Two Ventricles Are Not Better Than One in the Fontan Circulation: Equivalent Late Outcomes Supreet P. Marathe, MCh, Diana Zannino, MS, William Y. Shi, PhD, Karin du Plessis, PhD, Jascha Kehr, MD, Gopinath Perumal, MCh, Jessica Sun, MD, Thomas L. Gentles, FRACP, Julian Ayer, FRACP, PhD, Gary F. Sholler, FRACP, Charlotte E. Verrall, BS (Hons), Nelson Alphonso, FRACS, David Andrews, FRACS, Yves d’Udekem, MD, PhD, and David S. Winlaw, FRACS Heart Centre for Children, The Children’s Hospital at Westmead, Sydney, Australia; Murdoch Children’s Research Institute, Melbourne, Australia; Department of Cardiothoracic Surgery, Monash Medical Centre, Melbourne, Australia; Department of Pediatrics, Faculty of Medicine, University of Melbourne, Melbourne, Australia; Starship Green Lane Pediatric and Congenital Cardiac Service, Starship Children’s Hospital, Auckland, New Zealand; Queensland Pediatric Cardiac Service, Lady Cilento Children’s Hospital, Brisbane, Australia; Sydney Medical School, University of Sydney, Sydney, Australia; School of Child and Adolescent Health, Sydney Medical School, University of Sydney, Sydney, Australia; Department of Cardiothoracic Surgery, Princess Margaret Hospital, Perth, Australia; and Department of Cardiac Surgery, Royal Children’s Hospital, Melbourne, Australia
Background. A subset of patients who underwent Fontan operations has two adequate-sized ventricles, but an anatomic biventricular circulation cannot be achieved because of complex morphology or for technical reasons. This study sought to determine whether these patients with two-ventricle Fontan circulation had superior outcomes compared with those with a single ventricle. Methods. A binational Fontan Registry of patients (n [ 1,377) was analyzed to identify those patients with two adequate ventricles. This cohort was compared with patients with single-ventricle Fontan circulation. The primary end point was a composite end point called “Fontan failure” encompassing death, heart transplantation, Fontan takedown or conversion, proteinlosing enteropathy, plastic bronchitis, or New York Heart Association functional class III or IV. Results. A total of 79 Fontan patients with two adequate ventricles (2V) were compared with 1,291 single ventricle (1V) Fontan patients. Median follow-up for the entire cohort was 11.5 years (interquartile range, 5.1 to
18.8 years). There was no difference in unadjusted 15year freedom from Fontan failure (2V: 81% [95% confidence interval (CI), 69% to 94%] vs 1V: 86% [95% CI, 83% to 88%], p [ 0.4). Propensity-score matching for potential confounding factors yielded 75 two-ventricle Fontan patients matched with 604 single-ventricle Fontan patients, in which 15-year freedom from Fontan failure was also not different (2V: 79% [95% CI, 67% to 94%] vs 1V: 87% [95% CI, 84% to 91%], p [ 0.3). Conclusions. The two-ventricle Fontan circulation does not have better outcomes compared with the single-ventricle Fontan circulation. Late outcomes may depend more on other characteristics of the Fontan circulation. This finding is relevant when the Fontan procedure is being considered as an alternative to anatomic repair in patients with complex two-ventricle morphologies.
S
have yielded excellent survival for this complex group of patients in recent years [2]. However, the quality of life and freedom from adverse events remain far from ideal for Fontan survivors [3].
ince its introduction in 1971, the Fontan operation and the resultant circulation have comprised the final destination therapy for patients with a single functional ventricle [1]. Being inherently imperfect and nonphysiologic, the Fontan circulation can result in a myriad of complications. Nonetheless, refinement of technique and better understanding of Fontan physiology
(Ann Thorac Surg 2018;-:-–-) Ó 2018 by The Society of Thoracic Surgeons
Dr d’Udekem discloses a financial relationship with MSD and Actelion.
Accepted for publication Aug 13, 2018. Presented at the American Heart Association Scientific Sessions, Anaheim, California, Nov 11–14, 2017. Address correspondence to Dr Winlaw, Heart Centre for Children, The Children’s Hospital at Westmead, Cnr Hawkesbury Rd and Hainsworth St, Westmead NSW 2145, Australia; email:
[email protected].
