Portal de Periódicos da CAPES

Post‐transfusion purpura caused by anti‐HPA‐3a antibodies that are only detectable using whole platelets in the platelet immunofluorescence test

2009; Wiley; Volume: 20; Issue: 3 Linguagem: Inglês

10.1111/j.1365-3148.2009.00978.x

1365-3148

Pere Barba, Pedro Boix Pallares, Núria Nogués, Carme Saurina, M Bara, Immaculada Vinyets, Eduardo Muñiz‐Díaz,

Heparin-Induced Thrombocytopenia and Thrombosis

Alloantibodies against human platelet antigens (HPA) are involved in several entities such as neonatal alloimmune thrombocytopenia (NAT), platelet refractoriness (PR) and post-transfusion purpura (PTP) (Berry et al., 2007). To date, 17 HPA systems have been identified, of which 6 are biallelic (HPA-1, -2, -3, -4, -5 and -15 ) (Metcalfe, 2004). A wide range of techniques have been developed for HPA antibody detection over the last 25 years. The most widely used of these techniques are as follows: the platelet immunofluorescence test (PIFT), the monoclonal antibody immobilization of platelet antigens (MAIPA) assay, the solid-phase red cell adherence assay and a variety of enzyme-linked immunosorbent assay (ELISA) based techniques (Metcalfe, 2004). The molecular basis of the majority of serologically defined HPA antigens has been elucidated, and genotyping of the HPA polymorphisms is routinely used. The difficulty of detecting HPA antibodies is variable because some antibodies demand greater effort to be detected. A number of studies have identified anti-HPA-3a antibodies as the most problematic specificity to detect (Bessos et al., 2005; Berry et al., 2007). The HPA-3a antigen is determined by the presence of an isoleucine residue at the polymorphic position 843 of the glycoprotein IIb alpha chain (Harrison et al., 2003). Anti-HPA-3a antibodies have been involved in NAT, PR and more rarely in PTP. PTP is a rare complication of blood transfusion characterized by severe thrombocytopaenia within 5–10 days of transfusion of blood products (Taaning & Svejgaard, 1994; Lubenow et al., 2000; Welling et al., 2003). Its incidence has been estimated to be around 1/50 000–100 000 transfusions (Metcalfe, 2004). About 90% of the cases are women with a history of pregnancy or transfusion during which primary immunization occurred. The blood transfusion triggers an anamnestic response in boosting HPA antibodies in an already sensitized person. Thrombocytopenia in PTP is caused by antibody-mediated destruction. Why HPA alloantibodies destroy both the autologous platelets as well as the platelets of donor origin is not completely understood (Lubenow et al., 2000). Main symptoms include purpura, cutaneous bleeding, epistaxis and gastrointestinal haemorrhage with a considerably high mortality rate (10–20%). The most commonly involved antibody in PTP is anti-HPA-1a in an HPA-1a negative patient receiving blood products from an HPA-1a positive donor. Although less frequent, cases of PTP due to antibodies with other specificities have also been reported in the literature (Lucas et al., 1997; Allen et al., 2007). A 55-year-old woman was admitted to the Intensive Care Unit (ICU) because of septic shock. She had been diagnosed with acute myeloid leukaemia (AML) secondary to myelodysplastic syndrome and admitted to the haematology unit where she received secondary-AML chemotherapy with standard treatment idarrubicine, citarabine and etoposide (ICE). During the neutropenic period she developed fever, bacteremia due to Staphylococcus epidermidis and acute renal failure caused by amphotericin prophylaxis. Twenty-five days after chemotherapy, she presented oligury and dyspnoea and was admitted to the ICU. On admission, the haemoglobin level and WBC and platelet counts were, 80 g L−1, 3.65 × 109 L−1 and 159 × 109 L−1, respectively. A week after admission in the ICU, the patient developed acute severe and abrupt thrombocytopaenia (nadir 12 × 109 L−1) and purpuric exanthema appeared. At that time antibiotics and mechanical ventilation were discontinued because of improvement in the patient. She remained afebrile for several days and C-reactive protein levels decreased from 506 to 34 mg dL−1. During admission in the haematology unit, she had received several leukoreduced red blood cell units and platelet pools, the last one 7 days before developing thrombocytopaenia. The patient did not receive heparin during admission, and no other drugs accounting for thrombocytopaenia were found. PTP was suspected and whole blood and plasma samples were sent to our laboratory. The patient was sent back to the haematology unit and underwent treatment with conventional haemodialysis three times per week. Purpuric exanthema was resolved after 1 week and the