Transplantation as a Primary Treatment for Hypoplastic Left Heart Syndrome: Intermediate-Term Results

Transplantation as a Primary Treatment for Hypoplastic Left Heart Syndrome: Intermediate-Term Results Anees J. Razzouk, MD, Richard E. Chinnock, MD, Steven R. Gundry, MD, Joyce K. Johnston, RN, Ranae L. Larsen, MD, Marti F. Baum, MD, Neda F. Mulla, MD, and Leonard L. Bailey, MD Departments of Surgery and Pediatrics, Loma Linda University School of Medicine, and Children's Hospital, Loma Linda, California Background. Hypoplastic left heart syndrome is a lethal malformation. For the last 1 years, orthotopic cardiac transplantation has been our preferred treatment for infants with hypoplastic left heart syndrome. Methods. One hundred seventy-six infants with hypoplastic left heart syndrome were entered into a cardiac transplant protocol between November 1985 and November 1995. Interventional procedures to stent the ductus arteriosus or enlarge the interatrial communication were performed in 8 and 35 patients, respectively. Thirty-four patients (19%) died during the waiting period, and 142 infants underwent cardiac transplantation. Age at cardiac transplantation ranged from 1.5 hours to 6 months (median, 29 days). The majority of grafts were oversized, and the median graft ischemic time was 273 minutes (range, 6 to 576 minutes). The implantation procedure used a period of hypothermic circulatory arrest ranging from 23 to 11 minutes (median, 53 minutes). Repair of other significant defects included interrupted aortic arch (5) and total or partial anomalous pulmonary venous connection (7). Results. There were 13 early and 22 late deaths. Patient actuarial survival at I month and at 1, 5, and 7 years was 91%, 84%, 76%, and 7% respectively. Half of the late deaths were due to rejection. Severe graft vasculopathy was confirmed in 8 patients. Retransplantation was performed in 5 patients for graft vasculopathy (4) and rejection (1). Lymphoblastic leukemia developed in 1 patient 3 years after cardiac transplantation. Conclusions. Cardiac transplantation can be performed in infants with hypoplastic left heart syndrome with good operative and intermediate-term results. Improved survival can be achieved with increased donor availability, better management of rejection, and control of graft vasculopathy. (Ann Thorac Surg 1996;62:1-8) H ypoplastic left heart syndrome (HLHS) is one of the more common congenital heart defects diagnosed during neonatal life. It is the most common cardiac anomaly with one functional ventricle. Prior to the 198s, this complex malformation was uniformly fatal, with 95% of untreated patients dying within the first month of life. The pioneering efforts of Norwood and associates [1] and Bailey and colleagues [2] during the past decade introduced two surgical approaches to the newborn with HLHS, namely staged reconstruction and cardiac transplantation (CTx). For the last 1 years the preferred primary treatment for infants with HLHS at Loma Linda University Medical Center has been orthotopic cardiac transplantation. This approach was initially based on the successful results of neonatal heart transplantation in the goat model. The unpredictable outcome with open repair techniques for HLHS and the encouraging early results of cardiac allotransplantation in infancy from other insti- Presented at the Thirty-second Annual Meeting of The Society of Thoracic Surgeons, Orlando, FL, Jan 29-31, 1996. Address reprint requests to Dr Razzouk, Division of Cardiothoracic Surgery, Loma Linda University Medical Center, 11234 Anderson St, Loma Linda, CA 92354. tutions [3] prompted us to continue with the policy of offering CTx to those infants. Cardiac transplantation offers the potential of a normal cardiovascular physiology as compared with the limitations of a single-ventricle physiology achieved after multistaged reconstruction. The purpose of this report is to review the entire 1-year experience at Loma Linda University Medical Center with CTx for hypoplastic left heart syndrome, analyze the intermediate-term results, and discuss the major issues of this therapy. Material and Methods Hospital records and the heart transplant computer data base at Loma Linda University Medical Center were reviewed to identify all infants with HLHS who were evaluated between November 1985 and November 1995. During that period, a total of 19 infants with the diagnosis of HLHS were initially registered for CTx with the United Network for Organ Sharing. Fourteen patients were subsequently unlisted due to medical contraindications to transplantation (7) or to parents choosing alternative therapy (7). The remaining 176 infants were entered into a pediatric cardiac transplant protocol. Thir- 1996 by The Society of Thoracic Surgeons 3-4975/96/$15. Published by Elsevier Science Inc PII S3-4975(96)295-

