SELECTION OF PATIENTS FOR INITIAL CLINICAL TRIALS OF SOLID ORGAN XENOTRANSPLANTATION

Abstract

Several groups have reported extended survival of genetically-engineered pig organs in nonhuman primates, varying from almost 10 months for life-supporting kidney grafts and >2 years for nonlife-supporting heart grafts to less than 1 month for life-supporting liver and lung grafts. We have attempted to define groups of patients who may not have an option to wait for an allograft. These include kidney, heart, and lung candidates who are highly-allosensitized. In addition, some kidney candidates (who have previously lost at least 2 allografts from rapid recurrence of native kidney disease) have a high risk of further recurrence and will not be offered a repeat allotransplant. Patients with complex congenital heart disease, who may have undergone previous palliative surgical procedures, may be unsuitable for ventricular assist device implantation. Patients dying of fulminant hepatic failure, for whom no alternative therapy is available, may be candidates for a pig liver, even if only as a bridge until an allograft becomes available. When the results of pig organ xenotransplantation in nonhuman primates suggest a realistic potential for success of a pilot clinical trial, highly-selected patients should be offered participation.

Introduction

A major limiting factor in allotransplantation remains the availability of organs from deceased donors. In the USA, 22 patients on the United Network for Organ Sharing (UNOS) waiting list die each day (ie, >8,000 each year) because a suitable deceased human donor organ does not become available (http://optn.transplant.hrsa.gov/). There is therefore a need for an alternative source of organs for patients with terminal organ failure.¹

The use of pigs as sources of organs (xenotransplantation) could potentially address this problem. Substantial progress has been achieved in pig organ transplantation in nonhuman primates with (i) better understanding of the pathobiology,² (ii) innovations in genetic engineering,³ (iii) improvements in perioperative management, and (iv) the introduction of novel immunosuppressive agents.^4–7

A life-supporting kidney graft has functioned for almost 10 months^8,9, and a nonlife-supporting heart graft for >2 years.^5,6,9 Although the immunosuppressive regimens used are currently not clinically applicable, it is reasonable to assume that equivalent biologics will become available or further genetic engineering of the organ-source pig will allow conventional immunosuppressive therapy to prevent the adaptive immune response.⁴ Progress in liver and lung xenotransplantation has been slower because of the presence of additional factors that result in early graft failure or recipient demise.^9–11

There are several ethical concerns to be considered before embarking on clinical trials of xenotransplantation.¹² These will not be discussed fully here but, for example, kidney transplant candidates have the option to remain on renal replacement therapy, and many heart transplant candidates can be supported by ventricular assist devices (VADs) or total artificial hearts (TAHs). While these currently available technologies for supplemental organ support may be adequate and reasonable options for many patients, we believe there is a select subset who may not have an option to wait for an organ allograft. (Indications for clinical trials of tissue or cell xenotransplantation have been addressed by others^13,14 and are already underway.)

The US FDA suggests that xenotransplantation should be limited to “patients with serious or life-threatening diseases for whom adequately safe and effective alternative therapies are not available,” while limiting patients to those “who have potential for a clinically significant improvement with increased quality of life following the procedure” (http://www.fda.gov/cber/guidelines.htm).

In our opinion, the unifying factor is that these patients should be significantly disadvantaged or completely excluded from being offered a human organ. Although there may be occasional exceptions, prospective recipients should fulfill the general listing criteria that are already standard-of-care. They must pass (i) medical (eg, current negative cancer screening, negative infectious disease workup, adequate function of other organs), (ii) surgical (eg, adequate vascular targets for anastomosis), and (iii) psychosocial (eg, no active uncontrolled psychiatric disease, adequate history of compliance, adequate caregiver and social support) evaluations. Although there might be a temptation to offer xenotransplantation to patients who do not fulfill these criteria, this would almost certainly result in poor outcome that would neither provide a valid assessment of the potential of xenotransplantation nor be of benefit to the patient.

We here present our initial considerations as a basis for discussion among the transplant community.

