Background Due to increasing success with repair or palliation in childhood, there is a rapidly growing population of adult patients with complex congenital heart disease who may require transplantation. There remains little data on outcomes of cardiac transplantation in this group.
Methods 38 orthotopic cardiac transplants were performed in 37 patients (18 men) ≥18 years of age with congenital heart disease (CHD) from 1988 to 2009 in our institution. Outcomes were reviewed using medical records and transplant databases.
Results 15 patients (41%) had univentricular and 22 (59%) biventricular physiology. The biggest group was transposition of the great arteries following atrial switch in eight patients (22%). Six (16%) had no previous surgical intervention. Mean age at transplant was 33.5 years (range 19.1–59.9 years). 11 patients (30%) required additional surgical procedures at transplant. 16 (43%) died, 12 early and 4 late deaths (1.8, 2.4, 2.7 and 7 years). Survival was 70% at 30 days, 68% at 1 year, 58% at 5 years and 53% at 10 and 15 years. Outcome improved in later eras with reduction in 30-day mortality from 50% to 18% and increase in 5-year survival from 50% to 69%. Two patients developed post-transplant lymphoproliferative disease. None required long-term renal replacement therapy. One patient was re-transplanted for cardiac allograft vasculopathy.
Conclusions While operative mortality following cardiac transplantation for adult congenital heart disease is higher than for other diagnostic groups, long-term survival is good and comparable to patients without CHD. Disappointing early results are improved with increasing experience.
- Congenital heart disease
- cardiac transplantation
- heart transplant
- paediatric cardiology
Statistics from Altmetric.com
Improved outcomes for patients with congenital cardiovascular malformations have resulted in a growing population of adult patients with congenital heart disease (CHD).1 A significant proportion of these patients will require, or need to be considered for, transplantation in adult life.
Many patients with univentricular physiology will have undergone previous palliative procedures. Late complications following completion of Fontan procedure are significant and unfortunately relatively common.2 They include ventricular failure, progressive atrioventricular valve regurgitation, protein-losing enteropathy, pleural effusions, lymphopaenia, hypogammaglobulinaemia and atrial arrhythmias. Once these complications develop, they are often refractory to standard medical management, and transplantation may be the only option for therapy. Patients who have had a biventricular repair may also develop ventricular failure, especially those with a systemic right ventricle such as following the atrial switch procedure for transposition of the great arteries.
Patients with CHD provide specific challenges around the time of transplant which may significantly influence outcome. Complicated reconstructive anatomy needs to be taken into account, and this may increase operative mortality.3 There may be the presence of pre-formed human leucocyte (HLA) antibodies due to the previous use of homografts4 leading to difficulty in obtaining a suitable donor organ. Some of these patients are also at particular risk of elevated pulmonary vascular resistance (PVR), a well-recognised risk factor for early organ failure.5
In this study, we describe the experience, in a single institution, of cardiac transplantation in the adult population (≥18 years) with CHD. This should help inform discussion as to whether or not cardiac transplantation is a realistic and worthwhile option for the increasing population of adults with congenital cardiovascular malformations.
Freeman Hospital is one of six adult cardiac transplant centres in the UK and one of the two paediatric centres. The first adult cardiac transplant in this institution was in 1985, with the first paediatric transplant performed in 1987 and the first transplant for adult CHD in 1988.
Patients (aged ≥18 years) with CHD who underwent cardiac transplantation were identified using the cardiopulmonary transplant database and hospital medical records. Patients who underwent heart–lung transplantation were excluded from this study. Pre-transplant diagnosis and demographics, transplant procedure including additional surgical procedures at the time of transplant, survival and morbidity were reviewed. The data were subdivided into three eras, 1988–1994, 1995–2001 and 2002–2009. Long-term survival was evaluated using Kaplan–Meier survival curves.
Between January 1985 and July 2009, 658 primary orthotopic cardiac transplants were performed in adults (aged ≥18 years) in our institution. Of these, 38 (5.8%) were carried out for CHD. One of these was a re-transplant making 37 patients in total in this group.
Immunosuppression protocols were comparable to those in patients without CHD. Oral ciclosporin and azathioprine were given just before the transplant with intravenous methylprednisolone 500 mg given intra-operatively at cross-clamp release. Three further doses of intravenous methylprednisolone (125 mg 8 hourly) were then given and patients maintained subsequently on oral prednisolone, with dose weaned by 12 months to a maintenance of 0.1 mg/kg. Long-term maintenance therapy was with calcineurin inhibitors (ciclosporin or tacrolimus), an antimetabolite (azathioprine or mycophenolate mofetil) and oral prednisolone.
