Ventricular septal defect with pulmonary atresia: approaches, results, prognosticators and current status

Abstract

Ventricular septal defect with pulmonary atresia and major aortopulmonary collateral arteries is a complex congenital cardiac anomaly with a wide spectrum of anatomical variations. Akin to the same, the management options are also very diverse ranging from aggressive single-stage repair with unifocalisation to surgical palliation and/or staged repair and also heart transplant. There is no consensus on the best management option. This review aims at highlighting the various surgical options and proposing a management pathway suited for the subcontinent patients.

Introduction

Ventricular septal defect (VSD) with pulmonary atresia (PA) consists of a spectrum of disorders ranging from atresia of right ventricle to pulmonary vasculature connection and presence of confluent or non-confluent mediastinal branch pulmonary arteries to complete absence of mediastinal pulmonary arteries and major aorto-pulmonary collateral arteries (MAPCAs), being the only source of pulmonary blood flow. 20% to 40% of cases with VSD-PA are likely to have additional pulmonary blood supply from the systemic circulation in addition to the mediastinal branch pulmonary arteries [1, 2]. This variation in the anatomy leads to variable clinical presentation and thus opens a wide debate towards the possibilities of management of such patients. On one hand, the Stanford group advocates unifocalisation of haemodynamically significant MAPCAs and complete repair in single or as a staged procedure [3], while on the other hand, the Melbourne group concentrates on strategies for rehabilitation of the native pulmonary arteries rather than the utilisation of the MAPCAs. Only patients presenting with heart failure were offered single-stage complete repair and unifocalisation by the latter. In the following sections, we would like to discuss the various approaches, results and prognosticators and try to provide with the current status of the management of VSD-PA with MAPCAs. Heart-lung transplant has been mentioned as a modality of management in scattered reports, but will not be discussed below [4, 5].

Single-stage approach

Reddy et al. proposed the single-stage approach in the mid-1990s which laid the pathway for complex surgical unifocalisation of pulmonary blood supply and closure of VSD [6].

Anatomical considerations

Complete surgical repair for VSD-PA is achieved by VSD closure and providing single source of blood supply to the pulmonary bed. It is very essential to understand the anatomy of the MAPCAs and their anatomical and physiological relation with the mediastinal pulmonary arteries. The anatomy of the collateral arteries needs to be delineated from origin to their intra-parenchymal drainage to decide their fate. The key points to be considered before undertaking surgical repair can be summarised as:

1. MAPCAs:

   1. Origin: Site of origin is most commonly from the anterior surface of the descending aorta at the 4th to 6th thoracic vertebral level. In addition, they can be found arising from the under surface of the arch, the subclavian artery or the innominate artery. Rarely, they can originate from the distal descending aorta or the abdominal aorta [7, 8].
   2. Course: The mediastinal course of the collateral vessels can be straight or tortuous. They can be of variable length and calibre with variable degree of stenosis at the origin, at the distal end or at multiple levels. The anatomical relations to the carina, bronchi and oesophagus are important for planning the dissection, mobilisation and unifocalisation of these arteries. Retro-oesophageal MAPCAs are difficult to tackle and require different surgical technique as compared to those arising from the anterior surface.
   3. Drainage: The blood supply to the lung segments can be dual or solitary depending upon whether it is supplied by both collateral and native pulmonary arteries or by one of them.

Depending on the anatomical factors and the clinical status of the patient, repair of VSD-PA-MAPCAs is planned as a single-stage or a multi-staged procedure.

Hanley et al. from the Stanford group have been the forerunners of single-stage complete repair. In this strategy [3], all neonates are evaluated for anatomy of the MAPCAs and the mediastinal pulmonary arteries and the strategy is planned in correlation with the clinical condition of the patient. If the neonate is saturating < 75% or > 90%, the patient undergoes surgery irrespective of the anatomy. What type of surgery is done is determined by the anatomical details. If mediastinal pulmonary arteries (PAs) are present and confluent with good arborisation with dual supply of MAPCAs, surgical aorto-pulmonary window is created to allow for the growth of the native PAs. On re-evaluation at 6 months of age, pulmonary artery augmentation is done if required. If not, complete repair is performed in single stage without unifocalisation of MAPCAs. When the MAPCAs are found to be significant and favourable for surgery, the neonates undergo single-stage unifocalisation and complete repair. If the MAPCAs are unfavourable for complete repair, the patient undergoes a ductal stent or a systemic to pulmonary artery shunt and planned for unifocalisation and complete repair at 6 months of age.

