Heart–lung and lung transplantation in grown-up congenital heart disease: long-term single centre experience☆

Abstract

Objective: Because of considerable progress in paediatric cardiac surgery life expectancy of patients with congenital heart disease (CHD) has improved significantly over the years. There are a growing number of adults with CHD presenting with progressive decline of cardiopulmonary function and Eisenmenger’s syndrome. We analysed our experience with heart–lung and lung transplantation in this patient group. Methods: Since 1988, a total of 46 heart–lung transplantations and 5 double lung transplantations have been performed in adults with CHD at our institution. Underlying diagnoses were: ventricular septal defect, atrial septal defect, persistent ductus arteriosus and others. Pulmonary hypertension was present in all patients. Twelve patients had undergone previous cardiac procedures. All patients were included in this retrospective analysis. Mean follow up was 5.1 ± 4.7 years. Patient survival was estimated with the Kaplan–Meier method and analysed using the log-rank test. Results: Thirty-day mortality was 11.8% (n = 6). Survival was 80% at 1 year, 69% at 5 years and 53% at 10 years. Major causes of death were infection and sepsis, chronic rejection, initial graft failure and acute rejection. Compared to the overall mortality after lung and heart–lung transplantation for other indications at our institution there was no significant difference (1 year, 5 years, 10 years: 76%; 60%; 45%), but a tendency towards a better long-time survival of the CHD patients. Conclusions: Lung and heart–lung transplantation can be performed with an acceptable risk and a favourable long-term outcome in patients with grown-up CHD. Careful patient selection and planning of the surgical strategy is essential in this high-risk patient population.

1 Introduction

Because of considerable progress in paediatric cardiac surgery, life expectancy of patients with congenital heart (CHD) disease has improved significantly over the last 50 years [1,2]. Before the era of congenital heart surgery, less than 20% of children born with congenital heart defects survived to adult life. Now, most deaths from CHD occur in adults [2]. But even today most surgical procedures are palliative rather than curative procedures. As a consequence a growing number of adult patients with CHD present with progressive decline of cardiopulmonary function. For these patients cardiopulmonary transplantation becomes the only treatment option [2]. There are estimates that 10–20% of all patients with CHD will be potential candidates for heart, lung or heart–lung transplantation [3,4]. According to the ‘Guidelines for the Management of Grown up Congenital Heart (GUCH) Disease’ transplantation must be considered when short-term prognosis is reduced or quality of life is unacceptable [2].

A significant proportion of GUCH patients develop irreversible pulmonary hypertension [5,6]. In those patients isolated cardiac transplantation cannot be performed with an acceptable risk and thus, heart–lung and lung transplantation with simultaneous repair of the cardiac defect become the only treatment options. We analysed our experience with heart–lung and lung transplantation in adults with end stage CHD and Eisenmenger’s syndrome.

2 Patients and methods

2.1 Patients

A total of 76 heart–lung and 429 bilateral lung transplantations have been performed between January 1988 and June 2004 at our institution. A retrospective analysis of patient charts identified 51 adult patients (>16 years old) that underwent lung or heart–lung transplantation for congenital heart disease. Patients undergoing isolated heart transplantation and single lung transplantation have been excluded from this analysis.

2.2 Morbidity and mortality

Early death was defined as death before day 30. Underlying diagnoses, previous operations, associated procedures and survival have been documented. Long-term outcome was compared to the non-GUCH patients after heart–lung and bilateral lung transplantation.

2.3 Statistical analysis

Data were prospectively recorded and retrospectively analysed. Actuarial survival was estimated with the Kaplan–Meier method and analysed using the log-rank test. A two-sided p value of less than 0.05 was considered statistically significant. All data were analysed using the Statistical Program of Social Sciences (SPSS, version 13.0, SPSS, Inc., Chicago, Il).

3 Results

3.1 Patient population

Between January 1988 and June 2004 a total of 51 patients underwent heart–lung (HLTX) and double-lung transplantation (DLTX) for grown-up congenital heart disease at our institution. There were 29 women and 22 men with a mean age of 34 ± 10 years (range 17–52 years) with an average follow up of 5.1 ± 4.7 years. Forty-six patients received heart–lung transplantation and five patients received bilateral lung transplantation (Table 1 ).

