Heart Transplantation in Children for End-Stage Congenital Heart Disease Anees J. Razzouk and Leonard L. Bailey Heart transplantation (HT) as primary therapy for children with congenital heart disease (CHD) has become unusual. With improved early results of reconstructive surgery, the population of children and adults surviving with CHD is expanding. End-stage CHD related to myocardial dysfunction or circulation failure after prior surgery is becoming more common as an indication for HT. This heterogeneous group of CHD recipients referred for HT presents unique decisionmaking, technical, and physiologic challenges. Historically, a diagnosis of CHD has been a major risk factor for early mortality after HT. Rescue HT, especially in the setting of failing Fontan physiology, has the worst outcome. Early referral (before end-organ damage), proper selection, and optimization of recipients, as well as meticulous intra- and postoperative management are crucial to improving early outcomes of HT in this population. Beyond the early post-HT period, children with end-stage CHD experience long-term survival comparable to most other non-CHD recipients. Semin Thorac Cardiovasc Surg Pediatr Card Surg Ann 17:69-76 C 2014 Elsevier Inc. All rights reserved.
Introduction
I
mproved early outcomes of reconstructive surgery has resulted in an expanding population of children and adults who are presently living with their complex congenital heart disease (CHD), many of them living quite well. Primary heart transplantation (HT) has been irrelevant to their survival beyond early infancy. So, this once nearly doomed cohort of neonates, most of whom were born with lethal CHD, is now growing up with expectations of reaching the joys and vicissitudes of adult life, and beyond. In reality, however, data suggest that not all of these youngsters are destined to achieve these ideal expectations. Some will not reach childhood, much less adulthood, without additional surgical help. Instead, 10% to 20% will reach the end-stage of their CHD related to myocardial dysfunction or circulatory failure, and they will be referred for HT.1 They have become, and are, a heterogeneous group of potential HT recipients who present with unique decision-making, anatomic, and physiologic challenges. Historically, a congenital diagnosis has accounted for 59% of infants o1 year of age, 37% of children age 1 to 10 years, 23% Department of Cardiovascular and Thoracic Surgery, Loma Linda University Children’s Hospital, Loma Linda, CA. Address correspondence to Anees Razzouk, MD, 11175 Campus St., Suite 21121, Loma Linda, CA 92354. E-mail: [email protected]
http://dx.doi.org/10.1053/j.pcsu.2014.01.009 1092-9126/& 2014 Elsevier Inc. All rights reserved.
of adolescent patients age 11 to 17 years, and 3% of adults.2,3 HT for CHD has always been associated with worse surgical outcomes than HT for cardiomyopathy because of technical complexities and physiologic challenges of the operation. In addition, diagnosis of CHD remains a significant risk factor for 1-year and 5-year mortality after HT in both pediatric and adult recipients.2,3 Rescue HT, especially in the setting of failing Fontan circulation, has had the worst outcomes. Early referral (before advanced end-organ damage), proper selection and optimization of recipients, and meticulous intra- and postoperative management are crucial to improving operative outcomes of HT in this population. This review will address the unique challenges of HT in patients with CHD, particularly those who have had previous repair or palliation.
Evolving Indications for HT In recent years, the indications for HT in infants and children with CHD have changed significantly with fewer cardiac defects (e.g., complex single ventricle [SV] anomalies) requiring HT as primary therapy. HT is indicated for two-ventricle patients (e.g., tetralogy of Fallot, truncus arteriosus, Shone’s complex, d- and l-TGAs, Ebstein’s anomaly) with end-stage heart failure owing to severe myocardial dysfunction, irreparable valve disease, or intractable arrhythmias. However, failed 69
A.J. Razzouk and L.L. Bailey
70 surgical palliation is becoming the predominant indication for HT in children with CHD. Isomerisms, issues of situs, and visceral heterotaxia may add to surgical complexity but do not contraindicate HT. Single-ventricle patients, of variable anatomic substrates and at different stages of palliation, may require HT for progressive systolic or diastolic ventricular dysfunction, irreparable atrioventricular valve regurgitation, worsening hypoxemia, elevated pulmonary vascular resistance (PVR) or arrhythmias. Patients with systemic complications of Fontan physiology represent a unique and expanding group being referred for HT. Specific indications include protein-losing enteropathy, plastic bronchitis, stroke, thromboembolism, refractory ascites, and early cirrhosis of the liver. Potential candidates with flexible PVR r5 Wood units (wu) and/or transpulmonary gradient r10 mmHg usually have an acceptable risk for HT. Patients with PVR 4 9 wu and transpulmonary gradient 4 16 mmHg are not suitable for HT. Patients with elevated PVR (5 to 9 wu) require preoperative testing of pulmonary vasoreactivity and are at risk for early post HT graft failure. These guidelines may not necessarily apply to patients with Fontan circulation where a non-pulsatile flow state, low cardiac output, and presence of collaterals or a fenestration make the assessment of PVR impractical. In this unique circulation, PVR is often underestimated. Absolute anatomic contraindications to HT in CHD are rare, but may include severe diffuse hypoplasia of pulmonary arteries or irreparable pulmonary venous malformations. Combined heart–liver or heart–kidney transplants may be indicated in patients with irreversible single organ (liver or kidney) failure associated with chronic congestive heart failure.
Timing of Referral for HT The decision for HT in patients with CHD may be simple or remarkably complex. The population is heterogeneous with variability in age, genetic substrate, support systems, structural anomalies, dysrhythmias, non-cardiac risk factors, number and complexity of previous operations, symptoms, and endorgan function. The classic timing of listing for HT for any indication has been when the expected survival at 2 years is 50%. Such timing is difficult to predict in CHD patients who may “survive” for years, albeit with limited functional reserve and marginal cardiac output. While exercise protocols, metabolic testing, and chemical markers may provide significant prognostic information, they are not necessarily reliable in predicting the need for HT. The risks and benefits of additional conventional palliation must be considered and compared with those of HT. For high-risk candidates, mechanical circulatory support as a bridge or destination therapy is evolving as a viable option.4 The period on a HT waiting list can be long, and up to 20% of patients may die waiting, depending on their status and comorbidities.5,6 A diminished quality of life with intermittent symptoms and frequent hospitalizations should prompt referral for HT. Early listing (before end-organ dysfunction and inotropic or ventilatory
support) will reduce waiting list mortality and improve postHT outcomes.
Pre-HT Evaluation The complex anatomy of CHD, especially in the setting of multiple previous operations, requires detailed pre-HT imaging using echocardiography, cardiac magnetic resonance imaging, or computed tomographic angiography. Information regarding the proximity of the heart, great vessels, or conduits to the sternum, the anatomy of the systemic and pulmonary venous connections, the anatomy of the pulmonary arteries, the presence of aortic arch obstruction/coarctation, and the presence of important aorto-pulmonary collaterals is essential. Imaging may extend to the neck and groins to identify alternative sites for cannulation, and to the liver to detect cirrhosis or other liver lesions. Cardiac catheterization and full hemodynamic evaluation is performed in all patients. Coil embolization of significant AP collaterals may be performed pre or immediately post HT to reduce the risk of high-output graft failure. Multiple operations, with exposure to blood products and allograft patches or conduits, contribute to increased anti-HLA antibodies in patients with end-stage CHD. High levels of sensitization may increase waiting time and contribute to early graft failure and post-HT mortality. Panel reactive antibody (PRA) screening identifies sensitized patients and guides in pre- and post-HT immunomodulation therapy. Candidates with high immunologic risk (PRA 4 20%) receive pre HT intravenous immunoglobulin and plasmapheresis, both of which are continued postoperatively, when they also receive thymoglobulin and other immunosuppressive agents. A prospective cross-match is rarely necessary, and has been largely replaced by virtual cross-matching. Pre-HT coagulation abnormalities are often present as a result of liver dysfunction or chronic anticoagulation for arrhythmias, Fontan circulation, mechanical valve prostheses, or thromboembolism. Correction of coagulopathy, when possible, can reduce the need for post-HT blood transfusion and will decrease respiratory and metabolic morbidity. Other preoperative challenges that should be recognized and optimized before HT include compromised nutritional state (especially in the setting of protein-losing enteropathy or ascites), chronic anemia, susceptibility to infection with asplenia or DiGeorge syndrome, chronic hypoxemic state limiting neurologic reserve, and chronic low cardiac output state causing end-organ dysfunction. Acute or acute-on-chronic renal dysfunction is not a contraindication to HT, although its post-HT reversibility is difficult to predict. Infants and young children may require placement of a peritoneal dialysis catheter at the time of, if not before, HT. Older children and adults may require hemodialysis. Minor liver abnormalities (cholestasis, biochemical or coagulation disorders) are common in patients with chronic heart failure and are usually correctable during and after HT. However, SV patients with a Fontan circulation are at risk of developing major
Heart transplantation in children for end-stage CHD hepatic complications such as severe cirrhosis or hepatocellular carcinoma. These changes correlate with the duration of Fontan circulation and require special screening with liver biopsy and computed tomographic imaging. Bridging fibrosis on liver biopsy and hepatocellular carcinoma (after treatment with transarterial chemoembolization) may be acceptable indications for combined heart–liver transplant.7,8
Donor Considerations Major oversizing of the donor graft should be avoided in this population with a scarred, fixed mediastinum. A donor– recipient weight ratio of up to 2:1 is optimal, and the slightly larger graft may be advantageous for the recipient with elevated but responsive pulmonary artery pressure. Knowledge of the recipient’s cardiac anatomy and prior interventions is crucial in carrying out the donor graft recovery. Extra donor tissue is usually obtained en bloc with the donor heart and may include innominate vein, superior vena cava (SVC), full-length pulmonary artery branches (which may preclude lung donation), ascending aorta and the entire arch, and the pulmonary veins (Fig. 1). Except for closure of a patent foramen ovale, preparation of the donor heart is delayed until the time of implantation. Adequate graft preservation and efficient coordination of the donor and recipient operations are critical to the success of HT in this setting of potentially prolonged operative and graft ischemia times.
