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
Surgical management of various forms of double-outlet right ventricle uses a variety of approaches depending on the underlying anatomic form. In this study, we sought to determine the risk factors of mortality and reoperation in those with double-outlet right ventricle undergoing biventricular repair, according to anatomic characteristics and initial surgical strategy.
Methods
Between 1992 and 2013, 433 patients were included in the study. Double-outlet right ventricle was classified as double-outlet right ventricle with subaortic ventricular septal defect associated with subpulmonary obstruction in 33% of patients (n = 141), with subaortic ventricular septal defect without subpulmonary obstruction in 30% of patients (n = 130), with subpulmonary ventricular septal defect in 32% of patients (n = 139), and with noncommitted ventricular septal defect in 5% of patients (n = 23). Three types of repairs were performed: (1) intraventricular baffle repair, n = 149 (34%); (2) intraventricular baffle repair with right ventricular outflow tract reconstruction, n = 163 (38%); and (3) intraventricular baffle repair with arterial switch operation, n = 121 (28%).
Results
Thirty-day overall mortality was 7.4%. Early reoperation was needed in 6% of the cases. Early mortality was higher in the intraventricular baffle repair with arterial switch operation group (P = .01). Survival at 10 years was 86.2%, and freedom from reoperation at 10 years was 61.4%. At last follow-up (median, 5.7 years; 95% confidence interval, 4.5-6.6), mortality and reoperation rates were similar in the different surgical strategy groups. Late reoperation and late mortality were significantly higher in the double-outlet right ventricle with noncommitted ventricular septal defect group (P < .01). In multivariate analyses, risk factors for reoperation were concomitant surgical procedures (P = .03) and duration of cardiopulmonary bypass (P < .01). Risk factors for mortality were restrictive ventricular septal defect (P = .01), mitral cleft (P < .01), and associated coronary artery anomalies (P = .01).
Conclusions
Those with the anatomic type of double-outlet right ventricle with noncommitted ventricular septal defect were at higher risk for reoperation and mortality. Intraventricular baffle repair with arterial switch operation was the surgical strategy in patients at higher risk of early death. Initial surgical strategy did not influence the late outcomes.
The outcomes of DORV with biventricular repair remain unclear. Between 1992 and 2013, 433 consecutive patients were retrospectively included. Early mortality was higher in the IVR-ASO group. Late mortality and late reoperation rates were similar in the different biventricular surgical strategy groups, but were significantly higher in DORV with ncVSD. Multiple risk factors have been identified.
Double-outlet right ventricle (DORV) is a type of ventriculoarterial connection (or alignment) of the great arteries in which both great arteries arise entirely or predominantly from the right ventricle.
DORV is a heterogeneous group of congenital heart diseases subclassified by the International Nomenclature Committee for Pediatric and Congenital Heart Disease, according to the relationship of the ventricular septal defect (VSD) to the great arteries and to the presence or absence of pulmonary outflow tract obstruction (International Pediatric and Congenital Cardiac Code).
European Association for Cardio-Thoracic Surgery, Society of Thoracic Surgeons, List S. European Paediatric Cardiac Code (IPCCC Short List). April 1, 2012. 2012:1-14. Available at: http://ipccc.net/index.php/download-the-ipccc/. Accessed June 13, 2016.
reported a successful biventricular repair in a child with DORV, subaortic VSD, and concordant atrioventricular (AV) connections. Since then, complete correction through the use of a variety of surgical techniques has been achieved in more complex forms of DORV. The feasibility of a biventricular repair still remains a surgical challenge in the management of DORV.
Thus far, however, 2 points still remain unclear: which surgical strategy implies a higher risk of mortality or reoperation and when to abandon the ambition of biventricular repair. In this study, we sought to determine the risk factors for mortality and reoperation in those with DORV undergoing biventricular repair according to anatomic characteristics and initial surgical strategy.