Kidney Xenotransplantation

If kidney xenotransplantation is likely to provide benefit over chronic dialysis, there are several reasons why it would be an attractive option for initial clinical trials.¹⁵ Apart from renal failure, sufficient patients are likely to be in relatively good health so that the response to the xenograft can be studied without the confusion of co-morbid disease. If organ dysfunction develops, the patients could be supported by dialysis until xenograft function recovers or until loss of function is deemed irreversible. If indicated by the development of a life-threatening complication, eg, infection, malignancy, a kidney xenograft can be excised, allowing discontinuation of immunosuppressive therapy (in contrast, for example, to a pig heart graft [unless it is replaced by a TAH] or pig islets transplanted into the portal vein).

Patients who have the most limited access to kidney transplantation (determined by the longest estimated wait-time, those who have high mortality while on the wait-list, or those in whom listing has been declined) include those with (i) renal diseases that have a high likelihood of recurring, thus prohibiting the patients from repeat allotransplantation, (ii) 100% calculated panel-reactive antibodies (cPRA), or possibly (iii) loss of vascular access for chronic dialysis (Table 1).

Table 1.

Potential conditions for which initial clinical trials of pig kidney xenotransplantation may be justified*

Rapidly recurrent kidney disease Recurrent focal sclerosing glomerulo-sclerosis (FSGS) Leads to ESRD in about 12–50% of patients,59 who represent about 5% of transplant recipients,60 but the recurrence rate in the graft is 20–30% (reviewed in 61) and up to 50% in younger patients.62 Patients with rapid onset ESRD have higher recurrence rates.63 The recurrence rate in the second allograft can be between 80–100%.64–67 Recurrent immunoglobulin A (IgA) nephropathy Fifteen to 44% of patients with IgA nephropathy will progress to ESRD,68,69 and constitute up to 20% of all patients undergoing kidney transplantation in the USA.69 Histological recurrence of the disease occurs in as many as 60% of the recipients,62 but graft failure only occurs in 2–16%.70–73 After transplantation, rapid progression to ESRD may occur in patients with primary crescentic IgA nephropathy.74,75 Data suggest that the recurrence rate is higher in the second graft.76 Recurrent membranoproliferative glomerulonephritis (MPGN) Type II The 5-year survival of allografts in recipients with MPGN type II (n=78) is only 50%.77,78 In one study, of 29 patients, 12 (41%) developed a recurrence within 1 week to 14 months, of whom 5 (42% of patients with recurrence) lost their allografts.79 In a longitudinal study of 33 patients with MPGN (all types), 21 (49%) developed electron microscopy-proven recurrence in a mean of 4.7 years, of whom 14 (67%) lost their grafts.80 Other potentially-recurrent diseases For other rarer potentially-recurrent diseases, eg, light chain disease, anti-glomerular basement membrane antibody, hemolytic uremic syndrome-thrombotic thrombocytopenic purpura (HUS-TTP), recurrent oxalosis, cystinosis, and methylmalonic acidemia, there are insufficient data to quantify the rate of second allograft loss. However, in allograft recipients whose primary disease was mesangiocapillary glomerulonephritis, the recurrence rate leading to graft failure is 15%.81 High sensitization to HLA Loss of vascular access Elderly patients without significant concomitant disease

Rapidly recurrent kidney diseases include accelerated recurrence of focal sclerosing glomerulo-sclerosis, immunoglobulin A nephropathy, hemolytic uremic syndrome-thrombotic thrombocytopenic purpura (HUS-TTP), C3 glomerulonephritis, and light-chain disease. Taken together, in patients with end-stage renal disease (ESRD) from glomerulonephritis, recurrence of the original disease is the third most common cause of graft failure (following death with a functioning graft, or interstitial fibrosis and tubular atrophy, ie, ‘chronic rejection’).¹⁶ Retransplantation in pediatric recipients with recurrent disease, analyzed as a single group, had poorer survival of the second graft compared with the first (5.9 years vs. 8.3 years; p<0.01).¹⁷

An obstacle to obtaining information about the outcomes of repeat kidney transplantation is the lack of information on the etiology of ESRD in the Scientific Registry of Transplant Recipients (SRTR) database. At the University of Pittsburgh, however, we are reluctant to offer a third allograft to patients with 2 previous graft losses from recurrent primary disease within 4 years.

A guideline for offering a patient a pig kidney xenotransplant could therefore be rapid (within 2 years) loss of 2 sequential grafts without features of acute rejection or interstitial fibrosis and tubular atrophy on >2 biopsies per graft. An obvious question is whether there is a similar risk of disease recurrence in the xenograft. No data exist that can predict this. Nevertheless, even if recurrence occurs in the xenograft as rapidly as in an allograft, at least this would allow a kidney from a deceased human donor to be transplanted into another candidate in whom the allograft is likely to function for significantly longer.