If the neonate saturates > 90%, regardless of the anatomy, complete single-stage repair is performed. If saturation is between 75 and 90%, no neonatal intervention is performed and elective surgical repair is performed at 4–7 months of age. Intra-operative flow study is performed for these patients at this time. If the pulmonary artery pressures are found > 25 mmHg at flows of 3 l/min/m², right ventriculotomy is avoided and the patient undergoes a systemic to pulmonary artery shunt. If the pulmonary artery pressures are found normal, complete repair is performed.

The largest data published comes from the Stanford group of Frank Hanley et al. [3], who have presented an experience of 458 patients from November 2001 to April 2016. Fifty-four percent of the total patients underwent complete single-stage repair. They report an 88% repair rate at a median age of 8.6 months. Early and late mortality was 3.5% and 8.9% respectively. Median pulmonary artery to aortic pressure ratios were 0.40. The survival was around 85% at a 5-year follow-up. The Stanford group concludes that their strategy of early complete surgical repair and unifocalisation with inclusion of all lung segments along with extensive lobar and segmental pulmonary artery repairs has achieved very good results in patients with or without interventions done prior to undergoing repair at their centre.

The Birmingham group published data of 275 cases between 1988 and 2016 [1] where they achieved unifocalisation in 88% of their cohort. Early mortality was 2.8% with an estimated 10- and 15-year survival of 93% and 88% respectively. No statistical difference was found between survival of patients with absent or present mediastinal pulmonary arteries, a finding similarly concluded by Frank Hanley et al. in their experience.

Rehabilitation of native pulmonary arteries

In the 1970s, Kirklin et al. and Mc Goon et al. [9–11], in two separate experiences, utilised the strategy of rehabilitation of the native pulmonary arteries. The concept involves provision of forward flow into the mediastinal pulmonary arteries to assist in their growth, without unifocalising the MAPCAs and reassess later for complete surgical repair.

In 2005, the Melbourne group published their experience of unifocalisation procedures in management of these patients [12]. 82 patients, over a period of 20 years, underwent unifocalisation (n = 130), systemic to pulmonary shunt (n = 119) and MAPCAs ligation procedures (n = 76). 65% of the patients achieved complete repair. Twelve-year survival after complete repair was around 50%. Overall survival of patients to the age of 30 years was 58%. After a mean duration of around 3 years, among the 60 unifocalised MPPCAs, 26 got thrombosed, 12 developed stenosis more than 50% and 29 others showed no signs of growth. The shunt procedures performed in 29 patients recorded increased growth of the pulmonary arteries. They thus concluded that surgeries performed in patients with VSD-PA-MAPCAs should be aimed at promoting growth of the native pulmonary arteries rather than unifocalisation procedures.

Rehabilitation procedures comprise the following operative techniques:

1. Central shunt by Laks technique: An end-side anastomosis is done using 3 mm or 3.5 mm expanded Polytetraflouroethlene (ePTFE) tube graft between the native main pulmonary artery and ascending aorta (Fig. 1).

2. Melbourne shunt: This involves the direct implantation of the main pulmonary artery to the ascending aorta in an end-side fashion (Fig. 2). In contrast to the Laks technique, where there is minimal distortion of branch pulmonary arteries, in this technique, there can be stretching of the right pulmonary artery behind the aorta and it is difficult to adjust the flow in this type of shunt.

3. Palliative trans-annular patch: In this particular technique, the right ventricular outflow tract (RVOT) is slit open across the atretic pulmonary valve and extending proximally onto the right ventricle. The RVOT is the reconstructed using a patch after extensive muscle bundle resection.

