SEARCH

SEARCH BY CITATION

Keywords:

  • autoimmunity;
  • bleeding;
  • flow cytometry;
  • hemorrhage;
  • thrombocytopenia;
  • thrombopoiesis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Summary. Background: Severe thrombocytopenia is a major risk factor for hemorrhage, but platelet function and bleeding risk at low platelet counts are poorly understood, because of the limitations of platelet function testing at very low platelet counts. Objectives: To examine and compare platelet function in severely thrombocytopenic patients with acute myeloid leukemia (AML) or myelodysplasia (MDS) with that in patients with immune thrombocytopenia (ITP). Methods: Whole blood flow cytometric measurement of platelet activation and platelet reactivity to agonists was correlated with the immature platelet fraction (IPF) and bleeding symptoms. Results: Patients with AML/MDS had smaller platelets, lower IPF and substantially lower platelet surface expression of activated glycoprotein (GP)IIb–IIIa and GPIb, both with and without addition of ex vivo ADP or thrombin receptor-activating peptide, than patients with ITP. In both ITP and AML/MDS patients, increased platelet surface GPIb on circulating platelets and expression of activated GPIIb–IIIa and GPIb on ex vivo activated platelets correlated with a higher IPF. Whereas platelet reactivity was higher for AML/MDS patients with bleeding than for those with no bleeding, platelet reactivity was lower for ITP patients with bleeding than for those with no bleeding. Conclusions: AML/MDS patients have lower in vivo platelet activation and ex vivo platelet reactivity than patients with ITP. The proportion of newly produced platelets correlates with the expression of platelet surface markers of activation. These differences might contribute to differences in bleeding tendency between AML/MDS and ITP patients. This study is the first to define differences in platelet function between AML/MDS patients and ITP patients with equivalent degrees of thrombocytopenia.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

The degree of thrombocytopenia is a major determinant of bleeding risk [1]. However, whereas some patients exhibit little bleeding despite platelet counts < 10 000 μL−1, others suffer significant hemorrhages with platelet counts > 50 000 μL−1 [2–4]. Patients with extreme thrombocytopenia because of lack of platelet production (e.g. acute myeloid leukemia [AML] or myelodysplastic syndrome [MDS]) or primarily because of accelerated platelet destruction (e.g. immune thrombocytopenia [ITP]) may experience severe bleeding. Platelet function in patients with severe thrombocytopenia has never been adequately studied, because of methodological problems related to the assessment of platelet function in thrombocytopenia. The contribution of platelet function to bleeding risk in this setting is therefore poorly understood.

Platelet function in patients with AML or MDS may be impaired because of the underlying malignancy, chemotherapy or other drugs, or concurrent infections. It has traditionally been assumed that the larger, younger platelets of ITP patients are better functioning than normal platelets and [5–7], by inference, also better functioning than the platelets of patients with hematologic malignancies. However, these two types of thrombocytopenic condition are managed very differently with regard to platelet count thresholds for initiating treatment. It is recommended that adults with ITP are treated at platelet counts of 20 000–30 000 μL−1 [8,9], whereas prophylactic platelet transfusions are generally initiated at a platelet count of ≤ 10 000 μL−1 in patients with thrombocytopenia secondary to hematologic malignancy, bone marrow failure, or chemotherapy, in part to minimize alloimmunization and platelet refractoriness caused by more frequent transfusions [10]. Thus, the typical threshold for treatment of thrombocytopenia is often higher in ITP than in dysplastic or malignant disorders of the bone marrow.

There are very few data describing platelet function in severely thrombocytopenic patients, because, with the sole exception of flow cytometry, no platelet function assay can distinguish between the effects of thrombocytopenia and the effects of reduced platelet function. Whole blood flow cytometry can investigate platelet function independently of platelet count by examining the functional status of individual platelets [11]. Flow cytometry has been widely used and validated in patients with heart disease, in whom assessment of platelet function is considered to be important for monitoring of antiplatelet therapy [11–13]. Surprisingly, whole blood flow cytometry has infrequently been tested in the setting of marked thrombocytopenia, where it is uniquely able to assess platelet function [7,11,14].

