18, NUMBER 6
Platelet alloimmunization in multitransfused patients with haemato-oncological
Meenu Bajpai, babita Kaura, neelam marwaha, savita kumari, R. R. sharma, S.
Background. We studied the incidence of platelet alloimmunization in multitransfused
patients with haemato-oncological disorders and determined the factors influencing
alloimmunization. We also assessed the effect of alloimmunization on response
to platelet transfusion.
Methods. Fifty patients with haemato-oncological disorders who received multiple
transfusions were included. The patients were tested for antibodies before they
received any transfusion and then after 3–4 weeks of transfusion. Lymphocytotoxicity
and platelet immunofluorescence suspension tests were used to detect antiplatelet
antibodies. Symptomatic improvement was used to assess the response to platelet
Results. Thirty patients were positive by the lympho-cytotoxicity test, giving
an incidence of 60% for anti-HLA antibodies. The panel reactivity of the antibodies
ranged from 3% to 100%. Nineteen patients were positive by the platelet immunofluorescence
suspension test, 16 of whom were also positive by the lymphocytotoxicity test.
The overall incidence of antiplatelet antibodies was 66%. The number of transfusions
received and the underlying haemato-oncological disorder were not risk factors
for the development of antibodies. Patients with a past history of transfusions
and those with a positive obstetric history had a significantly higher incidence
of antibodies. The response to transfusion therapy was poor in patients with
antibodies, as 71.4% of patients with antibodies were non-responsive compared
to only 26.6% of antibody-negative patients.
Conclusion. A high percentage of multitransfused patients developed antiplatelet
antibodies. Previous sensitization was an important risk factor for the development
of antibodies. Patients with high panel reactivity (HLA) showed non-responsiveness
to platelet transfusions. Testing for the presence of antiplatelet antibodies
and provision of compatible platelets should be important components in the
management of patients with platelet transfusion refractoriness.
Natl Med J India 2005;18:134–6
Platelet transfusion is an essential component of supportive therapy in patients
with haemato-oncological disorders as many of them present with varying degrees
of thrombocytopenia leading to bleeding tendencies. Thrombocytopenia may be
either due to the underlying disease or due to the side-effects of therapy.
These patients usually require long term platelet transfusion support. However,
repeated platelet transfusions may fail to show the desired increment in platelet
counts. The cause for this refractoriness to platelet transfusion may be the
underlying clinical condition of the patient, e.g. fever, sepsis, drugs, hypersplenism,
or it may be immunologically mediated due to the development of alloantibodies.1 Platelets
express a variety of antigens such as human platelet antigens (HPAs), ABO antigens,
and human leucocyte antigens (HLAs). The problem of refractoriness
due to antibodies to ABO antigens can be overcome by transfusing ABO-compatible
platelets. Although alloimmunization to HPAs may occur, they are rarely the
sole antibodies responsible for platelet transfusion refractoriness. HLAs are
the major antigens implicated in alloimmunization and refractoriness. Data from
India on platelet alloimmunization in multitransfused patients are scarce. We
studied the incidence of platelet alloimmunization in multitransfused patients
with haemato-oncological disorders.
Fifty patients with haemato-oncological disorders admitted to the Department
of Internal Medicine at the Postgraduate Institute of Medical Education and
Research, Chandigarh, who received >4 transfusions of blood or blood components
during their hospital stay were included. Two blood samples were taken from
each patient. The first sample was collected before the patient received any
transfusion during the present admission and the second sample after 3–4
weeks of transfusion. Serum was separated and stored at –20 oC until testing.
The indirect platelet suspension immunofluorescence test (PSIFT) and lymphocytotoxicity
test (LCT) were used to detect antiplatelet antibodies; PSIFT detects antibodies
to both HPAs and HLAs as well as non-specific antibodies attached to platelets.
Normal reagent platelets were prepared by pooling platelets from 4 random donors
and fixing the platelets with paraformaldehyde after washing in EDTA-PBS. The
patients’ sera were incubated with normal pooled platelets followed by
incubation with fluorescein isothiocyanate (FITC)-labelled F(ab)2 fragment anti-IgG.
The platelets were then examined under ultraviolet illumination for the presence
of fluorescence. Sensitized platelets showed strong fluorescence compared to
non-sensitized ones.2 The LCT was used to detect
the presence of HLA antibodies and was done as described by Terasaki and McClelland.3 Briefly,
the patients’ sera
were incubated with a panel of lymphocytes from 30 random donors. Following
incubation, rabbit complement was added. Formalin was used to fix–stop
the reaction after addition of eosin dye. Lysis of >20% cells in a well was
considered positive. Cytotoxicity against even one of the panel lymphocytes
was considered positive. The panel reactivity of antibodies (the percentage
of panel cells to which the recipient formed cytotoxic antibodies) was calculated.