Operative Considerations Cardiac transplantation in patients with CHD may be technically challenging – especially after multiple prior operations,
71 and at 0200 hours during a weekend. Vascular access (for line placement) can be difficult and may require open cut-downs. Defibrillation pads are positioned on the chest and topical cooling measures are initiated. A repeat sternotomy is carefully accomplished to avoid massive hemorrhage or air embolism. Central cannulation of the aorta is usually possible; but, alternative sites such as both subclavian regions and both groin areas should be prepared as part of the sterile field. A single atrial cannula for venous drainage simplifies the cardiopulmonary bypass (CPB) circuit and allows for systemic cooling to 201C, an approach that offers protection to both recipient and graft. A crystalloid prime is especially desirable for hemodilution of lymphocytotoxic antibodies in pre-sensitized patients. Hypothermic low flow CPB (20 to 30 cc/kg/min) using active suckers for venous drainage facilitates explantation of the native heart and removal of extracardiac conduits and endovascular stents. A portion of the left pericardium is usually excised (to enlarge the mediastinal space and open it to the pleural cavity). Injury to the phrenic nerve is avoided by minimizing the dissection of structures adjacent to it. A variety of techniques have been described that permit orthotopic HT for even the most complex cardiac malformations.9,10 Using native atrial flaps and donor innominate vein, cavocaval connections can be performed even in the presence of abnormal situs and bilateral SVCs (Fig. 2). To avoid compression or stretching of the reconstructed innominate vein, the ascending aorta is kept long for retro-aortic, or is shortened for ante-aortic innominate vein placement (Fig. 3). Anatomic lesions in the native pulmonary arteries are corrected with donor pulmonary artery tissue. Lateralization of the main pulmonary artery anastomosis (Fig. 2) may be required in patients with dextrocardia,
Figure 1 Donor heart preparation for a recipient with situs inversus. The left pulmonary venous orifices are oversewn. The left atrium is opened vertically between the superior and inferior right pulmonary veins to be anastomosed to the recipient’s pulmonary atrial cuff. SVC, superior vena cava; IV, innominate vein; IVC, inferior vena cava. (Reprinted from the Journal of Thoracic and Cardiovascular Surgery, Vol 116; Vricella LA, Razzouk AJ, Gundry SR, et al, Heart transplantation in infants and children with situs inversus; pp 82-86, 1998, with permission from Elsevier.9)
A.J. Razzouk and L.L. Bailey
72
Figure 2 Orthotopic HT in a recipient with situs inversus. A tubular flap of redundant (remnant) systemic atrial tissue is constructed to redirect the left-sided inferior vena cava (IVC) rightward. Leftward lateralization of the main pulmonary artery is achieved by oversewing its orifice on the right and extending the arteriotomy to the left of midline. LSVC, left superior vena cava; LIVC, left inferior vena cava; MPA, main pulmonary artery; LPA, left pulmonary artery. (Reprinted from the Journal of Thoracic and Cardiovascular Surgery, Vol 116; Vricella LA, Razzouk AJ, Gundry SR, et al, Heart transplantation in infants and children with situs inversus; pp 82-86, 1998, with permission from Elsevier.9)
heterotaxy, or situs inversus. Recipient arch lesions (e.g., coarctation, arch obstruction or aneurysms) are reconstructed with donor aortic tissue. In patients with visceral heterotaxy or situs inversus, where the pulmonary atrium is midline or shifted to the right, the donor heart left pulmonary veins are oversewn, the left atrium is opened between the right pulmonary veins, and then anastomosed to the recipient’s right-sided left atrium. In rare cases when combined heart–liver transplantation is necessary, liver transplantation may be performed en bloc with the donor heart, or the procedures may be performed separately.8
Perioperative Considerations Multiple factors contribute to increased risk of bleeding after HT for CHD. These factors include: chronic anticoagulation, liver dysfunction, dense adhesions, multiple thoracotomies, collateral vessels, splanchnic venous congestion, prolonged operative and CPB times, cyanosis, and thrombocytopenia. In as much as possible, meticulous hemostasis is practiced before graft implantation. Transfusion of blood products is often necessary, but massive infusions of blood products will contribute to pulmonary hypertension and graft right ventricular dysfunction. Acute right ventricular failure rarely
Figure 3 Orthotopic HT in a recipient with situs inversus. Conduit formed by donor’s SVC and innominate vein (IV) lies anterior to the aorta after anastomosis to the left SVC. In recipients with bilateral SVCs or bilateral Glenn shunts, similar connections are performed with the right SVC anastomosed to this conduit at the ligature site of donor’s proximal SVC. SVC, superior vena cava. (Reprinted from the Journal of Thoracic and Cardiovascular Surgery, Vol 116; Vricella LA, Razzouk AJ, Gundry SR, et al, Heart transplantation in infants and children with situs inversus; pp 82-86, 1998, with permission from Elsevier.9)
Heart transplantation in children for end-stage CHD requires mechanical support, however, and can usually be managed with inhaled nitric oxide, pulmonary vasodilation, inotropic support, and, occasionally, an open chest. Aggressive dialysis is instituted in patients with renal failure and plasmapheresis is resumed postoperatively in highly sensitized recipients. For the rare patient with a phrenic nerve injury, early plication of the paralyzed hemidiaphragm facilitates weaning from ventilator support. Nutritional support and infection control, which are balanced with adequate immunosuppression, are critical in minimizing post-HT morbidity and mortality.
Results of HT in children with CHD: A single-center experience Four hundred ninety-six pediatric patients (age range, 1 day to 17.7 years) underwent orthotropic HT at Loma Linda University Children’s Hospital for the diagnosis of CHD (N ¼ 357), cardiomyopathy (CM; N ¼ 130), and other (N ¼ 9). Between 1985 and 1999 (early era), HT was the preferred primary therapy for infants with hypoplastic left heart syndrome (HLHS). There were 276 transplants for CHD (80.2%) and 68 transplants for CM (19.8%). The recent era (2000 to 2013) witnessed a decline in the overall number of recipients with CHD (N ¼ 81, 56.6%). There were 62 CM recipients (43.4%). Prior palliation or repair had become much more common among CHD recipients in the recent era. Indeed, the prevalence of prior surgical interventions within the CHD population increased from 27.5% (76 out of 276) in the early era to 58% (47 out of 81) in the recent era. In total, 123 patients with CHD, the prior surgery group (PSG), underwent 255 corrective or palliative operations prior to HT (Tables 1 and 2). The number of such prior surgical procedures was one in 52 patients, two in 41 patients, and three or more in 30 patients. The PSG had a significantly longer Table 1 Congenital Heart Disease Diagnosis Pre-HT No Prior Surgery Group (N ¼ 234) HLHS 165 Non-HLHS 69 Prior Surgery Group (N ¼ 123) Biventricular (N ¼ 49) Septal defects: 14 Severe Shone’s complex: 13 d- and l-TGAs: 9 Tetralogy: 5 Other: 8 Univentricular (N ¼ 74) DORV: 22 HLHS: 16 PA/TA: 11 Unbalanced AVSD: 11 Other 14 Abbreviations: HLHS, hypoplastic left heart syndrome; TGA, transposition of great arteries; DORV, double outlet right ventricle; PA, pulmonary atresia; TA, tricuspid atresia; AVSD, atrioventricular septal defect.
