http://www.jhltonline.org

1053-2498/$ - see fronhttp://dx.doi.org/10.10

E-mail address: gu

Reprint requests: Gtion Surgery, KarolinskTelephone þ48-8-585

INNOVATION

Intentional ABO-incompatible heart transplantation:A case report of 2 adult patients

Gunnar Tyd�en, MD, PhD,a Inger Hagerman, MD, PhD,b Karl-Henrik Grinnemo, MD, PhD,c

Peter Svenarud, MD, PhD,c Jan van der Linden, MD, PhD,c Gunilla Kumlien, MD, PhD,d andAnnika Wernerson, MD, PhDe

From the aDepartment of Transplantation Surgery, Karolinska University Hospital, Stockholm, Sweden; bDepartment of Cardiology,Karolinska University Hospital, Stockholm, Sweden; cDepartment of Cardiothoracic Surgery and Anaestesiology, Karolinska UniversityHospital, Stockholm, Sweden; dDepartment of Transfusion Medicine, Karolinska University Hospital, Stockholm, Sweden; and theeDepartment of Pathology, Karolinska University Hospital, Stockholm, Sweden.

KEY WORDS:ABO-incompatible;heart transplantation;outcomes;antigen-specificadsorption;immunoglobulin titers

t matter r 2012 I16/j.healun.2012.0

nnar.tyden@karoli

unnar Tyd�en, MDa University Hosp

-80000. Fax: þ46-

BACKGROUND: In 2009, we started to screen all patients on the heart transplant waiting list for thepresence of blood group anti-A or anti-B antibodies. From our experience with ABO-incompatiblekidney transplantation, we know that transplantation can safely be performed if the antibody level isreduced to a titer of immunoglobulin G (IgG) 1:8.METHODS: We decided to accept all patients with anti-A or anti-B antibody titer r 1:8 for ABO-incompatible heart transplantation without any special pre-treatment and patients with antibody titers ofIgG 1:16 and 1:32, provided 1 apheresis session could be performed immediately before transplantation.RESULTS: We found 6 of 13 patients were suitable for this program, and 2 ABO incompatible patientsunderwent successful transplantation with a follow-up of 1 year.CONCLUSION: In heart transplant candidates where there are problems obtaining a compatible heart andwho are not suitable for ventricular assist device support, ABO-incompatible heart transplantations canbe considered using our protocol, provided that the levels of anti-A or anti-B antibodies are low.J Heart Lung Transplant 2012; 31: 1307–10r 2012 International Society for Heart and Lung Transplantation. All rights reserved.

Traditionally, transplantation of hearts from ABO-incompatible donors is contraindicated because of the riskof hyper-acute rejection mediated by pre-formed antibodiesin the recipient of blood-group antigens of the donor.However, this contraindication does not apply if therecipient’s anti-ABO antibodies are removed before trans-plantation and if the rebound of antibodies is preventedpost-operatively. Lack of organs suitable for transplantationand thereby associated morbidity and death on the waitinglist have increased the interest for ABO-incompatible organ

nternational Society for Heart and Lung9.011

nska.se

, PhD, Department of Transplanta-ital, S-14186 Stockholm, Sweden.8-774-3191.

transplant, with improved results in living-donor kidneytransplantation, liver transplantation, and pediatric hearttransplantation.1

In 2001, we introduced a new protocol enabling safe ABO-incompatible kidney transplantation.2 The new protocol usesantigen-specific immunoadsorption (GlycoSorb, GlycorexTransplantation AB, Lund, Sweden), rather than plasmaexchange to remove existing anti-A or anti-B antibodies, andrituximab, rather than splenectomy, to prevent the rebound ofantibodies in combination with conventional tacrolimus,mycophenolate mofetil, and prednisolone immunosuppres-sion. We have used this protocol to successfully perform morethan 60 ABO-incompatible renal transplantations and reportedresults in graft survival equal to those in ABO-compatibleliving-donor renal transplant.3

Transplantation. All rights reserved.