Candidates for kidney transplantation who have been sensitized from previous human leukocyte antigen (HLA) exposure may develop anti-HLA antibodies that could lead to hyperacute rejection of an allograft. Before December 4^th 2014, these patients were disadvantaged and suffered very long wait-times for an organ. However, UNOS implemented a new kidney allocation system (KAS), which likely has significantly changed kidney allocation, with many more patients with a cPRA of 98–100% receiving an acceptable donor kidney.^18,19

At the University of Pittsburgh, during the first 6 months of 2014 (before the new KAS), 5 of 53 (9.4%) patients who underwent kidney transplantation had a cPRA >95%, while during the first 6 months of 2015 (after the new KAS), 17 of 55 (30.9%) had a cPRA >95%. It remains to be determined whether a subgroup of patients still exists who cannot or should not receive an allograft, but it is reasonable to assume that some patients will have anti-HLA antibodies in a configuration that will prohibit allotransplantation or delay it for many years.

Importantly, the current, albeit very limited, evidence is that allosensitization does not increase the risk of xenoreactivity nor does sensitization to a xenograft prevent future allotransplantation^20–22 (although recent unpublished studies by Tector’s group in 2016 questions this conclusion).

A very small group of patients exist who do not have the option of continuing chronic dialysis due to loss of vascular access. However, such a patient may have access to a human cadaveric graft on an emergent basis at the discretion of the local organ procurement organization.

Finally, elderly patients, who may not survive on the waitlist for a very long period of time (while a suitable deceased human donor becomes available), might also be considered as potential candidates for the initial clinical trials, as long as they do not have concomitant health problems that might significantly reduce the likelihood of a successful outcome. However, selection on the basis of age alone may prove controversial.

Heart Xenotransplantation

Cardiac transplantation remains the standard-of-care for end-stage heart failure.^23,24 Despite approaches to mitigate the disparity between potential recipients and deceased donors, eg, use of extended criteria donors, and left (L)VAD implantation as a bridge to allotransplantation or as destination therapy (with biventricular assist device support [BVAD] or a TAH being a less optimal alternative), the ongoing crisis remains unresolved.

Considerable advances in technology and the critical need to address donor organ shortage have made LVAD implementation an indispensable treatment option.^25–27 However, there are several adverse events associated with long-term VAD support. The incidence of coagulation dysfunction (device thrombosis, thromboembolism, hemorrhage [eg, gastro-intestinal] secondary to supratherapeutic anticoagulation), neurologic events, and infectious complications related to the power lines that traverse the skin all remain significant. Mechanical circulatory support is particularly problematic in the pediatric population due to distinct physiology and size constraints.^28,29 Additionally, the presence of, or development of, antibodies against HLA (allosensitization) may be linked to prior mechanical circulatory system device placement,³⁰ and is associated with worse outcomes after heart transplantation.³¹ Procuring a negative cross-matched organ for highly-sensitized patients is difficult.

Patients who are unable to be successfully supported by LVADs represent a special population that can be uniquely addressed with xenotransplantation (Table 2). Although additional consideration must be given to BVAD support or implantation of a TAH, the current results are less than optimal, and thus reduce the potential ethical dilemma between cardiac xenografts and mechanical support.

Table 2.

Potential conditions for which initial clinical trials of pig heart xenotransplantation may be justified*

Retransplantation Given the scarce resources and lower graft survival after retransplantation, some centers have preferentially directed organs to primary recipients. Xenotransplantation may be able to overcome this ethical predicament by providing an unlimited supply of organs to patients. High sensitization to HLA The presence of antibodies against HLA is associated with worse outcomes in heart transplant recipients, and is particularly relevant to retransplantation candidates as well as those with a prior VAD implantation. Early data suggest that the use of a xenograft does not increase the recipient’s degree of allosensitization. Complex anatomic or pathologic considerations preventing VAD implantation The presence of ventricular thrombus, mechanical prosthetic valves, postinfarction ventricular septal defect82 or the inability to ensure secure and hemostatic apical cannulation (either due to myocardial necrosis, expected ventricular perforation, infiltrative cardiomyopathy, or ischemic myocardial contracture) may limit the use of these devices. Restrictive cardiomyopathy In addition to a greater incidence of persistent right heart dysfunction, the use of VADs remains problematic in patients with restrictive cardiomyopathy due to reduced ventricular filling secondary to anatomical constraints of an undersized, thickened ventricle.83–85 As a result, heart transplantation currently remains the only therapeutic option.85,86 Size-mismatch between donor and recipient Substantial size-mismatch (beyond the 20–30% BMI ratio) can be a predictor of primary graft failure. These constraints further limit the availability of suitable organs. Xenotransplantation potentially offers the unique capability of selecting a graft that is appropriately size-matched to a particular recipient.

