4 Discussion
Recently, the results of the ASO for TBV have been excellent, with low hospital mortality rates as reported by Vergnat et al. (16) (5.8%), Soszyn et al. (6) (5.3%), and Sinzobahamvya et al. (7) (2.9%), among others. The risk factors for overall mortality are related to the coronary patterns, a higher weight at the time of the operation, preoperative PA banding, aortic arch anomaly and postoperative subneopulmonary obstruction (1, 6, 16, 17). Results from our study, which were achieved in LMIC show that, intraoperative ventricular septal defect enlargement [HR = 7.23, P < 0.001], secondary aortic cross clamping (HR 28.38, P< 0.001), post-operative pneumonia (HR 5.64, P = 0.023), and post-operative sepsis (HR 5.28, P = 0.017) were factors associated to overall mortality. These results reflect the reality of the learning curve faced by a congenital heart program in a LMIC as they develop their management of complex lesions, including dealing with missed diagnosis, late presentation and nosocomial infection.
According to many previous reports, TBV is a risk factor for RVOTO-related reoperation after the ASO compared to simple transposition of the great arteries (TGA) or even complex TGA (7, 8, 18, 19). Realizing this problem, we attempted to avoid RVOTO as much as possible by systematically checking the RVOT and totally resecting or dividing SPTs inferiorly up to the apex of the right ventricle (14). Interestingly, in 3 patients who required reoperation for RVOTO in our study, 2 of them did not undergo the RVOT procedure to resect SPTs during the ASO, as their procedures were performed in the early period of the study. The remaining patient who developed RVOTO was a TBV patient with type B aortic arch interruption, and her symptoms occurred 6 months after the ASO despite the attempt to resect the SPTs in the first operation. During the reoperation, the level of the RVOTO was far from the neo-PV, and we performed resection of the SPTs through the neo-PV and the tricuspid valve. We also noted that the right ventricle end diastolic volume may increase after SPT resection as the free wall of the right ventricle is now free from the ventricular septum.
Risk for RVOTO has been reported to occur in up to 53% of patients after ASO with TBV and remains a major cause of late morbidity (7) . Even when a preventative strategy is applied, reintervention due to RVOTO remains as high (20-35%). Only the experience from the Royal Melbourne Children’s Hospital reported an RVOT reintervention rate of less than 10%, specifically 2.2% (6). So, our rate of RVOT reintervention of only 6%, with a larger cohort is noteworthy, and the estimated freedom for RVOTO reoperation was 91.9% (SD=4.8) (95% CI, 0.7524 to 0.9749) at 9 years follow-up. A smaller native aortic valve annulus has been identified as a risk factor for RVOT reintervention (8). In contrast, we demonstrated growth of the neo-PV annulus as shown in the Fig. 4. We attribute this RVOT growth and the low incidence of RVOTO reintervention to our policy of aggressive division and/or resection of SPT.
Another common reason for reoperation after the ASO is supravalvular neo-PA stenosis (1-3, 17). However, in our series, we did not have any patients who developed neo-PA stenosis requiring either reoperation or reintervention. This complication was also rare in our ASO cohort (1/445 ASO patients). Attempts were made to size the autologous pericardial patch to be equal the height of the neo-PA, while the width of the neo-PA was equal to the diameter of the pulmonary bifurcation. We believe that a large patch does not guarantee the normal growth of the PA in the long term; instead, a size difference between the neopulmonary and pulmonary bifurcation may create turbulent flow at the anastomotic site and easily lead to stenosis. This approach has helped us avoid torsion of the neo-PA bifurcation and has allowed for normal growth of the neo-PA.