulmonary atresia/intact ventricular septum was converted to one and a half ventricle repair at Fontan takedown.
Late (n = 7)
Two patients died immediately after Fontan takedown. One patient had a successful transplant. Two patients are alive with an intermediate circulation, whereas 2 patients were successfully converted to a biventricular circulation. Of the 3 patients who presented with protein losing enteropathy, 1 patient died after Fontan takedown while in the remaining 2 patients the protein losing enteropathy improved after Fontan takedown.
Entire cohort
Death (n = 16). Sixteen (44%) patients died at a median of 57.5 days (IQR, 21.8, 76.8) after the Fontan procedure (Figure 1). Patients who had a takedown between 3 weeks and 6 months had the greatest mortality and patients who had the Fontan taken down within 48 hours had the lowest mortality (<2 days: 1/9, 11%; 2 days to 3 weeks: 2/6, 33%; 3 weeks to 6 months: 11/14, 79%; >6 months: 2/7, 28%; P = .007) (Figure 2). Transplant-free survival at 15 years was 47% (95% confidence interval [CI], 28-64). We found no association between any of the pre-takedown characteristics and death (Tables 3 and 4).
Figure 1Fate of all 36 Fontan takedown patients operated between 1975 and 2018. Overall mortality was 44%; 31% were alive with an intermediate circulation whereas 14% patients had a second Fontan. NYHA, New York Heart Association; BCPS, bidirectional cavopulmonary shunt; SPS, systemic to pulmonary artery shunt; AVF, arteriovenous fistula.
Figure 2Mortality and interval to Fontan takedown (n = 36, 1975-2018): patients who had Fontan takedown within 48 hours had the lowest mortality (11.1%), whereas patients who had takedown between 3 weeks to 6 months after Fontan had the greatest mortality (78.5%) (P = .007).
One patient with double inlet left ventricle underwent a heart transplant. The patient had a late Fontan takedown 14.3 years after an atriopulmonary Fontan. He presented with thrombosis of the Fontan circuit, which was taken down to a bidirectional cavopulmonary shunt, which was not tolerated. He successfully underwent a transplant 5 months after takedown. He was lost to follow-up at the time of the study.
Intermediate circulation (n = 11)
All 11 (31%) patients remained alive with an intermediate circulation after a median follow-up 9.4 years (IQR, 4.5, 15.3). The sources of pulmonary blood flow in this group included BCPS (n = 8), BCPS and systemic to pulmonary artery shunt (n = 1), BCPS and arteriovenous fistula (n = 1), and Hemi-Fontan, systemic to pulmonary artery shunt and arteriopulmonary fistula (n = 1). Ten of these patients were in New York Heart Association (NYHA) class I/II, with 1 patient in NYHA III. No patient in this group has had a heart transplant. Eight patients were deemed unsuitable for a second Fontan on repeat cardiac catheterization (pulmonary artery pressure >16 mm Hg [n = 6], elevated ventricular end-diastolic pressure [12 mm Hg; n = 1], and left pulmonary vein stenosis [n = 1]). The anatomic and hemodynamic parameters in these 8 patients were favorable before the first Fontan procedure and the reasons for the transition from favorable to unfavorable hemodynamics remain unclear. The reason for not attempting a second Fontan was unclear in the remaining 3 patients.
Second Fontan (n = 5)
Five (14%) patients underwent the Fontan for a second time after a median of 4.4 years (IQR, 1.9, 13). No remediable cause was identified in any patient before the second Fontan. One patient with DORV and hypoplastic left ventricle underwent an ECC fenestrated Fontan at 4 years of age. It was taken down within 24 hours for low cardiac output. He subsequently underwent a second Fontan after 2.1 years. He underwent a heart transplant approximately 7 years later for Fontan failure. The remaining 4 patients were alive in NYHA I/II at a median of 10.8 years IQR (3.1, 19.1). Before the second Fontan, cardiac catheter studies in these 4 patients demonstrated a median pulmonary artery pressure of 11 mm Hg (range, 9-13 mm Hg) and median ventricular end diastolic pressure of 9 mm Hg (range, 8-10 mm Hg). The hemodynamic parameters were similar to the findings on the cardiac catheter studies undertaken before the first Fontan in all 4 patients.
