OUP user menu

Successful Unintentional ABO-Incompatible Renal Transplantation
Blood Group A1B Donor Into an A2B Recipient

Emmanuel A. Fadeyi MD, Robert J. Stratta MD, Alan C. Farney MD, Gregory J. Pomper MD
DOI: http://dx.doi.org/10.1309/AJCPBGW81HQWBOZO 724-726 First published online: 1 May 2014


Objectives To report a successful unintentional transplantation of a deceased donor kidney from an “incompatible” A1B donor into a recipient who was blood group A2B with unsuspected preformed anti-A1 antibodies.

Methods The donor and recipient were both typed for ABO antigens. The recipient was tested for ABO and non-ABO antibodies. The recipient was typed for HLA class I and class II antigens, including HLA antibody screen. The T-and B-flow cytometry crossmatch test was performed using standard protocol.

Results The donor-recipient pair was a complete six-antigen human leukocyte antigen mismatch, but final T- and B-flow cytometry cross-match tests were compatible. The recipient was a 65-year-old woman with a medical history of end-stage renal disease secondary to diabetic nephropathy who underwent kidney transplantation from a 46-year-old brain-dead standard criteria donor. The recipient’s RBCs were negative with A1 lectin, and the recipient was thus typed as an A2 subgroup. Anti-A1 could be demonstrated in the recipient’s plasma. The donor’s RBCs were positive with A1 lectin, thereby conferring an A1 blood type.

Conclusions It is safe to transplant across the A1/A2 blood group barrier provided that the preformed antibodies are not reactive at 37°C and with anti–human globulin.

Key Words:
  • ABO incompatible
  • HLA mismatch
  • Antibody-mediated rejection
  • Renal transplant
  • Immunology
  • Special topics
  • Transfusion medicine

Because of the critical shortage of donor kidneys in the United States, a number of innovative techniques are being implemented, such as intentional transplantation of ABO-incompatible (ABOi) living or deceased donor kidneys. Initial attempts at transplantation across the ABO blood group barrier were associated with high rates of early graft loss secondary to hyperacute or accelerated antibody-mediated rejection.1 In studies of ABOi subgroups, analysis has primarily focused on transplants involving blood type A donors, comparing subtypes A1 and A2. Transplantation across the “minor” A2 barrier (A2 to O, A2 to B, or A2B to B) is believed to be associated with a lower risk of adverse outcomes than those that cross the “major” A1 barrier.2 Additional safeguards have been adopted recently that specifically focus on ABO subgroup typing of donors in preparation for transplantation across the minor ABO barrier. An update to the Organ Procurement and Transplantation Network3 policy stipulates dual confirmation of the donor A or AB subtype prior to listing the patient for organ transplant. However, less information exists regarding a recipient who is an A2 or A2B subtype. Confounding the issue is the counterintuitive occurrence of anti-A1 presenting in the plasma of a small percentage of A2 or A2B subtype individuals. In the event a recipient has an A2 or A2B subtype, O-type organs are compatible, but A1-subtype organs are compatible only if the recipient does not produce anti-A1.4 If anti-A1 antibody is produced but does not react in vitro at 37°C or at body temperature, then the anti-A1 would not be expected to cause transplant rejection of an A1 blood group organ. In this instance, the patient may unintentionally receive an organ challenging the major ABO barrier. To our knowledge, there has been only one reported case of ABOi transplant from an A1B donor into an A2B recipient; however, that patient demonstrated no preformed anti-A1 antibodies.4 We present the unusual case of a successful outcome in an elderly patient who received a deceased kidney from an ABOi donor (A1B to A2B donor-recipient pair) with preformed anti-A1 antibodies that demonstrated no reactivity at 37°C or with anti–human globulin (AHG), and in addition, the patient had detectable cold autoantibodies.

Case Report

Pre- and Peritransplant Clinical Data

A 65-year-old African American woman with a history of chronic kidney disease secondary to type 2 diabetes mellitus and long-standing hypertension dating back to the 1980s progressed to end-stage renal disease, requiring hemodialysis. She was otherwise fairly healthy except for morbid obesity, a remote history of sarcoidosis, prior right hip replacement, anemia of chronic disease, dyslipidemia, and secondary hyperparathyroidism. Prior to transplant, she had undergone a cardiac catheterization that demonstrated normal coronary arteries. She underwent standard criteria deceased donor kidney transplantation and tolerated the procedure well. She experienced a gradual return of graft function and was discharged 1 week following transplant with a declining serum creatinine level (from 8.1 to 2.2 mg/dL). Neither postoperative dialysis nor administration of blood products was required. Immunosuppression consisted of rabbit anti–thymocyte globulin induction in combination with half-dose mycophenolate mofetil, tapered corticosteroids, and the delayed administration of tacrolimus, which was started on the fifth postoperative day.