In this study, we used whole blood flow cytometry [15] and a standardized bleeding score [16] in patients with platelet counts ≤ 30 000 μL−1 to compare platelet function in severely thrombocytopenic AML/MDS patients with that in ITP patients. Platelet surface expression of P-selectin, activated glycoprotein (GP)IIb–IIIa (integrin αIIbβ3) and GPIb was measured with and without ex vivo agonist stimulation, in order to assess both the activation state of circulating platelets and platelet reactivity. In addition, we examined whether the proportion of newly synthesized platelets in the circulation contributed to platelet function.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Twenty-five patients with ITP and 21 patients with AML/MDS gave consent and were enrolled in this Institutional Review Board-approved study. Patients were selected on the basis of platelet counts ≤ 30 000 , age > 18 years, not having received platelet transfusions or antiplatelet agents within the previous 5 days, and not having any known disorders of hemostasis or platelet function. None of the patients received cytotoxic chemotherapy on the day of study.

A diagnosis of ITP was based on thrombocytopenia in the absence of another identifiable cause, normal or increased numbers of megakaryocytes (if a bone marrow examination had been performed), and/or response to intravenous immunoglobulin or steroids. Of the 21 patients in the AML/MDS group, 18 had a diagnosis of AML and three had MDS with circulating blasts.

Blood was drawn from patients by antecubital venipuncture into 4.5-mL 3.2% trisodium citrate Vacutainers (Becton Dickinson, Franklin Lakes, NJ, USA), a method previously shown not to induce ex vivo platelet activation [17]. Platelet counts were measured in a Bayer-Advia automated CBC counter immediately following the blood draw. The immature platelet fraction (IPF) and the immature platelet count (IPC) were measured for 17 of 25 ITP patients and 19 of 21 AML/MDS patients in a Sysmex XE-2100 autoanalyzer (Sysmex America, Inc, Mundelein, IL, USA) within 6 h of blood draw [18].

Twenty minutes after blood draw, aliquots of whole blood were incubated with fluorescently labeled mAbs and either 0.5 μm ADP, 20 μm ADP, 1.5 μm thrombin receptor-activating peptide (TRAP), 20 μm TRAP, or Hepes–Tyrode’s buffer (10 mm Hepes, 137 mm sodium chloride, 2.8 mm potassium chloride, 1 mm magnesium chloride, 12 mm sodium hydrogen carbonate, 0.4 mm sodium phosphate dibasic, 5.5 mm glucose, 0.35% w/v bovine serum albumin, pH 7.4) for exactly 15 min. The reaction was stopped with a 15-fold dilution in 1% formaldehyde in Hepes–saline buffer. Samples were maintained at room temperature and not agitated until fixation, to prevent handling activation.

The antibodies used were as follows: phycoerythrin (PE)-conjugated anti-P-selectin mAb (CD62P; clone 1E3; Santa Cruz Biotech, Santa Cruz, CA, USA); fluorescein isothiocyanate (FITC)-conjugated mAb PAC1 (Becton Dickinson Pharmingen, San Diego, CA, USA), which only binds to the activated conformation of GPIIb–IIIa [19]; and PE-Cy5-conjugated anti-CD42b (GPIb) mAb (clone HIP1; Becton Dickinson Pharmingen). PE-conjugated MIgG2a isotype (Santa Cruz Biotech) and FITC–PAC1, together with 2.5 μg mL−1 of the GPIIb–IIIa antagonist eptifibatide to block specific binding, served as the negative controls for P-selectin and PAC1, respectively.