Ó 2018 by The Society of Thoracic Surgeons Published by Elsevier Inc.
The Supplemental Tables and Figures can be viewed in the online version of this article [https://doi.org/10. 1016/j.athoracsur.2018.08.024] on http://www. annalsthoracicsurgery.org.
0003-4975/$36.00 https://doi.org/10.1016/j.athoracsur.2018.08.024
2
MARATHE ET AL FONTAN WITH TWO ADEQUATE VENTRICLES
A small subset of patients has two adequate-sized ventricles but must be treated using the single-ventricle (Fontan) pathway because biventricular septation is not possible for morphologic or technical reasons [4]. This group includes patients with double-outlet right ventricle, congenitally corrected transposition of the great arteries, and transposition of the great arteries with ventricular septal defect and left ventricular outflow tract obstruction, among others. We sought to examine whether this group of patients with a Fontan circulation but with two normal-size ventricles had better outcomes than conventional Fontan patients with a single ventricle. This an important issue because, for many patients in this group, the decision to proceed to a Fontan circulation as opposed to a biventricular anatomic repair involves balancing the long-term risks of the former with the higher early risks of the latter.
Patients and Methods The Australian and New Zealand Fontan Registry includes all patients who have undergone the Fontan operation in either country or have undergone their procedure elsewhere and are being followed up in the region. The details of creation, design, and maintenance of the Registry have been described elsewhere [5]. All Fontan patients with two adequate ventricles (defined later) were identified from the Registry. Patients who died before discharge after the Fontan procedure and those who underwent Fontan takedown during the same admission were excluded from this study.
Definition of Two-Ventricle Fontan Ventricles were considered to be adequate if ventricular size itself did not preclude anatomic biventricular repair. Only diagnoses where anatomic biventricular repair was conceptually feasible were included in the two-ventricle Fontan group (2V). Hence patients with atrioventricular valve atresia were excluded. All patients in this group were considered for an anatomic repair during decision making, but a Fontan pathway was pursued for reasons other than ventricular size. All data for these patients were reexamined using available records by investigators (S.P.M., G.P., J.K., and K.D.) to confirm the adequacy of both ventricles by means of expert consensus opinion using usual assessment of ventricular size, including adequacy of atrioventricular valve inflow, ventricular length, and proportion compared with the “systemic” ventricle.
End Points The primary end point was a composite end point termed “Fontan failure.” Fontan failure was defined as any of the following events: death, heart transplantation, Fontan takedown or conversion, protein-losing enteropathy, plastic bronchitis, or New York Heart Association functional class III or IV at follow-up. Secondary end points were prolonged effusions, time to first arrhythmia episode, time to first thromboembolic event, and first reintervention after the Fontan procedure.
Ann Thorac Surg 2018;-:-–-
Prolonged effusions were defined as effusions lasting for more than 30 days or requiring reoperation. Arrhythmia events were defined as a sustained episode of supraventricular tachycardia, including atrial fibrillation and flutter. Thromboembolic events were defined as thrombus within the Fontan circuit, pulmonary embolism, transient ischemic attack or reversible neurologic deficit (lasting 1 to 72 hours), or persistent stroke (lasting >72 hours).
Statistical Analysis Statistical analyses were performed using R software version 3.3.2 (R Foundation, Vienna, Austria). Categorical data were summarized as counts and percentages and compared using Fisher’s exact test. Continuous data were summarised as mean (SD) or median (interquartile range), depending on normality of distribution and compared using the t test or Wilcoxon rank sum test. Survival was assessed by the Kaplan-Meier method, and the cumulative incidence function was estimated for time to first arrhythmia, thromboembolic events, and reintervention with death and heart transplantation as a competing risk. Propensity-score analysis was performed to ensure that Fontan patients with two adequate ventricles were compared were appropriate single-ventricle patients. Matching was performed with the R-package “MatchIt” using the “nearest” method with a fixed calliper width of 0.01. Because of the large number of singleventricle patients available, a matching ratio of 10:1 was used, with the balance being adequate if the standardized mean difference across all variables was less than 0.1.