2 CHAMBERLAIN PAPER RAZZOUK ET AL Ann Thorac Surg CARDIAC TRANSPLANTATION FOR HLHS 1996;62:1-8 '85 '86 '87 '88 '89 '9 '91 '92 '93 '94 '95 Year of Trans#antatlon Fig 1. Yearly distribution of pediatric cardiac transplantation at Loma Linda University Medical Center (November 1985 to November 1995). (HLHS = hypoplastic left heart syndrome.) ty-four patients (group I) died during the waiting period (median, 28 days; range, 1 to 127 days). Nearly half of those who died (16 of 34 patients) were hospitalized outside our facility. Group II consisted of 142 infants who completed the transplant protocol by undergoing orthotopic CTx (Fig 1). Patient age at transplantation ranged from 1.5 hours to 183 days (mean, 4 _+ 33 days; median, 29 days). More than half of those patients who received transplants (74 of 142 patients) were less than 3 days of age. Thirty-eight patients in this series had fetal diagnosis of HLHS, and twenty-five were registered for transplantation in-utero. Twenty of those infants underwent CTx; 2 fetuses were delivered by cesarian section when a donor heart became available and received transplants at 1.5 hours and 3 hours of life. The waiting period for group II patients, not including the in-utero time, ranged from to 171 days (mean, 27 --- 27 days; median, 19 days). The diagnosis of HLHS was made by echocardiography in all patients; 14 patients had additional cardiac anatomic variations (Table 1). Pathologic examination of explanted hearts revealed aortic atresia in 52, mitral atresia in 21, both aortic and mitral atresia in 35, and other morphologic findings in 34 specimens. Pretransplantation Support and Management Prostaglandin E~ infusion (.25 to.5/~g/kg per minute) was maintained throughout the waiting period to ensure ductal patency. Eight infants with compromised systemic perfusion due to a restrictive ductus arteriosus underwent percutaneous (7) or open (1) placement of a ductal stent. Five of these patients successfully underwent transplantation, 2 of whom were completely weaned off prostaglandin E~ while awaiting transplantation. Progressive hypoxemia due to a severely restrictive interatrial communication necessitated intervention in 35 patients, of whom 27 received transplants. Such interventional procedures consisted of balloon atrial septostomy in 31 patients and blade septostomy or open surgical septectomy in 4 patients. Forty-one patients were mechanically ventilated (inspired Oxygen fraction =.21) at the time of transplantation, but only 4 required inotropic support.,,7, Organ Procurement and Operative Technique Donors and recipients were matched according to ABO blood group compatibility; preoperative serum cytotoxicity assays and tissue-typing were not routinely performed. Donors from all over North America including Alaska provided viable cardiac grafts for Loma Linda's recipients located in southern California. Donor:recipient weight ratio ranged from.6 to 4.2 with a median of 1.8. The longest graft ischemic time was 576 minutes. Donor organ preservation consisted of a single dose of crystalloid cardioplegia and cold storage at 4 C. Three transplant surgeons performed the donor procurement and recipient operations in this series. The operative technique used a period of hypothermic circulatory arrest as was originally described by Bailey and colleagues [4]. The donor aorta was used to reconstruct the recipient's hypoplastic aortic arch well beyond the ductal tissue. Modification of the transplantation technique was necessary when repair of other complex cardiovascular defects was also performed. Immunosuppression Cyclosporine has been the major immunosuppressive agent for all patients. Recipients were started on an intravenous cyclosporine infusion (.1 mg. kg -1. h -1) when a donor was identified. Subsequently, cyclosporine was given orally (1 to 2 mg. kg -1. day -1 in three divided doses) when oral intake was well established. Initial cyclosporine target levels were 25 to 3 ng/ml then were adjusted to 1 to 15 ng/ml at 12 months. Azathioprine administration was started immediately after transplantation at 3 mg. kg -1 day -1 and by I year was reduced to 1 mg" kg -~. day -~. Azathioprine was further adjusted to maintain the white blood cell count at greater than 4, cells//~l. Methylprednisolone was ad- Table 1. Clinical Features of 142 Infants With Hypoplastic Left Heart Syndrome Treated With Orthotopic Cardiac Transplantation a Variable Value Male:female ratio 87:55 Associated diagnosis IAA 5 TAPVC or PAPVC 7 Situs inversus 1 Dextrocardia 1 Weight (kg) 3.7 +.8 (range, 2.-6.4; median, 3.6) Graft ischemic time (min) Circulatory arrest time (min) Hospital stay after CTx (days) 264 + 12 (range, 6-576; median, 27) 53 + 1 (range, 23-11; median, 53) 25 +_ 2 (range, 8-111; median, 17) a Data are reported as mean -+ standard deviation. CTx = cardiac transplantation; IAA = interrupted aortic arch; TAPVC or PAPVC = total or partial anomalous pulmonary venous connection.

