Int

Introduction: Multiply transfused patients are frequently subjected to platelet alloimmunization. These platelet alloantibodies produced can result in refractoriness to platelet transfusion. Material and Methods: Ninety five thrombocytopenic (platelet count <100 x 10⁹/l) and multilply transfused (more than 2 times transfusion of any blood component with last transfusion more than 2 weeks prior to the study) patients were recruited prospectively. The blood samples were tested using a Solid Phase system (Capture P). Results: There were 45 males (47.4%) and 50 females (52.6%) recruited with ages from 3 to 90 years. The frequency of transfusions ranged from 10-168. Seven patients (7.4%) were detected to have platelet alloantibodies, predominantly anti-HPA-5b in 4 patients (4.2%). Three of them (3.2%) showed a non-specific pattern. Six patients (6.3%) had received packed cells less than 20 units and another 1 (1.1%) received more than 20 units of packed cells. Four patients (4.2%) received platelet transfusion of less than 20 units and another 3 patients (3.2%) received more than 20 unit platelets. Conclusion: The study may have future implications for the selection of platelet donors for alloimmunized recipients in HUSM.

Alloimmunization is an immune response that happens when body reacts to foreign antigen (from donor’s blood) and creates antibodies against them. Patients receiving leucocytes reduced blood product are at much lower risk for refractoriness to platelet transfusion than are recipients of blood that is not leucocytes reduced (Pisciotto, 1993).

Alloimmunization in polytransfused patient is a well known observation which very often results in refractoriness to platelet transfusions. These refractory patients may become resistant to subsequent platelet transfusion thus increasing the risk of spontaneous bleeding (Pisciotto, 1993). Alloantibodies are usually directed against Human Leucocyte Antigen (HLA) and their frequency is evaluated between 20 to 70 %. Alloantibodies may also be directed against platelet-specific antigens but their frequency is controversial (Uhrynowska and Zupanska, 1996). Alloantibodies against platelet membrane glycoprotein are a common cause for febrile non hemolytic transfusion reactions (Bierling, Fromont and Bettaleb, 1989).

A patient with previous platelet refractoriness secondary to alloimmunization may need HLA-matched or crossmatched platelets for transfusion support (Kiefel et al, 2001).

The aim of this study was to determine the prevalence and type of human platelet specific antibodies seen in our local population.

This was a cross sectional study conducted at Hospital Universiti Sains Malaysia (HUSM) from June 2003 to June 2004.This study was approved by the School of Medical Sciences Research and Ethical Committee. All subjects gave an informed written consent

Subjects were recruited from wards, medical clinic and transfusion records from Transfusion Medicine Unit. Subjects with history of multiple transfusions (more than twice of any of blood components, more than 2 weeks but not more than 2 years of their last transfusions) and thrombocytopenia (platelet count less than 100 x 10⁹/l). Patients who were on heparin or with autoimmune diseases such as immune-thrombocytopenic purpura (ITP), systemic lupus erythematosus (SLE) or autoimmune haemolytic anaemia (AIHA) were excluded. Selected patients were given informed written consent and fresh sample from the patients were taken. Seven mls of peripheral blood were taken and collected in EDTA bottles. Two mls of the peripheral blood for full blood picture (FBP) and 5 mls for antibody detection. The analysis was performed at the Haematology Laboratory, HUSM.

Full blood picture was done as a screening to confirm thrombocytopenia. Screening test to detect human platelet-specific antibody was performed by using Capture –P Solid phase system from Immunocor Inc.. Capture-P is a solid phase antibody detection system. Patient or donor platelets were first bound to the surfaces of polystyrene microplate wells. They were subsequently used to capture platelet antibodies from patient or donor sera. Serum was incubated briefly in platelet-coated wells to allow antibodies, if present, to bind to the platelets. Unbound immunoglobulins were then washed from the wells and replaced with a suspension of anti-IgG-coated indicator red cells. Centrifugation brings the indicator red cells in contact with antibodies bound to the immobilized platelets. In the case of positive tests, the migration of the indicator red cells to the bottom of the wells was impeded as anti-IgG bridges were formed between the indicator red cells and the platelet-bound antibodies. As a consequence of such bridging, the indicator red cells will cover the immobilized platelets in a confluent monolayer. In contrast, in the absence of platelet antigen-antibody interactions, ie, negative tests, the indicator red cells will not be impeded during their migration, and will pellet to the bottom of the wells as tightly packed, well-defined cell buttons.These samples were run with the positive, weak positive and negative control. Positive cases were confirmed by the Capture-P Ready Screen test which has the same principle with the screening test.

The data was analysed by using SPSS software, Descriptive analysis and Chi-square (to associate the presence of antibody with age, diagnosis, number of plasma and packed cells transfused).