3.2 Underlying diagnoses and previous operations

Underlying diagnoses and previous operations are listed in Table 1. The most common diagnosis was ventricular septal defect in 18 patients. Other diagnoses were: atrial septal defect (n = 4), persistent ductus arteriosus (n = 7), transposition of the great arteries (n = 2), double-outlet right ventricle (n = 2), atrioventricular septal defect (n = 3), single ventricle anatomy (n = 3), pulmonary atresia (n = 5), tricuspid atresia (n = 3), truncus arteriosus (n = 2) and others (n = 2).

Twelve patients had undergone previous surgical procedures. In the DLTX group, one VSD closure and one PA banding had been performed. In the HLTX group, two patients had received diagnostic sternotomy without any surgical correction. Four had received modified BT-shunt (2 of them combined with other complex procedures). VSD closure (n = 1), repair of aortic coarctation (n = 1), Mustard operation (n = 1) and PA banding (n = 1) had been performed in the other patients. Indication for transplantation was a significant decline of cardiopulmonary function combined with pulmonary hypertension in all patients (Table 1).

3.3 Organ preservation and surgical procedures

For cardioplegia St. Thomas solution was used in 11 patients, University of Wisconsin (UW) solution in 18 patients and histidine–tryptophan–ketoglutarate (HTK) solution in 17 patients according to the era of transplantation. For preservation of the lungs Euro-Collins solution was used in 25 patients, whereas 26 organs were preserved with Perfadex solution. Mean ischaemic time was 255 ± 44 min in the heart–lung group and 322 ± 65 min in the bilateral lung group (for the second lung).

All but one heart–lung transplantations were performed via median sternotomy. In one patient transverse thoraco-sternotomy (clamshell incision) was performed because of severe adhesions. Tracheal anastomosis was performed in all heart–lung transplants. Venous anastomoses were accomplished using either the atrial or bicaval technique. Bilateral lung transplantations were performed via clamshell incision in all patients. During all procedures cardiopulmonary bypass was used.

3.4 Additional surgical procedures

Six patients required additional surgical procedures at the time of transplantation. In three patients of the DLTX group repair of cardiac defects has been performed simultaneously: tricuspid valve repair and PFO closure in one patient, VSD closure and de-banding of the pulmonary artery in the second patient and ASD closure in the third patient. In the HLTX group one patient received repair of aortic coarctation, one patient underwent additional renal transplantation and in one patient atypical lung resection was performed because of size mismatch (Table 1).

3.5 Immunosuppression and postoperative management

Postoperative immunosuppression consisted of an induction therapy with anti-thymocyte globulin (ATG) followed by a triple maintenance therapy based on cyclosporine or tacrolimus in combination with azathioprine or mycophenolate mofetil and steroids. All patients were followed up for life in our transplant outpatient department.

3.6 Re-transplantations

Two patients after heart–lung transplantation underwent redo-transplantation (DLTX) because of severe bronchiolitis obliterans syndrome (BOS III) with significantly impaired pulmonary graft function. These re-transplantations have been performed 5.7 and 7.7 years after the initial transplantation. Unfortunately, both patients died within the first 30 days after re-transplantation. One of them died because of severe infection and the other because of early graft failure (Table 2 ). Two patients after HLTX are listed for redo lung transplantation because of severe bronchiolitis obliterans syndrome. Another patient suffers from BOS III but is not a candidate for re-transplantation because of severe co-morbidities. Redo heart–lung transplantation has not yet been performed.

3.7 Morbidity and mortality

Thirty-day mortality was 11.8% in all GUCH patients. There were 6 early and 13 late deaths. Causes of death are summarised in Table 2. Among the early deaths, one patient who underwent DLTX and simultaneous VSD closure and PA de-banding died early because of severe infection. Five patients died early after HLTX. Among them, there were two operative deaths: in one patient initial graft failure occurred, in the other patient severe bleeding resulted in biventricular graft failure. The other patients died of infection (n = 1), acute graft failure or rejection (n = 1) and other reasons (n = 1). Causes of late deaths were infection and sepsis in seven patients, one of them after re-transplantation. Acute graft failure occurred in one patient after re-transplantation. The other patients died of severe bronchiolitis obliterans syndrome (n = 2), pulmonary embolism (n = 1), pancreatitis (n = 1), and other reasons (n = 1) (Table 2, Fig. 1 ).