Ann Thorac Surg CHAMBERLAIN PAPER RAZZOUK ET AL 3 1996;62:1-8 CARDIAC TRANSPLANTATION FOR HLHS ministered intravenously during the transplant operation (25 mg/kg) and postoperatively every 12 hours for four doses only. Chronic oral steroid therapy was not part of the routine immunosuppressive regimen. The induction protocol for infant heart transplantation has gradually evolved over time. Early in our experience, murine monoclonal antibody (OKT-3 at.1 mg-kg -1. day -1) was used occasionally as an induction agent. However, since 199, a polyclonal antithymocyte serum (AMR, Inc, Nashville, TN) has been administered to all recipients older than 3 days at the time of transplantation. Antithymocyte serum was given intravenously (.5 ml kg -1 day -1) on the day of transplantation, and the dose was repeated for a total of 5 days. Surveillance and Follow-up All surviving recipients were followed up closely at Loma Linda University Medical Center for the first 6 months after transplantation. Afterwards, close communication was maintained with the referring cardiologist and primary pediatrician. Transplant clinic visits initially took place twice a week, and then at 3 months, evaluations were done weekly. Close monitoring included serial echocardiography with Doppler/color flow mapping, electrocardiography, physical examination, and measurement of cyclosporine levels. Cardiac catheterization and coronary angiography were performed 1 year after transplantation and every 2 years thereafter unless otherwise indicated. Cumulative follow-up has been complete as of December 1995 and includes 532 patient-years with a mean follow-up of 3.8 + 2.6 years. The diagnosis of cardiac graft rejection was based on a set of echocardiographic and clinical parameter s that have been previously described in detail [5]. Endomyocardial biopsy was not performed routinely but was reserved for clinically confusing situations when the diagnosis of rejection could not be made noninvasively. The diagnosis of graft coronary artery disease was based on coronary angiography, gross and histologic examination of explanted grafts (after retransplantation) or autopsy specimens, or a combination of angiography and examination. Graft coronary artery disease was considered severe when luminal obliteration was greater than 5%. Data regarding other transplantation related events have been progressively collected and entered into the Transplant Center database. Statistical Methods Survival and freedom from event probability curves were generated using the Kaplan-Meier method. Multivariate logistic regression analysis was used to determine which recipient- or donor-related variables were significant predictors of operative mortality (defined as death within 3 days of operation or death in the hospital before discharge after transplantation). For continuous data, variability was reflected by - standard deviation. Results Survival Pretransplantation mortality for this series was 19% (34/176). The most common cause of death among those who died waiting for a donor heart was cardiac failure in 14 patients. Other causes of death in this group included necrotizing enterocolitis or sepsis in 5 patients, restrictive atrial septal communication in 5 patients, and other miscellaneous causes in 1 patients. In the group that came to CTx, there were 13 early deaths with an operative mortality of 9.2%. Four of those deaths were due to technical issues, 3 to acute graft failure, 2 to acute rejection, 2 to pneumonia, I to pulmonary hypertension, and 1 to a perforated duodenal ulcer. Of the variables analyzed as possible risk factors for operative death, none was significant when multivariate analysis was performed: Age at CTx p =.88 Weight at CTx p =.66 Year of CTx p =.13 Sex p =.47 Graft ischemic time >4 h p =.7 Circulatory arrest time >5 min p =.89 Pretransplantation intervention p -=.84 Pretransplantation ventilator support p =.14 Donor:recipient weight ratio >2 p =.51 Postoperative peritoneal dialysis p =.11 Twenty-two deaths occurred late in this series. Causes of late death included rejection (acute or chronic) in 11 patients, infection in 5 patients, graft coronary artery disease in 2 patients, and 1 death each due to toxic epidermal necrolysis, hemorrhagic shock after circumcision, smoke inhalation, and abdominal aortic thrombosis. Currently 17 patients are alive. Actuarial survival estimates at 1, 5, and 7 years are 84%, 76%, and 7% respectively (Fig 2). Graft Vasculopathy Severe graft coronary artery disease was diagnosed in 8 patients. Coronary angiography confirmed the diagnosis of severe graft vasculopathy (GV) in 5 patients, 4 of whom underwent elective retransplantation with one death; the fifth patient experienced sudden death while being evaluated for retransplantation. Graft coronary artery disease was discovered on the autopsy specimen of 3 other patients whose late death was related to severe rejection. Actuarial freedom from angiographic or specimenproven GV at 5 and 7 years was 98% and 77%, respectively (Fig 3). Rejection Graft rejection was most commonly seen during the first 3 months after transplantation. There were 279 rejection episodes treated during 532 patient-years of follow-up. The overall incidence of treated rejection was.52 per patient-year. Of those who survived at least 12 months, the number of rejections in the first year averaged 1.6 +