Results A total of 1,377 patients who had undergone Fontan operations were available for analysis. Four patients had insufficient data to ascertain ventricular adequacy and were thus excluded. Of the 82 patients with two adequate-size ventricles, 3 had a diagnosis of tricuspid atresia and were excluded. Thus, 79 patients met criteria for the two-ventricle group (2V), thus leaving 1,291 patients in the one-ventricle group (1V). The clinical characteristics of both groups are presented in Table 1. The reasons for not pursuing anatomic repair were noncommitted ventricular septal defect through which it was difficult to route the left ventricle to the aorta, a very large area of septation required for ventricular septal defect closure, aberrant coronary artery anatomy, straddling of the atrioventricular valves, abnormal location of the cardiac mass, particularly dextrocardia, or a combination of these factors. A significantly higher number of patients in the 2V group had dextrocardia or mesocardia (20% vs 8%, p ¼ 0.001) which probably influenced decision making. Patients in both groups had equal number of procedures before the Fontan operation (p ¼ 0.2). Details of the Fontan procedure for the two groups are presented in Table 2. The extracardiac conduit was the most common type of Fontan operation in both groups,
Ann Thorac Surg 2018;-:-–-
MARATHE ET AL FONTAN WITH TWO ADEQUATE VENTRICLES
3
Table 1. Basic Characteristics and Pre-Fontan Data Variable Male Ventricle morphology
Cardiac position Isomerism (heterotaxy)
Primary diagnosis
Number of prior (pre-Fontan) cardiac procedures
Pre-Fontan interventions
Level/Statistic
Left Right Indeterminate Missing Normal Dextrocardia/mesocardia None Left Right Missing Tricuspid atresia Double-inlet left ventricle Double-outlet right ventricle Atrioventricular canal or AVSD (aka unbalanced AVSD or common AV valve) Pulmonary atresia HLHS ccTGA Other Missing Mean (SD) Missing None 1 2 3 4 5 6 BAS Atrial septectomy PA band Systemic-pulmonary shunt Norwood ASO DKS TAPVD repair Aortic arch/coarctation repair PA reconstruction BCPS BCPS with forward flow Kawashima Hemi-Fontan LVOTO resection/enlargement of BVF Tricuspid repair/replacement Mitral repair/replacement Common AV valve repair/replacement Others
One-Ventricle (n ¼ 1,291) 736 (57%) 797 (63%) 445 (35%) 32 (3%) 17 1,186 (92%) 105 (8%) 1,159 (93%) 34 (3%) 54 (4%) 44 306 (24%) 227 (18%) 164 (13%) 94 (7%)
134 (10%) 171 (13%) 66 (5%) 117 (9%) 12 2.0 (1.1) 45 103 (8%) 275 (22%) 545 (44%) 222 (18%) 81 (7%) 15 (1%) 5 (0%) 75 (6%) 342 (26%) 302 (23%) 481 (37%) 193 (15%) 12 (1%) 119 (9%) 16 (1%) 124 (10%) 91 (7%) 671 (52%) 779 (60%) 24 (2%) 32 (2%) 8 (1) 33 (3%) 15 (1%) 14 (1%) 524 (41%)
