HLA antibody sensitization - implications for heart transplantation

Steven Zangwill, MD, pediatric cardiologist, Herma Heart Center, Children's Hospital of Wisconsin, assistant professor, Pediatrics (Cardiology) Medical College of Wisconsin

The Major Histocompatibility Complex (MHC) is a genetically encoded series of molecules found on human cells that have far reaching implications for the immune system. They represent the basis for the body's recognition of "self" as well as function as presentation proteins that carry foreign cells (infections, donor blood or donor tissue) to other parts of the immune system for destruction. The immune system itself is very complex with numerous different types of cells interacting through a myriad of blood-borne mediators.

White blood cells (WBC) are perhaps the most immunologically active part of this system and lymphocytes are one type of WBC (and the type most frequently involved in organ transplant rejection). The two most important types of lymphocytes are B cells and T cells.  While there are many sub-types of T cells (helper, suppressor, cytotoxic), these cells are the basis for what is termed cell-mediated immunity. B cells are the cells that make antibodies to help attack foreign substances.  Organ rejection is caused most commonly by T cells (and therefore can be seen on a tissue biopsy) although rejection can, with less frequency, be caused primarily by B cells (so-called humoral rejection). This type of rejection can be more difficult to diagnose but just as critical to recognize and treat. The MHC therefore plays an integral role in immune surveillance and interacts with the cellular components and their inflammatory mediators to facilitate the complicated and critically important immune system functions. Found on chromosome 6, the MHC encodes three major classes of Human Leukocyte Antigen (HLA) molecules.

• Class I – proteins found on the surface of nearly all cells and primarily act as presenting cells for cytotoxic T cells.
• Class II – proteins found mainly on the surface of specialized immune cells and act primarily as presenting proteins.
• Class III – proteins involved in diverse immune systems including complement activation. Not a focus for transplantation.

MHC molecules themselves are fairly complex proteins and have significant heterogeneity from person to person. We have a nomenclature system to identify Class I and Class II alleles (gene types) with the major focus  on specificity at the A, B, C and DR loci (gene locations). Every person has one type for each locus inherited from their parents. For example, one person's HLA type may be classified as A7 A17, B6, B35, C22, - DR7, DR18. (This person is pure for C22). If this person receives blood, platelets, a kidney or a heart from a person who is A26, A28 . . . the recipient immune system may recognize the foreign HLA markers and make antibodies specifically against these HLA antigens. The B cells that make these antibodies become memory cells and will be ready to make an aggressive response if these antigens are introduced a second time. (This is the basis for the effectiveness of childhood immunizations). Many of our patients who come to heart transplantation have had previous exposures to foreign HLA types either from routine blood transfusions, previous surgeries, the use of assist devices or previous tissue grafts.

Before we list a patient for heart transplantation, we routinely perform a test called a Panel Reactive Antibody (PRA). With this we test a patient's blood against a series of antigen-coated beads representing typical HLA types seen in the general population. Using this test, we can identify the presence of pre-formed antibodies that the patient may have to potential donor HLA types. We can express the result as a numerical percentage, reflecting a probability of having antibodies directed against a percent of the general population; alternatively, we can report the specific HLA antigens to which the patient has demonstrable antibody.

Multiple studies have shown that patients with elevated PRA are at higher risk for morbidity and mortality following heart transplantation. This risk includes specifically acute and chronic rejection (both cellular and humoral) as well as an independent risk of long-term problems in addition to those caused by recognizable organ rejection. Additionally, a specific concern for this population is a problem called hyper-acute rejection which may occur when a sensitized patient receives an organ with HLA types to which the recipient already has antibodies. This can result in nearly immediate rejection of the heart with complete loss of graft function less than an hour after transplantation.

To reduce these risks, a number of strategies exist for patients who are highly sensitized (have a high level of antibodies).