The clinical response and role of previous sensitization due to pregnancy or
transfusion and response to platelet therapy were analysed. The clinical relevance
of antiplatelet antibodies was assessed by clinical response to platelet therapy.
The patients were considered responsive if there was improvement in their bleeding
manifestations. Prevention of bleeding in patients with low platelet counts
was also considered a response. Patients who had no improvement in their bleeding
manifestations with platelet therapy were labelled as non-responsive. Statistical
analysis was done by the chi-square test.
Of the 50 patients included, 31 were men and 19 women. The age of the patients
ranged from 13 to 75 years with a mean (SD) of 34.9 (17.8) years. Aplastic anaemia
(16) and acute leukaemias (23) were the main underlying disorders. The others
were chronic myeloid leukaemia (6), non-Hodgkin lymphoma (2) and myelo-fibrosis
(1). All patients with chronic myeloid leukaemia were in blast crisis. The number
of transfusions of blood and/or blood components received by the patients ranged
from 4 to 63 with a mean (SD) of 21.0 (13.7). Eleven women had a history of
previous childbirth while one was pregnant at the time of the study. A history
of transfusions in the past was present in 12 patients.
Fifty pairs of samples (before and after transfusion) were tested for HLA antibodies
by LCT. Forty-seven patients were antibody negative in their initial samples.
Of these, 27 became antibody positive in the second sample, suggesting the development
of antibodies during follow up. Three patients were antibody positive in both
samples suggesting the presence of pre-existing anti-bodies. However, the panel
reactivity of antibodies in the initial samples of these 3 patients was low,
with reactivity against 6%, 6% and 10% of the panel, respectively. The antibodies
in the subsequent samples of these patients showed a higher panel reactivity
of 16.6%, 26.7% and 100%, suggesting the appearance of new antibodies. The remaining
20 patients were antibody negative in both samples. The incidence of anti-HLA
antibodies by LCT was 60%. The panel reactivity of the antibodies ranged from
3% to 100%; 19 patients showed reactivity with more than 20% of the panel and,
of these, 2 samples showed reactivity with 100% of the panel.
PSIFT was positive in 19 patients. All of them showed antibodies only in their
second samples. Of the 19 positive samples, 16 were also positive by LCT. Three
samples showed positivity only for PSIFT. The incidence of antibodies using
PSIFT was 38%. The overall incidence of antibodies (detected by LCT and/or PSIFT)
Women with a positive obstetric history showed a higher incidence of antibodies
(83%) as compared to patients with no such history (60.5%), though this was
not statistically significant. The incidence of antibody positivity was higher
in patients with a previous history of transfusion (91.7%) as compared to patients
with no such history (57.9%), and this difference was statistically significant
(p<0.1). There was no correlation between the number of transfusions and
antibody formation. The underlying disorder also did not correlate with antibody
Table I. Antibodies and response to platelet therapy*
|* p<0.05 (presence of antibodies significantly reduces
the response to platelet therapy)
7 patients did not require platelet transfusions but received
red blood cells and fresh frozen plasma.
Seven patients did not require any platelet transfusions (they
received transfusions of red blood cells and fresh frozen plasma)
during the follow up and hence were not included while assessing
the response (Table I). The response to platelet therapy was
significantly less in patients with antibodies (p<0.05).
Seven of the 11 patients with a panel reactivity of <20% were
responsive to platelet therapy whereas only 1 of the 19 patients
with a panel reactivity of >20% were responsive.
Platelet transfusion therapy is life-saving for patients with
haemato-oncological disorders and bleeding due to thrombocytopenia.
Prior to the advent of platelet transfusion therapy, haemorrhage
contributed to the death of up to 53% of patients with leukaemia.4
However, platelet transfusion has its own problems. One of them
is alloimmunization to HPA and HLA antigens. Alloimmunization
depends upon a number of factors such as the type of platelet
products transfused, the number of transfusions received and
the immune status of the patient. At our centre, non-leucoreduced
random donor platelets pooled from 5–6 donors are the mainstay
of platelet transfusion therapy, although single-donor apheresis
platelets may occasionally be transfused. The former increases
the risk for alloimmunization due to greater donor exposure.