4 CHAMBERLAIN PAPER RAZZOUK ET AL Ann Thorac Surg CARDIAC TRANSPLANTATION FOR HLHS 1996;62:1-8 lu 1~- 9-8- 5- A 4-1- (19) - ~,~;... (5o) (16) --Transplanted (N=142) ---All Registered (N=176) 1 2 3 4 5 6 Years Fig 2. Actuarial survival of all infants with hypoplastic left heart syndrome registered for transplantation (dashed line) and receiving transplants (solid line), 1.6 (median = 1; mode = ). Actuarial freedom from graft rejection at 3 months, 1 year, 5 years, and 7 years was 3%, 28%, 19%, and 15% respectively (Fig 4). Two early deaths and 11 late deaths were directly attributed to rejection. Infection Approximately 5% of patients had an infection in the first year after transplantation. Viral and bacterial infections were most frequent. Early after transplantation, major infections were noted to correspond to times of most intense immunosuppression. The infectious organisms responsible for the five late deaths were parainfluenza, Aspergillus, adenovirus, respiratory syncytial virus, and Candida. Lymphoproliferative Disease Acute lymphocytic leukemia developed in only I patient in this series 3 years after transplantation. He was managed by reduction of immunosuppression as well as chemotherapy. He is still alive I year after the diagnosis, but mucormycosis has developed in him. t (142) go- 8o- 7- so 8 E E 3o 2o ~" 1 ~ (17) (83) (49) ' ' ' ' ' - - 4 ) i i i i i i i 1 2 3 4 $ 6 7 Years Post-Transplantation Fig 3. Actuarial freedom from graft vasculopathy. ]42) ~ 11111 "6" 8 e,, 7 2 u. 1 ~ v (~1), (~),_._...~(1) i, i i i i i 1 2 3 4 5 6 7 Years P(~t-Transplantatlon Fig 4. Actuarial freedom from rejection. Renal Status Perioperative peritoneal dialysis was instituted in 17 infants after transplantation. The duration of peritoneal dialysis ranged from I to 11 days. None of the survivors has required chronic dialysis. The mean serum creatinine level of survivors is.68 -+.23 mg/dl (range,.4 to 1.5 mg/dl) and the mean glomerular filtration rate is 78 -+ 26 ml min -1 1.7 m -2 (range, 25 to 146 ml min -1 1.7 m-2). Twenty patients (18.7%) are currently receiving antihypertensive therapy. Reoperations Five patients in this series underwent retransplantation with one operative death due to severe rejection, The indication for retransplantation was GV in 4 patients and severe acute rejection in I patient. Freedom from retransplantation at 1, 5, and 7 years was 99%, 98%, and 95%, respectively (Fig 5). Residual or recurrent coarctation was detected in 29 patients after transplantation; 4 patients required surgical coarctectomy and 25 patients were treated with balloon angioplasty without any mortality. Sick sinus syndrome and intermittent heart block warranted the placement of a permanent pacemaker in 2 children at 8 months and 4.5 years after transplantation. Both patients are alive, and 1 of them is no longer pacemaker dependent. Two patients required plication of the left hemidiaphragm, and 1 patient underwent 3 reoperations for recurrent pulmonary venous obstruction before he died of infection. Freedom from reoperation, including retransplantation, at 1, 5, and 7 years was 94%, 9%, and 87%, respectively (Fig 6). Growth and Quality of Life Transient perioperative seizure activity was recognized in 2% of patients in this series. Only 1 patient, who suffered an embolic stroke 2 years after CTx, is currently receiving anticonvulsant therapy. Neuropsychologic outcome was normal in nearly 9% of survivors. The growth curve percentiles for height and weight have followed normal progressive growth patterns for the majority of patients. During the first year after CTx, the number of hospital readmissions and the total number of hospital days averaged 1.9 (range, to 9) and 13.9 (range, to 188),

Ann Thorac Surg CHAMBERLAIN PAPER RAZZOUK ET AL 5 1996;62:1-8 CARDIAC TRANSPLANTATION FOR HLHS 1 9 P,,,,.- 8 E,:. 2'~ 7 ',-,~ 6 so ~m ~ 4 IJ. t- I-- 3 1 (142) (17) (83) (49) (16) + I i i i ~, + 1 2 3 4 S 6 7 Years Post.Transplantation Fig 5. Actuarial freedom from retransplantation. respectively. Thirty-four survivors are currently 6 years old or older and all are attending school. With the exception of 2 children who require special education, the psychosocial and scholastic performance of this group is average or above average. Over half of those "long-term" survivors are maintained on cyclosporine monotherapy for their immunosuppression at an average daily dose of 5 mg/kg. Only 8 patients receive corticosteroids, for reasons of asthma (1), retransplantation (1), and recurrent rejection (6). The majority of survivors have excellent cardiovascular status and normal left ventricular ejection fraction. In a preliminary study, 1 recipients (mean age, 6.7 years) and a size-matched group of normal children underwent treadmill testing. There was no significant difference in maximal heart rate (165 +- 17 versus 187 _ 19 beats/min; p ---.68) or oxygen consumption (34.8 + 8 versus 43.2 _.