IVIG: Ironically, monthly administration of pooled human antibodies has been found to reduce the level of HLA specific circulating antibodies in potential transplant recipients. This strategy can be used pre- as well as post-transplantation.

Immunosuppression: Medications including Mycopheno-late Mofetil (Cell Cept) and/or Cyclophosphamide (Cytoxan) have been shown to suppress Ab development and/or expression. While Cell Cept routinely is used post transplant, some centers have used it pre-transplant for sensitized patients with variable success. Cytoxan seems more effective although, perhaps, with a somewhat less favorable side effects profile.

Plasmapheresis: An extracorporeal process whereby blood is filtered to remove pre-formed antibodies. This can be performed pre-transplant, during the operation itself (on bypass), as well as post-transplant if needed.

Matching: Utilized in renal transplantation, national registry lists give priority to patients who have "perfect" HLA matches with specific donors. This unfortunately is not a tenable solution for heart transplantation given the limited availability of donor organs.

Cross-matching: We can test a recipient's blood directly with a specific potential donor. Due to time constraints, this usually is only possible with a local donor. By convention, many centers will require a prospective cross-match for any recipient with a PRA greater than 10 percent. This severely can limit the ability to find a suitable organ.

Virtual cross-matching: Recent preliminary data from the transplant center at Children's Hospital of Wisconsin suggests that we can predict actual cross-match results by comparing donor HLA types to pre-determined recipient HLA antibody specificity. Due to the complexity and heterogeneity of HLA molecules, this process requires further (ongoing) study.

Total lymphoid irradiation (TLI): TLI can be used to destroy sensitized lymphocytes post-transplant and may decrease the likelihood of both humoral as well as cell mediated rejection. 

Aggressive surveillance: Sensitized post-transplant patients may have augmented "routine" immunosuppression as well as more aggressive rejection surveillance. This may include increased frequency of catheterization with biopsy as well as utilizing special immunoflourescent stains to detect Ab activation in the cardiac allograft.

While the management of our sensitized candidates for heart transplant can be challenging, our armamentarium and experience continues to grow and we have good reason to remain optimistic for these children. Recent advancements in immunosuppression as well as rejection surveillance and histocompatibility likely will lead to continually improving outcomes.

Ventricular versus innominate artery shunt in the modified Norwood
procedure: Evaluating the evidence

Kathy Mussatto, RN, BSN, cardiothoracic research nurse,
Herma Heart Center, Children's Hospital of Wisconsin; Nancy Ghanayem, MD, pediatric critical care specialist, Children's Hospital, assistant professor, Pediatrics (Critical Care), Medical College.

"You cannot change where you've been, only where you are going." – Anonymous

The science of surgery for congenital heart disease (CHD) has evolved rapidly since its inception in the early 1950s.  Techniques and technology have progressed so significantly that what was once thought state-of-the-art – Dacron patches for repair of coarctation – has been found to be fallible and what was once thought impossible – the arterial switch procedure for transposition of the great arteries – has become routine. Changes in practice have been driven by individual innovation and knowledge shared through publications and research presentations. The strongest evidence to support modifications to existing procedures is derived from unbiased, hypothesis-driven clinical research.  Unfortunately, given the relative rarity of individual congenital heart defects, establishing irrefutable evidence often is difficult, if not impossible. Increasingly, collaboration between CHD treatment centers is being seen as the optimal method for effective testing of hypotheses and generation of data that truly can be used to drive evidence-based decisions in practice.

One of the greatest surgical challenges in congenital heart disease has been the management of infants with hypoplastic left heart syndrome (HLHS). The advent of the Norwood procedure along with important advances in medical management have, for the first time, allowed patients with this diagnosis to have a reasonable hope for survival. At Children's Hospital of Wisconsin, our "modern" experience with the Norwood procedure began in 1992 when we had our first stage I survivor and an overall early survival rate (30 days post-op and hospital discharge) of 40 percent (2/5 patients). In 1994, early survival was 55 percent (6/11 patients) and by 1999, after the implementation of several significant changes in management, 83 percent of patients (15/18) survived stage I palliation. Most recently, in 2002, 23 Norwood procedures were performed and 100 percent of these infants survived to hospital discharge.  These results currently are the best early survival rates reported in the world.