73 Table 2 Prior Surgery Group (N ¼ 123): Last Major Operation Pre-HT Type of Surgery
No. of Patients
Valve repair or replacement Systemic to pulmonary artery shunt Pulmonary artery banding Fontan Glenn Norwood Septal defect repair Other
27 25 17 13 11 9 8 13
average graft ischemia time (min) compared with the no prior surgery group (NoPSG): 304 þ 117 vs. 270 þ 122, P o .013. Similarly, CPB times were significantly longer for the PSG (P o .0001). Overall 30-day operative mortality for the entire CHD cohort was 11.6% and was not significantly different between the two eras. Early mortality for the NoPSG was 10.3% (24 out of 234) and for the PSG was 16.3% (18 out of 123); P o .22. Within the PSG, seven of 23 (30.4%) recipients with PRA Z 20% died early after HT (P o .021) and 10 of the 18 (55.6%) patients who had perioperative renal failure, and required dialysis, died postoperatively (P o .001). Recipients with three or more procedures before HT had a significantly higher operative mortality than those who had two or less such procedures (26.7% vs. 10.8%; P o .032). Although the operative mortality for the Fontan subgroup was high (30.8%, 4 out of 13), it was not significantly different when compared with other subgroups, perhaps because of the small number of patients. Within the PSG, recipients with SV had similar early mortality as those with two ventricles (13.5% vs. 16.3%; P o .66). Beyond the early post-HT period, the difference in longterm outcomes of CHD patients compared with CM patients is not significant. Survival at 15 years is 57.4% versus 66.6%, respectively (P o .304) (Fig. 4). However, the NoPSG of CHD recipients had a slight survival advantage, at 15 years, over the PSG of 61% versus 48.9%, respectively (P o .038) (Fig. 5). Primary causes of early death in the PSG (18 out of 123) were unmanageable pulmonary hypertension and technical problems. The major causes of early death in the NoPSG (24 out of 234) were management issues and acute graft failure.
Discussion Advances in medical, surgical, and transcatheter interventions have significantly improved the early survival of infants and children with CHD. Nevertheless, HT remains the only viable treatment option for some infants, children, and adults with end-stage CHD. In this category are those with previous reparative surgery and deteriorating ventricular function, as well as those with prior palliative reconstruction and failing circulatory physiology. This heterogeneous and often complicated population of potential HT candidates is rapidly growing. It will undoubtedly stretch the resources of heart transplant programs throughout the world. Results of pediatric HT have gradually improved in recent years, and the long-term survival
A.J. Razzouk and L.L. Bailey
74
Figure 4 Kaplan-Meier analysis of patient survival after HT by diagnostic category. Long-term survival is not significantly different in the two diagnostic groups. CHD, congenital heart disease; CM, cardiomyopathy
of HT for CHD, with some exceptions, now appears to parallel that of recipients transplanted for other indications. The diagnosis of CHD, nevertheless, remains an important risk for operative mortality after HT. The risk varies according to age, anatomic substrate, prior surgical history, stage of palliation, and physiologic state at the time of transplantation. While this population of potential HT recipients continues to expand, the decision for HT has become increasingly difficult, as the demand for donated organs far surpasses the limited supply. Comprehensive pre-HT evaluation, including full assessment of risk factors, is an important component of the decision-making process. Indeed, patient selection may be the key factor in reducing early mortality after HT for CHD. The decision for HT may be particularly vexing among patients with failing Fontan physiology. Clearly, rescue HT after acute failure of a Fontan procedure is unwarranted and has a poor prognosis, especially when systemic ventricular function is preserved. This situation is far better avoided than surgically or medically managed. Hence, among patients with SV who are high-risk candidates for Fontan completion, HT should be considered as an appropriate alternative to Fontan completion.11–13
As the population of Fontan survivors grows, this diagnostic subgroup is becoming the predominant category of CHD patients referred for HT. Technical challenges related to complex anatomy and multiple previous operations, along with various degrees of physiologic decompensation (such as chronically elevated central venous pressure, low cardiac output, renal and hepatic dysfunction, elevated PVR, and malnutrition) place this subgroup of CHD patients at considerable additional risk for HT. Numerous studies12–18 have identified recipient Fontan circulation as an independent risk factor for early phase mortality after HT. Both Kovach et al18 and, in a separate report, Davies et al17 found that Fontan completion within 6 months of listing for HT to be a significant risk factor. In a large multicenter study15 of 488 patients (median age, 12.4 years) transplanted for CHD, survival at 3 post-HT months was significantly worse in CHD recipients versus children with CM (86% vs. 94%). A previous Fontan procedure was identified as the highest independent predictor of mortality, with a relative risk of death of 8.6. In 2004, the group at Columbia University19 reported a 1-month mortality of 18.8% (20 out of 106) after HT for CHD.
Figure 5 Kaplan-Meier analysis of patient survival after HT for CHD. Survival of the No Prior Surgery group is statistically better than survival of the Prior Surgery group.
Heart transplantation in children for end-stage CHD The need for pulmonary artery reconstruction was identified as a strong predictor of early mortality. In another study20 from the same institution, which focused on HT after the Fontan operation (n ¼ 24), and the Glenn procedure (n ¼ 11), there were 10 deaths less than 2 months after transplantation. Nine of the deaths occurred in the Fontan group. The 1-year and 5-year survival for the Fontan group was 62.5% and 57% compared with 71.5% and 67.5% for the entire cohort. In their series of 73 pediatric patients who underwent HT for CHD, Simmonds and colleagues1 reported an operative mortality of 17.8% with similar 1-year survival for univentricular (n ¼ 38) and biventricular (n ¼ 35) circulations (75% vs. 78%). As in most other institutions, their HT operative mortality for Fontan recipients was 26.6% (4 out of 15). They noted improved outcomes of HT for CHD since the year 2000, with 1-year survival for CHD patients of 90% compared with 94% for CM recipients (P ¼ .756). A recent study from St. Louis Children’s Hospital12 reported 307 pediatric heart transplants performed for CHD (57%), CM (39%), and retransplant (4%). The CHD group had significantly worse 10-year survival compared with the CM group (66.7% vs. 83%; P o .007). However, the SV without palliation subgroup and the CM group had similar 10-year survival (76% vs. 83%; P o .696). Still, the SV recipients with prior palliation, particularly Fontan patients, had poorer 10-year survivals of 57.1% and 55.6%, respectively, compared with the SV subgroup without palliation (76%; P o .003). More recently, Kanter and colleagues21 from Emory University reported much better than usual early results of HT among 27 carefully selected Fontan patients. The 30day, 1-year, and 5-year actuarial survival was 96.3%, 81.5%, and 65.5%, respectively. Of 500 Fontan patients in their institution, 36 (7%) were listed for HT. One patient was later removed from the list, and eight others died while waiting. Further selection excluded from consideration for HT all patients with significant hepatic dysfunction. At the time of transplant, two thirds (67%) of the patients were United Network for Organ Sharing (UNOS) Status 1 and 22% were being mechanically ventilated. The only operative death was a child with heterotaxy who was in renal failure and supported with mechanical ventilation before HT. Despite the excellent operative survival for these Fontan recipients, a cluster of deaths was observed during the first 6 months. Other risk factors that appear to contribute to higher postHT mortality among CHD recipients include mechanical circulatory support, mechanical ventilation, visceral heterotaxy, and renal failure (CrCl o 40 mL/min/1.73m2).14,17,18,22 In the Loma Linda University HT series, within the subgroup of CHD patients who had prior surgery, half of the recipients with perioperative renal failure died early. In a study by Auerbach and colleagues23 of 189 pediatric recipients of HT, 37% of whom had CHD, multivariate analysis identified CHD (hazard ratio, 1.8) and mechanical ventilation (hazard ratio, 1.9) as significant predictors of graft loss. Moreover, the combination of mechanical ventilation and renal dysfunction was a strong predictor of poor early outcome after HT for CHD. When recipients had at least two of three significant risk
75 factors (CHD, mechanical ventilation, or renal failure), 1-year graft survival was significantly worse than for those without risk factors (50% vs. 90%). Infants with HLHS who fail palliation constitute another high-risk subgroup for HT. A recent publication from the Pediatric Heart Transplant Study24 identified HLHS infants with prior staged palliation as having the lowest 1-year survival after HT (70%) when compared with infants with CM (89%), those with CHD with and without surgery (79% and 82%), and those with HLHS without palliative surgery (79%). Moreover, there was no improvement in the current era in survival after HT for HLHS with prior palliative surgery. Jacobs and associates,25 who documented similar 5-year survival for pediatric recipients with CM and CHD, also observed that survival after HT for HLHS with prior palliative surgery was somewhat worse than survival after primary transplantation for HLHS. Allosensitization, most commonly seen among CHD recipients with prior cardiac operations, remains an immunologic challenge that contributes to increased morbidity and mortality after HT. Prior reports from our group26 and others27,28 identified recipient pre-sensitization as an important risk factor for early death after HT. In the report from the Pediatric Heart Transplant Study Group,28 an elevated PRA at listing was associated with a higher risk of death while waiting, and a higher risk of operative mortality. A prior Norwood operation was strongly associated with elevation of PRA. In addition, marked allosensitization (PRA Z50%) was associated with a greater than two-fold increase risk of death within the first post-HT year. Prospective crossmatching in pediatric HT, while a topic of discussion, is difficult to accomplish, and should be avoided. Its potential benefit toward reducing post-transplant mortality is largely neutralized by the increased attrition during longer waitlist times. The impact of recent virtual cross-matching strategy on waiting times and the benefits of directed desensitization protocols on future outcomes of HT has yet to be determined. Finally, pulmonary hypertension has long been recognized as a cause of early graft failure and recipient mortality during the HT process.29 In the Loma Linda series, pulmonary hypertension was the most common cause of early death after HT in the CHD group that had prior surgery. Some patients with CHD are more prone to pulmonary hypertension than others. Long-standing Shone’s complex, transposition of the great arteries, and truncus arteriosus each carry higher risks of pulmonary hypertension. Determination of PVR and pulmonary arterial capacitance in the CHD population is often complicated, which leads to underestimation, particularly in patients with SV. Moreover, among patients with latefailing Fontan circulations, pulmonary vascular disease may only become apparent after HT.30 Perioperative pulmonary hypertension unresponsive to pulmonary vasodilatation with inhaled nitric oxide or prostacyclin will demand the use of mechanical circulatory support, either in the form of extracorporeal membrane oxygenation or a ventricular assist device.