Materials and Methods
Patients
We reviewed all records of consecutive patients with DORV who underwent operation in 2 French centers (Necker Enfants Malades, Paris, France; Centre Chirurgical Marie Lannelongue, Le Plessis Robinson, France) between 1992 and 2013. Those with conotruncal anomalies with the presence of double conus but without both great arteries arising entirely or predominantly from the right ventricle
were not included in this cohort. Those with DORV associated with double-inlet ventricle, absent AV connection, or severely imbalanced ventricles precluding biventricular repair were excluded. The diagnosis of DORV with 2 balanced ventricles (or mild asymmetry not precluding biventricular repair) was confirmed after review of echocardiograms, cardiac magnetic resonance imaging or computed tomography scans, angiograms, and surgical reports. In all cases, the type of DORV was finally confirmed according to the anatomic description given by the surgeon. When biventricular repair was considered unattainable in light of both preoperative investigations or intraoperative evaluation, the patient was excluded. Demographic data included gender, prenatal or postnatal diagnosis, center, and associated genetic anomalies.
Four anatomic groups were defined according to the classification of the International Pediatric and Congenital Cardiac Code
European Association for Cardio-Thoracic Surgery, Society of Thoracic Surgeons, List S. European Paediatric Cardiac Code (IPCCC Short List). April 1, 2012. 2012:1-14. Available at: http://ipccc.net/index.php/download-the-ipccc/. Accessed June 13, 2016.
we defined the VSD in DORV according to the direction of the blood flow coming from the left ventricle through the VSD: directly into the aorta (subaortic VSD), the pulmonary artery (subpulmonary VSD), or both (doubly committed VSD). These “committed” VSDs always open in the outlet of the right ventricle and are cradled within the limbs of the septal band (outlet VSD). Conversely, the blood flow through an ncVSD is not directed toward the outflow tract, but directly toward the right ventricular cavity, running perpendicular to the ventricular septum. We include in this category all the VSDs that are not located between the 2 limbs of the septal band, that is inlet, muscular, and central membranous VSD.
Associated cardiac anomalies were noted: VSD (location, number, and restriction when significant pressure gradient between the ventricles was diagnosed), subpulmonary and subaortic obstruction, atrioventricular septal defect (AVSD), mitral valve cleft and other mitral valve anomalies, straddling of the mitral or of the tricuspid valve, and coronary artery anomalies.
Three surgical strategies were used: (1) intraventricular baffle repair (IVR); (2) IVR with right ventricular outflow tract (RVOT) reconstruction, including IVR with RVOT patch enlargement, réparation à l'étage ventriculaire (REV), and Rastelli procedures, IVR-RVOT; and (3) IVR with arterial switch operation (ASO), IVR-ASO.
Concomitant surgical procedures were reported: VSD enlargement, AV valve repair, correction of abnormal pulmonary venous return, coarctation or interrupted aortic arch repair, chordae mobilization, subaortic stenosis resection, and pulmonary artery branch stenosis repair.
Surgical data included prior palliative intervention, age and weight at time of repair, date of surgery, surgical strategy, concomitant surgical procedure, cardiopulmonary bypass (CPB) duration, duration of mechanical ventilation, reoperation, and delay between repair and reoperation.
Finally, we collected early mortality and reoperation occurring less than 30 days after repair and late mortality and reoperation occurring more than 30 days after repair. Follow-up information was available in 341 of 381 surviving patients (90%). This study was approved by our ethics board (Comité de Protection des Personnes–Ile-de-France VI).
Statistical Analysis
Continuous variables are presented as the mean value ± standard deviation when variables were normally distributed and median value with 95% confidence interval (CI) when they were not. Categoric variables are presented as percentages. Comparisons of categoric variables were made using the chi-square test or Fisher exact test when appropriate. Time to death and time to reoperation are shown in Kaplan–Meier curves and all actuarial curves start at the time of surgery. Cox proportional hazard regression analysis, with the date of repair used as the start date, was performed as univariate and multivariate analyses (Table E1 shows a list of variables considered for multivariable analyses) to investigate the risk factors for early death, late death, early reoperation, and late reoperation. When there were less than 10 outcome events per predictor variable, we used the chi-square or Mann–Whitney test to determine it as appropriate. We excluded early death and reoperation from the analysis when we evaluated late death or reoperation. All entered variables were selected on the basis of clinical experience, univariate analysis, and previously published data. Receiver operating curve analysis was used to study the ability of significant parameters to predict early death. Analysis was conducted using MedCalc (MedCalc Software, Mariakerke, Belgium).