Two-Ventricle (n ¼ 79) 45 (57%) 79 (100%)
63 16 67 4 7 0 0 26 15
4 0 15 19 1.8 8 18 36 12 4 0 0 3 18 14 44 1 0 1 4 4 7 38 54 2 1 0 1 0 0
(80%) (20%) (86%) (5%) (9%) 1 (0%) (0%) (33%) (19%)
(5%) (0%) (19%) (24%) 0 (1.0) 1 (10%) (23%) (46%) (15%) (5%) (0%) (0%) (4%) (23%) (18%) (56%) (1%) (0%) (1%) (5%) (5%) (9%) (48%) (68%) (3%) (1%) (0) (1%) (0%) (0%)
30 (38%)
p Value >0.99 -
0.001 0.054
<0.001
Total (N ¼ 1,370) 781 (57%) 797 (59%) 445 (33%) 32 (2%) 17 1,249 (91%) 121 (9%) 1,226 (93%) 38 (3%) 61 (5%) 45 306 (23%) 227 (17%) 190 (14%) 109 (8%)
0.6 0.5 0.3 0.002 <0.001 >0.99 0.012 0.024 0.2 0.5 0.6 0.2 0.7 >0.99 >0.99 0.7 >0.99 >0.99
138 (10%) 171 (13%) 81 (6%) 136 (10%) 12 2.0 (1.1) 46 111 (8%) 293 (22%) 581 (44%) 234 (18%) 85 (6%) 15 (1%) 5 (0%) 78 (6%) 360 (26%) 316 (23%) 525 (38%) 194 (14%) 12 (1%) 120 (9%) 20 (1%) 128 (9%) 98 (7%) 709 (52%) 833 (61%) 26 (2%) 33 (2%) 8 (1%) 34 (2%) 15 (1%) 14 (1%)
0.7
554 (40%)
0.2 0.96
(Continued)
4
MARATHE ET AL FONTAN WITH TWO ADEQUATE VENTRICLES
Ann Thorac Surg 2018;-:-–-
Table 1. Continued Variable Pre-Fontan oxygen saturation (%)
One-Ventricle (n ¼ 1,291)
Level/Statistic
Pre-Fontan mean PA pressure (mm Hg)
Mean (SD) Missing Mean (SD) Missing
Systemic-pulmonary collaterals
Missing
82.2 (6.6) 383 11.7 (3.6) 370 247 (27%) 385 12 (1%)
Pre-Fontan arrhythmia
Two-Ventricle (n ¼ 79)
p Value
83.7 (7.3) 15 11.5 (2.6) 21 5 (8%) 20 1 (1%)
0.1 0.6 0.001 0.5
Total (N ¼ 1,370) 82.3 (6.7) 398 11.7 (3.6) 391 252 (26%) 405 13 (1%)
Boldface indicates statistical significance. aka ¼ also known as; ASO ¼ arterial switch operation; AV ¼ atrioventricular; AVSD ¼ atrioventricular septal defect; BAS ¼ balloon atrial septostomy; BCPS ¼ bidirectional cavopulmonary shunt; BVF ¼ bulboventricular foramen; ccTGA ¼ congenitally corrected transposition of the great arteries; DKS ¼ Damus-Kaye-Stansel procedure; HLHS ¼ hypoplastic left heart syndrome; LVOTO ¼ left ventricular outflow tract obstruction; PA ¼ pulmonary artery; TAPVD ¼ total anomalous pulmonary venous drainage.
but patients who had undergone the atriopulmonary Fontan operation were more common in the 1V group. The Fontan procedure started being performed for 2V patients only after 1985. The creation of Fontan fenestration was similar in both groups; however, fenestration practices differ among the member institutes. There was no difference in cardiopulmonary bypass time,
cross-clamp time, and length of stay between the two groups. A significantly higher number of patients in the 2V group had a concomitant atrial septectomy along with a Fontan operation, thus probably reflecting a late decision to pursue a Fontan pathway. Early postoperative complications are presented in Supplemental Table 1.