76
Conclusion HT provides the ultimate palliation for CHD patients who develop severe heart failure or suffer from failing physiology after reconstructive surgery. The demand for this therapeutic option appears to be on the increase, as the population of CHD survivors expands. Recipients of HT for CHD with no prior surgical palliation have a survival advantage over those with previous palliative procedures, particularly over those with a failing Fontan circulation. The overall long-term survival after HT for CHD is good, and it parallels that for CM. However, the early mortality of HT for CHD remains higher than that for CM and is largely related to recipient’s pre-transplant status and cardiac surgical history. Early outcome improvements will require earlier referral and careful selection for HT, based on the assessment of all risk factors. Whereas “elective” HT, as a planned therapy, is more prudently offered to lower-risk candidates, CHD patients with multiple risk factors (high-risk category) may require an alternative treatment option, such as mechanical circulatory support.
Acknowledgments The authors wish to acknowledge and thank James Fitts, MA and Joyce Johnston-Rush, RN for their assistance in data collection and statistical analysis.
References 1. Simmonds J, Burch M, Dawkins H, et al: Heart transplantation after congenital heart surgery: improving results and future goals. Eur J Cardiothorac Surg 2008;34:313-317 2. Kirk R, Dipchand AI, Edwards LB, et al: The registry of the international society for heart and lung transplantation: fifteenth pediatric heart transplantation report – 2012. J Heart Lung Transplant 2012;31: 1065-1086 3. Stehlik J, Edwards LB, Kucheryavaya AY, et al: The registry of the society for international heart and lung transplantation: 29th official adult heart transplant report – 2012. J Heart Lung Transplant 2012;31:1052-1064 4. Kirklin JK, Pearce FB, Dabal RJ, et al: Cardiac transplantation and mechanical support for functional single ventricle. World J Pediatr Congenit Heart Surg 2012;3:183-193 5. Pincott ES, Burch M: Indications for heart transplantation in congenital heart disease. Curr Cardiol Rev 2011;7:51-58 6. Almond CSD, Thiagarajan RR, Piercey GE, et al: Waiting list mortality among children listed for heart transplantation in the United States. Circulation 2009;119:717-727 7. Baek JS, Bae EJ, Ko JS, et al: Late hepatic complications after Fontan operation; non-invasive markers of hepatic fibrosis and risk factors. Heart 2010;96:1750-1755 8. Hill AL, Maeda K, Bonham CA, et al: Pediatric combined heart-liver transplantation performed en bloc: A single-center experience. Pediatr Transplantation 2012;16:393-397 9. Vricella LA, Razzouk AJ, Gundry SR, et al: Heart transplantation in infants and children with situs inversus. J Thorac Cardiovasc Surg 1998;116: 82-86 10. del Nido PJ, Bailey LL, Kirklin JK: Surgical techniques in pediatric heart transplantation. ISHLT Monograph Series 2007;2:83-102
A.J. Razzouk and L.L. Bailey 11. Griffiths ER, Kaza AK, Wyler von Ballmoos MC, et al: Evaluating failing Fontans for heart transplantation: Predictors of death. Ann Thorac Surg 2009;88:558-564 12. Voeller RK, Epstein DJ, Guthrie TJ, et al: Trends in the indications and survival in pediatric heart transplants: A 24-year single-center experience in 307 patients. Ann Thorac Surg 2012;94:807-816 13. Michielon G, Parisi F, Squitieri C, et al: Orthotopic heart transplantation for congenital heart disease: An alternative for high-risk Fontan candidates? Circulation 2003;108(suppl II):II-140-II-149 14. Davies RR, Sorabella RA, Yan J, et al: Outcomes after transplantation for “failed” Fontan: A single-institution experience. J Thorac Cardiovasc Surg 2012;143:1183-1192 15. Lamour JM, Kanter KR, Naftel DC, et al: The effect of age, diagnosis, and previous surgery in children and adults undergoing heart transplantation for congenital heart disease. J Am Coll Cardiol 2009;54:160-165 16. Bernstein D, Naftel D, Chin C, et al: Outcome of listing for cardiac transplantation for failed Fontan: A multi-institutional study. Circulation 2006;114:273-280 17. Davies RR, Russo MJ, Mital S, et al: Predicting survival among high-risk pediatric cardiac transplant recipients: An analysis of the united network for organ sharing database. J Thorac Cardiovasc Surg 2008;135:147-155 18. Kovach JR, Naftel DC, Pearce FB, et al: Comparison of risk factors and outcomes for pediatric patients listed for heart transplantation after bidirectional Glenn and after Fontan: An analysis from the Pediatric Heart Transplant Study. J Heart Lung Transplant 2012;31:131-139 19. Chen JM, Davies RR, Mital SR, et al: Trends and outcomes in transplantation for complex congenital heart disease. Ann Thorac Surg 2004;78:1984-2004 20. Jayakumar KA, Addonizio LJ, Kichuk-Chrisant MR, et al: Cardiac transplantation after the Fontan or Glenn procedure. J Am Coll Cardiol 2004;44:2065-2072 21. Kanter KR, Mahle WT, Vincent RN, et al: Heart transplantation in children with a Fontan procedure. Ann Thorac Surg 2011;91:823-830 22. Ghandi R, Almond C, Sing TP, et al: Factors associated with in-hospital mortality in infants undergoing heart transplantation in the United States. J Thorac Cardiovasc Surg 2011;141:531-536 23. Auerbach SR, Richmond ME, Chen JM, et al: Multiple risk factors before pediatric cardiac transplantation are associated with increased graft loss. Pediatr Cardiol 2012;33:49-54 24. Everitt MD, Boyle GJ, Schechtman KB, et al: Early survival after heart transplant in young infants is lowest after failed single-ventricle palliation: A multi-institutional study. J Heart Lung Transplant 2012;31:509-516 25. Jacobs JP, Asante-Korang A, O’Brien SM, et al: Lessons learned from 119 consecutive cardiac transplants for pediatric and congenital heart disease. Ann Thorac Surg 2011;91:1248-1255 26. Dionigi B, Razzouk AJ, Hasaniya NW, et al: Late outcomes of pediatric heart transplantation are independent of pre-transplant diagnosis and prior cardiac surgical intervention. J Heart Lung Transplant 2008;27: 1090-1095 27. Jacobs JP, Quintessenza JA, Boucek RJ, et al: Pediatric cardiac transplantation in children with high panel reactive antibody. Ann Thorac Surg 2004;78:1703-1709 28. Mahle TW, Tresler MA, Edens E, et al: Allosensitization and outcomes in pediatric heart transplantation. J Heart Lung Transplant 2011;30: 1221-1227 29. Fukushima N, Gundry SR, Razzouk AJ, et al: Risk factors for graft failure associated with pulmonary hypertension after pediatric heart transplantation. J Thorac Cardiovasc Surg 1994;107: 985-989 30. Mitchell MB, Campbell DV, Ivy D, et al: Evidence of pulmonary vascular disease after heart transplantation for Fontan circulation failure. J Thorac Cardiovasc Surg 2004;128:693-702
Introduction
I
mproved early outcomes of reconstructive surgery has resulted in an expanding population of children and adults who are presently living with their complex congenital heart disease (CHD), many of them living quite well. Primary heart transplantation (HT) has been irrelevant to their survival beyond early infancy. So, this once nearly doomed cohort of neonates, most of whom were born with lethal CHD, is now growing up with expectations of reaching the joys and vicissitudes of adult life, and beyond. In reality, however, data suggest that not all of these youngsters are destined to achieve these ideal expectations. Some will not reach childhood, much less adulthood, without additional surgical help. Instead, 10% to 20% will reach the end-stage of their CHD related to myocardial dysfunction or circulatory failure, and they will be referred for HT.1 They have become, and are, a heterogeneous group of potential HT recipients who present with unique decision-making, anatomic, and physiologic challenges. Historically, a congenital diagnosis has accounted for 59% of infants o1 year of age, 37% of children age 1 to 10 years, 23% Department of Cardiovascular and Thoracic Surgery, Loma Linda University Children’s Hospital, Loma Linda, CA. Address correspondence to Anees Razzouk, MD, 11175 Campus St., Suite 21121, Loma Linda, CA 92354. E-mail: [email protected]
http://dx.doi.org/10.1053/j.pcsu.2014.01.009 1092-9126/& 2014 Elsevier Inc. All rights reserved.