Results
Population and Anatomic Characteristics
Overall, we identified 482 patients with DORV. Eighteen patients had only a palliative procedure (death before biventricular repair) and were excluded. Thirty-one patients (6%) followed a sequential cavopulmonary connection program and were excluded: 12 with DORV with ncVSD, 9 with multiple VSD, 5 with straddling mitral valves, and 5 with other severe mitral valve anomalies.
Finally, 433 DORV cases with 2 adequate ventricles underwent biventricular repair and were included in the study. Demographic and anatomic characteristics are summarized in Table 1. The surgical procedures are shown in the Video 1 (DORV with ncVSD with IVR-ASO).
Table 1Patient characteristics
Variable
No. (%)
Demographic characteristics
Male/female
260 (60)/173 (40)
Center (Necker/Plessis Robinson)
216 (50)/217 (50)
Prenatal diagnosis
171 (39)
Chromosomal/genetic abnormalities
50 (12)
IPCCC classification
Fallot type
141 (33)
VSD type
130 (30)
Taussig–Bing type
139 (32)
ncVSD type
23 (5)
Anatomic characteristics
VSD
No. of VSDs
1
373 (86)
Multiple
61 (14)
Restrictive VSD
35 (8)
Ventriculoarterial spatial relationship
D-malposition
343 (79)
D side-by-side
42 (10)
L-malposition
24 (5.5)
L side-by-side
2 (0.5)
Undetermined
22 (5)
Subpulmonary obstruction
184 (42)
Pulmonary atresia
19 (4)
Stenosis (infundibular/valvular/supravulvular)
165 (38)
Subaortic and supra aortic obstructions
101 (23)
Valvular or subvalvular aortic stenosis
21 (5)
Coarctation of the aorta
62 (14)
Interrupted aortic arch
18 (4)
AV valve anomalies
AVSD
12 (3)
Straddling
67 (16)
Mitral/tricuspid/bivalvular
17 (4)/42 (10)/8 (2)
Other mitral valve anomalies
36 (8)
Coronary artery anomalies
112 (26)
DORV Taussig–Bing type
67/139 (48)
IPCCC, International Pediatric and Congenital Cardiac Code; VSD, ventricular septal defect; ncVSD, noncommitted ventricular septal defect; AV, atrioventricular; AVSD, atrioventricular septal defect; DORV, double-outlet right ventricle.
The number and type of prior palliative interventions according to the anatomic groups are shown in Table 2. The 18 interventions listed in the group “Others” included 6 isolated pulmonary artery branch repair, 4 closures of the arterial duct, 3 partial cavopulmonary connections (converted secondarily to biventricular strategy), 2 corrections of partial anomalous pulmonary venous return, 1 tricuspid valve repair, 1 aberrant left pulmonary artery branch reimplantation, and 1 pacemaker implantation. The proportion of DORV with ncVSD requiring a palliative procedure before repair (12/23, 52%) was significantly higher than in the VSD type (41/130, 32%; P < .01), Fallot type (44/141, 31%; P < .01), and Taussig–Bing type (38/139, 27%; P < .01). Table 3 shows the surgical strategy in the 4 anatomic groups with concomitant surgical procedures.
Table 2Prior palliative interventions according to anatomic groups (180 procedures in 135 patients [31%])
Fallot type
VSD type
Taussig–Bing type
ncVSD type
Total
No. of patients
44
41
38
12
135
PA banding
2
25
10
12
49
PA banding + CoA repair
0
16
10
1
27
Blalock shunts
42
0
14
15
71
Atrial septectomy
2
1
8
4
15
Others
6
2
6
4
18
Total
52
44
48
36
180
VSD, Ventricular septal defect; ncVSD, noncommitted ventricular septal defect; PA, pulmonary artery; CoA, coarctation of the aorta.