4. Right ventricle to pulmonary artery connection: Antegrade flow to the native branch pulmonary arteries can be achieved by using a valved or a valveless conduit from the right ventricle. Valveless conduit (ePTFE tube graft) can be used of a restrictive size, and blood flow across a valved conduit can be adjusted by banding the conduit (Fig. 3). The Birmingham group favours this technique [1] in view of the advantages of avoiding the fall in diastolic blood pressures and preferential streaming of de-oxygenated blood into the pulmonary circulation which achieves better oxygenation than a systemic to pulmonary shunt. In addition, it provides easy access for any catheter interventions required during the interim period prior to complete surgical repair.

5. Branch pulmonary artery patch plasty: Extensive hilum to hilum patch augmentation of the central pulmonary arteries (Fig. 4) might be required as an interim procedure before complete repair to achieve favourable size and growth of the native vessels.

Fig. 1.

Laks technique for central shunt. (source: Yves d’Udekem. Rehabilitation of pulmonary arteries in pulmonary atresia, VSD and MAPCAs, Operative Techniques in Thoracic and Cardiovascular Surgery: A Comparative Atlas 2019)

Fig. 2.

Melbourne Shunt. (source: Yves d’Udekem. Rehabilitation of pulmonary arteries in pulmonary atresia, VSD and MAPCAs, Operative Techniques in Thoracic and Cardiovascular Surgery: A Comparative Atlas 2019)

Fig. 3.

Right ventricle to pulmonary artery connection using a non-valved e-PTFE tube and b valved conduit banded to avoid excessive blood flow. (source: Yves d’Udekem.Rehabilitation of pulmonary arteries in pulmonary atresia, VSD and MAPCAs, Operative Techniques in Thoracic and Cardiovascular Surgery: A Comparative Atlas 2019)

Fig. 4.

Hilum to hilum patch augmentation using e-PTFE patch. (source: Yves d’Udekem. Rehabilitation of pulmonary arteries in pulmonary atresia, VSD and MAPCAs, Operative Techniques in Thoracic and Cardiovascular Surgery: A Comparative Atlas 2019)

• 6.Central pulmonary artery replacement: Replacement of the pulmonary arteries which fail to grow proximally but are good sized distally can be replaced by ePTFE tube graft from hilum to hilum (Fig. 5).
• 7.Lobar branch re-implantation into central pulmonary artery: A significant segmental branch can be anastomosed in an end-side fashion onto the native pulmonary arteries at the time of initial stages or at the time of complete repair (Fig. 6). Alternatively, it can be achieved by an interposition graft between the segmental artery and the central pulmonary arteries.

Fig. 5.

Replacement of the proximal central pulmonary arteries with an e-PTFE tube graft. (source: Yves d’Udekem.Rehabilitation of pulmonary arteries in pulmonary atresia, VSD and MAPCAs, Operative Techniques in Thoracic and Cardiovascular Surgery: A Comparative Atlas 2019)

Fig. 6.

Re-implantation of lobar pulmonary artery on the central branch pulmonary artery a direct anastomosis and b using e-PTFE tube graft. (source: Yves d’Udekem. Rehabilitation of pulmonary arteries in pulmonary atresia, VSD and MAPCAs, Operative Techniques in Thoracic and Cardiovascular Surgery: A Comparative Atlas 2019)

Soquet et al. from the Royal Children’s Hospital, Melbourne, in their latest experience of rehabilitation strategies [13], concluded that this strategy can be implemented in 90% of the cases. From June 2003 to December 2014, among a total cohort of 37 patients, 33 had a mean of 2 procedures before complete repair. Median age at the time of shunt was 3.3 weeks and complete repair was achieved in 73% patients at a median age of 1.7 years. Mortality after the rehabilitation was 10% (n = 3) in their cohort. Five out of the 33 patients had failure of rehabilitation procedure.

The surgical team at the Royal Children’s Hospital, Melbourne, have adopted the strategy of utilisation of native pulmonary arteries as the major source of pulmonary blood supply rather than utilising the MAPCAs for the same after their experience in 2005 [12]. In their recent experience [13], they achieved a repair rate of 82%. There were no early deaths but 10% late mortality among the cohort of 33 patients. After a median follow-up of 4.5 years, the VSD could be closed in 73% of patients at a median age of 1.7 years. Median ratio of pressure in right ventricle and left ventricle (pRV/pLV) ratio was 0.64 at a median of 22 months after repair.