A growing number of patients with heart transplants are succumbing to chronic allograft dysfunction,³² and may ultimately be candidates for retransplantation. Allocating a second allograft reduces the number of available organs for primary transplant candidates and extends their time on the waiting list. Furthermore, retransplantation is associated with lower graft survival, particularly in patients with acute graft failure.^33,34 In contrast, xenotransplantation is not constrained by rationing concerns.

Despite successful palliative procedures, some patients with complex congenital heart disease remain vulnerable to late myocardial dysfunction, which is a major cause of death.^35,36 Adult patients with congenital heart disease are more likely to be designated with a lower UNOS urgency for heart transplantation, and may subsequently die on the waiting list.³⁶

Patients with single-ventricle physiology following the Fontan procedure, which encompass up to 80% of transplants for congenital heart disease, pose unique challenges both from the standpoint of significant structural and physiologic anomalies and their history of multiple previous operative procedures.³⁷ VAD implantation is frequently unfeasible or unsuitable in patients with Fontan failure where the development of complications, such as protein-losing enteropathy and desaturation secondary to venovenous collateralization, occurs in the setting of normal systolic function.³⁶ Xenotransplantation could be considered where the patient is not a candidate for VAD support.

Other pathologic conditions that may preclude VAD implantation are listed in Table 2. A TAH might be indicated for several of these conditions, but xenotransplantation is likely to offer a competitive alternative. Approximately 70% of patients with a TAH are successfully ‘bridged’ to cardiac allotransplantation, though the mortality on the device is >25%.^38,39 Systemic infection (incidence of 50%), driveline infection (25%), thromboembolic/hemorrhagic events (30%), and device malfunction ⁴⁰ remain major complications. Furthermore, the weight of the externalized device may be prohibitive for those with limited mobility and functional loss.

Obtaining an appropriate size-matched donor heart, particularly in the pediatric population, is important as a significant mismatch can be a predictor of primary graft failure.^41,42 Ideal donor/recipient body mass index (BMI) ratios within 20–30% are the generally accepted criteria.⁴³ The selection of a pig graft that is appropriately size-matched to a particular recipient, eg, one who is rapidly deteriorating, may thereby conserve the supply of allografts and potentially optimize posttransplant outcome. However, significant growth of pig kidneys within the first month after transplantation in nonhuman primates has been reported, although this may be associated in some cases with partial ureteric obstruction and the development of hydronephrosis.^7,44 However, rapid growth has not been reported after pig heterotopic heart transplantation in nonhuman primates^5,6,45 Infants with complex congenital heart disease might therefore prove candidates for pig heart transplantation.

Liver Xenotransplantation

There has been only one clinical case of pig liver transplantation,⁴⁶ but considerable experience with ex vivo pig liver perfusion as a bridge to allotransplantation.⁴⁷ In most cases, the immunologic challenges, including hyperacute rejection, resulted in only brief pig liver function, but the use of livers from genetically-engineered pigs has improved outcomes.^2,9

The Model for End-stage Liver Disease (MELD) scoring system is the basis for liver allocation in the USA, with the highest scores receiving offers of livers from deceased subjects based on local/regional allocation schemes. Patients with acute liver failure are given priority ahead of the MELD allocation system (recently updated – OPTN Policy 9.1: 01/21/16) with designation as Status 1. Nevertheless, overall, end-stage liver disease patients only have a 50% chance of receiving a transplant, with >1,500 patients on the UNOS liver waiting list dying each year, and a further >1,000 being removed from the waiting list because they have become too sick for transplantation.⁴⁸

Surprisingly, in a single-center analysis (between 2005 and 2012) a significant number of these patients had a low MELD score (≤22) at the time of listing.⁴⁹ In fact, 275 of 893 patients in this category died or were delisted as being too sick to undergo liver transplantation, while 245 patients underwent cadaveric donor liver transplantation.⁴⁹ If the risk factors for mortality or deterioration in these patients could be identified, and no living donor allograft is available, these patients might be candidates for liver xenotransplantation.