The median follow-up was 5.7 years (95% CI, 4.5-6.6). The overall survival was 87.8% at 1 year, 87.2% at 5 years, and 86.2% at 10 years. The overall freedom from reoperation rate was 86.5% at 1 year, 74.1% at 5 years, and 61.4% at 10 years. At last follow-up, 351 of 381 survivors (92%) were in New York Heart Association functional class I, and 30 of 381survivors (8%) were in New York Heart Association functional class II.
Outcomes according to double-outlet right ventricle type
Those with noncommitted VSD had a higher risk of mortality (P = .02) (Figure 1, A) and reoperation (P = .04) (Figure 1, B) in comparison with those with other anatomic types. Mortality was higher (hazard ratio, 2.18; 95% CI, 1.10-4.30) and reoperation was more frequent (hazard ratio, 1.68; 95% CI, 1.07-2.63) in the Taussig–Bing type than in the VSD type. No difference was found in mortality or reoperation rate between the Taussig–Bing type and Fallot type, and between the VSD type and the Fallot type.
Figure 1Outcomes according to anatomic type of DORV. A, Survival according to anatomic type of DORV. B, Freedom from reoperation according to anatomic type of DORV. CI, Confidence interval; VSD, ventricular septal defect.
No significant statistical difference was found in late mortality (P = .20) (Figure 2, A) and late reoperation rate (P = .13) (Figure 2, B) according to the 3 surgical strategies.
Figure 2Outcomes according to surgical strategy. A, Survival according to surgical strategy. B, Freedom from reoperation according to surgical strategy. CI, Confidence interval; IVR, intraventricular baffle repair; ASO, arterial switch operation; RVOT, right ventricular outflow tract.
A total of 32 of 433 patients died within 30 days after repair (7.4%). The causes of death were as follows: 12 during surgery (5 during IVR + ASO, with no significant statistical difference compared with the other surgical strategies; odds ratio [OR], 1.7; 95% CI, 0.4-5.7; P = .47), 4 during IVR with concomitant surgical procedure (AV valve repairs and coarctation repairs), and 3 during IVR-RVOT with concomitant surgical procedures (AV valve repairs, correction of APVR, and subaortic stenosis resection); 7 sepsis cases, 4 low cardiac output syndromes, 4 pulmonary hypertension crises, 2 electromechanical dissociations, 1 hemorrhage, and 2 unclearly explained. Early postoperative characteristics are shown in Table 4. The IVR + ASO was the surgical strategy with the highest rate of early death (OR, 3.1; 95% CI, 1.4-7.1; P = .01). The causes of early death after IVR + ASO (15 patients) were 5 during surgery, 4 sepsis cases, 2 low output cardiac failures, 1 pulmonary hypertension crisis, 1 electromechanical dissociation, 1 hemorrhage, and 1 unclearly explained. Early mortality was significantly higher in DORV with ncVSD (P = .04), low weight at operation (P = .04), left ventricular outflow tract obstruction (P = .04), and associated surgical procedure (P = .03). By using receiver operating characteristics, weight less than 4000 g predicted early mortality with a sensitivity of 80% and a specificity of 68%.
Table 4Early surgical and postoperative characteristics (<30 days) according to surgical strategies
Early cardiac reoperation was needed in 26 patients (6%) with no significant statistical differences between the anatomic types of DORV and the surgical strategies. The most frequent cause of early reoperation for subaortic VSD (Fallot type and VSD type) was residual VSD (35%, 9/26) (Table 5).