Combined strategy

Iyer and Mee in 1991 [14] proposed the concept of combined strategy for management of VSD-PA-MAPCA where rehabilitation of native vessels was followed by unifocalisation and complete repair. There is an overlap between the strategies of unifocalisation and rehabilitation. In the unifocalisation strategy, it includes both the MAPCAs and the native pulmonary arteries which are involved in providing pulmonary blood supply. In the rehabilitation strategy, the MAPCAs with dual supply are also the ones being utilised while the native pulmonary arteries are being rehabilitated. It is quite evident by the review of recent literature regarding management strategies for VSD-PA-MAPCAs, that most of the centres adopt more of a combined strategy in which the patients may undergo a preparatory procedure for diminutive pulmonary arteries, with or without unifocalisation, followed by complete repair.

Gupta et al. from University of California at Los Angeles (UCLA) medical centre, Los Angeles [15], adopted the strategy of rehabilitation as the first stage followed by unifocalisation in the second and complete repair as the final stage. A total of 104 patients were operated, of whom 63 underwent stage 1 and 92 were unifocalised. Fifty-eight patients achieved complete surgical repair at a median age of 5.2 years. Median pRV/pLV intraoperatively was 0.5. Early mortality was close to 11% and late mortality was 5%. Babliak et al. reported an overall survival rate of 92% among cohort of 83 patients. Twenty-eight patients were managed by a single-stage strategy. More than 80% of these underwent complete surgical repair. Patients with 22q11 deletion had poor survival as compared to ones with no chromosomal abnormality [16]. Song et al. [17] from Korea reported a significant decrease in size of MAPCAs on serial angiography( from 5.2 ± 2.9 to 4.1 ± 2.9 ) after a mean duration of 20 months. they also proposed an integerated approach for better long term survival. 

Catheter interventions

Re-interventions on the pulmonary arteries or MAPCAs were required to be done in the event of stenosis or thrombosis. Balloon dilatation or percutaneous stenting is required post unifocalisation for the native pulmonary arteries and the unifocalised MAPCAs. To avoid overcirculation, collateral vessels with dual supply need to be coil embolised either before or after the surgery. According to Dragulescu’s experience [18], all of their patients had interventional catheterisations consisting of 36 pulmonary angioplasties and 11 stent implantations after the initial RV-PA connection. In another experience of Chen et al. [19], among a cohort of 69 patients, 16 patients (23.9%) with inadequate native PA growth underwent percutaneous PA re-intervention to address PA stenosis. Twenty-two patients (32.8%) with adequate native PA growth had surgical pulmonary angioplasty at complete repair. Comparing the patients with and without PA re-intervention, pulmonary stenosis requiring angioplasty at RV-PA connection was the only risk factor associated with PA re-intervention.

Palliation

Patients with unfavourable anatomical attributes like poor or absent central pulmonary arteries, or multiple solitary supply collaterals, or the ones who are high risk for post-operative suprasystemic right ventricular pressures after complete repair, are palliated with systemic to pulmonary shunt or RV-PA continuity in the form of a ePTFE graft or a palliative trans-annular patch. The VSD is left open and these patients are subjected to complete repair later in life. The Melbourne group [13] have successfully achieved unifocalisation in these patients several years after initial palliation. Similar experience was reported by the Birmingham group where in almost 40% of patients, VSD was left open at the time of initial surgery and subsequently closed at the time of complete repair in almost 50% of these patients [1]. In the Stanford experience, in patients who could not undergo complete repair at presentation, they reported central shunts to be a more predictable and protected source of pulmonary blood supply rather than an RV-PA continuity using a conduit. In addition, they reported the formation of RVOT pseudoaneurysm in kids with RV-PA conduit [3].

Prognosticators

1. Chromosomal abnormalities: Stanford data demonstrated that patients with chromosome 22q11 deletion had poor survival as compared to kids with no chromosomal abnormalities. In addition, they reported 5 deaths among 13 patients with Alagille syndrome. Almost all of these patients needed re-intervention on the pulmonary arteries. The results were better in cohort of patients with Alagille syndrome but no MAPCAs.