Blood of 95 patients, demographic data is shown in Table 1. Platelet alloantibodies were detected in 7 (7.4%) patients. All the patients were Malay. There were 2 males and 5 females. The age of the patients was between 12 years old to 75 years old. The number of packed cells that had been transfused was between 1 unit to 25 units. The number of platelet concentrate that had been transfused was between 1 unit to 139 units. The data of patients with platelet alloantibodies as shown in Table II.

There was no significance relationship between presence of platelet alloantibody and age of patient (P value > 0.05). This is probably due to small number of patient recruited in this study.

There was also no significant relationship between presence of platelet alloantibody with sex and race of the patients.

The number of packed cells and platelet concentrate transfused also showed that it was not associated with the presence of platelet alloantibody Table III.

The purpose of this prospective study was to determine the frequency and specificity of platelet alloantibody in these multiply transfused patients and also to correlate with any underlying disease, age of patients, sex of patients, race of patients, and also the numbers and types of blood components given to the patients with the development of platelet antibodies.

The finding of anti-human platelet antigen (HPA)-5b as the highest frequency in this study was similar with the finding by Kiefel et al. which had shown that the antibody specificity found with the highest frequency was anti-HPA-5b (Br^a)(Kiefel et al, 2001). Other specificities found (in decreasing order of frequency) were anti-HPA-1b (Pl^A2), anti-HPA-5a (Br^b), anti-HPA-2b (Ko^a), and anti-HPA-1a (Pl^A1).

Reports about frequency and specificity of platelet-specific alloantibodies in multitransfused patients are rare. This is due to the fact that in such patients platelet-specific antibodies will occur rarely without HLA antibodies, thus complicating their identification and especially their specificities (Murphy and Waters, 1990). The possibility of having the non-specific pattern of platelet alloantibodies in this study (3.2%) may be due to the presence of anti-HLA especially anti-HLA 1 which is the commonest to be presence in leukaemia patients.

In the United States, HLA class 1 antibodies were involved in majority of the alloimmunized cases, whereas platelet specific antigens (e.g. HPA) were involved in approximately 10-20% of refractory cases. Both types of antibodies were involved in approximately 5% of cases. A single random RBC or platelet transfusion induces anti-HLA antibodies in fewer than 10% of recipients (most likely related to the tolerogenic effect of blood transfusions). If patients have more than 20 transfusions, they become sensitized in increasing proportions; after 50 transfusions, the majority (as many as 70%) of patients have anti-HLA antibodies. The presence of HLA antibodies showed better correlation with platelet refractoriness than antibodies directed against platelet-specific antigens. The most common platelet-specific antibody was directed against the HPA-1a antigen present on glycoprotein 11a (Sepulveda et al, 2001).

Study done by Uhrynowska et al, 1996, showed that refractoriness to platelet transfusions due to platelet-specific antibodies was often difficult to assess since, if they developed anti-HLA antibodies, such patients must be transfused with HLA-matched platelets. Nevertheless, platelet-specific antibodies (HPA-1a, HPA-1b, HPA-2b, HPA-3a) have already been found to be responsible for platelet refractoriness (Kicklert et al, 1990). Uhrynowska et al, 1996 also showed that the efficacy of platelet transfusions was difficult to assess due to other non haematological factors which could contribute to refractoriness ⁽⁹⁾.

Alloantibody specificities encountered in multiply transfused patients differ considerably from those in patients with post-transfusion purpura (PTP) and maternal alloantibodies in neonatal alloimmune thrombocytopenia (NAIT). Two latters conditions, anti-HPA-1a was the platelet antigen, most frequently encountered among white patients ⁽¹⁰⁾, whereas anti-HPA-1b and anti-HPA-5b were prevalent among multiply transfused patients. In a large series of multiparous blood donor, anti-HPA-5b was the most common alloantibody ⁽¹¹⁾. However, anti-HPA-1a was the second most frequent alloantibody in that donor group, which reflects the different mode of immunization in that study population (pregnancy).

The frequency of platelet-specific antibodies in multitransfused patients and their implication in platelet transfusion refractoriness has been a matter of controversy for years. The main reasons are the difficulties encountered in differentiating between HLA- and platelet-specific antibodies and, within platelet-specific antibodies, between auto- or alloantibodies, using the available platelet antiglobulin test ⁽¹²⁾. Thus, data suggesting a frequency of platelet-specific antibodies in about 20% of patient were obtained indirectly, such as different reaction patterns of sera tested on lymphocytes and platelets of the same individual as well as transfusion failures with HLA-matched platelets ⁽¹³⁾. In the recent years, capture assays were developed, using immobilized platelet membrane glycoprotein (GP) carrying the antigens under investigation to avoid using whole platelets. These assays overcome the difficulties in differentiating platelet-reactive antibody mixtures and allow determination of antibody specificity.

