An unusual case of red blood cell immunisation following liver transplantation

Introduction

Antibodies to erythrocyte antigens can be observed in some recipients following haematopoietic stem cell or solid organ transplantation. This is most commonly due to a primary immune response after transfusion of red blood cells (RBC), although in some cases it may depend on the passenger lymphocyte syndrome in which donor’s viable B lymphocytes, transferred with the graft, become responsive to the recipient’s or transfused RBC antigens and produce antibodies to these antigens. Moreover a secondary immune response can be observed in individuals sensitised by previous transfusions or pregnancies and later re-exposed to the same foreign antigens^1–3.

Liver transplantation has been notoriously considered as a potential challenge for the transfusion service as the need for blood components may be substantial, although improvements in surgical techniques, anaesthesia and organ preservation have sharply reduced overall transfusion requirements and systems of intraoperative blood salvage and autologous blood transfusion have significantly reduced the need for allogeneic blood components.

Here we report an unusual case of red cell alloimmunisation, with production of two different RBC antibodies, following liver transplantation in a patient transfused exclusively with one apheresis platelet concentrate during transplantation.

Case report

A 44-year-old woman with autoimmune hepatitis underwent liver transplantation with a liver graft from a 67-year-old, brain dead, ABO and RhD type identical donor.

The liver was perfused in situ through the aorta with 4 L of histidine-tryptophan-ketoglutarate (Custodiol, Fresenius, Koehler Chemie GmbH, Alsback-Hähnlein, Germany); on the back table it was flushed through the portal vein with an additional 0.5 L. The liver was transplanted using the piggy-back technique after 7 hours of cold ischaemia⁴. The patient received 1.8 L of autologous blood, which was collected using an intraoperative blood salvage system (Cell Saver^®, Haemonetics, Braintree, MA, USA), and one unit of apheresis platelets. The immunosuppressive regimen consisted of basiliximab induction (Simulect^®, Novartis, Basel, Switzerland) followed by tacrolimus (trough levels 5–8 ng/mL), mycophenolate mofetil 1 g b.i.d. and prednisolone. The post-operative course was uneventful, the patient did not require further transfusions and was discharged on post-operative day 16. A summary of her liver function tests and haemoglobin during the admission is shown in Figure 1.

Figure 1.

The time course of haemoglobin and transaminases during the admission for liver transplantation.

Dotted line: aspartate aminotransferase (AST, normal value <0.75 microkatal/L); broken line: alanine aminotransferase (ALT, normal value <0.60 microkatal/L); continuous line: haemoglobin.

Preop: preoperative; PTD: post-transplant day.

Before the transplantation both blood group typing and antibody screening were performed using an automated instrument, IH-1000 (Bio-Rad Laboratories, DiaMed GmbH, Cressier, Switzerland), at our emergency laboratory. The patient’s blood group was O RhD positive and the antibody screening with the indirect antiglobulin test (IAT), using a Coombs anti-IgG gel card (Bio-Rad Laboratories, DiaMed GmbH) and two validated in-house test cells, was negative. Type and screen was performed as a compatibility test on another sample, using a DiaClon BAS-test IgG gel card (Bio-Rad Laboratories, DiaMed GmbH) to recheck the ABO typing and four in-house test cells. No RBC antibodies were detected.

Four months later the patient returned for a follow up during which packed RBC units were reserved. The gel card type and screen was unexpectedly positive and an investigation was initiated.

The blood group was controlled and an antibody screening in solid phase with Capture R Ready Screen 3 (Immucor Inc., Atlanta, GA, USA) was performed using an automated instrument, NEO (Immucor, Rödemark, Germany), according to standard procedures in our laboratory for routine blood group typing. The test confirmed the presence of RBC antibodies.

The antibody identification clearly showed two specificities: the first one was anti-E reactive in gel IAT, but not in polyethylene glycol (PEG) IAT and the second one was anti-Jk^b reactive in all antigen-positive cells in PEG IAT, and just weakly with some but not all cells expressing double dose (homozygous) Jk^b antigen in gel IAT. The direct antiglobulin test (DAT) was negative. The patient’s phenotype was E−, Jk^b−.

Further investigations were performed to understand the origin of the immunisation. The transfused apheresis platelet concentrate, collected by a Trima Accel cell separator (TerumoBCT, Lakewood, CO, USA), came from a donor whose RBC phenotype was registered in the computer system and known to be E+, Jk^b+ (the extended phenotype is shown in Table I).

Table I.

The extended blood types of the platelet donor and the liver donor.

	ABO/RhD type	Other RBC antigens Phenotype/predicted phenotype
Apheresis platelet donor	O RhD+	ccDEe, Jk(a+b+), Fy(a+b+), S+, s+, K−, Lea−
Liver donor	O RhD+	Ee, Cw−, Jk(a−b+), Fy(a+ b+), K−, k+, Kp(a–b+), Js(a−b+), M+, N+, S−, s+, U+, Di(a−b+), Do(a+b+), Hy+, Joa+, Co(a+b−), Yt(a+b−)

The platelet donor’s RBC antigens were determined by serology (phenotype); the liver donor’s RBC antigens were determined by retrospective genetic analysis of previously frozen DNA (predicted phenotype).

RBC: red blood cells.