The Journal of Heart and Lung Transplantation, Vol 31, No 12, December 20121308

An important lesson learned from our experience withABO-incompatible transplantations is that many patientshave quite low anti-A or anti-B antibody titers and thereforedo not necessarily need several preoperative apheresissessions. Instead, pre-operative and post-operative intensivetreatment suffices. For that cohort of patients where a livingdonor not is possible, the protocol would be useful also forABO-incompatible heart transplantations. Not every patientwith severe heart failure is suitable for ventricular assistdevice support, and the ability to perform heart transplanta-tion across blood-group barriers would not only decreasetime on the waiting list but also increase optimal use ofevery possible donor.

Methods

The protocol for ABO incompatible organ transplant has beenapproved by the Karolinska University Hospital Ethical Commit-tee. All eligible patients were informed about the treatmentprotocol and potential complications of an ABO-incompatibletransplantation.

In 2009, we started to screen all patients on the heart waiting listfor the presence of anti-A or anti-B immunoglobulin (Ig) M andIgG antibodies. We learned from our experience with ABO-incompatible kidney transplantations that that transplantation canbe safely performed if the antibody level is reduced to a titer of IgG1:8. We therefore decided to accept all patients with an equivalentor lower anti-A or anti-B titer for incompatible heart transplanta-tion without any special pre-treatment. Because 1 apheresis session(antigen-specific adsorption or plasma exchange) processing a1.5 plasma volume will reduce the antibody titer by 2 titer steps,4

we decided to also accept patients with antibody titers of IgG 1:16and 1:32, provided 1 apheresis session could be performedimmediately before transplantation.

Antibody titers of IgM and IgG were quantified with gel hemagglutination technique5 postoperatively, and preemptive apheresissessions with antigen-specific adsorption were performed everythird day for 9 days according to our protocol for ABO-incompatible kidney transplantation. Also, according to theprotocol, all patients received 1 dose of rituximab (375 mg/m2)and 1 dose of intravenous Ig (0.5 g/kg) immediately beforetransplantation. Induction was made with anti-thymocyte globulinaccording to our standard protocol for heart transplantation.

Post-operatively, the patients were maintained on a standardtriple-drug immunosuppressive protocol of tacrolimus with targettrough level of 4 to 12 mg/ml, mycophenolate mofetil with areaunder the curve 95–190 mmol/liter � h and prednisolone indecreasing doses to 5 mg/day at 6 months. Cytomegalovirus,toxoplasmosis, and pneumocystis prophylactic treatment weregiven according to routine protocol. IgM and IgG titers of anti-A oranti-B antibodies were monitored closely during the first monthand less frequently thereafter. Endomyocardial biopsy specimenswere taken according to a standard protocol of once weekly for thefirst month, every second week for Months 2 to 3, every fourthweek in Months 4 to 6, and every 8 weeks during Months 7 to 12,or earlier, if clinically indicated.

Results

When we screened patients on the waiting list for hearttransplantation, 6 of 13 patients (46%) had anti-A or anti- Bantibody levels low enough for them to be included in this

protocol, and 2 of those successfully received transplants.The remaining 4 patients received ABO compatible hearts,which were the first choice.

Patient 1

The first patient was 46-year-old man with severe heartfailure due to previous myocarditis. He was otherwisehealthy and had been receiving optimal heart failuretreatment for more than 1 year but remained in New YorkHeart Association (NYHA) class IIIb and at the time wasreceiving intermittent inotropic support. He was negative incross-test and without human leukocyte antigen antibodies.The transplantation, which took place after 1 month on thewaiting list, was an AB to B combination. The pre-operativeantibody titer against AB test erythrocytes was IgM 1:4 andIgG 1:4. The titers remained unchanged postoperatively inrepeated controls according to the protocol, and the clinicalcourse during the patient’s first year was uneventful, exceptfor a cytomegalovirus infection and recurrent gingival viralinfection.

Two of 17 endomyocardial biopsy specimens during thefirst year showed signs of moderate cellular rejection, withgrade 2R, and were resolved with intravenous Solu-Medrol(0.5 g; Pfizer, New York, NY). The rejection episodes werediagnosed within the first 3 months, and no signs of cellularrejection were found thereafter. Cardiac function remainednormal on every echocardiography performed after allbiopsies, and there were no clinical symptoms indicatingrejection. Staining for C4d was positive in all biopsyspecimens during the first 5 months, without morphologicsigns of antibody-mediated rejection. In the remainingspecimens, C4d staining was negative. There were no signsof hemolysis during the first year and after the firstpostoperative period, and the hemoglobin levels were stableat �120 g/liter.

