Uncommon Presentation of Post-Transfusion Purpura in an Elderly Male: A Case Report and Unique Alloantibody Identification

Introduction

Post-transfusion purpura (PTP) is a rare phenomenon that can develop 5–10 days after transfusion, marked by severe thrombocytopenia (<10 000 platelets/µL) and often associated with varying degrees of mucosal bleeding and purpura [1,2]. The true incidence of PTP is unknown, with reported incidences ranging from 1 in 24 000 transfusions to 1 in 100 000 transfusions, with suspected underreporting given the difficulty in distinguishing PTP from other causes of immune-mediated thrombocytopenia [1,2]. Identified patients with PTP are typically multiparous women who have been previously alloimmunized with prior blood transfusions or pregnancy [3,4]. PTP is thought to be mediated by alloantibodies against platelet antigens, with human platelet antigen 1a (HPA-1a) being the most common culprit [5]. This report is of a unique case and challenging diagnosis of a 69-year-old man with refractory cardiogenic shock and thrombocytopenia 4 days following blood transfusion, diagnosed with post-transfusion purpura.

Case Report

A 69-year-old man with past medical history of coronary artery disease and chronic obstructive pulmonary disease presented in cardiogenic shock secondary to ST-elevation myocardial infarction (STEMI). He received emergency coronary artery stenting with initial clinical improvement. Six days after stenting, he developed medically refractory cardiogenic shock, for which he was transferred to a tertiary medical center for venoarterial extra-corporeal membrane oxygenation (ECMO). His hospital course was notable for transfusion of 1 unit of packed red blood cells (pRBCs) on the day of admission and transfusion of 2 units of pRBCs and 2 units of platelets on day 4 of admission. Initial pRBC transfusion was indicated after the patient arrived with Hgb 8.7 g/dL and experienced peri-operative bleeding during emergency ECMO cannulation, after which Hgb was found to be 8.3 g/dL (reference range: 13.5–17.5 g/dL). Our hospital policy states that standard transfusions consist of leuko-reduced apheresis platelets and leuko-reduced pRBCs.

Hematology was consulted for acute onset thrombocytopenia when the patient’s platelet count plummeted from 147 k to <2 k (reference range: 150–399 k/uL) on day 4 of hospital admission. Repeat complete blood count, citrated platelet count, and peripheral blood smear confirmed severe thrombocytopenia with an estimated platelet count of 1000 k/uL, with very few clumps and no platelet satellitism. Physical exam was remarkable for gross hematuria in the setting of an indwelling Foley catheter and stable scattered ecchymosis, which were present on admission. Three days after acute thrombocytopenia onset, he developed mild oozing from the site of his femoral catheter. His personal and familial histories were negative for any hematologic disorders or malignancies.

Coagulation parameters were checked (fibrinogen, international normalized ratio, partial thromboplastin time) and were within normal limits, thus excluding disseminated intravascular coagulation. Hemolysis evaluation was unremarkable, with a normal haptoglobin and fractionated bilirubin, and a reticulocyte index of 0.89%, consistent with a hypo-proliferative bone marrow response, likely in the setting of acute critical illness. Lactate dehydrogenase was consistently elevated throughout admission, in accordance with recent STEMI. Heparin-induced thrombocytopenia (HIT) antibody and direct antiglobulin testing were negative, ruling out HIT and decreasing initial suspicion of an autoimmune hemolytic process. Further work-up demonstrated a low folic acid (4.3 ng/mL, reference range: 7.0–31.4 ng/mL), normal vitamin B12, and negative Helicobacter pylori stool antigen. Imaging performed several days prior to and after thrombocytopenia showed no signs of lymphadenopathy or hepatosplenomegaly, which was reassuring that the thrombocytopenia was not caused by a hematologic malignancy. Given these results and the history of recent transfusion, suspicion for post-transfusion purpura (PTP) was high. ECMO was believed to be contributing to some degree to anemia and thrombocytopenia but would not entirely account for the development of acute severe thrombocytopenia. Folate deficiency did not provide a satisfactory clinical explanation for the patient’s cytopenia.