Table 5Indications for early reoperation according to anatomic group
A total of 20 patients died, 16 (80%) within the first year after repair. Causes of death were end-stage heart failure in 8 patients, complications of a mechanical valve in 3 patients, sudden death in 3 patients, death after reoperation in 2 patients (1 valve replacement after endocarditis, 1 RVOT enlargement), noncardiac death in 2 patients, and 1 missing information. No significant difference in late mortality was found between the anatomic types or the surgical strategies. Factors associated with late mortality for each surgical strategy in univariate analysis are shown in Table E2. In multivariate analysis, risk factors for late mortality were IVR: mitral valve abnormality (P = .01) and concomitant procedure (P < .01); IVR-RVOT reconstruction: subaortic obstruction (P = .02); and IVR-ASO: concomitant procedure (P < .01).
In the overall analysis, the factors associated with late mortality in univariate analysis are shown in Table E3. To determine the anatomic characteristics that affected late mortality, we performed a multivariate backward analysis (Table E4) including restrictive VSD, mitral cleft, and coronary artery anomalies. Risk factors for late mortality were restrictive VSD (OR, 2.7; 95% CI, 1.2-6.3; P = .01), mitral cleft (OR, 3.5; 95% CI, 1.6-5.6; P < .01), and coronary artery anomalies (OR, 2.9; 95% CI, 1.5-3.9; P = .01).
Late reoperation
A total of 97 patients underwent 136 reoperations at a median of 3.1 years after repair (95% CI, 0.1-13) (Tables 6 and 7). A total of 71 patients underwent only 1 reoperation, 16 patients underwent 2 reoperations, 8 patients underwent 3 reoperations, 1 patient underwent 4 reoperations, and 1 patient underwent 5 reoperations. Freedom from reoperation was not different according to anatomic group or surgical strategy. Factors associated with late reoperation for each surgical strategy in univariate analysis are shown in Table E5. In multivariate analysis, risk factors for late reoperation were IVR: restrictive VSD (P = .04) and subaortic obstruction (P = .04); IVR-RVOT reconstruction: duration of CPB (P < .01); and IVR-ASO: subaortic obstruction (P = .03), concomitant procedure (P < .01), and days of intubation (P < .01).
Table 6Indications for late reoperation according to anatomic group
In the overall analysis, risk factors associated with reoperation in univariate analysis are shown in Table E3. In multivariate analysis (Table E4), including IPCC classification, surgical strategy, concomitant surgical procedure, and duration of CPB, concomitant surgical procedure (OR, 1.6; 95% CI, 1-2.5; P = .03) and duration of CPB (OR, 1.004; 95% CI, 1.004-1.009; P < .01) were the 2 factors associated with late reoperations.
Discussion
We report a large series of patients with DORV who underwent biventricular repair over a period of 21 years in 2 surgical centers. The early mortality rate was 7.4%, and 10-year overall mortality rate was 13.8%. The reoperation rate was 6% for early reoperation, and 10-year overall reoperation rate was 38.6%. Anatomic type of DORV influenced the outcomes, because ncVSD was associated with a higher risk of death and reoperation compared with the other groups. Mortality was higher in the Taussig–Bing type than in the VSD type. These differences may be related to the fact that the anatomic type of DORV is the main determinant of the surgical strategy. However, late survival and freedom from reoperation were not different for the 3 surgical cohorts.
recently reported a similar series with a shorter follow-up, with early death in 4.4%. The contemporary character of the study and lower proportion of Taussig–Bing type (20%) in their series may explain the difference with the present report. Kleinert and colleagues
We observed that DORV with ncVSD was the only anatomic subgroup in which early and late mortality rates were significantly higher compared with other anatomic groups. Belli and colleagues
Biventricular repair of double outlet right ventricle with non-committed ventricular septal defect (VSD) by VSD rerouting to the pulmonary artery and arterial switch.
reported biventricular repair in a high proportion of DORV with ncVSD, the choice of this approach as opposed to a univentricular palliation program should be carefully analyzed, because the risk of death and reoperation remains high. The distance between the VSD and the great arteries,
and the evaluation of the outflow tracts, AV valves, and global anatomy are the main parameters that should be analyzed to indicate a biventricular repair or a univentricular strategy in DORV with ncVSD. Redirecting the ncVSD to the unobstructed RVOT associated with the ASO procedure, performed in 5 cases in our series, appeared to be a valuable alternative in this unfavorable anatomic subset.