Currently, approximately 1,500 patients develop fulminant liver failure each year, of whom <20% receive a liver transplant. Recovery with medical therapy occurs in approximately 45%, and death without transplantation in the majority of the remainder. ^48,50 Successful bridging with a pig liver may, therefore, be life-saving (Table 3).

Table 3.

Potential conditions for which initial clinical trials of pig liver xenotransplantation may be justified*

Acute/fulminant liver failure Survival until either recovery of the native liver or until a suitable allograft can be obtained may be achieved by ex vivo pig liver perfusion (or auxiliary pig liver transplantation) Primary liver allograft failure or acute decompensation of chronic liver disease These conditions may warrant consideration for destination therapy by orthotopic pig liver transplantation. Ekser et al suggested that suitable patients might include those with (i) primary liver allograft failure as the result of primary nonfunction (incidence 5%) or hepatic artery thrombosis (incidence 3%) within 7 days of allotransplantation,50 or (ii) acute decompensation of chronic liver disease caused by variceal hemorrhage secondary to longstanding portal hypertension. However, these patients are almost always allocated an allograft as an emergency and so liver xenotransplantation, although a potential alternative, would be considered controversial by many.

Pig liver xenotransplantation is likely to be first introduced into the clinic as a form of ex vivo temporary support of patients who are likely to die before a suitable deceased human donor liver becomes available or who would otherwise die because they are not acceptable as candidates for liver allotransplantation. Adequate temporary support may provide time for sufficient recovery of native liver function to enable a further period of life or to bridge to allotransplantation.

Presently, there is not an effective mechanical support system (comparable to dialysis, VADs, or ECMO) for patients with either acute or chronic liver disease. The molecular adsorbent recirculating system (MARS), which is US FDA approved for acute toxic injury to the liver and refractory encephalopathy, has not been applied on any significant scale and is not likely to provide a level of mechanical support comparable to options for other organs. Thus, there is a strong and justifiable rationale for liver xenografts to be utilized as biologic support for patients with life-threatening complications of liver failure.

Patients with acute liver failure from acetaminophen toxicity, and other less common agents, may have a high likelihood of recovery, but significant periods of liver dysfunction exist, and these patients may benefit from pig liver support during the critical period of native liver recovery. This would not commit the medical team or patient to a liver transplant, nor would it require significant immunosuppressive therapy. It would take the form of daily ex vivo blood perfusion through an isolated genetically-engineered pig liver. A theoretical alternative in those with prolonged, but potentially reversible, liver dysfunction would be temporary auxiliary (heterotopic) liver xenotransplantation, which would provide support until the native liver had recovered, at which time the pig liver could be excised.^51,52

Destination therapy by pig liver xenotransplantation would probably only be considered if initial trials of ex vivo pig liver support are associated with results that suggest moderately long-term graft survival can be anticipated (Table 3).

The question can reasonably be asked as to whether the pig liver will provide sufficient metabolic support for the patient. Data in this respect are limited, but current evidence supports the likelihood that relatively good function can be obtained by a pig liver graft, including parameters of coagulation.⁵³ The current major problem is the rapid development of thrombocytopenia,⁵⁴ which will need a solution by further genetic engineering of the source pig^3,55 before any clinical trial can be contemplated.

Lung Xenotransplantation

Progress in experimental pig lung transplantation in nonhuman primates has been limited by numerous problems not seen after pig kidney or heart xenotransplantation. ^9,11 Considerable further progress will be required before a clinical trial will be seriously contemplated. Nevertheless, we believe it is not too early to begin to consider what form a clinical trial would take. Initial clinical trials might include (i) patients who require prolonged extracorporeal membrane oxygenator (ECMO) support, (ii) patients with chronic rejection (bronchiolitis obliterans syndrome), and (iii) those requiring combined transplantation of the lung and another organ (Table 4).

Table 4.