Biventricular/1½ ventricle repair (n = 3)
Two (5.5%) patients had a biventricular repair, of whom 1 patient was alive in NYHA 1 after 3.1 years and 1 patient was lost to follow-up. One patient had a 1½ ventricle repair and was alive in NYHA 1 after 19 years. Freedom from death/transplant after Fontan takedown was 59% at 1 year (CI, 41%-74%), 56% at 5 years (CI, 38%-71%) and 52% at 10 years (CI, 34%-68%) (Figure 3).
Figure 3Freedom from death/transplant after Fontan takedown was 59% at 1 year (CI, 41%-74%), 56% at 5 years (CI, 38%-71%), and 52% at 10 years (CI, 34%-68%) (n = 36, 1975-2018).
From a clinical standpoint, having to take down the Fontan is disappointing. For patients undergoing single-ventricle “palliation,” and their families, it is a catastrophe. Fontan takedown has been performed since the early years of our experience with this procedure and has been successful in salvaging the acutely unstable patient. However, the role of Fontan takedown is difficult to assess, as it is rarely performed and risk factors remain unknown. Furthermore, most patients have favorable pre-Fontan hemodynamics. There are only a limited number of reports on this rare procedure and long-term outcomes remain poorly defined.
The largest and the most comprehensive study to date has been reported by Almond and colleagues,
who described 53 Fontan takedowns over 27 years from 1979 to 2006. Takedown was performed intraoperatively in 12 (22%) patients, within the first postoperative month in 31 (58%) and between 1 month and 1 year in the remaining 10 (18%) patients.
The incidence of Fontan takedown is similar to our study (2.3%) The incidence of takedown for atriopulmonary, lateral tunnel and extracardiac conduit Fontan procedures was also similar in both studies.
takedown was performed during the Fontan procedure itself in 12 patients (22%) with a further 31 patients (58%) occurring within the first postoperative month and the remaining 10 (18%) occurring between 1 month and 1 year.
8 (57%) of the takedowns were performed within 2 weeks, 4 (28%) between 2 weeks and 6 months with the remaining 2 occurring between 1 and 2 years after the index procedure. The report from Murphy and colleagues
Takedown was performed during the Fontan procedure itself in 2 patients (12%) and within the first 2 postoperative months in the remaining 16 patients (88%). These reports are similar to our study, in which the majority of takedowns occurred early after the Fontan procedure with only 20% performed after 6 months.
The reported mortality for early Fontan takedown varies between 36% and 66%.
This is in contrast to our study, where the mortality was comparatively low (11%) when the Fontan was taken down within 48 hours. The highest mortality occurred between 3 weeks and 6 months (79%). This is similar to the report by Trezzi and colleagues
wherein 17 (94%) of the 18 Fontan takedowns to a BCPC survived the early post-takedown period. There is limited information on the outcomes of patients who have a takedown beyond 3 to 4 weeks of the Fontan. In the study by Almond and colleagues,
In our study the group of patients who underwent a Fontan takedown between 3 weeks and 6 months of age commonly presented with persistent pleural effusions and had a disproportionally high risk of mortality (79%). Most of these patients were subject to multiple procedures in an attempt to treat the effusions before taking down the Fontan circulation. Similar to other reports, our data suggest that there is an early window of opportunity to perform the Fontan takedown with a low mortality. Beyond this period, there is a substantial increase in mortality. We recommend that early takedown be considered sooner than later in patients with high-volume pleural drainage in the early postoperative period.
It is highly unlikely that meaningful risk prediction for Fontan takedown will ever be possible, given the small number of patients in every study. Previously identified risk factors include atriopulmonary Fontan and right dominant ventricular morphology.
In our study, there was no statistical association between any pre-takedown characteristic and mortality. Furthermore, pre-Fontan hemodynamic data were available in 86% of patients. The mean pulmonary artery pressure was <15 mm Hg in the majority of patients and no patient had greater than moderate ventricular dysfunction or greater than moderate atrioventricular valve regurgitation. In our opinion, the factors contributing to the need to take down the Fontan circulation remain unknown and poorly understood. As technical imperfections cannot be ruled out as a contributing factor, we suggest that these should be aggressively ruled out or surgically addressed in the early postoperative period.