Patient 2

The second patient was a 67-year-old man with severe heartfailure due to a previous myocardial infarction and was inNHYA class IIIa. Besides optimal heart failure medication,he was receiving intermittent inotropic support and had beenon the waiting list for 20 months. He was negative in cross-test and without human leukocyte antigen antibodies. Thetransplantation was an A1 to 0 combination, and the pre-operative antibody titer against A1 test erythrocytes wasIgM 1:4 and IgG 1:8. The titer successively increased aftertransplantation to a maximum of IgG 1:32. After 2 apheresissessions (1 antigen-specific adsorption and 1 plasmapher-esis) the antibody titer decreased to a stable IgG 1:8 level.

Signs of cellular rejection (grade 2R) developed inspecimens of 5 of 19 performed endomyocardial biopsiesduring the first year. The rejections were diagnosed withinthe first 6 months and were resolved with 3 intravenouspulses Solu-Medrol (0.5 g). There were never any signs ofclinical impairment or depressed cardiac function onechocardiography. The protocol biopsy specimens stained

Figure 1 (Top left) Cellular infiltrations at 4 weeks after ABO-incompatible heart transplantation. (Top right) Diffuse positive C4dstaining in capillaries at 4 weeks after ABO-incompatible heart transplantation. (Bottom left) Normal myocardial biopsy specimen is seen1 year after ABO-incompatible heart transplantation. (Bottom right) Staining for C4d is negative at 1 year after ABO-incompatible hearttransplantation.

Tyd�en et al. ABO-Incompatible Heart Transplantation 1309

diffusely positive for C4d, without any morphologic signs ofantibody-mediated rejection, and the positive stainingtended to be weaker over time (Fig 1). As in patient 1 therewere no signs of hemolysis, and the hemoglobin levels werestable at �110 g/liter. The patient’s clinical course duringthe first year was uncomplicated, except for a cytomegalo-virus infection and norovirus gastroenteritis 8 months aftertransplant.

At follow-up at 12 months, both patients were in NYHAclass I with normal cardiac function on echocardiographyand coronary angiograms, and exercise performance resultswere within normal reference ranges. Endomyocardialbiopsy specimens at 12 months were considered normal(grade 0R, AMR0) in both patients, and there were norebound increases in anti-A or anti-B antibody levels.

Discussion

Previous reports on ABO-incompatible heart transplanta-tions have mostly been on unintentional cases. Thus,Cooper6 described 8 patients, of which 5 transplants werehyper-acutely rejected and only 1 patient survived for morethan 1 year. However, many successful ABO-incompatibleheart transplants have been reported in infants,7 and a recentreport from the pediatric heart transplant study showedequivalent 1-year survival and freedom from rejectioncompared with ABO-compatible transplantation.8 Theexplanation for the favorable outcome with ABO-incompatible heart transplants in infants is that childrenaged up to approximately 1 year do not have any A or Bantibodies and therefore do not need any special treatment.

An intentional ABO-incompatible heart transplantationwas recently reported as an emergency procedure in a

desperate case.9 Plasmapheresis was performed pre-operatively and post-operatively, and the patient was treatedaccording to a protocol very much based on the protocolused in our patients.

In the present report, we describe the full implementationof an ABO-incompatible program in heart transplantationbased on acceptance of certain anti-A or anti-B antibodylevels in patients on the waiting list. In accordance with ourprevious knowledge from ABO-incompatible renal trans-plantation, this protocol seems to have the potential of asuccessful and safe procedure. That any blood group can beaccepted, theoretically increases by 25% the chance for eachpatient to receive a transplant, calculated from the distribu-tion in the population of the different blood groups. Thiswould imply a substantial reduction in time on the waitinglist and optimal use of every possible donor.