Potential conditions for which initial clinical trials of pig lung xenotransplantation may be justified*

Prolonged ECMO support ECMO is a potentially life-saving therapy for patients with respiratory failure,87–89 and has been successfully used as a bridge to lung allotransplantation in highly-selected patients.87,88,90,91 However, in most series posttransplant survival is inversely correlated to duration of ECMO support 87,88 and timely access to successful lung xenografts might substantially shorten the duration of ECMO support and improve outcomes. Patients with chronic rejection of an allograft (bronchiolitis obliterans syndrome) Bronchiolitis obliterans syndrome currently accounts for between 5–10% of lung transplants,92,93 and is associated with a high incidence of sensitization to donor-specific HLA; less often, antibodies against a broader HLA panel of potential donors are observed, presumably consequent to other sensitizing events or ‘epitope spreading’. Identifying a ‘negative cross-match’ against prospective human donors becomes increasingly difficult with increasing degrees of allosensitization. Consequently, lung xenotransplantation may first be offered to patients with deteriorating lung allograft function who have high titers of antibody reactive to a high proportion of the donor HLA panel. Patients requiring combined transplantation of the lung and another organ Simultaneous access to a kidney graft from the same source pig could address the currently unmet needs of the growing population of lung recipients with failing allografts whose renal dysfunction is caused or aggravated by chronic exposure to nephrotoxic immunosuppressive drugs. Patients requiring replacement of both the heart and the lungs (en-bloc heart-lung transplantation) are disadvantaged by current organ allocation algorithms, and are statistically more likely to die while waiting than to receive a transplant.94 This problem is amplified for patients who have been broadly sensitized to HLA alloantigens, eg, following prior repair of congenital cardiac anomalies. Because these patients are not usually chronically immunosuppressed, chronic renal insufficiency and infectious risks are likely to be less problematic relative to lung allograft recipients suffering from chronic rejection. As such, patients listed for heart-lung allotransplantation might also be considered for heart-lung xenotransplantation, which would provide information regarding the short- and long-term performance of both organs simultaneously.

Discussion

Because of the availability of dialysis, VADs, TAHs, and ECMO, selection of patients for the initial clinical trials of organ xenotransplantation is more complex than might be anticipated. The initial clinical trials cannot be carried out on patients in whom allotransplantation is contraindicated by such problems as chronic infection or malignancy, which would likely be associated with a very poor outcome, as this would not be a fair trial of the potential of xenotransplantation nor help the patient. Furthermore, the intensity of the immunosuppressive therapy that will be required (or that has already been administered to a patient awaiting retransplantation) is a factor that will need to be considered in the suitability of the patient for a xenograft. A thorough psychological assessment of the patient will also be very important.

Initial clinical trials of any pig organ will obviously require consistent demonstration of sustained life-supporting xenograft function in a preclinical model, as affirmed by convened expert international panels.^56,57 Ultimately, we believe that, with the appropriate genetic modifications, pig organ transplantation will be offered to all patients with terminal organ failure using clinically-applicable immunosuppressive therapy. With the current rate of progress in pig-to-nonhuman primate kidney and heart transplantation, we anticipate that clinical trials will certainly be undertaken within the next 5 years.

Nevertheless, several topics need to be clarified. These include the issue of obtaining truly informed consent for what is a novel form of therapy of which most patients and their families will have little knowledge. In particular, the ethical dilemma exists of how a critical or comatose patient would consent for liver xenograft support. Preemptive consent prior to decompensation has been suggested,⁵⁰ although emergent consent would not be very different from that made by the legal representatives of patients in clinical trials of trauma and resuscitation. Other ethical considerations have been discussed previously.¹² Furthermore, Institutional Review Boards will need education in the potential benefits and possible risks of xenotransplantation.

Although specific topics have been discussed in the past,^50,58 we are unaware of any recent reports in which the selection of patients for initial clinical trials of pig organ xenotransplantation has been discussed, and so we offer our initial deliberations as a basis for comment and discussion.

Abbreviations

cPRA

calculated panel-reactive antibodies

ECMO

extracorporeal membrane oxygenator

ESRD

end-stage renal disease

FDA

US Food and Drug Administration

HLA

human leukocyte antigen

KAS

kidney allocation system

MELD

Model for End-stage Liver Disease

MPGN

membranoproliferative glomerulonephritis

TAH

total artificial heart

UNOS

United Network for Organ Sharing

VAD

ventricular assist device