Despite these phenomenal improvements in survival, both late mortality and morbidity remain a significant problem and unfortunately, the results achieved here at Children's Hospital have not been consistently reproducible at other centers. Recently, data reported from the Congenital Heart Surgeon's Society involving 29 centers and 710 infants with HLHS demonstrated an actuarial survival of 72 percent at 1 month, 60 percent at 1 year, and 54 percent at 5 years (Ashburn, 2003). Late attrition commonly is reported to range from 5 to 15 percent. In addition, significant morbidity is common. Evaluation of a series from Children's Hospital of Philadelphia identified length of hospital stay > 30 days in 30 percent, seizures in 13 percent, NEC – 10 percent, CPR – 10 percent, and need for reintubation in 34 percent (Gaynor, In press). Identified risk factors for mortality include: native ascending aorta < 2.5 mm, weight at operation < 2.5 kg, older age at stage I, significant tricuspid valve regurgitation, extracardiac defects or syndromes, and residual lesions following stage I. While the exact cause of death frequently is unknown, coronary insufficiency from diastolic run-off has been implicated. In a post-mortem study, 27 percent of deaths were attributed to impaired coronary perfusion, 19 percent due to excessive pulmonary blood flow and 17 percent to obstruction to pulmonary blood flow (Bartram, 1997). The group at the University of Michigan Congenital Heart Center conclusively demonstrated reduced coronary flow and myocardial oxygen delivery resulting from diastolic run-off via the Blalock-Taussig shunt (Donnelly, 1998). They speculate this may have important implications for both the early post-op course and long-term ventricular performance.

An integral part of the Norwood procedure is placement of a synthetic shunt to provide a pathway for blood flow to the pulmonary circuit. Traditionally this shunt has been a modified Blalock-Taussig shunt (MBTS) placed from the innominate or other central artery to the pulmonary arteries. Because the shunt is placed distal to the aortic valve it allows perfusion to the pulmonary circuit during both systole and diastole. This creates the potential for coronary artery steal, the phenomenon of blood flowing away from the coronary arteries and into the low-pressure pulmonary arteries during diastole.

Recently, several groups have presented data on the use of a shunt placed directly from the right ventricle to the pulmonary arteries (RV-PA). The technique involves making a small right ventriculotomy, undermining of the RV muscle and placement of a Goretex shunt directed to the PA confluence. First described by Kishimoto et al. (1993), the technique was popularized by Dr. Sano in 2001 when he reported an improvement in early survival from 50 percent with the MBTS to 89 percent with the RV-PA shunt. Other groups have reported similar results including Dr. Norwood's current practice in Delaware where survival improved from 70 percent to 92 percent (Pizarro, 2003) and Malec (2002) in Poland showed an improvement from 65 to 89 percent with the RV-PA shunt. All groups reported a smoother early postoperative course and decreased morbidity.

Some of these disadvantages, including issues with shunt stenosis or thrombosis and inadequate pulmonary artery growth, also are encountered with the classic MBTS.