Acquired RVOT obstruction was the main cause of reoperation in the Taussig–Bing type compared with the VSD type. Hayes and colleagues
hemodynamically significant subaortic obstruction (valvar or subvalvar aortic stenosis) and anatomic substrate for subaortic obstruction (expressing itself as coarctation of the aorta or interrupted aortic arch) were the only anatomic factors, with VSD location associated with early mortality.
The anatomic factors associated with late mortality (except ncVSD) were restrictive VSD, mitral cleft, and coronary artery anomalies. Restrictive VSD is a major anatomic parameter that influences preoperative clinical tolerance, timing of surgical intervention, duration of procedure, and intracardiac tunnel geometry. The enlargement of VSD, which was indicated in front of a diagnostic of restrictive VSD during the preoperative evaluation or the surgical procedure (when the surgeon deemed it necessary to intraventricular baffle), was the most frequent additional procedure (23/45, 51%). In the series by Li and colleagues,
the most frequent additional procedure (23/45, 51%) was the enlargement of the VSD, which was indicated when restrictive VSD was diagnosed during the preoperative evaluation or during the surgical procedure (when the surgeon considered it necessary). This was performed in only 35 of 122 patients (29%) in our series. Nevertheless, the IVR, with or without enlargement of VSD, carries a high risk for reoperation for VSD residual (or modification of baffle repair) ahead of DORVs or all types of conotruncal abnormalities.
Intramural ventricular septal defect is a distinct clinical entity associated with postoperative morbidity in children after repair of conotruncal anomalies.
We believe this surgical parameter requires a particularly precise surgical management (size and localization of septal resection) to avoid risks in the short term (conduction disorders, residual VSD) and long term (subaortic obstruction). In regard to reoperation for left ventricular outflow obstruction, one may conclude that an aggressive approach with almost routine VSD enlargement may decrease late left ventricular outflow obstruction. Nevertheless, because it was difficult for us to demonstrate this approach statistically, we have to consider these conclusions and suggestions cautiously. A prospective study would be appropriate to investigate this.
Outcomes According to Surgical Strategy
IVR-ASO was the surgical strategy that carried the highest risk of early mortality independently of the anatomic subtype of DORV. Of note, IVR-ASO was performed primarily in infants with lower weight and who required higher duration of CPB (Table 4). These factors are known to be risk factors for early mortality after surgery for congenital heart diseases.
Cardiac surgery in infants with low birth weight is associated with increased mortality: analysis of the Society of Thoracic Surgeons Congenital Heart Database.
The need to build an intracardiac baffle from the VSD to the neoaorta, the higher incidence of abnormal coronary artery origin/course, and the frequent association of aortic arch obstruction, resulting in a longer procedure, may explain this trend.
In this study, we could not demonstrate any differences in late mortality according to the initial strategy. The small number of deaths in the first year postsurgery (4/433, 1%) certainly limits this conclusion. Li and colleagues
reported 2 deaths (2/151, 1%) in their series, with a follow-up of 4.3 years. Late mortality is low after biventricular repair in DORV, and adverse events are mainly reoperations. Indeed, approximately one third of the patients in our series had to undergo a reoperation (Table 7). Some 43% of these reoperations were related to left heart residual or acquired anomalies. The types of left heart anomalies observed were different according to the initial surgical strategy. Subaortic obstruction related to baffle stenosis was the main cause of reoperations in the IVR and IVR-RVOT groups (90% of indications), whereas it occurred in only one third of patients undergoing reoperation after IVR-ASO. Conversely, aortic arch repair or ascending aorta and aortic valve repair were frequent in the IVR-ASO group and rare in the 2 other surgical strategies. These results are similar to those of Losay and colleagues,
who reported that aortic arch reconstruction (15/39, 38%) and aortic valve repair (13/39, 33%) were the major indications of late reoperation for left heart anomalies in a large series of 1200 transpositions of the great arteries after ASO. Belli and colleagues
also showed that subaortic valve obstruction developed with time and was the main cause of reoperation (73%) after IVR or IVR-RVOT. Of note, abnormal coronary artery distribution was a risk factor for late mortality in multivariate analysis. We were not able to demonstrate that this was related only to the IVR-ASO procedure.