For flow cytometric analysis of platelet count, anticoagulated blood was labeled with FITC-conjugated anti-GPIIIa (CD61) mAb (clone Y2/51; Dako Cytomation, Carpinteria, CA, USA), PE-conjugated anti-GPIIb (CD41) mAb (clone 5B12; Dako Cytomation), and PE-Cy5-conjugated anti-CD42b mAb (clone H1P1; Becton Dickinson Pharmingen).

Fixed samples were stored at 4 °C and sent by overnight courier to the Center for Platelet Function Studies at the University of Massachusetts Medical School for analysis. A known quantity of RFP-30-5 calibration beads (Spherotech, Lake Forest, IL, USA) was added to allow cell counts to be calculated. Analysis was performed in a Becton Dickinson FACSCalibur flow cytometer, which was calibrated daily to ensure proper instrument functioning and consistent fluorescence measurements over time. Platelet surface P-selectin, activated GPIIb–IIIa and GPIb expression levels were measured relative to the isotype control as mean fluorescence intensity (MFI). For GPIb, the magnitude of change following agonist stimulation was calculated by subtracting the GPIb MFI value with added agonist from the MFI value with no added agonist. Platelets for flow cytometric counting were identified by characteristic forward light scatter (FLS) and side light scatter, and CD61, CD41 and CD42b expression. The platelet count was calculated relative to the number of internal standard calibration beads that had been added to the sample in a known quantity prior to flow cytometric analysis. Mean FLS of platelets was also recorded as an approximation of platelet size. A control sample from a healthy donor was also analyzed in parallel with each study sample to ensure correct sample handling.

At the time of blood draw, bleeding was assessed by history and by physical examination by one of three trained assessors (J.B.B., B.B., or C.T.), using a standardized bleeding scale that quantifies bleeding at nine anatomic sites and has been shown to have good interuser reliability [16]. Bleeding scores were later summarized into three categories: any bleeding vs. no bleeding; skin/oral bleeding vs. no skin/oral bleeding; and any non-skin/oral bleeding vs. no non-skin/oral bleeding.

The flow cytometry data were analyzed with a mixed model repeated measures anova. For all variables, the anova models were checked for Gaussian residuals and equality of variance. If a significant disease (ITP vs. AML/MDS) by treatment (platelet agonist condition) interaction was found, five Bonferroni-adjusted pairwise comparisons of AML/MDS vs. ITP for each of the five treatment conditions were performed. A comparison was considered to be statistically significant if P < 0.01 (Bonferroni adjustment to maintain an overall 5% significance level). The association between disease and the categorical bleeding score variables was examined with the chi-square test. The Kruskal–Wallis test was used to compare the bleeding scores and the continuous variables, IPF and IPC. The Wilcoxon rank sum test was used to compare IPF data for subjects with AML/MDS and those with. ITP. Spearman correlations were used to examine the association between platelet function flow variables and IPF values. Results were considered to be significant if P < 0.05. Graphs were drawn with graphpad prism 5.0a software (2007).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

There was no difference in age between the two patient groups (mean age of the ITP patients was 53 years [range: 21–80 years]; mean age of the AML/MDS patients was 59.3 years [range: 38–76 years]; P = 0.06). A higher proportion of the ITP patients were female (68% females for ITP vs. 19% females for AML/MDS, P = 0.001). At the time of study, none of the patients were receiving treatment with thrombopoietin receptor agonists, and 60% of the ITP patients had undergone splenectomy.

Mean platelet counts were similar between the ITP and the AML/MDS patient groups (15 850 ± 2338 vs. 20 938 ± 3729 μL−1, P = 0.2; Fig. 1A). However, the relative and absolute numbers of immature platelets (IPF and IPC) were lower in AML/MDS patients than in ITP patients, reflecting a lower rate of platelet production (IPF, 11.4% ± 1.2% vs. 25.0% ± 2.9%, P = 0.0002; IPC, 2357 ± 403 vs. 4000 ± 60 μL−1, P = 0.03; Fig. 1C,D). Platelets were much smaller in AML/MDS patients than in ITP patients (69.0 ± 4.7 vs. 211.2 ± 23.9 arbitrary FLS units, P < 0.0001; Fig. 1B).