Thirty-seven patients ≥18 years of age with CHD received an orthotopic cardiac transplant, 18 men and 19 women. There was an increase in the number of transplants performed over time with nine patients (24%) transplanted in 1988–1994, 12 (33%) in 1995–2001 and 16 (43%) in 2002–2009. Mean age at transplant was 33.5 years (range 19.1–59.9 years). Mean follow-up post-transplant at the time of study was 7.9 years (range 0.4–21.5 years). There was one re-transplant.
Fifteen patients (41%) had univentricular physiology; 22 patients (59%) had biventricular physiology. The proportion of patients with univentricular physiology fluctuated over time from 44% in 1988–1994 to 58% in 1995–2001 and 25% in 2002–2009. Diagnoses and previous surgical interventions for each patient are shown in table 1. Twelve patients had previously undergone an aorto-pulmonary (AP) shunt procedure as initial palliation (one had a Pott's shunt, three had Waterston shunts, the remainder were Blalock–Taussig shunts).
The three biggest diagnostic groups were simple transposition of the great arteries (TGA), double inlet left ventricle (DILV) and tricuspid atresia (TA).
Eight patients (22%) had TGA (seven previous Mustard's procedure and one previous Senning). Six patients (16%) had DILV—two of these patients had previously undergone Fontan completion and one had no previous interventions. Five patients (14%) had TA—four of these patients had been palliated with an AP shunt, and one was a previous Fontan.
Three patients had previously undergone completion of Fontan circulation (as outlined above, two of the patients with DILV and one of the TA group).
Six patients had no previous surgical or trans-catheter intervention, and these had diagnoses of congenitally corrected TGA (2), Ebstein's anomaly (1), DILV (1), pulmonary stenosis/atrial septal defect (1) and hypoplastic right ventricle(1).
Eleven patients had atrial arrhythmias, and nine had permanent pacemakers.
Five patients had severe renal insufficiency (serum creatinine >2.5 mg/dl or 221 μmol/l) pre-transplant. An additional two patients were receiving peritoneal dialysis for chronic renal failure (table 2). Four of these seven patients had biventricular physiology, and three had single ventricles. One patient (patient C in table 2) had an immune complex glomerulonephritis on renal biopsy. In the remaining six patients with renal impairment, no underlying cause other than low cardiac output was diagnosed.
Cardiac catheterisation results were available for review in 19 patients. Catheterisation results for nine patients were not available. Mean PA pressure for those measured was 20 mmHg (range 13–40 mmHg), mean transpulmonary gradient 9 mmHg (range 7–14 mmHg) and mean PVR 2.6 (range 1.6–5.0) Wood U/m2. In 10 patients not catheterised, PVR was assumed to be low based on history, physical examination and physiology.
HLA antibody testing was not routinely done post-transplant in our population. Pre-transplant panel reactive antibodies (PRAs) were not available in 12 patients. In the 25 patients screened, the majority (84%) had 0% PRAs. Four patients had positive PRAs (55%, 87%, 5% and 25%). Retrospective crossmatch was negative in all patients.
Thirty-three of the 37 (89%) patients were transplanted from home. Four patients were in hospital on inotropic support at the time of transplant. No patients were on ventricular assist device support before the transplant.
Eleven patients (30%) needed additional surgical procedures at the time of transplant (table 3). Mean ischaemic time was 210 min (125–285 min). No patient required a thoractomy in addition to midline sternotomy.
One patient required insertion of an intra-aortic balloon pump in order to be weaned off bypass. This was subsequently removed, and he was successfully discharged home.
Outcomes and survival
Early (30-day) postoperative survival in the whole population was 70.3% (11 deaths). There were five late deaths after 30 days (table 4), one at 86 days in a patient with pre-existing renal failure who died of gastro-intestinal ischaemia and sepsis without discharge from hospital. There was one late death at 1.8 years due to an acute respiratory infection. The other three late deaths were at 2.4, 2.7 and 7 years, all due to cardiac allograft vasculopathy (CAV). One of these late deaths due to CAV was in a patient who had been re-transplanted for CAV in the original graft.
Overall, four of the 15 patients (27%) with univentricular physiology died. One death was in a Fontan patient, one following cavo-pulmonary (Glenn) shunt and two after AP shunts. Three of these deaths were early with the fourth at 86 days. Twelve of the 22 patients with biventricular physiology (55%) died (eight of these deaths within 30 days).