Although fairly compelling data has been presented in terms of improved early survival, reports to date have been from small numbers of patients and single institutions that previously had struggled with outcomes following the traditional Norwood procedure. Many long-term issues remain unresolved including the impact on cardiac output, ventricular function and interim mortality. To systematically compare the outcomes of the two surgical techniques a collaborative trial has been proposed to involve Children's Hospital of Wisconsin, University of Michigan Congenital Heart Center, Children's Hospital of Philadelphia and Boston Children's Hospital. The proposed study has been named, The VISIONS Collaborative Trial: Ventricular versus Innominate Shunt: Impact on Norwood Survival. The specific aim of the study will be to determine the incidence of a composite primary end-point at 12 months post-surgery and compare this incidence between the MBTS and the RV-PA shunt. The primary end-points include: death, listing for transplantation, need for ECMO, need for re-operation related to the shunt or cross-over to the alternative shunt after placement of the initial shunt, and failure to achieve stage II palliation by 12 months of age. Secondary aims are to compare post-op ICU morbidity and mortality, invasive hemodynamics and pulmonary artery growth by cardiac cath prior to stage II, early post-op and pre-stage II RV function by ECHO, nutritional parameters and somatic growth, neurodevelopmental outcome and the incidence of the apolipoprotein E?4 gene, and coronary artery perfusion and flow reserve. To achieve adequate statistical power, 210 patients will be randomized over a two-year period to either the RV-PA or innominate artery shunt procedure.  Randomization will be stratified by surgeon, presence or absence of aortic atresia, and presence of obstructed pulmonary venous return or restrictive atrial septum.  Diagnosis of a single, morphologic RV, planned Norwood procedure and written informed consent will constitute inclusion criteria. Families choosing not to participate will be informed that their child will receive the planned Norwood procedure with the placement of the shunt to be determined by the surgeon; nonparticipation will not influence shunt placement technique. A data safety monitoring board (DSMB) composed of an independent panel of clinical experts including a surgeon, two cardiologists, a statistician and an ethicist, will evaluate data from the trial every six months and will advise the study group on adverse events and the value of study continuation.

Currently, the study is under review by the respective Institutional Review Boards at each cardiac center. Plans are to begin enrolling patients in the trial in early 2004.  At Children's Hospital of Wisconsin, the study will be led by co-principal investigators, Nancy Ghanayem, MD, and James Tweddell, MD, medical director, Cardiothoracic Surgery, Children's Hospital and associate professor, Surgery (Cardiothoracic), Medical College of Wisconsin.  The study offers a rare opportunity to scientifically evaluate the unique advantages and/or disadvantages of two different surgical techniques. It will demand the collaboration and sharing of knowledge between four major cardiac centers but promises to create an exceptional cohort of children and families facing life with complex congenital heart disease. Their willingness to participate in this landmark study will add substantially to our understanding of the management and overall impact of congenital heart disease.

Comparison of intravenous morphine analgesia with caudal morphine
analgesia in single ventricle patients

Eckehard Stuth, MD, pediatric anesthesiologist, Children's Hospital of Wisconsin, associate professor, Anesthesiology, Medical College of Wisconsin; Richard Berens, MD, pediatric anesthesiologist, Children's Hospital, associate professor, Anesthesiology, Medical College; George Hoffman, MD, medical director, Anesthesiology, Children's Hospital, associate professor, Anesthesiology, Medical College.

In late 2002 the clinical protocol at Children's Hospital titled "A randomized, double-blind study of regional vs. intravenous analgesia to promote early endotracheal extubation in children undergoing single ventricle palliation" was approved (CHW 02/132, HRRC 402-02). This is a prospective clinical trial of patients undergoing Stage II (bidirectional Glenn or HemiFontan) or Stage III (Fontan-type) single ventricle palliation. The rationale of this study is based on three premises: 

1. Expedient postoperative extubation is beneficial in patients without a pulmonary ventricle, because spontaneous breathing promotes pulmonary blood flow at a lower venous pressure and thus improves the cardiovascular performance of the single ventricle patient.
2. Optimal respiratory mechanics, including low work of breathing and maintenance of lung volume (especially functional residual capacity), most likely will occur with the combination of good analgesia and preservation of airway reflexes.
3. The scientific benefits of single dose caudal epidural morphine are poorly documented in this setting. 