+ 6 ml kg -t min-1; p =.98) between the two groups. Comment Nearly 1,5 to 2, babies born in the United States and Canada each year have HLHS. Current management of those infants varies from supportive care only, concluding in death, to staged palliative reconstruction and cardiac transplantation. Successful repair of HLHS by staged reconstruction was first reported by Norwood and associates in 1983 [1]. Since then there have been various modifications of this approach. Currently reconstructive repair consists of three carefully planned operations that culminate in the separation of the systemic and pulmonary circulations by the application of the Fontan procedure. Although Iannettoni and associates [6] reported in recent years improved results of first-stage palliation for HLHS with 85% hospital survival, cumulative experience with first-stage operation from other institutions reflects a 3% to 4% operative mortality [7, 8]. Moreover, the cumulative risk associated with multistage reconstruction and interstage attrition remains high. Newborns with HLHS and significant tricuspid valve regurgitation have been known to be poor candidates for palliative reconstruction. Moreover, Boston Children's Hospital's 1-year experience with palliative operations for HLHS indicated that infants with certain anatomic subtypes (eg, mitral atresia or aortic atresia) face significant early mortality after reconstruction and perhaps are best treated with cardiac transplantation [8]. Early attempts to treat HLHS patients with transplantation were made in 1984 by Yacoub in London (unreported case) and Bailey and associates [9], who implanted a baboon heart in the case of Baby Fae. Successful neonatal cardiac allotransplantation for HLHS was first performed by Bailey in November 1985; the recipient is now 1 years old and currently the oldest survivor of such therapy. One of the major issues with transplantation is the shortage of heart donors. The management of HLHS infants during the waiting period can be extremely challenging. They are prone to the development of intractable heart failure or systemic organ dysfunction as the waiting period lengthens. Prostaglandin Ea infusion is maintained at the minimal dose to avoid the potential side effects of its prolonged administration. Infants with a restrictive foramen ovale tend to experience progressive hypoxemia during the waiting period and therefore require appropriate intervention to enlarge the interatrial communication and reduce pulmonary venous congestion. Although some restriction in blood shunting at the atrial level is desirable in infants with HLHS to prevent pulmonary overcirculation, a severely restrictive atrial septai communication has been shown to be a significant negative risk factor for death before transplantation [1]. As institutions, neonatologists, and cardiologists develop more experience in caring for these infants, the pretransplantation mortality, which now ranges from 2% to 4%, can be reduced significantly [11-14]. The short-term results of CTx for HLHS have been excellent and the operative mortality, in our experience, has remained around 1% over the 1-year period. Backer and colleagues [15] reported a 15% operative mortality in their early experience but no in-hospital deaths in their later series of 1 FILHS patients who received transplants. Others, whose reports included a smaller group of patients, had an 18% to 37% operative mortality [12, 13]. I: Q. 9 O* o~ E E +o @ ++ II. (142) luu,.,.,_(~) 9 8 7 6.5 4 3 2 1 1 (75) (42) (12) i ~ i i i I 2 3 4 5 6 7 Years Post-Transplantation Fig 6. Actuarial freedom from reoperation.

6 CHAMBERLAIN PAPER RAZZOUK ET AL Ann Thorac Surg CARDIAC TRANSPLANTATION FOR HLHS 1996;62:1-8 Early death is usually related to technical issues, graft failure, infection, or rejection. Management and technical failures can be minimized as the experience of the transplant team grows and the learning curve flattens. Four operative deaths in this series were related to surgical technique or an error in judgement. Patchy ischemic necrosis of the myocardium developed in one donor heart and led to the recipient's death 5 days after CTx, At autopsy, the small vessels as well as the major coronary arteries were clear. There was no apparent cause of necrosis, but perhaps myocardial preservation may not have been adequate. Another patient was discovered at the time of transplantation to have anomalous drainage of the right pulmonary veins and a persistent left superior vena cava. The right pulmonary venous connection was ambiguous and could not be clearly identified. The infant died a day later with marked congestion of the right pulmonary vasculature. When other complex cardiovascular defects are associated with HLHS, echocardiography alone may not be sufficient to define the anatomy properly. In these instances pretransplantation cardiac catheterization may be prudent. One neonate with HLHS and interrupted aortic arch received an oversized heart from a 13-month-old child (donor: recipient weight ratio greater than 4). The baby died a day later. Autopsy revealed a hypoplastic native descending aorta and a restrictive anastomosis between donor and recipient aorta. The combination of interrupted aortic arch, small descending aorta, and an oversized graft proved lethal in that case. The fourth management-related death was due to a mural thrombus of the distal aorta in a 9-day-old baby. The thrombus formed along an umbilical artery catheter and occluded the renal arteries. It has also been proposed that some neonates with HLHS may have pulmonary vascular disease at birth negatively influencing survival. In fact, Turrentine and associates' series [12] of 13 transplants for HLHS had two operative deaths secondary to pulmonary hypertension, and 1 of those 2 patients demonstrated Heath-Edwards grade 4 pulmonary vascular changes on postmortem examination. The only death in our series directly related to pulmonary hypertension occurred early in our experience when a 3-month-old infant died on the second day after