platelet count reached normal levels within 3 weeks after the diagnosis. No specific treatment was administered to the patient. Patient serum and ethylenediaminetetraacetic acid (EDTA)-anticoagulated blood were used for all studies. A platelet specific antibody only reactive with HPA-3a positive platelets was detected in PIFT. The investigation of HLA class I antibodies proved to be negative with an ELISA test (Quick Screening®, GTI Diagnostics, Wisconsin, USA). Platelet specific antibody investigation proved to be negative in solid-phase assay (PAK12®, GTI Diagnostics, Wisconsin, USA) and in the MAIPA test (Kiefel et al., 1987) when immobilizing the IIb–IIIa, Ib–IX, Ia–IIa platelet glycoprotein complexes and the CD 109 molecule. Subsequently, a PIFT was performed again using another panel of platelets with a known genotype. We obtained positive results using platelets with genotype HPA-3a/3a, whereas negative results were obtained with HPA-3b/3b genotype platelets (Table 1). An anti-HPA-3a specificity was suspected and thus a MAIPA test was performed using eight different anti-GPIIb/IIIa and anti-GPIIIa monoclonal antibodies (CLB-tromb/7,6C9 by Menarini® (CLB, Valkenswaard, The Netherlands), P2 by Immunotech® (Coulter, Marseille, France), 962 and 672 by Hospital Clinic, Barcelona, Spain, PL2–73 and CLB/tromb/1 from the 14th International Platelet Immunology Workshop and C17 and HIP8 by Immunostep® (Salamanaca, Spain)). Negative results were also obtained with all these monoclonal antibodies. All these tests were repeated 1 month later with a second plasma sample, and same results were obtained. Finally, we performed HPA genotyping analysis using polymerase chain reaction with allele-specific primers (PCR-SSP) as described by Cavanagh et al. (1997). The following results were obtained: HPA-1a/1a, HPA-2a/2a, HPA-3b/3b, HPA-5a/5a and HPA-15a/15b. We report here one of the few cases of PTP caused by an anti-HPA-3a antibody. The patient was a woman who had received several platelet transfusions during admission, the last one 7 days before developing the clinical features of PTP. No other cause of thrombocytopaenia was detected and heparin-induced thrombocytopaenia, infection and pharmacological causes were specifically excluded (Taaning & Svejgaard, 1994; Lubenow et al., 2000). HLA class I antibodies were excluded using an ELISA assay. We were only able to demonstrate the presence of HPA alloantibodies with an anti-HPA-3a specificity with the PIFT technique. The solid-phase technique (PAK12®) and the MAIPA test using different monoclonal antibodies failed to demonstrate this specificity despite repeating all the tests with two different samples. However, the genotyping analysis supported the result of the PIFT technique as the patient was HPA-3b/3b. Recent published data from 35 laboratories participating in the quality assurance exercises organized by the National Institute for Biological Standards and Control showed that anti-HPA-3a is the most problematic specificity to be detected (Allen et al., 2007). The percentage of laboratories that correctly identified anti-HPA-3a specificity ranged from 21.2 to 90.9%. Most of the errors were due to false negative results, whereas false positive results had a much lower frequency. Interestingly, only 44% of the laboratories used intact platelet assays such as PIFT. In our study we only detected the specificity using a whole platelet assay (PIFT), whereas tests using fractioned platelet membrane (MAIPA test and other solid-phase assays) consistently failed to detect HPA-3a specificity. In agreement with other authors (Glade-Bender et al., 2001; Harrison et al., 2003), our study suggests that the solubilization of the platelet membrane alters the conformation of the HPA-3a epitope leading to false negative results in the antibody investigation. Some authors suggested that the delay in performing the serological studies could also affect their specificity and sensitivity, especially after freezing the samples and after more than 14 days of storage (Socher et al., 2008). We kept the samples at 4°C from the moment of receipt and we performed all tests within 14 days, so it is unlikely that our results were influenced by these factors. In conclusion, we report a case of PTP exclusively caused by an anti-HPA-3a antibody that is only detectable with whole platelet assays. Despite the low incidence of PTP caused by anti-HPA-3a, we strongly recommend the use of whole platelet assays in all PTP cases in order to avoid false negative results that could lead to a misdiagnosis and an inadequate therapeutic approach. P. Barba, *‡ P. Pallarés, † N. Nogués, * C. Canals, * M. Gracia, * I. Vinyets * & E. Muñiz-Diaz *