Table 2. Fontan Characteristics Variable Age at Fontan (years) Type of Fontan
Era of Fontan
Fenestration CPB time ACC time Length of hospital stay (days) Concomitant procedure
Level/Statistic Mean (SD) AP LT ECC Missing 1975–1989 1990–1999 2000–2009 2010–2016 Yes Missing Mean (SD) Missing Median [range] Missing Median [range] Missing None Aortic valve repair/replacement Tricuspid valve repair/replacement Mitral valve repair/replacement PA reconstruction LVOT resection Pacemaker insertion Maze procedure Atrial septectomy Others
One-Ventricle (n ¼ 1,291) 5.7 205 247 832
(3.8) (16%) (19%) (65%) 7 172 (13%) 309 (24%) 455 (35%) 355 (27%) 467 (37%) 26 117.4 (51.7) 567 29 [0–179] 674 14 [1–148] 370 822 (64%) 16 (1%) 21 (2%) 7 (0%) 73 (6%) 4 (0%) 8 (1%) 4 (0%) 38 (3%) 235 (18%)
Two-Ventricle (n ¼ 79) (3.3) (6%) (17%) (77%) 1 3 (4%) 17 (22%) 32 (41%) 27 (34%) 29 (37%) 1 113.0 (40.7) 30 26 [0–113] 34 16 [6–73] 21 51 (65%) 1 (1%) 1 (1%) 1 (0%) 6 (8%) 0 (0%) 2 (3%) 0 (0%) 7 (9%) 17 (22%)
p Value
6.0 5 13 60
0.4 0.035
0.04
>0.99 0.9 0.9 0.8 0.9 >0.99 >0.99 0.4 0.5 >0.99 0.1 >0.99 0.012 0.5
Total (N ¼ 1,370) 5.7 210 260 892
(3.8) (15%) (19%) (65%) 8 175 (13%) 326 (24%) 487 (36%) 382 (28%) 496 (37%) 27 117.1 (51.1) 597 28 [0–179] 708 14 [1–148] 391 873 (64%) 17 (1%) 22 (2%) 8 (0%) 79 (6%) 4 (0%) 10 (1%) 4 (0%) 45 (3%) 252 (18%)
Boldface indicates statistical significance. ACC ¼ aortic cross clamp; AP ¼ atriopulmonary; CPB ¼ cardiopulmonary bypass; LVOT ¼ left ventricular outflow tract; PA ¼ pulmonary artery.
ECC ¼ extracardiac conduit;
LT ¼ lateral tunnel;
Ann Thorac Surg 2018;-:-–-
MARATHE ET AL FONTAN WITH TWO ADEQUATE VENTRICLES
5
Supplemental Tables 2 and 3 and Supplemental Figures 2A and 2B. Reinterventions apart from those considered as components of Fontan failure are depicted in Table 4 and Figure 3.
Propensity Score Model
Fig 1. Freedom from Fontan failure. Kaplan-Meier analysis showing freedom from Fontan failure between the two-ventricle and oneventricle Fontan groups. (Components presented in Table 3.)
The propensity score model performed well, with 694 patients matched (Supplemental Table 4). Patients were matched for cardiac position, isomerism, previous arch repair, age at Fontan, type of Fontan operation, and era of Fontan operation. In the matched cohort, there was similar freedom from late Fontan failure, as presented in Figure 4 (2V: 79% [95% CI, 67% to 94%] vs 1V 87% [95% CI, 84% to 91%], p ¼ 0.3). Univariable Cox regression analysis of the propensity-matched groups did not demonstrate any association between ventricular morphology and late Fontan failure (hazard ratio, 1.31 [95% CI, 0.69 to 2.48], p ¼ 0.4).
Fontan Failure Median follow-up for the entire cohort was 11.5 years (interquartile range, 5.1 to 18.8 years). There was no difference in freedom from Fontan failure between the two groups (estimates at 15 years post Fontan operation, (2V: 81% (95% confidence interval [CI], 69% to 94%] vs 1V: 86% [95% CI, 83% to 88%], p ¼ 0.4) (Fig 1). On multivariable Cox regression, there was no association between ventricular morphology and late Fontan failure (hazard ratio, 1.18 [95% CI, 0.64 to 2.16], p ¼ 0.6). The components of Fontan failure are presented in Table 3. When the 1V group was separated into right dominant and left dominant groups and compared with the 2V group, freedom from Fontan failure was not different (left vs 2V, p ¼ 0.2; right vs 2V, p ¼ 0.7) (Fig 2; Supplemental Figs 1A, 1B). The right dominant 1V group had a higher incidence of failure compared with the left dominant 1V group but not compared with the 2V group. None of the patients with isomerism in the 2V group had a Fontan failure event.