image

Figure 1.  Platelet parameters in ITP and AML/MDS. (A) Platelet count, (B) Platelet size, (C) Immature platelet count (IPC) and (D) Immature platelet fraction (IPF). n = 25 for ITP, n = 21 for AML/MDS. Mean values ± SEM are shown. * indicates < 0.05; *** indicates P < 0.001. Abbreviations: AML, acute myeloid leukaemia; FLS, forward light scatter; ITP, immune thrombocytopenic purpura.

Download figure to PowerPoint

When platelets were examined for surface markers of activation with no addition of agonists ex vivo, levels of surface activated GPIIb–IIIa were significantly lower in AML/MDS platelets than in ITP platelets (MFI, 6.06 ± 0.5 vs. 10.38 ± 1.3, P = 0.007; Fig. 2A). There was no difference in the surface expression of P-selectin between ITP platelets and AML/MDS platelets (MFI, 6.5 ± 0.6 vs. 7.3 ± 1.2, P = 0.5; Fig. 2B).

image

Figure 2.  Platelet surface markers of activation in ITP and AML/MDS with and without ex vivo agonists ADP and TRAP. Expression of platelet surface activated GPIIb/IIIa (panel A), P-selectin (panel B) and GPIb (panel C) in ITP and AML/MDS patients with and without addition of platelet agonists ex vivo. n = 25 for ITP, n = 21 for AML/MDS. Mean ± SEM is shown. * indicates P < 0.05; ** indicates P < 0.01; *** indicates P < 0.001. Abbreviation: MFI, mean fluorescence intensity.

Download figure to PowerPoint

In response to ex vivo stimulation with ADP and TRAP, each at lower and higher concentrations, increases in surface activated GPIIb–IIIa and P-selectin expression and a decrease in surface GPIb expression were observed, as expected, in both AML/MDS and ITP platelets (Fig. 2A–C). However, expression of surface activated GPIIb–IIIa remained significantly lower for AML/MDS platelets than for ITP platelets (P = 0.025; Fig. 2A). Expression of surface P-selectin was also significantly lower in AML/MDS platelets in response to high TRAP stimulation (P = 0.002), although this difference was not significant in the other agonist conditions (Fig. 2B). Expression of GPIb was significantly lower in AML/MDS patients than in ITP patients, both with no added agonist (MFI, 187.3 ± 9.4 vs. 420.0 ± 30.4, P < 0.0001) and following ex vivo stimulation, presumably reflecting the relationship between platelet surface expression of GPIb and platelet size (Fig. 2C). There was no significant difference between ITP and AML/MDS patients with respect to the percentage decrease in platelet surface GPIb expression. In ITP patients, the percentage decreases in platelet surface GPIb expression as compared with no agonist were 7.5% with 0.5 μm ADP, 22.5% with 20 μm ADP, 6.1% with 1.5 μm TRAP, and 25.6% with 20 μm TRAP. In AML/MDS patients, the percentage decreases in platelet surface GPIb expression as compared with no agonist were 10.4% with 0.5 μm ADP, 21.5% with 20 μm ADP, 7.2% with 1.5 μm TRAP, and 22.3% with 20 μm TRAP.

There were no significant differences in platelet count, platelet size or IPF between splenectomized and non-splenectomized ITP patients (Fig. 3A); however, expression levels of platelet surface markers of activation with and without agonist stimulation were higher in those ITP patients who had undergone splenectomy than in non-splenectomized ITP patients, particularly for activated GPIIb–IIIa (Fig. 3B).

image

Figure 3.  Platelet parameters and platelet function in non-splenectomized vs. splenectomized patients with ITP. (A) Platelet count, platelet size, immature platelet count (IPC) and immature platelet fraction (IPF) in non-splenectomized (black columns) and splenectomized patients (grey columns) with ITP. n = 10 for non-splenectomized and n = 15 for splenectomized patients. (B) Expression of platelet surface activated GPIIb/IIIa, P-selectin and GPIb in non-splenectomized and splenectomized patients with ITP with and without addition of platelet agonists ex vivo. Mean values ± SEM are shown. * indicates P < 0.05; ** indicates P < 0.01.