An interesting finding was that the post-operative biopsyspecimens from both patients stained positive for C4d. Asstated by Fedrigo et al,10 histologic parameters of earlyantibody-mediated rejection, for instance signs of endothe-lial swelling and interstitial edema, should be combinedwith C4d staining, which is a more sensitive test. PositiveC4d staining has previously been described in ABO-incompatible renal graft recipients,11 and the phenomenonis probably explained by a continuous low-grade degrada-tion of A or B antibodies within the organ that seems toevoke no harm as long as the antibody titer levels are equalto or below a cutoff level of IgG 1:8. This is in accordancewith the outcome of our 2 patients, who remained clinicallystable with normal cardiac function despite positive stainingfor C4d in several endomyocardial biopsy specimens. Bothpatients were negative in histologic parameters, and thepositive C4d staining diminished over time and remained

The Journal of Heart and Lung Transplantation, Vol 31, No 12, December 20121310

negative at follow-up after 1 year. Deposition of C4d incardiac transplants has been associated with the develop-ment of coronary artery vasculopathy and poor out-come.12–14 Therefore, long-term follow-up of the patientsis extremely important. With ABO-compatible transplanta-tion as the first choice and ABO-incompatible transplanta-tion as the second choice in highly selected patientsaccording to our strategy, an optimal match of everyavailable organ may be achieved. We believe this would beethically justified and result in no harm to patients indifferent blood groups who are waiting for an organ.

We conclude that in heart transplant candidates wherethere are problems obtaining a compatible heart and who notsuitable for ventricular assist device support, ABO-incompatible heart transplantations can be considered usingour protocol, provided that the levels of anti-A or anti-Bantibodies are low. Pre-operative antibody levels maychange due to infections and other events and shouldtherefore be monitored regularly during waiting time.

Disclosure statement

None of the authors has a financial relationship with a commercialentity that has an interest in the subject of the presented manuscriptor other conflicts of interest to disclose.

References

1. Warner PR, Nester TA. ABO-incompatible solid-organ transplantation.Am J ClinPathol 2006:125(Suppl):S87-94.

2. Tyd�en G, Kumlien G, Genberg H, et al. ABO-incompatible kidneytransplantation without splenectomy, using antigen specific immunad-sorption and rituximab. Am J Transplant 2005;5:145-8.

3. Tyd�en G, Donauer J, Wadstrom J, et al. Implementation of a protocolfor ABO-incompatible kidney transplantation—a three-center-experience with 60 consecutive transplantations. Transplantations2007;83:1153-5.

4. Kumlien G, Ullstrom L, Loswall A, Persson L-G, Tyd�en G. Clinicalexperience with a new apheresis filter that specifically depletes ABOblood group antibodies. Transfusion 2006;46:1568-75.

5. Kumlien G, Wilpert J, Safwenberg J, Tyd�en G. Comparing the tubeand gel techniques for ABO antibody titration, as performed in threeEuropean Centers. Transplantation 2007;84:S17-9.

6. Cooper D. A clinical survey of cardiac transplantation between ABOblood group incompatible recipients and donors. Transpl Proc1990;22:1457.

7. West L, Pollock-Barziv S, Dipchand A, et al. ABO-incompatible hearttransplantation in infants. N Engl J Med 2001;344:793-800.

8. Hendersson H, Canter C, Mahle W, et al. ABO-incompatible hearttransplantation: analysis of the Pediatric Heart Transplant Study(PHTS) database. J Heart Lung Transplant 2012;31:173-9.

9. Bucin D, Johansson S, Malm T, et al. Heart transplantation across theantibodies against HLA and ABO. Transpl Int 2006;19:239-44.

10. Fedrigo M, Gambino A, Benazzi E, et al. Role of morphologicparameters on endomyocardial biopsy to detect sub-clinical antibody-mediated rejection in heart transplantation. J Heart Lung Transplant2011;30:1381-8.

11. Setoguchi K, Ishida H, Shimmura H, et al. Analysis of renal transplantbiopsies in ABO-incompatible kidney transplantation. Am J Transpl2008;8:86-94.

12. Fedrigo M, Gambino A, Tona F, et al. Can C4d Immunostaining onendomyocardial biopsies be considered a prognostic biomarker in hearttransplant recipients? Transplantation 2010;90:791-8.

13. Moseley EL, Atkinson C, Sharples LD, et al. Deposition of C4d andC3d in cardiac transplants: a factor in the development of coronaryartery vasculopathy. J Heart Lung Transplant 2010;29:417-23.

14. Wu GW, Kobashigawa JA, Fishbein MC, et al. Asymptomaticantibody-mediated rejection after heart transplantation predicts pooroutcomes. J Heart Lung Transplant 2009;28:417-22.