transplantation. However, neither he nor any of the other patients who had early death showed significant pulmonary vascular disease on postmortem histologic examination. For the last 5 years, it has been our protocol to maintain all HLHS recipients on a prostaglandin E 1 regimen after transplantation and gradually wean them from the drug over a 1-week period. Perhaps this practice, along with our tendency to accept oversized donor hearts, has contributed to the low incidence of postoperative complications related to pulmonary hypertension. Size matching of the pediatric donor heart to the infant recipient can be a complex issue because of the severe shortage of donors. Undersized donor hearts can be problematic, especially in the presence of pulmonary hypertension. Our data regarding oversized grafts did not support suggestions by Turrentine and associates [12] that a donor-to-recipient weight ratio of greater than 1.9 had an adverse effect on survival. In our series, grafts from donors up to four times the weight of the recipient have been well tolerated with minimal morbidity. In cases of oversized grafts, the pericardium along the left pleural cavity was excised to the level of the left phrenic nerve. On occasion, the sternum was kept open and delayed primary sternotomy closure was done a few days later. This approach was used in 16 patients in this series with no increase in morbidity. Excessive cardiac output and systemic hypertension seen with larger donor hearts may contribute to cerebral edema and visceral and renal vasospasm. In such cases, inotropic agents are usually avoided after transplantation and vasodilator drugs are used instead to control hypertension. Rejection is a major cause of death in pediatric heart transplant recipients, partly because early diagnosis of rejection is often challenging. The routine use of surveillance endomyocardial biopsy in infants is not very practical and, in the absence of cardiac symptoms, has not been helpful in detecting rejection early [16]. Half of the late deaths in the present series were directly related to graft rejection. In a collected series of 151 children who underwent CTx at different institutions, 8% of late deaths were due to rejection [15]. The rate of rejection in the present series is very similar to that reported by others (1.6 to 1.95 episodes per patient), despite avoidance of chronic steroids [12, 13, 15]. Turrentine and associates [12] noted a lower frequency of rejection among neonate recipients, and 4% of their HLHS patients were not treated for rejection. In our series, more infants than newborns had a long-term rejection-free history, and the freedom from rejection at 5 years was only 2%. This finding may be related to antithymocyte serum induction immunotherapy in our older infants and the longer period of follow-up. Accelerated graft vasculopathy is a major limitation of long-term survival of patients after CTx. The incidence of GV among pediatric heart transplant recipients has ranged from 2% to 4% at 3 years [17]. The earliest documentation of GV in our series was 18 months after CTx. Backer and colleagues [15] reported histologic evidence of intimal proliferation at 13 months and severe GV at 15 months after CTx in 2 separate children. Graft coronary vasculopathy was identiffed as the immediate cause of two late deaths in our series. It was also a major contributing factor to one late death due to severe rejection. Although we have previously noted an association between multiple rejection episodes and GV, the etiology of this lesion is probably multifactorial [18]. It is not known yet whether all pediatric heart recipients are at risk of having GV develop if they live long enough. Therefore, close and long-term follow-up of all patients is indicated. Coronary angiography, dobutamine stress echocardiography, and intracoronary ultrasound imaging may have complementary roles in detecting this lesion. Graft coronary artery disease at~ects the major epicardial vessels and branches along their entire length, and the only effective treatment, therefore, is retransplantation. Another major issue with CTx in infants has been the

Ann Thorac Surg CHAMBERLAIN PAPER RAZZOUK ET AL 7 1996;62:1-8 CARDIAC TRANSPLANTATION FOR HLI-IS side effects of life-long immunosuppressive therapy. The incidence of hypertension in this group of patients has varied from 2% to 5% [12, 15, 19]. Backer and associates [15] reported severe renal insult from combination cyclosporine and amphotericin in 1 neonate who required continuous ambulatory peritoneal dialysis. The other 1 infants in that report had good renal function with a mean serum creatinine level of.73 _.14 mg/dl. Ninety percent of survivors in our population had a creatinine level less than I mg/dl at an average period of 4.6 -+ 1.7 years after CTx. Lymphoproliferative disease in children is usually associated with Epstein-Barr virus infection and with a more intense immunosuppressive therapy. We strongly believe that avoiding the use of chronic steroids in infants and maintaining a level of immunosuppression as low as is compatible with good allograft function will reduce the incidence of neoplasms, hypertension, and renal dysfunction. Concerns have been raised regarding the growth potential and the quality of life of children surviving CTx. We had observed delayed bone age maturation and growth impairment in children taking prednisone. However, when steroid therapy was