of adolescent patients age 11 to 17 years, and 3% of adults.2,3 HT for CHD has always been associated with worse surgical outcomes than HT for cardiomyopathy because of technical complexities and physiologic challenges of the operation. In addition, diagnosis of CHD remains a significant risk factor for 1-year and 5-year mortality after HT in both pediatric and adult recipients.2,3 Rescue HT, especially in the setting of failing Fontan circulation, has had the worst outcomes. Early referral (before advanced end-organ damage), proper selection and optimization of recipients, and meticulous intra- and postoperative management are crucial to improving operative outcomes of HT in this population. This review will address the unique challenges of HT in patients with CHD, particularly those who have had previous repair or palliation.
Evolving Indications for HT In recent years, the indications for HT in infants and children with CHD have changed significantly with fewer cardiac defects (e.g., complex single ventricle [SV] anomalies) requiring HT as primary therapy. HT is indicated for two-ventricle patients (e.g., tetralogy of Fallot, truncus arteriosus, Shone’s complex, d- and l-TGAs, Ebstein’s anomaly) with end-stage heart failure owing to severe myocardial dysfunction, irreparable valve disease, or intractable arrhythmias. However, failed 69
A.J. Razzouk and L.L. Bailey
70 surgical palliation is becoming the predominant indication for HT in children with CHD. Isomerisms, issues of situs, and visceral heterotaxia may add to surgical complexity but do not contraindicate HT. Single-ventricle patients, of variable anatomic substrates and at different stages of palliation, may require HT for progressive systolic or diastolic ventricular dysfunction, irreparable atrioventricular valve regurgitation, worsening hypoxemia, elevated pulmonary vascular resistance (PVR) or arrhythmias. Patients with systemic complications of Fontan physiology represent a unique and expanding group being referred for HT. Specific indications include protein-losing enteropathy, plastic bronchitis, stroke, thromboembolism, refractory ascites, and early cirrhosis of the liver. Potential candidates with flexible PVR r5 Wood units (wu) and/or transpulmonary gradient r10 mmHg usually have an acceptable risk for HT. Patients with PVR 4 9 wu and transpulmonary gradient 4 16 mmHg are not suitable for HT. Patients with elevated PVR (5 to 9 wu) require preoperative testing of pulmonary vasoreactivity and are at risk for early post HT graft failure. These guidelines may not necessarily apply to patients with Fontan circulation where a non-pulsatile flow state, low cardiac output, and presence of collaterals or a fenestration make the assessment of PVR impractical. In this unique circulation, PVR is often underestimated. Absolute anatomic contraindications to HT in CHD are rare, but may include severe diffuse hypoplasia of pulmonary arteries or irreparable pulmonary venous malformations. Combined heart–liver or heart–kidney transplants may be indicated in patients with irreversible single organ (liver or kidney) failure associated with chronic congestive heart failure.
Timing of Referral for HT The decision for HT in patients with CHD may be simple or remarkably complex. The population is heterogeneous with variability in age, genetic substrate, support systems, structural anomalies, dysrhythmias, non-cardiac risk factors, number and complexity of previous operations, symptoms, and endorgan function. The classic timing of listing for HT for any indication has been when the expected survival at 2 years is 50%. Such timing is difficult to predict in CHD patients who may “survive” for years, albeit with limited functional reserve and marginal cardiac output. While exercise protocols, metabolic testing, and chemical markers may provide significant prognostic information, they are not necessarily reliable in predicting the need for HT. The risks and benefits of additional conventional palliation must be considered and compared with those of HT. For high-risk candidates, mechanical circulatory support as a bridge or destination therapy is evolving as a viable option.4 The period on a HT waiting list can be long, and up to 20% of patients may die waiting, depending on their status and comorbidities.5,6 A diminished quality of life with intermittent symptoms and frequent hospitalizations should prompt referral for HT. Early listing (before end-organ dysfunction and inotropic or ventilatory
support) will reduce waiting list mortality and improve postHT outcomes.
Pre-HT Evaluation The complex anatomy of CHD, especially in the setting of multiple previous operations, requires detailed pre-HT imaging using echocardiography, cardiac magnetic resonance imaging, or computed tomographic angiography. Information regarding the proximity of the heart, great vessels, or conduits to the sternum, the anatomy of the systemic and pulmonary venous connections, the anatomy of the pulmonary arteries, the presence of aortic arch obstruction/coarctation, and the presence of important aorto-pulmonary collaterals is essential. Imaging may extend to the neck and groins to identify alternative sites for cannulation, and to the liver to detect cirrhosis or other liver lesions. Cardiac catheterization and full hemodynamic evaluation is performed in all patients. Coil embolization of significant AP collaterals may be performed pre or immediately post HT to reduce the risk of high-output graft failure. Multiple operations, with exposure to blood products and allograft patches or conduits, contribute to increased anti-HLA antibodies in patients with end-stage CHD. High levels of sensitization may increase waiting time and contribute to early graft failure and post-HT mortality. Panel reactive antibody (PRA) screening identifies sensitized patients and guides in pre- and post-HT immunomodulation therapy. Candidates with high immunologic risk (PRA 4 20%) receive pre HT intravenous immunoglobulin and plasmapheresis, both of which are continued postoperatively, when they also receive thymoglobulin and other immunosuppressive agents. A prospective cross-match is rarely necessary, and has been largely replaced by virtual cross-matching. Pre-HT coagulation abnormalities are often present as a result of liver dysfunction or chronic anticoagulation for arrhythmias, Fontan circulation, mechanical valve prostheses, or thromboembolism. Correction of coagulopathy, when possible, can reduce the need for post-HT blood transfusion and will decrease respiratory and metabolic morbidity. Other preoperative challenges that should be recognized and optimized before HT include compromised nutritional state (especially in the setting of protein-losing enteropathy or ascites), chronic anemia, susceptibility to infection with asplenia or DiGeorge syndrome, chronic hypoxemic state limiting neurologic reserve, and chronic low cardiac output state causing end-organ dysfunction. Acute or acute-on-chronic renal dysfunction is not a contraindication to HT, although its post-HT reversibility is difficult to predict. Infants and young children may require placement of a peritoneal dialysis catheter at the time of, if not before, HT. Older children and adults may require hemodialysis. Minor liver abnormalities (cholestasis, biochemical or coagulation disorders) are common in patients with chronic heart failure and are usually correctable during and after HT. However, SV patients with a Fontan circulation are at risk of developing major
Heart transplantation in children for end-stage CHD hepatic complications such as severe cirrhosis or hepatocellular carcinoma. These changes correlate with the duration of Fontan circulation and require special screening with liver biopsy and computed tomographic imaging. Bridging fibrosis on liver biopsy and hepatocellular carcinoma (after treatment with transarterial chemoembolization) may be acceptable indications for combined heart–liver transplant.7,8
Donor Considerations Major oversizing of the donor graft should be avoided in this population with a scarred, fixed mediastinum. A donor– recipient weight ratio of up to 2:1 is optimal, and the slightly larger graft may be advantageous for the recipient with elevated but responsive pulmonary artery pressure. Knowledge of the recipient’s cardiac anatomy and prior interventions is crucial in carrying out the donor graft recovery. Extra donor tissue is usually obtained en bloc with the donor heart and may include innominate vein, superior vena cava (SVC), full-length pulmonary artery branches (which may preclude lung donation), ascending aorta and the entire arch, and the pulmonary veins (Fig. 1). Except for closure of a patent foramen ovale, preparation of the donor heart is delayed until the time of implantation. Adequate graft preservation and efficient coordination of the donor and recipient operations are critical to the success of HT in this setting of potentially prolonged operative and graft ischemia times.