Download figure to PowerPoint

To investigate whether the lower surface expression of platelet GPs in AML/MDS patients than in ITP patients (Fig. 2) was related to the lower proportion of immature platelets in AML/MDS patients than in ITP patients (Fig. 1), we explored the association between IPF and platelet function. Considering both ITP and AML/MDS patients combined, a higher IPF was significantly correlated with increased platelet surface expression on circulating platelets (i.e. no added agonist) of GPIIb–IIIa (P = 0.001), P-selectin (P < 0.01), and GPIb (P < 0.0001) (Fig. 4A). The magnitude of change in platelet surface GPIb expression following ex vivo agonist activation was also strongly correlated with IPF, regardless of agonist or agonist concentration (Fig. 4B). Similarly, IPF was positively correlated with GPIIb–IIIa expression on high-dose ADP-stimulated platelets (P = 0.0484, data not shown). IPF was not correlated with platelet surface GPIIb–IIIa or P-selectin expression for any of the other agonist conditions (data not shown).

image

Figure 4.  Relationship between IPF and platelet markers of activation. (A) Correlation between markers of activation expressed on the surface of circulating platelets (i.e. with no added agonist) and immature platelet fraction (IPF%). Each dot represents an individual patient (ITP and AML/MDS combined). (B) IPF%vs. agonist-induced change in platelet surface GPIb expression for low and high concentrations of ADP and TRAP. The data shown are the magnitude of change in GPIb MFI following agonist stimulation as compared to the level of expression with no added agonist.

Download figure to PowerPoint

There were no major differences in the overall incidence of bleeding between AML/MDS and ITP patients. However, there was a tendency for non-skin/oral bleeding, i.e. more clinically significant bleeding, to occur more commonly among AML/MDS patients than among ITP patients (P = 0.118; Table 1). To explore the relationship between platelet function and bleeding, we categorized patients according to types of bleeding symptom. There were no significant differences in platelet count, IPF and IPC between the three bleeding categories: bleeding (any site) vs. no bleeding; any skin/oral bleeding vs. no skin/oral bleeding; and any non-skin/oral bleeding vs. no non-skin/oral bleeding. However, for all three bleeding categories, ITP patients who experienced bleeding had lower platelet reactivity (e.g. lower TRAP-stimulated platelet surface activated GPIIb–IIIa and platelet surface P-selectin expression) than ITP patients who had no bleeding (Fig. 5, left column); there was no difference in platelet activation levels with no added agonist. In contrast, AML/MDS patients who experienced bleeding had higher platelet activation levels with no added agonist and also higher platelet reactivity (e.g. higher TRAP-stimulated platelet surface activated GPIIb–IIIa and platelet surface P-selectin expression) than those who had AML/MDS but no bleeding (Fig. 5, right column).

Table 1.   Bleeding in ITP and AML/MDS patients
 ITP (n = 23)AML/MDS (n = 20)
  1. AML/MDS, acute myeloid leukemia/myelodysplasia; GI, gastrointestinal; ITP, immune thrombocytopenia.