2. pRV/pLV: Kirklin et al. [9] demonstrated the significance of early post-operative right ventricular (RV)/left ventricular (LV) systolic pressure ratio post complete repair in patients with Tetralogy of Fallot (TOF) and pulmonary atresia. Similar correlation can also be made in patients with VSD-PA-MAPCAs, as experienced by Hanley et al. [20] who reported that patients with RV/aortic pressure ratios more than 0.4 after complete repair were at increased risk of mortality and associated with higher rate of collateral and branch pulmonary artery revision. Gupta et al. [15] reported a pRV/pLV ratio of more than 0.55 as an indicator of poor outcome.

3. Absent central pulmonary arteries: In the Melbourne experience, patients with the smallest central vessels at birth, with a diameter inferior to 1.5 mm, are more likely to fail a rehabilitation strategy. Native vessels with a dual supply seem to grow less if the collateral feeding this territory was large [13]. Mainwaring et al. from Stanford group [20], however, states that the absence of central pulmonary arteries was associated with the same outcome measures compared with patients who had central pulmonary arteries. Nevertheless, re-interventions were noted to be more in number and frequency in patients with absent central pulmonary vessels. Patients with absent central pulmonary arteries usually have large-sized MAPCAs for pulmonary blood supply which makes it more favourable for complete repair as a single-stage procedure.

4. Pulmonary blood flow: The patients with limited pulmonary blood flow have the least suitability for complete repair, and the chances of survival improve with increase in pulmonary blood flow at presentation. Best results have been reported in patients with generous pulmonary blood supply as compared to those with scant pulmonary blood supply [21]. The Melbourne group comes to state that the growth of the native vessel is less if a large collateral supplying the same segment is unifocalised. Thus, they suggest that patients with multiple extensive small MAPCAs and resulting in dual supply to the pulmonary segments should be left palliated with rehabilitation strategy and achieving complete repair at a later stage.

5. RV-PA conduit versus central shunt: As stated previously, the Stanford group have emphasised the placement of central shunt to be a more haemodynamically favourable option for rehabilitation of native pulmonary arteries in kids where VSD is left open. The reason being that RV-PA conduit leads to pulmonary overcirculation and more chanced of a pseudoaneurysm formation in the RVOT.

6. Others: Some other factors which may play a role as a prognostic indicator and a factor for re-intervention are Jagged -1 (JAG 1) chromosome mutation, choice of prosthetic/biological material for branch pulmonary artery repair and choice of conduit (homograft, Contegra, Hancock, ePTFE valved tube grafts, Bovine pericardial valved conduits). None of the studies mentioned state the above as a factor responsible for mortality, but do play a role in the rate of re-interventions the patient undergoes.

7. Long-term prognosticators: Important factors, affecting the long-term prognosis of the surgical procedure performed, are branch pulmonary artery stenosis, dilatation of the aortic root, right ventricular dysfunction, rhythm abnormalities, conduit stenosis and residual or additional intra-cardiac shunts.

The Indian experience

One of the largest experience from India is reported by Murthy et al. [22] who published their experience in 2010 with 124 patients, over a span of 13 years, who underwent single-stage complete unifocalisation with or without complete repair. Median age was 3.2 years. They were able to achieve a complete repair rate of 60% in their cohort. There were 16 early deaths (13%) and 3 late deaths (2.4%). They concluded that single-stage unifocalisation reduced the number of operations and hospitalisations, and was thus less costly than a multistage procedure. Tissue to -tissue anastomosis can promote the growth of the vessels as the children grow. In a personal communication with Rajesh Sharma et al., in their experience of 102 patients over a period of 10 years, two management strategies were followed namely single-stage complete repair and staged unifocalisation and repair. Management strategy was decided on the basis of clinical presentation and calculation of Nakata index and total neo-pulmonary artery index (TNPAI). Early mortality was around 12% for the full cohort. Late mortality was around 7%. Complete repair and septation could be achieved in almost 82% of the patients.