Univariate and multivariate analysis for the identification of risk factors for mortality was performed. However, none of the variables or any combination of age, gender, diagnosis, ischaemic time, previous operations (p = 0.054) or era of transplantation reached statistical significance. Nevertheless, previous operative procedures seem to have a negative impact on survival after transplantation: mortality in the previously operated group was 6/12 patients (4 early and 2 late deaths) whereas mortality in the patients without previous surgical intervention was 13/39 (2 early and 11 late deaths).

The actuarial survival rate was 80 ± 6% at 1 year, 69 ± 7% at 5 years and 53 ± 9% at 10 years. Compared to the overall mortality of the non-GUCH patients after lung and heart–lung transplantation at our institution there was no significant difference: 1 year survival was 76 ± 2%, 5 year survival was 60 ± 3% and 10 year survival was 45 ± 4% (Fig. 2 ).

4 Discussion

This study describes our experience with heart–lung and lung transplantation in adults with end stage congenital heart disease and Eisenmenger’s syndrome. Although patients with unrecognised cardiac malformations resulting in severe and irreversible elevation of pulmonary vascular resistance have become less common nowadays, there is a growing number of patients with end-stage CHD presenting with failing cardiopulmonary function after one or more palliative procedures. Some of these patients develop pulmonary hypertension despite early correction. For this group cardiopulmonary transplantation becomes the only treatment option [5–7].

According to the ‘Guidelines for the Management of Grown-up Congenital Heart Disease’, published in 2003, there are some important differences in the GUCH patients compared to other indications for cardiopulmonary transplantation [2]:

Timing of transplantation is difficult because risk stratification scores that are available for end-stage congestive heart failure may not apply to grown-up patients with congenital heart disease. This makes decisions regarding timing of transplantation difficult. Patients are well adapted to their limited working capacity and thus, often present very late to our transplantation units. Hosseinpour et al. [3] point out that survival of adult patients with Eisenmenger’s syndrome is hard to predict. Therefore, deteriorating quality of life and increasing number of hospital admissions may indicate the optimal time for listing [3]. In this context waiting list mortality has to be considered. According to Smits et al. waiting list mortality for patients with Eisenmenger’s syndrome on the elective heart–lung list is 17.6% per year within the Eurotransplant area. Compared to patients suffering from end-stage pulmonary hypertension and other end-stage diseases with a 1-year waiting list mortality of 44% and 33%, respectively, survival of the Eisenmenger patients is significantly better. Probability of transplantation is 41% per year in the Eisenmenger patients [8].

Another significant issue in this patient group is technical considerations. The surgical procedure may be complicated by multiple previous operations and complex anatomic variations. Especially in cyanotic cardiac malformations multiple pleuro-pulmonary collateral vessels may cause profuse bleeding [3,4]. Therefore, Pigula and co-workers advocate access by bilateral thoraco-sternotomy (clamshell incision) rather than sternotomy in order to facilitate haemostasis [6]. We also perform clamshell incisions in patients where severe bleeding from adhesions is anticipated. In addition, complex anatomic variations may preclude implantation of the organ by the conventional technical approach. Technical adjustments, including cannulation techniques, may become necessary [3].

Among our patients, only 24% had undergone previous cardiac surgery, two of them received merely explorative sternotomies without any further surgical intervention. This is due to the fact that 34 years ago, which is the mean age of our study group, paediatric cardiac surgery was still an evolving field. Today, there are only few cardiac defects that are not eligible for early surgical repair or palliation [1,5]. As a consequence, the development of severe and irreversible elevation of pulmonary vascular resistance can be prevented in the majority of patients and thus, isolated cardiac transplantation can be performed, if necessary. On the other hand, previous surgical interventions increase the operative risk. Severe bleeding may occur during re-entry to the chest and anatomic variations may require additional reconstructive surgery. Thus, longer operation and bypass times may lead to prolonged organ ischaemia and impaired postoperative organ function [5]. As Hosseinpour et al. emphasise, a close communication between the donor and the recipient surgical team is essential, because the donor team must allow time for the recipient team to prepare for the implantation of the donor organ. In particular, meticulous haemostasis must be achieved before implantation [3].