Patients who are taken down to an intermediate circulation appears to have an acceptable survival. In the study from Boston, 8 survivors (27%) were alive at a median of 9.7 years post-takedown.
reported 8 survivors (80%) at a median of 6.3 years post takedown, all with reasonable hemoglobin oxygen saturations (median 84%). In our study, there were 11 patients (31%) with an intermediate circulation at a median of 9.4 years post takedown, 10 of whom were in NYHA I/II. Previous reports have suggested that a cavopulmonary shunt or aortopulmonary shunt or both are an acceptable form of long-term palliation in a patient with single ventricle physiology, with 10- and 20-year survival between 80% and 90% and 50%, respectively.
a high proportion of patients were identified with correctable abnormalities that may have contributed to the initial Fontan failure. The longest follow-up in this group has been reported by Trezzi and colleagues.
In our study 5 (14%) of patients underwent a second Fontan and 4 of them were alive in NYHA I/II at a median follow up of 10.8 years. Similar to other authors, we recommend that patients undergo investigation to identify and treat potentially remediable factors including branch pulmonary artery stenosis, arrhythmias, thrombus, pulmonary arteriovenous malformations, arch obstruction or hemi-diaphragmatic paresis prior to an attempt at a second Fontan.
Anecdotal reports of enlarging the fenestration in patients with early Fontan failure have been previously described, but the long-term outcome of these patients remains uncertain.
have used this approach in 6 patients with 1 intra-operative death and 5 mid-term survivors (6-81 months).
Limitations
Our study bears all the limitations of a retrospective study spanning over 4 decades. Even though we did not find an era effect, the substantial changes in the approach to the Fontan procedure, decision making, surgical techniques and postoperative management over the past decades are highly likely to have influenced outcomes.
Conclusions
The incidence of Fontan takedown is low, but overall mortality is high. Factors influencing Fontan takedown remain unclear. The majority of Fontan takedowns occurred within 6 months with the greatest incidence between 3 weeks and 6 months. The mortality of this rescue procedure was 11% when performed within 2 days and increased to 33% and 79% when performed between 2 days and 3 weeks and between 3 weeks and 6 months respectively (Figure 4). Long-term outcomes after takedown to an intermediate circulation are acceptable. A small proportion of survivors can undergo a successful second Fontan. We should have a low threshold for Fontan takedown in patients who demonstrate signs of early failure within days of the procedure.
Figure 4Fate of all 36 Fontan takedown patients operated between 1975 and 2018. Overall mortality was 44%, 31% stayed alive with an intermediate circulation, whereas 14% patients had a second Fontan. Mortality after early Fontan takedown was low whereas it was highest for takedowns between 3 weeks to 6 months. Freedom from death/transplant at 10 years was 52%.
Professor d’Udekem is a consultant with MSD and Actelion. All other authors reported no conflicts of interest.
The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.
The authors acknowledge their research teams for their support in maintaining the Australian & New Zealand Fontan Registry and the support provided to the Murdoch Children's Research Institute by the Victorian Government's Operational Infrastructure Support Program.
References
d'Udekem Y.
Iyengar A.J.
Galati J.C.
Forsdick V.
Weintraub R.G.
Wheaton G.R.
et al.
Redefining expectations of long-term survival after the Fontan procedure: twenty-five years of follow-up from the entire population of Australia and New Zealand.
The Australia & New Zealand Fontan Registry is partly funded by a National Health and Medical Research Council partnership grant (1076849). Professor D'Udekem is a Clinician Practitioner Fellow of the NHMRC (1082186).
Marathe and colleagues1 queried the Australia and New Zealand Fontan Registry to evaluate early and late outcomes of patients who had undergone a Fontan takedown procedure.
The Fontan physiology since its inception was never meant to be ideal for the human heart. Nevertheless, it has become the final step in the path of palliating many patients with single-ventricle disease.1 Currently, the total cavopulmonary connection or Fontan procedure places the pulmonary circulation like a dam between the systemic venous return and the systemic ventricle, creating, like any dam, upstream congestion and downstream decreased flow. These 2 features are the root cause of all early and late Fontan complications.
32672-6&id=fx1.jpg){kind=link}
32672-6&id=fx2.jpg){kind=link}
32672-6&id=gr1.jpg){kind=link}
32672-6&id=gr2.jpg){kind=link}
32672-6&id=gr3.jpg){kind=link}
32672-6&id=gr4.jpg){kind=link}
32672-6&id=fx3.jpg){kind=link}