Surgical complications leading to early deaths occurred in five patients. In one of these cases, death was due to severe intra-operative bleeding. Two patients died of bleeding post-operatively, one patient suffered an intra-operative air embolism (although cardiopulmonary bypass had been obtained via the femoral route, injury to the systemic ventricle occurred causing an air embolism). In the fifth patient, there was difficulty with the pulmonary venous anastomosis leading to pulmonary venous obstruction.
Donor organ failure occurred in five patients. In two of these cases, primary right ventricular (RV) failure occurred, thought due to undiagnosed high PVR. The other three patients developed biventricular failure. One of these patients died on day 12, and in this case, the donor heart was small (donor weight 15 kg less than recipient weight). Another died in surgery, and in this case, a donor with marginal function had been accepted due to the recipient's poor state at time of transplant. The remaining death was also due to primary graft failure with both LV and RV failure.
Kaplan–Meier survival was 70.3% at 30 days, 67.6% at 1 year, 58.2% at 5 years, 52.9% at 10 years and 52.9% at 15 years (figure 1a). There was no survival difference between men and women (p=0.5) or age at transplant (p=0.85).
The data were subdivided into three eras, 1988–1994, 1995–2001 and 2002–2009. Survival was improved in the later eras with a reduction in 30-day mortality from 50% in 1988–1994 to 18% in 2002–2009. Five-year survival increased from 50% to 69% (figure 1b).
Morbidity and other outcomes
Two patients (5%) developed post-transplant lymphoproliferative disease, at 4.1 and 8.2 years post-transplant. Both of these patients subsequently died of CAV.
Seven patients (19%) required dialysis post-transplant (two of whom were on dialysis pre-transplant). Of these seven patients, four died within 30 days. No long-term survivors have developed severe renal insufficiency (defined according to the International Society for Heart and Lung Transplantation (ISHLT) criteria of either chronic dialysis, renal transplant or creatinine >2.5 mg/dl). One patient had a successful pregnancy 2.9 years post-transplant.
One patient was re-transplanted at 9.3 years following the first transplant for CAV and epicardial constriction. She was inotrope dependent at the time of re-transplantation. She developed recurrent CAV in her second graft and died 2.7 years following re-transplantation.
Assessment for cardiac transplantation
Data on number of referrals to our institution and assessments for cardiac transplantation in adults with CHD were available for a 10-year period (Jan 2000 to Dec 2009). During this time, 67 patients were assessed and 29 (43%) listed. Of the 29 patients listed, 4 were subsequently de-listed (all four remain alive), 4 died while waiting, 2 remain listed and 19 were transplanted. Median time on the waiting list for those patients transplanted was 223 days (range 10–3121 days). Thirty-eight patients were not listed following assessment (11 of these patients have subsequently died). Numbers assessed per year have fluctuated over time with approximately 10 patients assessed per year in the last 2 years.
Improvements in diagnosis of congenital cardiovascular malformations over the last 20 years, as well as progress in medical treatment, surgical repair and palliation has led to a significant increase in the number of children surviving to adult life with CHD.1 7 8 This population has previously been estimated as increasing by 1600 adult cases per year in the UK.1 There is understandable concern that many patients will develop late problems with cardiac function and need consideration for transplantation. The perceived high risk in patients with complex congenital heart disease may, however, reduce their likelihood of being transplanted, particularly with the current relative scarcity of donor organs.
Our data show that early mortality following cardiac transplantation in this group is high—30%—and there are a number of reasons for this. An increased risk in patients with univentricular physiology has been previously described9 10 and although our mortality figures do not demonstrate this, this may be partly a reflection of the small numbers of patients with univentricular physiology in our series. Only three patients in our series had undergone Fontan palliation; therefore, we acknowledge that we are unable to make firm conclusions based on our data about this patient group. Despite the risks for patients with failing Fontan circulations, transplantation does offer some hope, and complications such as protein losing enteropathy have been shown to recover following transplant.11 12 Additional co-morbidities including the presence of aorto-pulmonary collaterals and liver impairment are important and need evaluation. In our series, there was one early death in a Fontan patient due to RV failure. This was a patient with borderline PVR in whom a smaller donor was accepted and was before our current practice of attempting to use oversized donors in the setting of higher PVR. The other two Fontan patients remain well suggesting that good long-term outcome is achievable.