Secondary End Points (Pleural Effusion >30 Days, Arrhythmic or Thromboembolic Events, Reintervention) There was no difference between the two groups with regard to secondary end points and freedom from reintervention. Secondary end points are shown in
Comment Despite its inherent imperfections, the Fontan circulation is the destination pathway for all patients with a single ventricle. In rare instances, even in the presence of two adequate-size ventricles, patients are committed to the Fontan pathway because a biventricular circulation cannot be achieved, for anatomic or technical reasons. This situation is encountered most commonly in cases of complex double-outlet right ventricle, congenitally corrected transposition of the great arteries, or transposition of the great arteries. It is often argued that these patients could be better off with a Fontan circulation rather than a complex, high-risk anatomic repair [6–11]. This argument is partly based on the unproven premise that when two ventricles work together as a single systemic ventricle, we could expect lower end-diastolic pressures resulting in lower Fontan pressures and fewer complications. In this study we attempted to determine whether a “two-ventricle” Fontan circuit translated into better Fontan outcomes. The role of ventricular morphology in outcomes after Fontan palliation has been closely examined, with most studies concluding that right dominant morphology results in poorer outcomes at different stages of single ventricle palliation [2, 12], although recent reports
Table 3. First Fontan Failure Event Event Type
One-Ventricle, n (%)
Two-Ventricle, n (%)
Total (n ¼ 210), n (%)
70 (5.4) 10 (0.7) 39 (3) 3 (0.2) 40 (3) 35 (2.7) 2 (0.1) 199/1291(15)
1 (1.2) 2 (2.5) 2 (2.5) 1 (1.2) 2 (2.5) 2 (2.5) 1 (1.2) 11/79 (14)a
71 12 41 4 42 37 3 210/1370(15)
Death Transplant NYHA III/IV Fontan takedown Fontan conversion PLE PB Total a
No patients with isomerism (heterotaxy).
NYHA ¼ New York Heart Association (functional class);
PB ¼ plastic bronchitis;
PLE ¼ protein-losing enteropathy.
6
MARATHE ET AL FONTAN WITH TWO ADEQUATE VENTRICLES
Fig 2. Freedom from Fontan failure. Kaplan-Meier analysis showing freedom from Fontan failure between the right dominant oneventricle Fontan versus left dominant one-ventricle Fontan versus two-ventricle Fontan groups.
challenge this concept [13, 14]. However, the twoventricle Fontan subset has not been exclusively studied before. Our study proves that presence of two ventricles in the Fontan circuit does not confer any advantage according to the outcomes examined. There are important implications of this result for cardiologists and surgeons across the spectrum of both complex anatomic repair and single-ventricle patients. The overall question whether complex anatomic repair is better than Fontan circulation remains unanswered because follow-up beyond 20 years is required, together with detailed assessments of exercise capacity and quality of life. Importantly, our study suggests that a decision to follow the Fontan pathway should not be based on the presumption that a two-ventricle Fontan circulation is a better or more favorable substrate than the conventional (1V) Fontan circulation for patients. Our study informs consideration of biventricular repair strategies in cases where technical complexity or borderline-size ventricles make anatomic repair challenging, although a direct comparison of approaches in this patient group is yet to be made [15–17].
Ann Thorac Surg 2018;-:-–-
One potential reason for the lack of superiority of the two-ventricle Fontan group despite having a larger myocyte mass could be the fundamental shortcoming of the Fontan circulation—preload deficiency [18]. Because the pulmonary circulation acts as a resistor in the circuit, having additional myocardium (a “stronger pump”) distal to the lungs may make no difference and will have no impact on systemic venous hypertension, which will continue to pose problems. Few studies in the last century can point toward other probable reasons of nonefficient performance of two ventricles in the Fontan circulation. In 1991, Penny and colleagues [19] demonstrated abnormal and asynchronous intraventricular flow and diastolic filling in this group of patients by using echocardiographic studies. These investigators attributed these findings to a combination of preload reduction and acquired ventricular hypertrophy [19]. In 1994, Yamamura and associates [20] demonstrated similar findings using catheterization studies. These anomalous mechanics may not actually lower enddiastolic pressures, thereby negating any potential advantage of a larger myocyte mass. In addition, the suboptimal response of the right ventricle to chronic pressure load is well described [21]. Coupling the right ventricle with a left ventricle in a two-ventricle Fontan circulation may not circumvent this problem, and ventricular interaction could mean that right ventricular dysfunction has a negative influence on the mechanics of both ventricles. To explore our finding further, clinical outcomes need to be correlated with hemodynamic data: ventricular end-diastolic pressures, pulmonary artery pressures, and transpulmonary gradients. Exercise testing is necessary to assess the performance of the circuit under conditions of higher cardiac output. Moreover, diastolic volume distributions between the two ventricles need to be studied using magnetic resonance imaging. Magnetic resonance imaging or computational fluid dynamic studies can also reveal blood flow patterns from two ventricles through a common systemic outflow and help evaluate whether this process is energy efficient.