At our institution thoracic computed tomography is routinely performed in all patients presenting for lung and heart–lung transplantation. Multidetector computed tomography has now been widely accepted as an appropriate non-invasive diagnostic tool for preoperative assessment of complex cardiac anatomy in re-operative heart surgery [9–11]. Three-dimensional reconstructions are particularly helpful in patients after operative correction of congenital heart disease, because they provide a clear visualisation of cardiac anatomy which is easily understandable not only for specialised paediatric cardiac surgeons. A clear definition of the anatomy is crucial for the surgical strategy. Moreover, the proximity of the heart and vessels to the sternum and chest wall can be estimated. Thus, cannulation of the femoral vessels may be indicated. Finally, the existence of pleuro-pulmonary collaterals can be assessed. Because of our experience with fatal bleeding from such collaterals (patient no. 2, Table 2), we consider the existence of extensive pleuro-pulmonary collaterals a contraindication for heart–lung transplantation. Fig. 3 shows a CT scan of a patient with pulmonary atresia and multiple aorto-pulmonary collateral arteries (MAPCAS) who presented for heart–lung transplantation because of progressive cardiopulmonary failure and Eisenmenger reaction. Unfortunately, this patient had developed multiple pleuro-pulmonary collaterals and has therefore not been accepted as a candidate for heart–lung transplantation at our institution (Fig. 3).

Among the presented transplant recipients, there were 46 patients that underwent heart–lung transplantation and 5 patients that underwent bilateral lung transplantation. In contrast to other groups, we do not support single lung transplantation in combination with cardiac surgery in patients with Eisenmenger’s syndrome. Waddel and colleagues point out that SLTX combined with cardiac repair might be a therapeutic option for these patients, although the results of single lung transplantation in pulmonary hypertension are not convincing [5]. Our experience with SLTX in Eisenmenger’s syndrome consists of three patients who were not included in this study. Two of them required re-transplantation in terms of double-lung transplantation because of severe organ dysfunction, one of them within the fourth postoperative week. Both patients recovered after re-transplantation. One patient died early after single lung transplantation. Although single lung transplantation is theoretically feasible in these patients and although there are several theoretical advantages such as shorter bypass time, less bleeding b.o., less invasive procedure and organ sparing procedure, the postoperative course is significantly more difficult and exposes the patients to a high risk. In five patients in this study, repair of the cardiac defect has been performed prior or simultaneously to double lung transplantation. Thus, the donor heart was preserved for another recipient. This therapeutic approach has also been successfully performed by other groups [5,6,12]. Unfortunately, most patients in this study suffered from severely impaired cardiac function or anatomic variations that excluded isolated lung transplantation combined with cardiac repair.

Another crucial issue in this young patient population is long-term survival after cardiopulmonary transplantation. Chronic rejection and infection still remain major causes of death despite improved immunosuppressants and antibiotics. Redo lung transplantation after heart–lung transplantation is technically feasible, but was not successful in both patients in this study. They died within 30 days after re-transplantation, one because of severe infection and the other one because of early graft failure (Table 2). Ten-year survival of patients with congenital heart disease after cardiopulmonary transplantation was 53% in our patients and 37% according to the ISHLT registry report [13]. Compared to the overall mortality of the non-GUCH patients after lung and heart–lung transplantation at our institution there was no significant difference [14,15]. Therefore, we consider cardiopulmonary transplantation a reasonable therapeutic option for this patient group despite the above-mentioned difficulties. Detailed knowledge about anatomical variations, exclusion of multiple pleuro-pulmonary collaterals and careful planning of the operation may reduce the operative risk in these patients. However, considering the young age of this patient group and the significant shortage of donor organs, transplantation should always be considered as the final therapeutic option in patients with congenital heart disease. Early diagnosis of cardiac defects and adequate surgical or interventional treatment remain the most effective strategies to avoid cardiopulmonary transplantation or postpone it as long as possible.

5 Conclusion

Lung and heart–lung transplantation can be performed with an acceptable risk and a favourable long-time outcome in GUCH patients compared to other indications. Nevertheless, it remains the final therapeutic option for CHD and should be postponed as long as possible because of the shortage of donor organs and the limited long-term prognosis.

References