Operative Considerations Cardiac transplantation in patients with CHD may be technically challenging – especially after multiple prior operations,
71 and at 0200 hours during a weekend. Vascular access (for line placement) can be difficult and may require open cut-downs. Defibrillation pads are positioned on the chest and topical cooling measures are initiated. A repeat sternotomy is carefully accomplished to avoid massive hemorrhage or air embolism. Central cannulation of the aorta is usually possible; but, alternative sites such as both subclavian regions and both groin areas should be prepared as part of the sterile field. A single atrial cannula for venous drainage simplifies the cardiopulmonary bypass (CPB) circuit and allows for systemic cooling to 201C, an approach that offers protection to both recipient and graft. A crystalloid prime is especially desirable for hemodilution of lymphocytotoxic antibodies in pre-sensitized patients. Hypothermic low flow CPB (20 to 30 cc/kg/min) using active suckers for venous drainage facilitates explantation of the native heart and removal of extracardiac conduits and endovascular stents. A portion of the left pericardium is usually excised (to enlarge the mediastinal space and open it to the pleural cavity). Injury to the phrenic nerve is avoided by minimizing the dissection of structures adjacent to it. A variety of techniques have been described that permit orthotopic HT for even the most complex cardiac malformations.9,10 Using native atrial flaps and donor innominate vein, cavocaval connections can be performed even in the presence of abnormal situs and bilateral SVCs (Fig. 2). To avoid compression or stretching of the reconstructed innominate vein, the ascending aorta is kept long for retro-aortic, or is shortened for ante-aortic innominate vein placement (Fig. 3). Anatomic lesions in the native pulmonary arteries are corrected with donor pulmonary artery tissue. Lateralization of the main pulmonary artery anastomosis (Fig. 2) may be required in patients with dextrocardia,
Figure 1 Donor heart preparation for a recipient with situs inversus. The left pulmonary venous orifices are oversewn. The left atrium is opened vertically between the superior and inferior right pulmonary veins to be anastomosed to the recipient’s pulmonary atrial cuff. SVC, superior vena cava; IV, innominate vein; IVC, inferior vena cava. (Reprinted from the Journal of Thoracic and Cardiovascular Surgery, Vol 116; Vricella LA, Razzouk AJ, Gundry SR, et al, Heart transplantation in infants and children with situs inversus; pp 82-86, 1998, with permission from Elsevier.9)
A.J. Razzouk and L.L. Bailey
72
Figure 2 Orthotopic HT in a recipient with situs inversus. A tubular flap of redundant (remnant) systemic atrial tissue is constructed to redirect the left-sided inferior vena cava (IVC) rightward. Leftward lateralization of the main pulmonary artery is achieved by oversewing its orifice on the right and extending the arteriotomy to the left of midline. LSVC, left superior vena cava; LIVC, left inferior vena cava; MPA, main pulmonary artery; LPA, left pulmonary artery. (Reprinted from the Journal of Thoracic and Cardiovascular Surgery, Vol 116; Vricella LA, Razzouk AJ, Gundry SR, et al, Heart transplantation in infants and children with situs inversus; pp 82-86, 1998, with permission from Elsevier.9)
heterotaxy, or situs inversus. Recipient arch lesions (e.g., coarctation, arch obstruction or aneurysms) are reconstructed with donor aortic tissue. In patients with visceral heterotaxy or situs inversus, where the pulmonary atrium is midline or shifted to the right, the donor heart left pulmonary veins are oversewn, the left atrium is opened between the right pulmonary veins, and then anastomosed to the recipient’s right-sided left atrium. In rare cases when combined heart–liver transplantation is necessary, liver transplantation may be performed en bloc with the donor heart, or the procedures may be performed separately.8
Perioperative Considerations Multiple factors contribute to increased risk of bleeding after HT for CHD. These factors include: chronic anticoagulation, liver dysfunction, dense adhesions, multiple thoracotomies, collateral vessels, splanchnic venous congestion, prolonged operative and CPB times, cyanosis, and thrombocytopenia. In as much as possible, meticulous hemostasis is practiced before graft implantation. Transfusion of blood products is often necessary, but massive infusions of blood products will contribute to pulmonary hypertension and graft right ventricular dysfunction. Acute right ventricular failure rarely
Figure 3 Orthotopic HT in a recipient with situs inversus. Conduit formed by donor’s SVC and innominate vein (IV) lies anterior to the aorta after anastomosis to the left SVC. In recipients with bilateral SVCs or bilateral Glenn shunts, similar connections are performed with the right SVC anastomosed to this conduit at the ligature site of donor’s proximal SVC. SVC, superior vena cava. (Reprinted from the Journal of Thoracic and Cardiovascular Surgery, Vol 116; Vricella LA, Razzouk AJ, Gundry SR, et al, Heart transplantation in infants and children with situs inversus; pp 82-86, 1998, with permission from Elsevier.9)
Heart transplantation in children for end-stage CHD requires mechanical support, however, and can usually be managed with inhaled nitric oxide, pulmonary vasodilation, inotropic support, and, occasionally, an open chest. Aggressive dialysis is instituted in patients with renal failure and plasmapheresis is resumed postoperatively in highly sensitized recipients. For the rare patient with a phrenic nerve injury, early plication of the paralyzed hemidiaphragm facilitates weaning from ventilator support. Nutritional support and infection control, which are balanced with adequate immunosuppression, are critical in minimizing post-HT morbidity and mortality.
Results of HT in children with CHD: A single-center experience Four hundred ninety-six pediatric patients (age range, 1 day to 17.7 years) underwent orthotropic HT at Loma Linda University Children’s Hospital for the diagnosis of CHD (N ¼ 357), cardiomyopathy (CM; N ¼ 130), and other (N ¼ 9). Between 1985 and 1999 (early era), HT was the preferred primary therapy for infants with hypoplastic left heart syndrome (HLHS). There were 276 transplants for CHD (80.2%) and 68 transplants for CM (19.8%). The recent era (2000 to 2013) witnessed a decline in the overall number of recipients with CHD (N ¼ 81, 56.6%). There were 62 CM recipients (43.4%). Prior palliation or repair had become much more common among CHD recipients in the recent era. Indeed, the prevalence of prior surgical interventions within the CHD population increased from 27.5% (76 out of 276) in the early era to 58% (47 out of 81) in the recent era. In total, 123 patients with CHD, the prior surgery group (PSG), underwent 255 corrective or palliative operations prior to HT (Tables 1 and 2). The number of such prior surgical procedures was one in 52 patients, two in 41 patients, and three or more in 30 patients. The PSG had a significantly longer Table 1 Congenital Heart Disease Diagnosis Pre-HT No Prior Surgery Group (N ¼ 234) HLHS 165 Non-HLHS 69 Prior Surgery Group (N ¼ 123) Biventricular (N ¼ 49) Septal defects: 14 Severe Shone’s complex: 13 d- and l-TGAs: 9 Tetralogy: 5 Other: 8 Univentricular (N ¼ 74) DORV: 22 HLHS: 16 PA/TA: 11 Unbalanced AVSD: 11 Other 14 Abbreviations: HLHS, hypoplastic left heart syndrome; TGA, transposition of great arteries; DORV, double outlet right ventricle; PA, pulmonary atresia; TA, tricuspid atresia; AVSD, atrioventricular septal defect.