It is known that previous pregnancies predispose to the production of multispecific HLA antibodies during transfusion therapy ⁽¹⁴⁾. However, there is no evidence to suggest for pre-immunization due to pregnancies as a main cause for the presence of platelet-specific alloantibodies in patients presenting with multispecific HLA antibodies ⁽¹²⁾.

According to Schnaidt et al, 1996, multispecific HLA antibodies are a prerequisite for transfusion-induced platelet-specific alloantibodies. Since only a small percentage of haematologic-oncologic patients undergoing platelet transfusion therapy develop such multispecific HLA-antibodies (Brand et al, 1998) have concluded that frequency of platelet-specific alloantibodies and their impact on platelet refractoriness with regard to all patients transfused was low. They also believed that, nonimmunological factors were more often implicated in platelet refractoriness than platelet-specific alloantibodies. For the group of highly HLA-immunized patients in need for HLA-matched single donor platelets, however, there are only as many as 25% will have additional platelet-specific alloantibodies. These platelet-specific alloantibodies will prohibit satisfactory transfusion results if HLA-matched but HPA-incompatible platelets are transfused.

Patients who were transfused on multiple occasions with red cells or platelets may develop platelet-reactive alloantibody and experience decreased clinical responsiveness to platelet transfusions¹⁶. Platelet transfusion refractoriness is a major complication of long-term platelet supportive care. Refractoriness lead to fatal bleeding complications in thrombocytopenic patients.

Although HLA alloimmunization is the most frequent immunological complication of platelet transfusion, significant numbers of patients receiving multiple platelet and red cell transfusions do not develop anti-HLA antibodies. In this study, there was no anti-HLA antibody detected. The reason was, anti-HLA was best detected using other method such as Lymphocytotoxicity test (LCT). It is therefore of interest to learn which clinical factors influence transfusion-induced platelet alloimmunization. Variables that have been identified include antigenic load, underlying disease and chemotherapy regimen. Although antigenic load appears to be an influencing factor, the relationship between number of platelet transfusions and HLA alloimmunization remains controversial. There are probability due to the heterogeneity of patient populations in terms of underlying disease, immunosuppressive treatment and prior sensitization to HLAs by transfusions and/or pregnancies¹⁷.

Alloimmunization in polytransfused patients is a well known observation which often results in decreased responses to subsequent platelet transfusions and failure to achieve haemostatic levels of platelets that may preclude these patients from important procedures, including bone marrow transplantation. Our data showed a significant rate of platelet alloimunization (7.4%) in multiply transfused thrombocytopenic patients. Perhaps more samples are needed to analyse the full spectrum of platelet alloimmunizations, especially platelet specific alloantibody.

From our data, the most frequent platelet alloantibody was anti-HPA-5b. For future transfusion, we recommend that patients receive compatible platelets after ruling out nonimmune, autoimmune and drug-related causes of platelet refractoriness since compatible platelets can significantly improve the platelet recovery.

ACKNOWLEDGEMENT

We thank the Universiti Sains Malaysia for the short term grant provided for this project and also to Transfusion Medicine staff, Hospital Universiti Sains Malaysia for their endless support and help.

Table I Demographic data of patients with multiply transfused thrombocytopenia

Demographic data

Number of patients (n)

(%)

Diagnosis

Haematological disorders

Chronic Liver Disease

End stage renal disease

Others

Gender

Female

Male

Ethnic

Malay

Chinese

Indian

Age

0-20 years

21-40 years

41-60 years

61-80 years

>80 years

66.3%

11.6%

3.2%

19.0%

52.6%

47.4%

92.6%

6.4%

1.1%

29.5%

15.8%

33.7%

17.9%

3.2%

	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5	Patient 6	Patient 7
Age	75	46	12	17	50	50	39
Diagnosis	Non-Hodgkin lymphoma	Aplastic anaemia	Familial thrombocy-topenia	Motor-vehicle accident	Aplastic anaemia	Aplastic anaemia	Acute Myeloid Leukaemia
Number of packed cells transfused	25	0	11	10	6	5	7
Number of platelet transfused	39	8	25	16	0	24	139
Type of alloantibody	Non specific pattern	HPA-5b	HPA-5b	HPA-5b	No specific pattern	No specific pattern	HPA-5b

Table III Association between presence of platelet alloantibody with age, sex, race, number of packed cells and plasma transfusion.

Independant

Absence of alloantibody

Presence of alloantibody

Statistical Tests

P value

Age

0-40

>40

0.06

0.807

Gender

Female

Male

1.19

0.275

Ethnic

Malay

Non Malay

0.00

0.999

Number of packed cells transfused

<20

>20

0.87

0.350

Number of platelet transfused

<20

>20

0.008

0.927