A retrospective investigation on both the patient’s and the liver donor’s serological status was started. A plasma sample from the patient, frozen prior to transplantation, was analysed with Capture-R Ready Screen 3 as a double check to exclude that the negative result previously obtained in gel IAT was dependent on a method-related false negative reaction. We confirmed that no RBC antibodies were detectable in the patient at the time of the transplantation on the basis of the standard preoperative investigation programme. A previously frozen plasma sample from the liver donor was also tested with Capture- R Ready Screen 3 for the presence of RBC antibodies with a negative result.

Red blood cell genotyping was performed on the DNA from the liver donor, purified before the transplantation using magnetic particle technology (EZ1 DNA Blood 200 μL Kit; Qiagen, Hilden, Germany) and a BioRobot EZ1 Qiagen (1–6 samples per run) instrument and stored a −20 °C. The sample was analysed by multiplex polymerase chain reaction using BLOODChip ID Core+ (Progenika Biopharma, Derio, Spain) and XMAP Luminex^® technology (Luminex Corporation, Austin, TX, USA). The predicted phenotype was E+, Jk^b+ (the extended blood type is shown in Table I). A summary of the immunohaematological investigations is presented in Table II.

Table II.

Summary of the immunohaematological investigations performed on the patient’s and liver donor’s samples.

Patient
Pre-transplant investigations	Blood group typing RBC antibody screening (2-cell screen gel card)	O RhD+ No RBC antibodies detected
	Type and screen (ABO control + 4-cell screen gel card)	No RBC antibodies detected
Post-transplant: 4 months after transplantation	Type and screen (ABO control + 4-cell screen gel card)	RBC antibodies detected
Post-transplant: further investigations of samples taken 4 months after transplantation	Blood group typing Antibody screening (3-cell screen solid phase)	O RhD+ RBC antibodies detected
	Antibody identification	Anti-E + anti-Jkb
	DAT	Negative
	Phenotype	CcDee, Jk(a+b−)
Post-transplant: retrospective investigations of plasma samples frozen before transplantation	RBC antibody screening (3-cell screen solid phase)	No RBC antibodies detected
Liver donor
Retrospective investigations of plasma and DNA samples frozen before transplantation	Antibody screening (3-cell screen solid phase)	No antibodies detected
	RBC genotyping predicted phenotype	E+, Jkb+

RBC: red blood cells; DAT: direct antiglobulin test.

Briefly, the results showed that both the platelet donor and the liver donor were positive for the red cell antigens the patient had antibodies to.

Since no samples from the recipient had been tested for RBC antibodies in the period immediately following the transplantation, it could not be demonstrated whether this was a primary immunisation or a secondary immune response in a patient with a history of pregnancies, although without previously known alloimmunisation.

Discussion

Although the RBC content in apheresis platelet units is notoriously very low, especially if compared with platelet concentrates obtained by pooling buffy coats, it has been suggested that residual RBC and RBC- derived microparticles with high immunogenicity could be sufficient to initiate a primary immune response or to boost a secondary immune response.

Cases of anti-D, but also non-D Rh immunisation have been reported as a consequence of transfusion of platelets concentrates^5,6. Moncharmont et al. performed an extensive study comparing pooled platelet concentrates and apheresis platelet concentrates showing in both cases, although with different frequencies, the occurrence of RBC immunisation even with combinations of multiple antibody specificities⁷.

It is relatively rare that a liver transplant can be performed without requiring RBC concentrates and fresh-frozen plasma. Considering that in the case reported here, the only blood component transfused was an apheresis platelet concentrate, this case opens the way to speculations about the possible mechanisms and the dose of antigen necessary to trigger immunisation.

It is critical to underline that modern technologies for blood group genotyping play a key role in the retrospective investigation of a donor’s antigenic set up and can represent a fascinating opportunity for understanding the way by which red cell immunization may occur. In this case residual erythrocytes in the platelet concentrate and in the transplanted organ, alone or in combination might have stimulated the patient’s immune system.

It has not been possible to establish whether RBC antibodies had developed as a primary or as an anamnestic response in a patient immunised through previous pregnancies; the second possibility might be true in particular in the case of anti-Jk^b. It is known that antibodies in the Kidd blood group system can weaken and eventually disappear over time.

The effluent from the liver after the in-situ perfusion in the donor often shows macroscopic traces of blood. We tried to quantify the red blood cell concentration by testing a sample taken at the end of another similar liver procurement procedure and found that the haematocrit was approximately 1%. This suggests that the presence of residual red cells in the organ can trigger or boost an immune response in recipients after transplantation and that blood should be thoroughly washed out of the organ before transplantation.

Another aspect to consider is that some blood group antigens, such as Duffy and Kidd antigens, are expressed not only on erythrocytes, but also on other tissues and organs. According to Lerut et al. they might act as minor histocompatibility antigens. This could contribute to organ dysfunction after transplantation in patients with corresponding RBC alloantibodies⁸, showing the possible clinical relevance of red cell immunisation also in the transplantation setting.

In conclusion we believe this case is an interesting illustration of unusual post-transplant RBC immunisation occurring in an immunosuppressed patient; it suggests that despite advanced, high quality techniques for collecting apheresis platelet concentrates and for organ perfusion, even a very small amount of residual red cells can be clinically relevant in stimulating the immune system.