discontinued, growth velocity normalized and the majority of infants have ultimately shown normal growth. The Bayley scales of infant development used for cognitive assessment found the majority of our infants with a normal mental developmental index and psychomotor developmental index. Neurologic abnormalities occurred in 1% of our patients, with dystonia being the most common disability. Numerous factors may contribute to poor neurologic outcome, including pretransplantation hypoxic ischemic brain injury and low output state, preexisting central nervous system abnormalities, intraoperative cerebral insults, cyclosporine neurotoxicity, and other metabolic and fluid derangements. A study by Rogers and associates [2] evaluated the neurodevelopmental outcome of 2 infants with HLHS who underwent staged surgical repair at three different Children's Hospitals. Of 11 survivors, 7 patients (64%) had varying degrees of mental retardation and 2 patients (18%) had severe cerebral palsy. Eight of 9 children with cognitive delays had acquired microcephaly. In conclusion, CTx in infants is associated with a 7% survival at 7 years of patients who receive transplants. The majority of recipients enjoy a good quality of life. Older survivors attend school and participate in ageappropriate activities. It is our belief that passive euthanasia for babies with HLHS is seldom acceptable. Surgical therapy, either staged reconstruction or CTx, provides hope for survival and is a better alternative to death. Although the long-term prognosis of CTx in this age group is unknown, we believe that CTx can be performed with good operative and intermediate-term results and an acceptable quality of life. Expansion of the donor pool improvements in the diagnosis and treatment of rejection, advances in immunomodulations, and control of graft vasculopathy can make CTx a more available and durable therapy and thus positively affect the survival of infants with HLHS. We acknowledge the important contributions of the neonatologists, pediatric cardiologists, cardiac surgical fellows, intensive care unit nurses, transplant coordinators, infectious disease specialists, and cardiac anesthesiologists at Loma Linda University Medical Center, Loma Linda, California. References 1. Norwood WI, Lang P, Hansen DD. Physiologic repair of aortic atresia--hypoplastic left heart syndrome. N Engl J Med 1983;38:~6. 2. Bailey LL, Nehlsen-Cannarella SL, Doroshow RW, et al. Cardiac allotransplantation in newborns as therapy for hypoplastic left heart syndrome. N Engl J Med 1986;315:949-51. 3. Mavroudis C, Harrison H, Klein JB, et al. Infant orthotopic cardiac transplantation. J Thorac Cardiovasc Surg 1988;96: 912-24. 4. Bailey LL, Concepcion W, Shattuck H, Huang L. Method of heart transplantation for treatment of hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 1986;92:1-5. 5. Chinnock RE, Baum MF, Larsen R, Bailey LL. Rejection management and long-term surveillance of the pediatric heart transplant recipient: the Loma Linda experience. J Heart Lung Transplant 1993;12:$255-64. 6. Iannettoni MD, Bove EL, Mosca RS, et al. Improving results with first-stage palliation for hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 1994;17:934-4. 7. Norwood WI, Jacobs ML, Murphy JD. Fontan procedure for hypoplastic left heart syndrome. Ann Thorac Surg 1992;54: 125--3. 8. Forbess JM, Cook N, Roth SJ, Serraf A, Mayer JE, Jonas RA. Ten-year institutional experience with palliative surgery for hypoplastic left heart syndrome. Circulation 1995;92(Suppl 2):262-6. 9. Bailey LL, Nehlsen-Cannarella SL, Concepcion W, Jolley WB. Baboon-to-human cardiac xenotransplantation in a neonate. JAMA 1985;254:3321-9. 1. Canter CE, Moorhead S, Huddleston CB, Spray TL. Restrictive atrial septal communication as a determinant of outcome of cardiac transplantation for hypoplastic left heart syndrome. Circulation 1993;88(Suppl 2):456-6. 11. Backer CL, Zales VR, Harrison HL, Idriss FS, Benson DW, Mavroudis C. Intermediate term results of infant orthotopic cardiac transplantation from two centers. J Thorac Cardiovasc Surg 1991;11:826-32. 12. Turrentine MW, Kesler KA, Caldwell R, et al. Cardiac transplantation in infants and children. Ann Thorac Surg 1994;57:546-54. 13. Webber SA, Fricker FJ, Michaels M, et al. Orthotopic heart transplantation in children with congenital heart disease. Ann Thorac Surg 1994;58:1664-9. 14. Chiavarelli M, Gundry SR, Razzouk AJ, et al. Cardiac transplantation for infants with hypoplastic left-heart syndrome. JAMA 1993;27:2944-7. 15. Backer CL, Zales VR, Idriss FS, et al. Heart transplantation in neonates and in children. J Heart Lung Transplant 1992;11: 311-9. 16. Balzer DT, Moorhead S, Saffitz JE, Huddleston CB, Spray TL, Canter CE. Utility of surveillance biopsies in infant heart transplant recipients. J Heart Lung Transplant 1995;14:195-11. 17. Penn I. Incidence and treatment of neoplasia after transplantation. J Heart Lung Transplant 1993;12:$328-36. 18. Bailey LL, Zuppan CW, Chinnock RE, et al. Graft vasculopathy among recipients of heart transplantation during the first 12 years of life. Transplant Proc 1995;27:1921-5. 19. Starnes VA, Griffin ML, Pitlick PT, et al. Current approach to hypoplastic left heart syndrome--palliation, transplantation, or both? J Thorac Cardiovasc Surg 1992;14:189-95. 2. Rogers BT, Msall MR, Buck GM, et al. Neurodevelopmental outcome of infants with hypoplastic left heart syndrome. J Pediatr 1995;126:496-8.