Laboratory Findings

The donor and recipient were both typed as AB positive. The recipient’s RBCs were negative with A1 lectin (Dolichos biflorus), and anti-A1 antibody was present in her plasma. The antibody screen was negative, thus ruling out non-ABO antibodies. A cold panel tested with her plasma using a panel of RBCs demonstrated nonspecific cold autoantibodies reacting at 4°C only. Nonspecific cold autoantibodies reacting at 4°C or below are considered not clinically significant. The donor’s RBCs were positive with A1 lectin, confirming A1 RBC antigen specificity. The patient’s panel reactive antibody (PRA) level was 0% before transplant and has remained 0% with a recent PRA level at 0% six years following transplant. The donor and recipient were a complete six-antigen human leukocyte antigen (HLA) mismatch at both class I and class II, but the final T- and B-flow cytometry cross-match tests were compatible. The donor and recipient were both cytomegalovirus (CMV) positive.

Posttransplant Data

Seven years following transplant, there has been no evidence of rejection, hemolysis, de novo donor-specific antibody production, or other complications attributable to the unsuspected antibody “incompatibility.” She received erythropoiesis-stimulating agent therapy in the first 6 weeks after the transplant, but subsequently, her hemoglobin level stabilized in the 10- to 11-g/dL range. During 7 years of follow-up, she has been successfully treated for CMV viremia with retinitis and osteonecrosis of the left hip. She is currently on tacrolimus monotherapy and remains active and functional. The renal allograft has stable function with a serum creatinine level of 1.4 mg/dL and a steady-state hemoglobin level of 11 g/dL.


Clinically, the two most common blood type A subgroups encountered are A1 and A2. A1 represents most group A donors (80%), who possess approximately 1 million A antigen epitopes per RBC, whereas A2 donors are the second most common subgroup (20%), possessing one-fifth (2.2 × 105) the number of A antigen sites compared with A1.5 Therefore, the A1 subgroup tends to be more antigenic and, if transplanted to a blood type O or B recipient, will result in hyperacute or accelerated antibody-mediated rejection. However, A2 grafts transplanted into O- or B-type recipients have better outcomes because of the decreased number of A antigen epitopes. Anti-A1 occurs as an alloantibody in the serum of blood type A2 and A2B individuals in proportions of 1% to 8% and 22% to 35%, respectively.5 Anti-A1 is usually of the IgM isotype and reacts best at room temperature or below. However, anti-A1 can be considered clinically significant if it reacts at 37°C or with AHG.

In a previous study, Albornoz et al,4 who reported successful transplantation from an A1B donor into an A2B recipient, the recipient was tested and found to be negative for anti-A1 antibodies. In another study of ABOi transplants and three to five HLA mismatches in elderly patients older than 60 years, the results demonstrated successful outcomes with no severe complications or rejection.6 While all patients in this series were conditioned to receive ABO-incompatible organs, none were A2 and received an A1 graft. Our case is unique in that it demonstrates that successful kidney transplantation can be achieved across the A1/A2 blood group mismatch even with recipient preformed antibodies and without the need for prior desensitization. It is important to note that the anti-A1 antibody in our case did not react at 37°C or with AHG. According to the usual practice of kidney transplantation, one might expect that a patient with anti-A1 antibodies would be ineligible to receive an A1 organ. However, our case demonstrates that this finding may not be absolute across the A1/A2 blood group mismatch barrier.

Recent studies have questioned the mechanism of antibody-mediated rejection in ABOi kidney transplantation when comparing ABO blood antigens with ABO antigens found on vascular endothelial cells of the kidney, resulting in production of anti-A and or anti-B antibodies. ABO blood antigens are found on the surface of erythrocytes. For antibody-mediated rejection to occur, the recipient must be sensitized to the ABO antigens on the vascular endothelial cells of the graft, resulting in the new production of anti-A/anti-B de novo antibodies. Anti-ABO blood antibodies and anti-ABO antibodies formed as a result of the new graft presumably would be very similar in structure.7 The acceptance of the renal allograft could also be due to a phenomenon called accommodation, which is defined as the survival and function of a transplanted organ in the presence of donor antigens expressed on the endothelial cells of the graft despite antibodies that can bind to the donor antigens and a fully functional complement pathway in the recipient.8 This phenomenon of accommodation could have played a role in this case; however, the preformed antibodies demonstrated were most likely of the IgM isotype, which is usually clinically insignificant. Cross-match–incompatible patients appear to be at greater risk for antibody-mediated rejection compared with those receiving ABOi transplantation, particularly when looking at outcomes comparing both in several studies.9

The successful outcome reported here could lead to increased availability of A1 or A1B allografts being transplanted into A2 or A2B (or other A subgroup) recipients with preformed antibody without the need for pretransplant desensitization or concern for antibody-mediated rejection. In conclusion, we believe that it is safe to transplant across the A1/A2 blood group barrier in the setting of preformed antibodies and expect satisfactory results with acceptable graft survival, provided that the cross-match is compatible and the preformed antibodies are not reactive both at 37°C and with AHG.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
View Abstract