73 Table 2 Prior Surgery Group (N ¼ 123): Last Major Operation Pre-HT Type of Surgery
No. of Patients
Valve repair or replacement Systemic to pulmonary artery shunt Pulmonary artery banding Fontan Glenn Norwood Septal defect repair Other
27 25 17 13 11 9 8 13
average graft ischemia time (min) compared with the no prior surgery group (NoPSG): 304 þ 117 vs. 270 þ 122, P o .013. Similarly, CPB times were significantly longer for the PSG (P o .0001). Overall 30-day operative mortality for the entire CHD cohort was 11.6% and was not significantly different between the two eras. Early mortality for the NoPSG was 10.3% (24 out of 234) and for the PSG was 16.3% (18 out of 123); P o .22. Within the PSG, seven of 23 (30.4%) recipients with PRA Z 20% died early after HT (P o .021) and 10 of the 18 (55.6%) patients who had perioperative renal failure, and required dialysis, died postoperatively (P o .001). Recipients with three or more procedures before HT had a significantly higher operative mortality than those who had two or less such procedures (26.7% vs. 10.8%; P o .032). Although the operative mortality for the Fontan subgroup was high (30.8%, 4 out of 13), it was not significantly different when compared with other subgroups, perhaps because of the small number of patients. Within the PSG, recipients with SV had similar early mortality as those with two ventricles (13.5% vs. 16.3%; P o .66). Beyond the early post-HT period, the difference in longterm outcomes of CHD patients compared with CM patients is not significant. Survival at 15 years is 57.4% versus 66.6%, respectively (P o .304) (Fig. 4). However, the NoPSG of CHD recipients had a slight survival advantage, at 15 years, over the PSG of 61% versus 48.9%, respectively (P o .038) (Fig. 5). Primary causes of early death in the PSG (18 out of 123) were unmanageable pulmonary hypertension and technical problems. The major causes of early death in the NoPSG (24 out of 234) were management issues and acute graft failure.
Discussion Advances in medical, surgical, and transcatheter interventions have significantly improved the early survival of infants and children with CHD. Nevertheless, HT remains the only viable treatment option for some infants, children, and adults with end-stage CHD. In this category are those with previous reparative surgery and deteriorating ventricular function, as well as those with prior palliative reconstruction and failing circulatory physiology. This heterogeneous and often complicated population of potential HT candidates is rapidly growing. It will undoubtedly stretch the resources of heart transplant programs throughout the world. Results of pediatric HT have gradually improved in recent years, and the long-term survival
A.J. Razzouk and L.L. Bailey
74
Figure 4 Kaplan-Meier analysis of patient survival after HT by diagnostic category. Long-term survival is not significantly different in the two diagnostic groups. CHD, congenital heart disease; CM, cardiomyopathy
of HT for CHD, with some exceptions, now appears to parallel that of recipients transplanted for other indications. The diagnosis of CHD, nevertheless, remains an important risk for operative mortality after HT. The risk varies according to age, anatomic substrate, prior surgical history, stage of palliation, and physiologic state at the time of transplantation. While this population of potential HT recipients continues to expand, the decision for HT has become increasingly difficult, as the demand for donated organs far surpasses the limited supply. Comprehensive pre-HT evaluation, including full assessment of risk factors, is an important component of the decision-making process. Indeed, patient selection may be the key factor in reducing early mortality after HT for CHD. The decision for HT may be particularly vexing among patients with failing Fontan physiology. Clearly, rescue HT after acute failure of a Fontan procedure is unwarranted and has a poor prognosis, especially when systemic ventricular function is preserved. This situation is far better avoided than surgically or medically managed. Hence, among patients with SV who are high-risk candidates for Fontan completion, HT should be considered as an appropriate alternative to Fontan completion.11–13
As the population of Fontan survivors grows, this diagnostic subgroup is becoming the predominant category of CHD patients referred for HT. Technical challenges related to complex anatomy and multiple previous operations, along with various degrees of physiologic decompensation (such as chronically elevated central venous pressure, low cardiac output, renal and hepatic dysfunction, elevated PVR, and malnutrition) place this subgroup of CHD patients at considerable additional risk for HT. Numerous studies12–18 have identified recipient Fontan circulation as an independent risk factor for early phase mortality after HT. Both Kovach et al18 and, in a separate report, Davies et al17 found that Fontan completion within 6 months of listing for HT to be a significant risk factor. In a large multicenter study15 of 488 patients (median age, 12.4 years) transplanted for CHD, survival at 3 post-HT months was significantly worse in CHD recipients versus children with CM (86% vs. 94%). A previous Fontan procedure was identified as the highest independent predictor of mortality, with a relative risk of death of 8.6. In 2004, the group at Columbia University19 reported a 1-month mortality of 18.8% (20 out of 106) after HT for CHD.
Figure 5 Kaplan-Meier analysis of patient survival after HT for CHD. Survival of the No Prior Surgery group is statistically better than survival of the Prior Surgery group.
Heart transplantation in children for end-stage CHD The need for pulmonary artery reconstruction was identified as a strong predictor of early mortality. In another study20 from the same institution, which focused on HT after the Fontan operation (n ¼ 24), and the Glenn procedure (n ¼ 11), there were 10 deaths less than 2 months after transplantation. Nine of the deaths occurred in the Fontan group. The 1-year and 5-year survival for the Fontan group was 62.5% and 57% compared with 71.5% and 67.5% for the entire cohort. In their series of 73 pediatric patients who underwent HT for CHD, Simmonds and colleagues1 reported an operative mortality of 17.8% with similar 1-year survival for univentricular (n ¼ 38) and biventricular (n ¼ 35) circulations (75% vs. 78%). As in most other institutions, their HT operative mortality for Fontan recipients was 26.6% (4 out of 15). They noted improved outcomes of HT for CHD since the year 2000, with 1-year survival for CHD patients of 90% compared with 94% for CM recipients (P ¼ .756). A recent study from St. Louis Children’s Hospital12 reported 307 pediatric heart transplants performed for CHD (57%), CM (39%), and retransplant (4%). The CHD group had significantly worse 10-year survival compared with the CM group (66.7% vs. 83%; P o .007). However, the SV without palliation subgroup and the CM group had similar 10-year survival (76% vs. 83%; P o .696). Still, the SV recipients with prior palliation, particularly Fontan patients, had poorer 10-year survivals of 57.1% and 55.6%, respectively, compared with the SV subgroup without palliation (76%; P o .003). More recently, Kanter and colleagues21 from Emory University reported much better than usual early results of HT among 27 carefully selected Fontan patients. The 30day, 1-year, and 5-year actuarial survival was 96.3%, 81.5%, and 65.5%, respectively. Of 500 Fontan patients in their institution, 36 (7%) were listed for HT. One patient was later removed from the list, and eight others died while waiting. Further selection excluded from consideration for HT all patients with significant hepatic dysfunction. At the time of transplant, two thirds (67%) of the patients were United Network for Organ Sharing (UNOS) Status 1 and 22% were being mechanically ventilated. The only operative death was a child with heterotaxy who was in renal failure and supported with mechanical ventilation before HT. Despite the excellent operative survival for these Fontan recipients, a cluster of deaths was observed during the first 6 months. Other risk factors that appear to contribute to higher postHT mortality among CHD recipients include mechanical circulatory support, mechanical ventilation, visceral heterotaxy, and renal failure (CrCl o 40 mL/min/1.73m2).14,17,18,22 In the Loma Linda University HT series, within the subgroup of CHD patients who had prior surgery, half of the recipients with perioperative renal failure died early. In a study by Auerbach and colleagues23 of 189 pediatric recipients of HT, 37% of whom had CHD, multivariate analysis identified CHD (hazard ratio, 1.8) and mechanical ventilation (hazard ratio, 1.9) as significant predictors of graft loss. Moreover, the combination of mechanical ventilation and renal dysfunction was a strong predictor of poor early outcome after HT for CHD. When recipients had at least two of three significant risk
75 factors (CHD, mechanical ventilation, or renal failure), 1-year graft survival was significantly worse than for those without risk factors (50% vs. 90%). Infants with HLHS who fail palliation constitute another high-risk subgroup for HT. A recent publication from the Pediatric Heart Transplant Study24 identified HLHS infants with prior staged palliation as having the lowest 1-year survival after HT (70%) when compared with infants with CM (89%), those with CHD with and without surgery (79% and 82%), and those with HLHS without palliative surgery (79%). Moreover, there was no improvement in the current era in survival after HT for HLHS with prior palliative surgery. Jacobs and associates,25 who documented similar 5-year survival for pediatric recipients with CM and CHD, also observed that survival after HT for HLHS with prior palliative surgery was somewhat worse than survival after primary transplantation for HLHS. Allosensitization, most commonly seen among CHD recipients with prior cardiac operations, remains an immunologic challenge that contributes to increased morbidity and mortality after HT. Prior reports from our group26 and others27,28 identified recipient pre-sensitization as an important risk factor for early death after HT. In the report from the Pediatric Heart Transplant Study Group,28 an elevated PRA at listing was associated with a higher risk of death while waiting, and a higher risk of operative mortality. A prior Norwood operation was strongly associated with elevation of PRA. In addition, marked allosensitization (PRA Z50%) was associated with a greater than two-fold increase risk of death within the first post-HT year. Prospective crossmatching in pediatric HT, while a topic of discussion, is difficult to accomplish, and should be avoided. Its potential benefit toward reducing post-transplant mortality is largely neutralized by the increased attrition during longer waitlist times. The impact of recent virtual cross-matching strategy on waiting times and the benefits of directed desensitization protocols on future outcomes of HT has yet to be determined. Finally, pulmonary hypertension has long been recognized as a cause of early graft failure and recipient mortality during the HT process.29 In the Loma Linda series, pulmonary hypertension was the most common cause of early death after HT in the CHD group that had prior surgery. Some patients with CHD are more prone to pulmonary hypertension than others. Long-standing Shone’s complex, transposition of the great arteries, and truncus arteriosus each carry higher risks of pulmonary hypertension. Determination of PVR and pulmonary arterial capacitance in the CHD population is often complicated, which leads to underestimation, particularly in patients with SV. Moreover, among patients with latefailing Fontan circulations, pulmonary vascular disease may only become apparent after HT.30 Perioperative pulmonary hypertension unresponsive to pulmonary vasodilatation with inhaled nitric oxide or prostacyclin will demand the use of mechanical circulatory support, either in the form of extracorporeal membrane oxygenation or a ventricular assist device.