Given concern for immune thrombocytopenic purpura (ITP) and PTP, the patient was started on dexamethasone 40 mg daily for 4 days and received intravenous immunoglobin (IVIg) 1 g/kg daily for 2 days. The platelet count gradually improved with supportive transfusions and medical therapy (Figure 1). After completion of dexamethasone and IVIg, as well as supportive transfusion of 4 units of platelets, cell counts recovered to platelets >100 k/uL for the remainder of his admission.

The initial platelet antibody screen sent to an American Red Cross Reference Laboratory to test for immunoglobulin G (IgG) antibodies against platelets was negative. However, suspicion for PTP remained high. Post-transfusion purpura panel testing, including anti-human platelet antigen (HPA) antibody testing, was performed at the nearest regional blood center using a sample drawn 3 days after thrombocytopenia onset. PTP testing consisting of monoclonal antibody immobilization of glycoproteins, platelet antibody bead array (PABA), and whole platelet flow cytometry assays, and the results were available 2 weeks later. Platelet antibody screening identified IgG antibodies. PABA was utilized to further identify the IgG antibodies as platelet-specific HPA-5b antibodies, indicating the presence of antibodies directed against HPA-5b platelet antigens. Platelet antigen genotyping was remarkable for HPA 5a/5a platelet antigens, thus demonstrating that the isolated HPA-5b antibodies were directed against non-native platelet antigens likely encountered from prior transfusion of blood products. Therefore, testing results were consistent with post-transfusion purpura. The patient was discharged from the hospital and continues to have normal platelet counts with no recurrence of thrombocytopenia or transfusions since discharge.

Discussion

Patients should be closely monitored after transfusion of blood products, given the wide spectrum of transfusion reactions that can occur. Transfusion reactions are adverse events associated with the transfusion of blood products, with onset ranging from immediately to weeks after transfusion. The most commonly observed transfusions reactions are febrile non-hemolytic, due to cytokine release from donor leukocytes, or a mild allergic reaction, from foreign donor proteins [6]. Given the often non-specific signs and symptoms during transfusion, the type of reaction can be difficult to determine. Delayed transfusion reactions, including post-transfusion purpura, are the result of an immune response to a previously encountered foreign antigen [6]. This report highlights the importance of monitoring patients for post-transfusion adverse events. Although PTP is rare, rapid diagnosis and management are critical given the associated morbidity and mortality.

Post-transfusion purpura is a delayed transfusion reaction that typically develops 5–10 days after transfusion, marked by severe thrombocytopenia (<10 000 platelets/µL) [2]. Typical characteristics include mucosal bleeding, petechiae, ecchymoses, and even intracranial hemorrhage, conferring an overall 20% risk of mortality [1,2]. Risk is highest with platelet transfusions and increases with each unit transfused [4]. It is important to note that PTP can occur after transfusion of platelets, pRBCs, and plasma, although there has been a distinct decrease with widespread use of leukocyte-reduced blood products [1,3]. The true incidence of PTP is unknown, with reports ranging from 1 in 24 000 to 100 000 transfusions [2,4]. PTP is likely under-diagnosed given the difficulty in distinguishing PTP from other immune-mediated thrombocytopenias such as HIT, ITP, and thrombotic thrombocytopenic purpura [1,2].

In many of the available case reports regarding PTP, HIT was initially suspected as the most likely diagnosis, which may have delayed PTP treatment [1,5,7]. In our patient, the HIT antibody was negative, excluding HIT and allowing clinicians to consider alternative diagnoses. Most of the patients in case reports, including our own, were already critically ill at the time of PTP development, further increasing the risk for morbidity and mortality. Given that the treatment for PTP and other thrombocytopenias vary greatly, it is important to consider PTP in any person developing thrombocytopenia after receiving blood products.

PTP is thought to be mediated by alloantibodies against platelet antigens, with HPA-1a being the most common culprit [2,8]. HPA are polymorphic glycoproteins found on the platelet cell membrane surfaces that allow platelets to engage in ligand-receptor interactions integral to platelet function [9]. It is believed that pregnancy exposes females to foreign platelet antigens, causing alloantibody sensitization. When next exposed to the offending antigen during transfusion of blood products, immune-mediated destruction of both autologous and donor platelets results [2,3]. Nulliparous females and males sensitized to foreign platelet antigens during prior transfusions are also observed to develop PTP [2].