Bleeding at any site, no. (%)17/23 (74)15/20 (75)
Grade 2 bleeding (any site), no. (%)3/23 (13)5/20 (25)
Skin or oral bleeding, no. (%)16/23 (70)14/20 (70)
Non-skin/oral bleeding, no. (%)2/23 (8)6/20 (30)
Description of non-skin/oral bleeding eventsGrade 1 GI (n = 1) Grade 1 epistaxis (n = 1)Grade 2 pulmonary (n = 1) Grade 2 GI (n = 1) Grade 1 epistaxis (n = 3) Grade 1 urinary (n = 1)
image

Figure 5.  Platelet function in patients with vs. without bleeding symptoms. Expression of platelet activation markers with and without ex vivo agonist stimulation in ITP patients (left column) and AML/MDS patients (right column) with vs. without bleeding symptoms. Dark bars: no bleeding (n = 6 for ITP, n = 5 for AML/MDS). Pale bars: bleeding any site (n = 17 for ITP, n = 15 for AML/MDS). * indicates P < 0.05; ** indicates P<0.01. Abbreviation: PSEL, P-selectin. MFI, mean fluorescence intensity.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Platelet count alone does not always predict bleeding in severely thrombocytopenic patients. Although major hemorrhage is more likely at platelet counts below 10 000 μL−1, significant bleeding may also occur at higher platelet counts, and the large majority of patients with even severe thrombocytopenia do not suffer from spontaneous bleeding [3,4]. Therefore, although severe thrombocytopenia is permissive for bleeding to occur, it is not sufficient.

In a study of patients with hematologic malignancy undergoing stem cell transplantation or chemotherapy, bleeding of World Health Organization grade 2 or above was found to occur at rates of 25% at platelet counts ≤ 5000 μL−1 and 17% at platelet counts 20–80 000 μL−1 [4]. In a study of pediatric ITP patients, although 75% of intracranial hemorrhages (ICHs) occurred at platelet counts ≤ 10 000 μL−1, 10% of ICHs occurred at platelet counts > 20 000 μL−1 [3]. External events, e.g. trauma or infection, and/or differences in platelet function may contribute to differences in bleeding risk among patients with comparable degrees of thrombocytopenia. Moreover, the types of bleeding may differ as well. Consistent with previous observations [16,20], in this study bleeding in ITP patients was characterized by petechiae, ecchymoses, and mouth bleeding, whereas AML/MDS patients more frequently experienced clinically significant gastrointestinal, genitourinary and pulmonary hemorrhage. A better understanding of platelet function in thrombocytopenia may be one important factor in the evaluation of bleeding risk and, if so, could lead to improved clinical management.

Few studies have examined platelet function in thrombocytopenic individuals. Studies by Karpatkin in 1978 [6] indicated that platelet aggregation was positively correlated with platelet volume in healthy individuals with normal platelet counts, implying that platelet function might be relatively high in ITP patients who have a high proportion of large, young platelets. Other groups have also reported a correlation between platelet size and/or platelet function with bleeding symptoms in thrombocytopenic patients [21,22]. Examining platelet surface P-selectin expression and platelet aggregation in ITP patients and healthy controls, Panzer et al. [14] reported that platelet P-selectin expression was higher in ITP patients than in healthy controls, and that this correlated with the size of platelet aggregates formed in vitro although not with bleeding symptoms. Furthermore, baseline platelet surface P-selectin expression ex vivo was found to be inversely correlated with its relative increase after exogenous addition of agonists, suggesting that in vivo activated platelets have reduced capacity for further activation [7]. However, these studies [7,14] used platelet-rich plasma and standardized the platelet count by dilution, both of which introduce the possibility of artefactual in vitro platelet activation. In contrast, in the present study, we examined platelets in whole blood without any separation or dilution of platelets, and did not agitate the blood during incubation, thereby minimizing the possibility of artefacts.

The main findings of the present study are: (i) platelets from AML/MDS patients were smaller, less immature, and had lower surface expression of GPIb and activated GPIIb–IIIa, both with and without addition of the agonists ADP or TRAP, than platelets from ITP patients with comparable degrees of thrombocytopenia; (ii) in both ITP and AML/MDS patients, increased platelet surface GPIb expression on circulating platelets and GPIb and activated GPIIb–IIIa expression on ex vivo activated platelets were associated with increased IPF; and (iii) AML/MDS patients with bleeding had higher TRAP-stimulated platelet surface activated GPIIb/IIIa and platelet surface P-selectin expression than AML/MDS patients with no bleeding, whereas ITP patients with bleeding had lower TRAP-stimulated platelet surface activated GPIIb–IIIa and platelet surface P-selectin expression than ITP patients with no bleeding.