76
Conclusion HT provides the ultimate palliation for CHD patients who develop severe heart failure or suffer from failing physiology after reconstructive surgery. The demand for this therapeutic option appears to be on the increase, as the population of CHD survivors expands. Recipients of HT for CHD with no prior surgical palliation have a survival advantage over those with previous palliative procedures, particularly over those with a failing Fontan circulation. The overall long-term survival after HT for CHD is good, and it parallels that for CM. However, the early mortality of HT for CHD remains higher than that for CM and is largely related to recipient’s pre-transplant status and cardiac surgical history. Early outcome improvements will require earlier referral and careful selection for HT, based on the assessment of all risk factors. Whereas “elective” HT, as a planned therapy, is more prudently offered to lower-risk candidates, CHD patients with multiple risk factors (high-risk category) may require an alternative treatment option, such as mechanical circulatory support.
Acknowledgments The authors wish to acknowledge and thank James Fitts, MA and Joyce Johnston-Rush, RN for their assistance in data collection and statistical analysis.
References 1. Simmonds J, Burch M, Dawkins H, et al: Heart transplantation after congenital heart surgery: improving results and future goals. Eur J Cardiothorac Surg 2008;34:313-317 2. Kirk R, Dipchand AI, Edwards LB, et al: The registry of the international society for heart and lung transplantation: fifteenth pediatric heart transplantation report – 2012. J Heart Lung Transplant 2012;31: 1065-1086 3. Stehlik J, Edwards LB, Kucheryavaya AY, et al: The registry of the society for international heart and lung transplantation: 29th official adult heart transplant report – 2012. J Heart Lung Transplant 2012;31:1052-1064 4. Kirklin JK, Pearce FB, Dabal RJ, et al: Cardiac transplantation and mechanical support for functional single ventricle. World J Pediatr Congenit Heart Surg 2012;3:183-193 5. Pincott ES, Burch M: Indications for heart transplantation in congenital heart disease. Curr Cardiol Rev 2011;7:51-58 6. Almond CSD, Thiagarajan RR, Piercey GE, et al: Waiting list mortality among children listed for heart transplantation in the United States. Circulation 2009;119:717-727 7. Baek JS, Bae EJ, Ko JS, et al: Late hepatic complications after Fontan operation; non-invasive markers of hepatic fibrosis and risk factors. Heart 2010;96:1750-1755 8. Hill AL, Maeda K, Bonham CA, et al: Pediatric combined heart-liver transplantation performed en bloc: A single-center experience. Pediatr Transplantation 2012;16:393-397 9. Vricella LA, Razzouk AJ, Gundry SR, et al: Heart transplantation in infants and children with situs inversus. J Thorac Cardiovasc Surg 1998;116: 82-86 10. del Nido PJ, Bailey LL, Kirklin JK: Surgical techniques in pediatric heart transplantation. ISHLT Monograph Series 2007;2:83-102
A.J. Razzouk and L.L. Bailey 11. Griffiths ER, Kaza AK, Wyler von Ballmoos MC, et al: Evaluating failing Fontans for heart transplantation: Predictors of death. Ann Thorac Surg 2009;88:558-564 12. Voeller RK, Epstein DJ, Guthrie TJ, et al: Trends in the indications and survival in pediatric heart transplants: A 24-year single-center experience in 307 patients. Ann Thorac Surg 2012;94:807-816 13. Michielon G, Parisi F, Squitieri C, et al: Orthotopic heart transplantation for congenital heart disease: An alternative for high-risk Fontan candidates? Circulation 2003;108(suppl II):II-140-II-149 14. Davies RR, Sorabella RA, Yan J, et al: Outcomes after transplantation for “failed” Fontan: A single-institution experience. J Thorac Cardiovasc Surg 2012;143:1183-1192 15. Lamour JM, Kanter KR, Naftel DC, et al: The effect of age, diagnosis, and previous surgery in children and adults undergoing heart transplantation for congenital heart disease. J Am Coll Cardiol 2009;54:160-165 16. Bernstein D, Naftel D, Chin C, et al: Outcome of listing for cardiac transplantation for failed Fontan: A multi-institutional study. Circulation 2006;114:273-280 17. Davies RR, Russo MJ, Mital S, et al: Predicting survival among high-risk pediatric cardiac transplant recipients: An analysis of the united network for organ sharing database. J Thorac Cardiovasc Surg 2008;135:147-155 18. Kovach JR, Naftel DC, Pearce FB, et al: Comparison of risk factors and outcomes for pediatric patients listed for heart transplantation after bidirectional Glenn and after Fontan: An analysis from the Pediatric Heart Transplant Study. J Heart Lung Transplant 2012;31:131-139 19. Chen JM, Davies RR, Mital SR, et al: Trends and outcomes in transplantation for complex congenital heart disease. Ann Thorac Surg 2004;78:1984-2004 20. Jayakumar KA, Addonizio LJ, Kichuk-Chrisant MR, et al: Cardiac transplantation after the Fontan or Glenn procedure. J Am Coll Cardiol 2004;44:2065-2072 21. Kanter KR, Mahle WT, Vincent RN, et al: Heart transplantation in children with a Fontan procedure. Ann Thorac Surg 2011;91:823-830 22. Ghandi R, Almond C, Sing TP, et al: Factors associated with in-hospital mortality in infants undergoing heart transplantation in the United States. J Thorac Cardiovasc Surg 2011;141:531-536 23. Auerbach SR, Richmond ME, Chen JM, et al: Multiple risk factors before pediatric cardiac transplantation are associated with increased graft loss. Pediatr Cardiol 2012;33:49-54 24. Everitt MD, Boyle GJ, Schechtman KB, et al: Early survival after heart transplant in young infants is lowest after failed single-ventricle palliation: A multi-institutional study. J Heart Lung Transplant 2012;31:509-516 25. Jacobs JP, Asante-Korang A, O’Brien SM, et al: Lessons learned from 119 consecutive cardiac transplants for pediatric and congenital heart disease. Ann Thorac Surg 2011;91:1248-1255 26. Dionigi B, Razzouk AJ, Hasaniya NW, et al: Late outcomes of pediatric heart transplantation are independent of pre-transplant diagnosis and prior cardiac surgical intervention. J Heart Lung Transplant 2008;27: 1090-1095 27. Jacobs JP, Quintessenza JA, Boucek RJ, et al: Pediatric cardiac transplantation in children with high panel reactive antibody. Ann Thorac Surg 2004;78:1703-1709 28. Mahle TW, Tresler MA, Edens E, et al: Allosensitization and outcomes in pediatric heart transplantation. J Heart Lung Transplant 2011;30: 1221-1227 29. Fukushima N, Gundry SR, Razzouk AJ, et al: Risk factors for graft failure associated with pulmonary hypertension after pediatric heart transplantation. J Thorac Cardiovasc Surg 1994;107: 985-989 30. Mitchell MB, Campbell DV, Ivy D, et al: Evidence of pulmonary vascular disease after heart transplantation for Fontan circulation failure. J Thorac Cardiovasc Surg 2004;128:693-702