The overall incidence of platelet alloantibodies (both HLA and
HPA) in our study was 66%. HLA antibodies are the major antibodies
involved in platelet refractoriness in multitransfused patients
requiring continued platelet support. The incidence of HLA antibodies
was 60% in our study. Three reviews on HLA alloimmunization have
mentioned incidences ranging from 25% to 95%, 30% to 100% and
30% to 70% in patients receiving non-leucoreduced blood components.5–7 Various workers have used different criteria and cut-off values
to define HLA alloimmun-ization. Currently there are no standard
criteria to define alloimmunization. In our study, sera causing
cell death of >20% against any panel cell was considered a
positive result, whereas in another study >10% cell death
against any panel cell was considered positive. Other authors
have considered positive results when the panel reactivity exceeded
5% or 20% of the total panel (lymphocyte killing per panel cell
not mentioned).8–10 Various studies have reported alloimmunization
rates of 34%–60%.9–14
Kiefel et al.10 in a study on 252 multitransfused patients with
haemato-oncological disorders reported an incidence of 42.9%
of HLA alloimmunization using the LCT and monoclonal antibody
immobilization of platelet antigen (MAIPA). However, they used
a cut-off value of >20% panel reactivity while defining HLA
alloimmunization. A Spanish study using MAIPA reported a relatively
lower incidence (17%) of HLA antibodies in multitransfused patients
with haemato-oncological disorders and they attributed it to
racial differences.15 Transfusion of leucoreduced blood components
leads to a much lower incidence of development of HLA antibodies.
A multicentric prospective study by ‘The Trial to Reduce
Alloimmunization to Platelets Study Group’ (TRAP) reported
an incidence of lymphocytotoxic antibodies ranging from 17% to
21% in patients receiving leucoreduced or leucocyte inactivated
blood components as compared to 45% in patients receiving non-leucoreduced
In our study, antibody positivity was detected in 19 patients
using PSIFT. Of these, 16 were also positive by the LCT. The
antibody positivity in these cases could be due to HLA and/or
HPA antibodies. Three samples showed positivity only in PSIFT.
Anti-bodies in these three cases could be HPA antibodies, non-complement
activating HLA antibodies, auto-antibodies, or rarely due to
HLA antibodies against an antigen that was represented in the
platelet pool but was missed in the panel of lymphocytes used.
Previous sensitization due to transfusions was a risk factor
for the development of platelet antibodies (p<0.1). Kurz et
al.11 also found that patients with a history of previous transfusion
had a higher rate of alloimmunization (p<0.013). In our study
there was no definite association between the number of transfusions
received during the same admission or the underlying diagnosis
with development of antibodies. This could be due to the small
sample size or variable donor–patient heterogeneity.
Holohan et al.17 showed a higher frequency of HLA alloimmunization
(80%–90%) in patients with aplastic anaemia compared to
those with haematological malignancies (40%–60%). Lee and
Schiffer18 found that patients with acute myeloid leukaemia were
more likely to develop HLA alloimmunization (44%) than patients
with acute lymphoblastic leukaemia (18%). These differences were
attributed to varying degrees of immuno-suppression and altered
immune status resulting from the disease process or from immunosuppressive
A previous study from India20 used PSIFT to determine the incidence
of platelet antibodies in 29 multitransfused patients with haemato-oncological
disorders, of whom 2 were antibody positive. These 2 were parous
females with acute myeloid leukaemia who had received multiple
blood transfusions. The low prevalence of antibodies was probably
due to the use of only PSIFT for detecting antibodies.
Platelet refractoriness due to HLA alloimmunization may be overcome
by providing HLA-compatible platelets in patients with low panel
reactivity, but the problem becomes intractable in patients with
high panel reactivity as compatible platelets are difficult to
find. In such patients, alternative strategies such as the use
of intravenous immunoglobin, treatment with cyclosporin, immunoadsorption
using staphylococcal protein A columns and plasmapheresis may
be used. In our study, 19 of 30 alloimmunized patients had >20%
panel reactivity and all except 1 of these patients were refractory
to random-donor platelets. However, 7 of 11 patients with <20%
panel reactivity were responsive to random platelet transfusions.
Therefore, the panel-reactive antibody level is an important
predictor of response to platelet transfusion therapy.6 HPA antibodies
are rarely the sole cause of refractoriness to platelet transfusion,
but should be sought and HPA-compatible platelets transfused
if the patient is not responding to HLA-compatible platelets.
Major advances have occurred in the treatment of patients with
haemato-oncological disorders. Platelet transfusion is a vital
support for these patients. However, keeping in view the risks
of alloimmunization, leucoreduced components and single-donor
rather than random-donor platelets should be transfused. In view
of the advances in the treatment of patients with haemato-oncological
disorders, transfusion support services in India need to provide
leucoreduced blood components to reduce the incidence of HLA
alloimmunization. As of now, non-leucoreduced components are
being used in most centres due to the cost of leucoreduction.