Reoperations involving the right heart obviously were more frequent in patients who had a RVOT reconstruction at initial repair. In this IVR-RVOT group, 60% (28/47) of late reoperations were related to RVOT anomalies with half of reoperations related to replacement of a conduit.
Residual VSD closure was required in 10% of the cases with no difference in the 3 strategies.
The initial strategy is determined by the underlying anatomy of the DORV. The VSD location is the major determinant to choose between the available options. Decisional algorithms have been designed to choose the optimal surgical procedure according to different anatomic factors.
When the initial repair is performed, late outcomes are similar in the different anatomic subtypes of DORV. This is certainly reassuring for the counseling of families, particularly when a prenatal diagnosis of DORV has been made.
Study Limitations
The main limitation of this study is in the anatomic classification of DORV, which is insufficiently complete to allow the anatomic groups to coincide with surgical strategies. The relationship of the VSD to the great arteries and the presence or absence of subpulmonary obstruction are necessary but not sufficient elements to decide on the proper surgical strategy for a DORV. This is the reason why various surgical strategies are presented for each anatomic group in our study.
Our surgical classification for 3 main groups has forced us to assemble different strategies for IVR-RVOT enlargement, such as the Lecompte or Rastelli procedures. We were unable to analyze separately the different substrategies in the IVR-RVOT group. For patients with the Taussig–Bing type undergoing a Lecompte or Rastelli procedure, we classified them in the IVR-RVOT surgical group because the aorta was not mobilized during surgery (no arterial switch by definition). That is why in our series the Taussig–Bing type does not automatically imply a classification for the IVR-ASO strategy. Risk factor analysis is by nature a predictor of outcome, which may, in turn, be influenced by varying treatment options.
Conclusions
DORV is a complex and heterogeneous anatomic group of congenital heart anomalies. Despite constant improvements in outcomes, mortality and need for reoperation remain significant because of the clinical challenge imposed by the “most complex” anatomic forms. The ncVSD type carries a higher risk of events compared with the other anatomic types of DORV, and the biventricular (vs univentricular) strategy in this setting needs to be evaluated by specific studies. The IVR-ASO group requiring a neonatal procedure has the highest risk of early death. One third of the patients had to undergo reoperation, and the type of reoperation depends on the initial repair. The absence of difference in late mortality or reoperation among the 3 surgical strategies is certainly an important element, particularly for prenatal counseling.
Conflict of Interest Statement
Authors have nothing to disclose with regard to commercial support.
Appendix
Table E1List of variables considered for multivariable analyses in overall analysis
European Association for Cardio-Thoracic Surgery, Society of Thoracic Surgeons, List S. European Paediatric Cardiac Code (IPCCC Short List). April 1, 2012. 2012:1-14. Available at: http://ipccc.net/index.php/download-the-ipccc/. Accessed June 13, 2016.
Biventricular repair of double outlet right ventricle with non-committed ventricular septal defect (VSD) by VSD rerouting to the pulmonary artery and arterial switch.
Intramural ventricular septal defect is a distinct clinical entity associated with postoperative morbidity in children after repair of conotruncal anomalies.
Cardiac surgery in infants with low birth weight is associated with increased mortality: analysis of the Society of Thoracic Surgeons Congenital Heart Database.
30457-3&id=fx1.jpg){kind=link}
30457-3&id=gr1.jpg){kind=link}
30457-3&id=fx2.jpg){kind=link}
30457-3&id=gr2.jpg){kind=link}