The mechanism of autologous platelet destruction in PTP is still not fully understood, with Shulman et al proposing 3 mechanisms when first describing PTP in 1961: 1) transfused soluble platelet antigen and autoantibodies form immune complexes that bind nonspecifically to autologous platelets causing indiscriminate macrophage platelet destruction; 2) transfused soluble platelet antigens coat autologous platelets, priming them for autoantibody-mediated destruction; or 3) autoantibodies and alloantibodies are concomitantly produced, with autoantibodies responsible for autologous platelet destruction [2,10]. The third mechanism seems to be best supported by current evidence [2].

Diagnosis of PTP requires high clinical suspicion given patient presentation and development of acute, severe thrombocytopenia 5–10 days following transfusion. Diagnostic testing involves identification of the involved HPA platelet antibodies and platelet-specific antigen identification, along with autologous platelet genotyping [2,4]. Alloantibodies against HPAs will be detected in the sera from PTP patients via direct or indirect testing including flow cytometry, platelet antibody bead array (PABA), and monoclonal antibody immobilization of platelet antigen (MAIPA) [2]. Patient HPA genotyping is performed via polymerase chain reactions to distinguish it from autoantibodies and confirm HPA antibody specificity, which can be useful when considering future transfusions [2,4].

The standard of treatment is IVIg 400–500 mg/kg/day for 1–10 days or 1–2 gm/kg/day for 2–5 days [2]. In addition to IVIg, steroids have also successfully recovered platelet counts, as well as plasmapheresis [5]. Our patient was treated with both IVIg and steroids, with excellent clinical response. Accurate diagnosis of PTP can help guide future transfusion management. While recurrence appears to be uncommon, patients with a history of PTP should receive screened HPA-compatible blood products given the significant morbidity and mortality of acute severe thrombocytopenia [2]. A simple and cost-effective enzyme-linked immunosorbent assay has been used to identify donor platelets lacking the most commonly implicated antigen, HPA-1a [11]. Such screening assays on donor blood products in affected patients can help mitigate recurrence of PTP with future transfusions.

Our case is unique in several ways. Notably, the patient was a male with no prior transfusion history. It is rare for PTP to develop in males, with 95.5% of cases documented in females previously [12]. In our patient, the onset of thrombocytopenia occurred more rapidly than typically observed in PTP – 4 days after index transfusion and 12 h after his second transfusion. Prior documented cases of PTP note thrombocytopenia developing within 5–10 days after initial transfusion [2,3].

The characteristic bruising of PTP was absent on physical exam, although the patient did develop hematuria and later had mild catheter oozing, which resolved with treatment. Several other case reports lacked skin findings on physical exam but are notable for significant mucocutaneous bleeding including gastrointestinal, subconjunctival, and retroperitoneal hemorrhages. In contrast, our patient had only subtle mucocutaneous bleeding via hematuria, demonstrating that the clinical presentation is perhaps broader than previously thought [7,13,14].

The identified antibody, HPA-5b, is somewhat novel, as it accounts only for 15.7% of thrombocytopenia’s PTP cases documented to date [2]. The first documented report of a male with anti-HPA-5b differs significantly from our report in that the patient had a prior history of transfusion and required significantly more blood products, increasing the risk of foreign antigen exposure [14]. In our patient, initial platelet antigen screening, drawn 12 h after onset of acute thrombocytopenia, was negative. Given the high suspicion for PTP, a more expansive platelet antigen screening test collected 3 days later was positive for the HPA-5b antibody and genotyped the patient as HPA-5a/5a. This suggests that a detectable level of antibodies may lag behind symptom onset and laboratory evidence of thrombocytopenia.

Conclusions

The acute onset of thrombocytopenia and male sex of our case emphasizes the need to consider PTP as a cause of acute severe thrombocytopenia in patients who have received blood products. Treatment should not be delayed given its high mortality, and concurrent treatment for more common thrombocytopenic disorders can be pursued to optimize patient outcomes. This report highlights the importance of monitoring patients for post-transfusion adverse events. Although PTP is rare, rapid diagnosis and management are required to control this potentially life-threatening condition.