Facilities for the testing of HLA and HPA antibodies and provision
of compatible platelets are required urgently at centres treating
patients with haemato-oncological disorders.
VM. Prevention and management of platelet transfusion refractoriness.
Vox Sang 1999;76:1–13.
- Knowles SM. Blood cell
antigens and antibodies: Erythrocytes, platelets and granulocytes.
In: Lewis SM, Bain BJ, Bates
I (eds). Dacie and Lewis
hematology. 9th ed. London:Churchill Livingstone; 2000:429–69.
- Terasaki PI, McClelland
JD. Microdroplet assay of human serum cytotoxins. Nature 1964;204:998–1000.
- Rintels PB, Kenney RM,
Crowley JP. Therapeutic support of the patient with thrombocytopenia.
Hematol Oncol Clin North Am 1994;8:1131–57.
A. White cell depletion: Why and how? In: Nance SJ (ed).
Transfusion medicine in the 1990s. Arlington:American
Association of Blood Banks;
- Curtis BR, Gottschall
JL, McFarland JG. Platelet immunology and alloimmunization.
In: Simon T, Dzik W, Snyder. E,
Stowell C, Strauss
RG (eds). Rossi’s principles
of transfusion medicine. Philadelphia:Williams and Wilkins; 2002:203–17.
- Kao KJ, del Rosario
ML. Platelet alloimmunization. In: Anderson KC, Ness
PM (eds). Scientific basis of
transfusion medicine: Implications for clinical
practice. Philadelphia:WB Saunders; 2000:409–19.
- Killick SB, Win
N, Marsh JC, Kaye T, Yandle A, Humphries C, et al. Pilot
study of HLA alloimmunization after transfusion
products in aplastic anaemia. Br J Haematol 1997;97:677–84.
- Klingemann HG, Self
S, Banaji M, Deeg HJ, Doney K, Slichter SJ, et al. Refractoriness
to random donor platelet transfusions
patients with aplastic anaemia: A
multivariate analysis of data from 264 cases. Br J
- Kiefel V, Konig C, Kroll
H , Santoso S. Platelet alloantibodies in transfused patients.
- Kurz M, Greinix H, Hocker
P, Kalhs P, Knobl P, Mayr WR, et al. Specificities of anti-platelet
disorders. Br J Haematol 1996;95:564–9.
B, Fromont P, Seror T, Duedari N, Bierling P. Platelet
alloimmunization after multiple transfusions:
study of 50 patients. Br
J Haematol 1992;81:395–400.
- Chow MP, Yung
CH, Tzeng JL, Hu HY, Lin WM. Platelet antibody
screening in patients with leukemia and
Zhonghua Yi Xue Za
Zhi (Taipei) 1991;47:237–41.
- Murphy MF, Metcalfe
P, Ord J, Lister TA, Waters AH. Disappearance of HLA
antibodies in acute leukaemia
patients alloimmunized by multiple
transfusions. Br J Haematol 1987;67:255–60.
- Pereira J, Bronfman
L, Bertin P, Marzouka E, Hidalgo P, Amaya S, et al. Platelet
transfusions: Prospective study in adults and
Rev Med Chil 1997;125:1305–12.
Trial to Reduce Alloimmunization to Platelets
Study Group. Leukocyte reduction
B irradiation of platelets
to prevent alloimmunization
to platelet transfusions. N Engl J Med 1997;337:1861–9.
- Holohan TV, Terasaki
PI, Deisseroth AB. Suppression of transfusion-related alloimmunization
cancer patients. Blood 1981;58:122–8.
- Lee EJ, Schiffer
CA. Serial measurement of lymphocytotoxic
antibody and response
patients. Blood 1987;70:1727–9.
- Novotny VM, van
Doorn R, Witvliet MD, Claas FH, Brand
A. Occurrence of
non-HLA antibodies after transfusion
pre-storage filtered platelets
and red blood cells: A prospective
study. Blood 1995;85:1736–41.
- Nanu A, Taneja A. Alloimmunisation
to platelet transfusions in the Indian
J Med Res 1992;96:112–14.
|Postgraduate Institute of Medical Education and Research,
Chandigarh 160012, India
MEENU BAJPAI, NEELAM MARWAHA, R. R. SHARMA,
S. K. AGNIHOTRI Department of Transfusion Medicine
BABITA KAURA Department of Immunopathology
SAVITA KUMARI Department of Internal Medicine
Correspondence to NEELAM MARWAHA; firstname.lastname@example.org