Table 4. Components of First Reintervention Type Reintervention Type Fontan revision Creation of fenestration Fenestration closure Pacemaker insertion Maze procedure Catheter ablation AV valve repair/replacement PA reconstruction Pleurodesis Others Total AV ¼ atrioventricular;
PA ¼ pulmonary artery.
One-Ventricle, n (%)
Two-Ventricle, n (%)
Total, n (%)
13 (1) 3 (0.2) 101 (8) 70 (5.4) 10 (0.7) 32 (2.4) 15 (1) 1 (0.1) 9 (0.7) 112 (8.6) 366/1,291(28)
0 (0) 0 (0) 8 (10) 2 (2.5) 2 (2.5) 0 (0) 0 (0) 1 (1.2) 0 (0) 8 (10) 21/79(26)
13 3 109 72 12 32 15 2 9 120 387/1,370(28)
Ann Thorac Surg 2018;-:-–-
MARATHE ET AL FONTAN WITH TWO ADEQUATE VENTRICLES
7
the lack of benefit from having two ventricles supporting the Fontan circulation. Ventricular size is a continuous variable, and the 1V group may have a spectrum of ventricular sizes ranging from diminutive ventricles to half ventricles.
Conclusion
Fig 3. Time to first reintervention. Cumulative incidence curves comparing time to first reintervention with death or heart transplantation as competing risks between one-ventricle and twoventricle Fontan groups. (Components presented in Table 4.)
Study Limitations Our study has the usual limitations of a retrospective study. The study groups are unequal and heterogeneous. Because the study covers an extensive period, an era effect cannot be excluded. We have focused on adequacy of ventricular size with classification by expert re-review and not true quantitative analysis. The comparison groups include patients with varying underlying morphology, and this may influence outcome. It is also possible that differences may become evident between the two groups over a longer follow-up period. Patients with heterotaxy syndromes who undergo the Fontan procedure are known to have poorer outcomes [22], and inclusion of these patients in the 2V group could worsen the outcome of the 2V Fontan group and partly explain the lack of difference in outcome between one- and two-ventricle Fontan procedures. However, in our series none of the patients with heterotaxy in the 2V group had a Fontan failure event, a finding suggesting that that the presence of heterotaxy alone does not explain
Fig 4. Freedom from Fontan failure (propensity matched). KaplanMeier analysis showing freedom from Fontan failure between propensity-matched one-ventricle and two-ventricle Fontan groups. (CI ¼ confidence interval; HR ¼ hazard ratio.)
A two-ventricle Fontan circulation does not have better outcomes when compared with the single-ventricle Fontan circulation. Late outcomes may depend more on other characteristics of the Fontan circulation. It also focuses the contribution of problems upstream of the heart in outcomes after the Fontan operation. This finding is relevant when the Fontan procedure is being considered as an alternative to anatomic repair in patients with complex two-ventricle morphologies. The authors wish to thank their research assistants for their support in maintaining the Australian and New Zealand Fontan Registry and the support provided to the Murdoch Children’s Research Institute by the Victorian Government’s Operational Infrastructure Support Program. This work was supported in part by the (1) Wishaw Trust, Heart Centre for Children, The Children’s Hospital at Westmead and (2) the National Health and Medical Research Council (NHMRC) Partnership grant (1076849). Dr d’Udekem is a Clinician Practitioner Fellow of the NHMRC (1082186).