Summary

On retrospective review of latest studies published on VSD-PA-MAPCAs, there has been found a wide variation in the age at first surgery for the patients with VSD-PA-MAPCA (range 3.3 weeks to 6.3 years). 22q11 microdeletion has been seen in around 35 to 40% of the patients and having a negative prognostic implication on the outcome of treatment. This could be due to variation in pulmonary vasculature and the MAPCAs anatomy in kids with this chromosomal anomaly. The prevalence of absent mediastinal pulmonary arteries was 15 to 20%. Single-stage repair has been extensively followed with good results by several groups. Early unifocalisation and complete repair is currently the most common strategy followed, but in a specific subset of patients (around 15%), whose clinical presentation is severely cyanotic and poor anatomic substrate with absent/poor mediastinal pulmonary arteries or underdeveloped intrapulmonary vascular tree, the unifocalisation strategy cannot be adopted. These subset of patients undergo rehabilitation procedures for the native pulmonary vessels or unilateral unifocalisation procedure with a rehabilitation surgery. Hanley et al. achieved complete repair rate of around 85% at an average age of 8.6 months with an early mortality rate of 3.5%. The Birmingham group reported an early mortality rate of 2.8% for the 249 patients who underwent single strategy repair. Udekem et al. [13] of the Melbourne group have achieved an 82% repair rate at an average age of 1.7 years with the staged approach. Utilising the combined strategy, Babliak and colleagues [16] have had a repair rate of 60% with no early mortality in a cohort of 60 patients. The anatomy of the aorto-pulmonary collaterals is different from the mediastinal pulmonary vessels and the outcomes after unifocalisation of these vessels is variable as suggested strongly by the Melbourne group. The rehabilitation of the native pulmonary arteries is the cornerstone to achieve good long-term results. Decision regarding the management of the patients has been largely based upon clinical presentation and computerised tomography (CT) or angiographic assessment of the pulmonary tree for the number of segments perfused though some groups [15, 18] have suggested use of various indices [23, 24, 25] for the pulmonary arteries and the collaterals to decide suitability for repair. Intra-operative flow study was utilised by some groups for the same [3, 5, 26]. Most of the groups achieved a repair rate ranging from 40 to 95%. The early mortality rate was reported to be less than 5% in the bigger cohort. Survival rates ranged from 78 to 85%.

Proposed management strategy

As is evident from the aforementioned review of various management strategies for the cardiac disorder with wide spectrum of presentation, it is very difficult to segregate the different management routes. Combined strategy following a common theme is more likely to provide a better outcome rather than following a single strategy rigidly. Based on the extensive review of a number of publications and keeping in mind the scenario in developing countries, We would like to propose the following management strategies (Figs.7, 8).

Fig. 7.

Management algorithm for VSD-PA-MAPCAs at age of presentation less than 9 months

Fig. 8.

Management algorithm for VSD-PA-MAPCAs at late presentation

Conclusion

The surgical management of VSD-PA-MAPCAs is presently divided into three main categories, namely unifocalisation, rehabilitation and combined strategy rather than stating as single-stage or multistage pathway. With the exception of the neonates undergoing complete repair early in life with unifocalisation and VSD closure, the rest of the patients are treated with a combined strategy. Best results with unifocalisation have been seen in the subset of patients with absent mediastinal pulmonary arteries and/or solitary supply MAPCAs.

The overlap between different management plans is evident in the finding that the patients with small native branch pulmonary arteries undergoing unifocalisation of MAPCAs need to have a systemic to pulmonary shunt or an RV-PA conduit for augmenting the growth of the native vessels. On the other hand, patients in the rehabilitation plan of management would need unifocalisation of the MAPCAs which forms the sole supply to the particular lung segment or segments. Catheter angiography with computerised tomography scan is required to accurately plan the procedure according to the anatomy of the pulmonary tree and the MAPCAs. Complete repair rates of around 70 to 80% have been achieved in both the rehabilitation and the unifocalisation pathways. Repeated surgical or catheter re-interventions might be required for both the unifocalised collaterals, branch pulmonary arteries and RV-PA conduit with or without pulmonary valves. Much more detailed, structured and uniform data collection is necessary to compare large multicentre data reviews to achieve more information and draw substantial conclusions.

Funding

None.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethics committee approval

Not required being a review article.

Informed consent

Not required being a review article.

Human and animal rights statement

Not required being a review article. No animals were involved.