PVR in these patients needs to be evaluated at the time of listing. In general, a transpulmonary gradient of <12 mmHg and PVR of ≤5 Wood U/m2 would be considered acceptable, and if higher values are found at catheterisation, pharmacological reversibility may provide reassurance that cardiac transplantation instead of heart–lung transplant may be suitable.13 Improvements in post-transplant care and support of the right ventricle means that in the current era, moderate elevations in PVR are not necessarily a contraindication to transplant.14
Patients with CHD may have complex reconstructed anatomy or have undergone previous palliative procedures requiring modification of surgical techniques, and increased length of donor vessels may be needed to correct structural or positional abnormalities of the atria, great vessels or systemic and pulmonary venous drainage.3 This may lead to increased ischaemic times when compared to transplantation for cardiomyopathy, but there has not been shown to be a significant difference in some published reports.14 15 The largest patient group with biventricular physiology in our series was TGA following atrial switch. These patients require atrial septal reconstruction using excess donor tissue. Furthermore, in this condition, the aorta is usually anterior and to the right as it exits the heart requiring complete mobilisation of recipient vessels and adequate lengths of donor arteries for correction.16 Pulmonary artery reconstruction was the commonest additional procedure in our series, carried out in 19% of our patients, and this may be especially needed in the setting of previous AP shunts and PA distortion. Previous studies have shown an increase in both short- and long-term mortality in patients requiring PA reconstruction which does not necessarily correlate with the degree of reconstruction required.17
An often significant concern is the presence of pre-existing HLA antibodies which may be associated with previous reconstruction using homograft material.4 Increased HLA antibodies may lead to decreased likelihood of obtaining a suitable donor organ and may significantly increase the amount of time these patients spend waiting. Interestingly, in our population, this was not found to be a major problem in the majority, although in the current era with increasingly sensitive screening techniques, this may become a more significant factor.
Currently, transplants for adult CHD make up only 3% of recent procedures reported to the ISHLT.18 Mean recipient age for adult heart transplants is 51 years,19 whereas in the CHD group, this is significantly lower14 20 and was 33 years in our series. Furthermore, although CHD is identified as a strong risk factor for 1-year mortality following cardiac transplantation, in those who survive 3 years, CHD actually has a 10-year survival advantage.13
Early and mid-term survival in our institution has improved in the more recent era which is in keeping with other reports,14 21 although longer-term survival remains uncertain due to more limited follow-up periods. It is difficult to compare results of transplantation for CHD with those transplanted for other causes due to confounding factors (including median age and sex differences). Thirty-day survival for non-congenital patients in our institution was 87.5% over the period studied which, although better than for congenital patients, is difficult to compare. Furthermore, just because figures are worse does not mean that patients with CHD should be excluded from transplantation. Results are changing, and improved survival now seen in the CHD group can largely be related to reduced peri-operative mortality with increasing surgical experience as well as improvements in intensive care support. We acknowledge the fact that it is difficult to extrapolate our results to settings such as emergency transplants or patients on ventricular assist device support, and the impact of possible increased numbers of patients with more complex anatomy being referred in subsequent years remains to be seen.
Transplantation does increase life span in patients with CHD and heart failure. In our study, we saw an increase in numbers with single ventricle physiology transplanted in the middle era, but in more recent years, this group has only comprised one quarter of the patient numbers. Absolute numbers, however, are relatively unchanged. More recently, we have seen an increase in patients with biventricular physiology, mainly a reflection of patients with TGA following Mustard or Senning's procedures presenting with systemic ventricular failure. With more patients now undergoing Fontan palliation, this group could be postulated to increase in future eras.
In our experience, assessment and transplantation of patients with adult CHD carries a major workload. Preoperative assessment of PVR is potentially difficult and probably a major contributor to early organ failure. Considerable input is needed from other health professionals, in particular the renal, hepatic and microbiology teams. The need for transplants to be carried out by surgeons experienced in congenital heart surgery is also clear as evidenced by the need for adjunctive surgery on the pulmonary arteries, aorta and systemic and pulmonary veins.
In conclusion, early mortality in the adult CHD group is higher than for adult patients transplanted for other causes but with comparable or increased late survival. Our data would support the contention that these early results will continue to improve with greater experience by specialised multidisciplinary teams. Although patient numbers are likely to continue to increase, the overall numbers will remain small relative to the adult cardiac transplant population. The organisation of transplantation services for adults with CHD, therefore, needs ongoing discussion at national and international level with continued focus. As these patients require specialised surgical expertise to ensure the continued improvement in short-term outcome, they should ideally be managed as part of an adult heart and lung transplant programme at large national centres, which have the appropriate support structures and experience in order to optimise outcomes.
The authors would like to thank Vaughan Carter of the Department of Histocompatibility and Immunogenetics, National Health Service Blood and Transplant, Newcastle upon Tyne for his assistance with obtaining pre-transplant antibody data.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.