References 1. Fontan F, Baudet E. Surgical repair of tricuspid atresia. Thorax 1971;26:240–8. 2. d’Udekem Y, Iyengar AJ, Galati JC, 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. Circulation 2014;130(Suppl 1): S32–8. 3. Rychik J. The relentless effects of the Fontan paradox. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2016;19: 37–43. 4. Freedom RM, Van Arsdell GS. Biventricular hearts not amenable to biventricular repair. Ann Thorac Surg 1998;66: 641–3. 5. Iyengar AJ, Winlaw DS, Galati JC, et al. The Australia and New Zealand Fontan Registry: description and initial results from the first population-based Fontan registry. Intern Med J 2014;44:148–55. 6. Delius RE, Rademecker MA, De Leval MR, et al. Is a highrisk biventricular repair always preferable to conversion to a single ventricle repair? J Thorac Cardiovasc Surg 1996;112: 1561–9. 7. Villemain O, Bonnet D, Houyel L, et al. Double-outlet right ventricle with noncommitted ventricular septal defect and 2 adequate ventricles: is anatomical repair advantageous? Semin Thorac Cardiovasc Surg 2016;28:69–77. 8. Marathe SP, Jones MI, Ayer J, et al. Congenitally corrected transposition: complex anatomic repair or Fontan pathway? Asian Cardiovasc Thorac Ann 2017;25:432–9. 9. Delius RE. 2-V or not 2-V: that is the question.plus some musings on thinking out of the box. J Thorac Cardiovasc Surg 2017;154:583–4. 10. Alsoufi B. Is there a limit to how far we should push the envelope in pediatric cardiac surgery? J Thorac Cardiovasc Surg 2016;152:707–8. 11. Backer CL. double outlet right ventricle: where are we now? Semin Thorac Cardiovasc Surg 2016;28:79–80.
8
MARATHE ET AL FONTAN WITH TWO ADEQUATE VENTRICLES
12. McGuirk SP, Winlaw DS, Langley SM, et al. The impact of ventricular morphology on midterm outcome following completion total cavopulmonary connection. Eur J Cardiothorac Surg 2003;24:37–46. 13. Atz AM, Zak V, Mahony L, et al. Longitudinal outcomes of patients with single ventricle after the Fontan procedure. J Am Coll Cardiol 2017;69:2735–44. 14. Alsoufi B, Gillespie S, Kim D, et al. The impact of dominant ventricle morphology on palliation outcomes of single ventricle anomalies. Ann Thorac Surg 2016;102:593–601. 15. Emani SM, del Nido PJ. Strategies to maintain biventricular circulation in patients with high-risk anatomy. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2013;16:37–42. 16. Emani SM, Bacha EA, Mcelhinney DB, et al. Primary left ventricular rehabilitation is effective in maintaining twoventricle physiology in the borderline left heart. J Thorac Cardiovasc Surg 2010;138:1276–82. 17. Kottayil BP, Sunil GS, Kappanayil M, et al. Two-ventricle repair for complex congenital heart defects palliated towards
Ann Thorac Surg 2018;-:-–-
18. 19.
20.
21. 22.
single-ventricle repair. Interact Cardiovasc Thorac Surg 2014;18:266–71. Gewillig M, Brown SC. The Fontan circulation after 45 years: update in physiology. Heart 2016;102:1081–6. Penny DJ, Rigby ML, Redington AN. Abnormal patterns of intraventricular flow and diastolic filling after the Fontan operation: evidence for incoordinate ventricular wall motion. Br Heart J 1991;66:375–8. Yamamura H, Nakazawa M, Park I, Nakanishi T, Momma K, Imai Y. Asynchronous volume changes of the two ventricles after Fontan operation in patients with a biventricular heart. Heart Vessels 1994;9:307–14. Piran S, Veldtman G, Siu S, Webb GD, Liu PP. Heart failure and ventricular dysfunction in patients with single or systemic right ventricles. Circulation 2002;105:1189–95. Bartz PJ, Driscoll DJ, Dearani JA, et al. Early and late results of the modified Fontan operation for heterotaxy syndrome. 30 years of experience in 142 patients. J Am Coll Cardiol 2006;48:2301–5.