This is a local pilot study and includes investigators from pediatric anesthesia, cardiovascular surgery, critical care and cardiology. The study also benefits from the strong cooperation of the operating room, cardiovascular, preoperative clinic and intensive care unit nursing staffs, as well as the support of pharmacists, anesthesia technicians, respiratory care services staff and EKG laboratory staff.
Patients with single ventricle physiology (hypoplastic right or left heart syndrome or other severe ventricular hypoplasia preventing two-ventricle repair) who are candidates for early extubation ("fast tracking") in the operating room are eligible for participation in this study. The fast track assessment is made jointly by the anesthesiologist and attending cardiac surgeon. Major exclusion factors are few but include contraindications to performance of a single shot epidural (such as significant bleeding disorder or skin infection at caudal area), ventilatory support before surgery, anticipated need for ventilatory support after surgery because of additional complex repairs such as relief of residual aortic arch obstruction, and evidence of prolonged QT syndrome.

The purpose of the study is to find out whether a single dose of caudal epidural morphine is superior to intravenous morphine in terms of ease of extubation, intraoperative and early postoperative cardiopulmonary stability and postoperative pain control and sedation requirements. In short operations with no need for cardiopulmonary bypass, a large single dose of epidural morphine (100 mg/kg) given in this study appears to provide better early postoperative pain relief than a similar dose of intravenous morphine (150 mg/kg) but such an advantage never has been demonstrated for moderately long cardiac operations such as stage II and III single ventricle palliation. In addition, when a caudal is contraindicated, good pain control often can be achieved with intravenous doses of morphine of similar size (150 mg/kg). Both types of pain relief are used routinely for these patients and we want to assess if one method provides a distinct advantage.

The most reliable way to objectively compare the relative benefits of single dose epidural morphine versus intravenous morphine is by random concealed assignment of the study analgesic. All consented patients are given a general anesthetic that is compatible with early postoperative extubation and receive a dose of droperidol after cardiopulmonary bypass for initial sedation. The general anesthetic is performed with the volatile agents sevoflurane and isoflurane before, during and after cardiopulmonary bypass supplemented with moderate intravenous doses of the opioid fentanyl. All patients receive a dose of morphine considered appropriate for early postoperative pain control, which either is given as caudal morphine before incision (100 mg/kg) or as intravenous morphine (150 mg/kg) after cardiopulmonary bypass and heparin reversal. Thus every patient will receive an effective dose of opioid analgesia regardless of treatment assignment, which will remain concealed. Additional rescue doses of fentanyl and morphine can be given at any time when clinically indicated, and no waiting period is required for additional analgesia after extubation. The intensive care unit (ICU) staff may order any analgesic or sedative that is indicated, but all drugs and dosages given within the first 12 hours after admission to the ICU should be carefully recorded on the analgesia/sedation study sheets. Some recommendations are given as to the initial doses of morphine, diphenhydramine (Benadryl) and other sedatives, but actual dosing is at the discretion of the attending staff. 

It is planned that the treatment assignments only will be revealed (unblinded) and data analysis performed after completion of the entire study (our target is to enroll about 60 patients age 2 months to 5 years over a two- to three- year period) unless safety concerns arise. The study began in 2003 and parents of 28 patients have been asked to participate. As of December 2003, 22 patients have been enrolled. Two study patients failed extubation, (had to be kept intubated or had to be reintubated within the first 12-24 hours postoperatively). This rate of failed extubation is  relatively normal for this patient population, and no unexpected outcomes have occurred to date. We hope the data from this study will give us more insight into optimal care for these patients, and are thankful for the cooperation and support of everyone who helps care of them. 

Pediatric cardiology fellowship program

Herma Heart Center accepts two physicians annually into a four-year cardiology fellowship training program. This training provides two years of clinical work and two years of research in pediatric cardiology.

The fellowship program is sponsored in conjunction with the Cardiovascular Research Center at the Medical College of Wisconsin to maximize exposure to multiple research opportunities. In addition, it is anticipated that graduates will have an opportunity to obtain a master of science degree as part of the fellowship.

For more information, contact Steve Zangwill, MD, at (414) 266-2380.