A staged approach for truncus arteriosus: proceed with caution

Primary neonatal repair has become the established standard of care for truncus arteriosus (TA) over the past 30 years [1]. Prior to this, in the 1970s and 1980s, it was common to advocate for medical management with the aim for repair at 3–6 months. The argument for neonatal repair was easily made due to the very high mortality in delaying surgery attributed to uncontrolled congestive heart failure, truncal valve regurgitation and pulmonary hypertensive injury [2, 3]. As part of a palliative staged approach, pulmonary artery banding (both internal and external bands) was also attempted early on in surgical management of TA, but difficulty balancing the circulation resulted in extremely disappointing outcomes with mortality of 50–70% [4, 5]. The manuscript from Ota et al. [6] in this issue of the Journal is an intriguing return to the concept of pulmonary artery banding (PAB) in TA and a philosophy of staged repair. The report comes from the national database of the Japanese Cardiovascular Surgical Database, which is close to a mandatory reporting system and so gives a valuable ‘allcomers’ overview of national practice. To those of us fixed on the concept of neonatal repair, it is a fascinating opportunity to look at a revisited older approach in the current era. At face value, it appears that the outcomes of a staged approach are encouraging, with the cumulative operative risk appearing to be similar to primary repair (11.9% vs 16%)—suggesting that avoiding major bypass surgery in the neonate could be a valid option. Looking into the data in more detail, the justification for staged approach is less clear. There is no record of any interstage mortality, yet 20% of the banded patients never achieved complete repair. It is certainly possible that some patients are simply en route to planned repair, but the majority of these ‘missing’ cases are already >18 months post-PAB [6, see Supplementary figure], and so well beyond even the longest time that they would be expected to survive with this palliation. The authors have no way of tracking these patients and the Japanese Cardiovascular Surgical Database should have flagged them up, even if they had undergone repair at a different institution. The inevitable conclusion is that many of these ‘lost’ patients may not have survived, and until there is complete follow-up, there is no way of knowing the true outcomes of a staged approach, which could have a combined mortality of up to 30% without this data. TA is a notoriously fragile circulation, especially in the presence of truncal regurgitation. Due to the proximity of the pulmonary arteries to the aortic root, the physiology of pulmonary artery banding can be associated with significant diastolic run off, which is exacerbated in the presence of truncal regurgitation. A further weakness in this study is the paucity of information on the grading and prevalence of truncal regurgitation—or whether the degree of regurgitation influenced the decision to offer primary repair or staged approach. Furthermore, PAB in the setting of interrupted aortic arch requires either inpatient care on prostaglandin or attempting PDA stenting, neither of which are attractive options in a group of patients who did not undergo their definitive repair until a median of 5 months of age. Only 1 patient received a PDA stent, so we presume the remainder of cases remained under inpatient care on prostaglandin infusion. Stepping away from the detail for a moment, it is important to ask how an entire national strategy appears to have evolved that challenges the accepted standard practice around the world. The answer may lie in the fact that (i) the advent of bilateral PA banding for hypoplastic left heart syndrome (the ‘hybrid Norwood’) has taught us the skills of branch pulmonary artery banding far more successfully than was done 30 years ago and (ii) low volume surgery for rare conditions leads to the resurgence of palliative strategies that avoids the need for complex neonatal repairs. The latter is not meant to be a criticism; it is a very pragmatic approach to the recognition that neonatal TA repair is complex surgery in low volume centres. However, it inevitably raises the well-rehearsed arguments in the support of centralization of complex neonatal surgery into large volume centres that can benefit from programmatic expertise and experience. It is difficult to support the authors’ recommendations that a staged approach is preferable, given that there was wide variation in practice and no consistency in how or why each approach was chosen in each center—nor what criteria favoured 1 approach over the other. There are also very limited data on the clinical status and comorbidities associated with the inter-stage period, the degree of cyanosis that was tolerated and how the timing for definitive repair was chosen. Having said this, there is evidence that the staged approach was valuable in the low birth-weight cohort (albeit a small sub-group) and avoiding major surgery and cardiopulmonary bypass in the <2-kg patients may be a valuable suggestion [6]. The authors need to provide us with complete follow-up of the entire cohort and supplement this with clear actuarial survival curves for each strategy before we can make any valid conclusions on this intriguing insight into the management of TA. Conflict of interest: none declared.