How may these findings be interpreted? In ITP, platelet production is variable, as a result of the effects of autoantibody-mediated inhibition of platelet production [23,24]. Even when partially inhibited, ITP patients make more platelets than do AML/MDS patients, as reflected by higher IPF and platelet size in ITP patients than AML/MDS patients. The activation state of circulating platelets was also higher in ITP patients than in AML/MDS patients, and ITP platelets also showed higher reactivity in response to the agonists ADP and TRAP. The correlations between IPF and both baseline platelet expression of activation markers and platelet reactivity in response to agonist stimulation suggest that the younger, larger platelets typically found in greater numbers in ITP patients have greater function, as was first reported in the 1960s [5].

ITP patients with bleeding had lower platelet reactivity in response to agonists than those who had no bleeding, suggesting that platelets from ITP patients with bleeding symptoms have undergone a degree of in vivo activation and therefore have a lower capacity for further ex vivo agonist-stimulated activation. Paradoxically and unexpectedly, AML/MDS patients who experienced bleeding had significantly higher ex vivo agonist-stimulated platelet reactivity than those who had no bleeding. One interpretation of this finding is that ITP patients with bleeding may have antiplatelet antibodies that interfere with platelet function [25] or that the platelets are partially activated by the bleeding, whereas in AML/MDS patients, although their platelets still have the capacity to respond if stimulated ex vivo, they are not appropriately activated in vivo – suggesting that generation of platelet activators at sites of damaged blood vessels may be impaired or dysregulated. Moreover, other factors in addition to the platelet activation state are necessary to prevent bleeding in severe thrombocytopenia. These findings and hypotheses need to be confirmed and addressed in additional studies.

Platelet surface P-selectin expression is a marker of degranulation, and the platelet surface binding of PAC1 (the reporter mAb used in the present study) measures the conformational change in platelet surface GPIIb–IIIa that allows fibrinogen to bind and platelet aggregation to occur [11]. These two markers are therefore sensitive indicators of platelet functionality [11]. GPIb is the receptor for von Willebrand factor, and platelet surface expression of GPIb decreases upon activation, as the receptor is proteolyzed and/or internalized [26–28]. GPIb expression is strongly related to platelet surface area and therefore to platelet size. As PAC1 binding measures a conformational change in GPIIb–IIIa receptors rather than the absolute number of receptors, and P-selectin is not expressed on the surface of the resting platelet [11], the present findings for platelet surface activated GPIIb–IIIa and P-selectin expression in ITP patients are unlikely to be attributable to differences in platelet size.

Only a moderate difference was observed in bleeding symptoms between the patient groups, reflecting the relatively small number of patients with bleeding in this study. In addition to platelet function and count, other factors contribute to bleeding, including fever, hemoglobin level, sepsis, uremia, hypoalbuminemia, coagulopathy, leukopenia or leukocytosis, vascular integrity, and iatrogenic factors [2]. These factors were not explored in this study.

Bleeding in severely thrombocytopenic patients is potentially devastating, and platelets constitute a commonly transfused blood product that is often in short supply [29]. It has long been postulated that thrombocytopenia caused by bone marrow failure carries a higher risk of bleeding than the same degree of thrombocytopenia in ITP, although this hypothesis has never been specifically tested. Although flow cytometric assessment of platelet activation is unlikely to be widely available as a routine test in the clinical setting, a better understanding of bleeding risk in thrombocytopenia is likely to have major clinical implications. If bleeding can be anticipated, it may be more feasible to intervene early to ameliorate or prevent it. The present study is the first to examine and define differences in platelet function in patients with equivalent degrees of thrombocytopenia caused by ITP or AML/MDS. Further studies are needed to explore why the sites of bleeding are different in ITP and AML/MDS, how platelet activation/function is regulated in each patient group, e.g. the role of the endothelial cell, and which type of testing is optimal for the study of platelet function in each of these patient groups, with a view to predicting and preventing clinical bleeding.