8 CHAMBERLAIN PAPER RAZZOUK ET AL Ann Thorac Surg CARDIAC TRANSPLANTATION FOR HLHS 1996;62:1-8 DISCUSSION DR CONSTANTINE MAVROUDIS (Chicago, IL): I congratulate the Program Committee for naming this deserving report for the 1996 J. Maxwell Chamberlain Memorial Paper. I also compliment Dr Razzouk and his colleagues for presenting a succinct and informative analysis of their 1-year unprecedented experience with orthotopic cardiac transplantation for hypoplastic left heart syndrome. Laudatory comments are common and deserving for Dr Bailey; he has unselfishly shared his vision and experience with all. At this 1-year mark, some reflections are in order. First, the improved results with cardiac transplantation and staged repair make the expectant care option, which results in death, a much lesser option. Even so, some neonatologists, pediatric cardiologists, and pediatricians have other ideas. In a recent published survey [1] of neonatologists, pediatric cardiologists, and pediatricians, most, but not all, agreed that orthotopic cardiac transplantation is reasonable for untreatable cardiac disease in infants. When asked if orthotopic cardiac transplantation should be presented negatively to protect parents, the following percentage of groups agreed: pediatric cardiologist, 14%; neonatologists and pediatricians, 48%; and health maintenance organization physicians, 61%. High-risk patients are treated in all subspecialties often with poorer results than presented here for therapy of hypoplastic left heart syndrome. We must insist on consistency and not single out certain diseases for criticism. Second, it is safe to say that increased experience with recent modifications of the Norwood procedure have led to improved survival at a number of institutions. The early bidirectional Glenn shunt en route to a Fontan operation has also improved the outcome during this time period. Moreover, the 19% pretransplantation mortality can be avoided in most patients because no obligatory wait for the donor organ is necessary. In fact, we and other colleagues at other institutions have converted to staged repair because of dwindling donor availability. In any case, what seems to be a therapeutic option for all patients must be looked at with the inevitability of two more operations and the possible ravages of the Fontan operation, which include ventricular failure, stroke, and protein-losing enteropathy. Third, the transplant option, as presented by the Loma Linda group, has a 7-year actuarial survival of 7%. Our 5-year survival is more or less 6% for patients with hypoplastic left heart syndrome. To put this in perspective, we recently lost a patient because of rejection 91,~ years postoperatively. No program has achieved the success of the Loma Linda group. The transplant option either as primary therapy or rescue therapy for failed staged repair remains an important therapeutic arm in the treatment of patients with hypoplastic left heart syndrome and must be continued by experienced centers. There are clear-cut advantages and disadvantages for both therapeutic arms. Breakthroughs in xenotransplantation and more acceptable immunotherapy will probably favor the transplant option. Two ventricles are clearly better than one. I have two questions. Have you experienced any significant variance in obtaining donors for your patients over the 1-year period? And second, we have recently shown a decrease in rat transplant vasculopathy with captopril administration and have instituted this regimen in our patients. Have you any experience with this mode of therapy? DR JOHN R. BENFIELD (Sacramento, CA): I wonder if you would comment about the psychological profile of these children, and if you could also give us some idea about the size and constitution of the team that renders the care. DR RAZZOUK: Thank you, Dr Mavroudis, for your kind remarks. I will answer President Benfield's question first regarding the psychological outcome of those recipients. It is very difficult to assess the mental and psychological state of infants in their developing stages, but we have used the Bayley scale for infant development to assess the cognitive function of the recipients and found out that about 9% of them have normal mental developmental index and psychomotor developmental index. As we looked at those survivors who are 6 years or older, all of them are attending school; only 2 are requiring special education, but they are still in an age-appropriate grade. Transplantation requires a team effort and certainly constant commitment and dedication by an institution. We have a large infrastructure at Loma Linda that consists of various specialists in different areas, including neonatologists, immunologists, pediatric infectious disease specialists, intensivists, and specially trained nurses. We have had the habit of sending a staff surgeon on almost all the procurement operations except for three occasions in the last 1 years. To answer Dr Mavroudis' question regarding the availability of donors over the last 1 years, we have gotten used to traveling all over the northern continental states looking for donors. Donor shortage is a real problem in transplantation. That is why we have had to use oversized grafts. The pattern has not changed very much lately. Our pretransplantation mortality has averaged 19% as we showed here. It has fluctuated between 15% and 25% over the years. We have used captopril on occasion to control patients in whom hypertension develops in the perioperative and immediate postoperative period, but we have not used captopril as a chronic drug to control graft vasculopathy. Thank you again very much. Reference 1. Higgins SS, Paul SM, Hardy CE, Ternullo-Retta C, Affonso D. Infant heart transplantation: a survey of physician beliefs. J Heart Lung Transplant 1994;13:59-65.