Addendum

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

B. Psaila, J. B. Bussel, A. L. Frelinger, and A. D. Michelson: participated in study design, sample collection and processing, and data analysis, and co-wrote the manuscript; B. Babula and C. Tate: recruited patients and participated in data analysis; M. D. Linden, M. R. Barnard, and Y. Li: processed samples and participated in data analysis; E. J. Feldman: participated in study design and recruited patients for the study.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

The authors thank M. Lesser and M. Kline at the Biostatistics Unit, Feinstein Institute, North Shore LUK for assisting with statistical analysis and data interpretation. B. Psaila is the recipient of a Kay Kendall Leukaemia Fund traveling fellowship and a Fulbright Scholarship for Cancer Research. J. B. Bussel receives funding from NIH grant 1U01 HL72196-01-05.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

This study was supported in part by research funding from GSK and Sysmex America Incorporated. Bussel currently receives clinical research support from the following companies: Amgen, Cangene, GlaxoSmithKline, Genzyme, IgG of America, Immunomedics, Ligand, Eisai, Inc., Shionogi, and Sysmex. J. B. Bussel’s family owns stock in Amgen and GlaxoSmithKline, and he has participated in Advisory Boards and/or consults for Amgen, GlaxoSmithKline, Ligand, Shionogi, Eisai, and Portola. The services of M. Lesser and M. Fine were funded in part by GlaxoSmithKline.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  • 1
    Gaydos LA, Freireich EJ, Mantel N. The quantitative relation between platelet count and hemorrhage in patients with acute leukemia. N Engl J Med 1962; 266: 9059.
  • 2
    Friedmann AM, Sengul H, Lehmann H, Schwartz C, Goodman S. Do basic laboratory tests or clinical observations predict bleeding in thrombocytopenic oncology patients? A reevaluation of prophylactic platelet transfusions. Transfus Med Rev 2002; 16: 3445.
  • 3
    Psaila B, Petrovic A, Page LK, Menell J, Schonholz M, Bussel JB. Intracranial hemorrhage (ICH) in children with immune thrombocytopenia (ITP): study of 40 cases. Blood 2009; 114: 477783.
  • 4
    Slichter SJ, Kaufman RM, Assmann SF, McCullough J, Triulzi DJ, Strauss RG, Gernsheimer TB, Ness PM, Brecher ME, Josephson CD, Konkle BA, Woodson RD, Ortel TL, Hillyer CD, Skerrett DL, McCrae KR, Sloan SR, Uhl L, George JN, Aquino VM et al. Dose of prophylactic platelet transfusions and prevention of hemorrhage. N Engl J Med 2010; 362: 60013.
  • 5
    Karpatkin S. Heterogeneity of human platelets, II: functional evidence suggestive of young and old platelets. J Clin Invest 1969; 48: 10836.
  • 6
    Karpatkin S. Heterogeneity of human platelets. VI. Correlation of platelet function with platelet volume. Blood 1978; 51: 30716.
  • 7
    Panzer S, Hocker L, Rieger M, Vormittag R, Koren D, Dunkler D, Pabinger I. Agonist-inducible platelet activation in chronic idiopathic autoimmune thrombocytopenia. Eur J Haematol 2007; 79: 189204.
  • 8
    Godeau B, Provan D, Bussel J. Immune thrombocytopenic purpura in adults. Curr Opin Hematol 2007; 14: 53556.
  • 9
    Neunert C, Lim W, Crowther M, Cohen A, Solberg L Jr, Crowther M. The American Society of Hematology 2011 evidence-based practice guideline for immune thrombocytopenia. Blood 2011; 117: 4190207.
  • 10
    Stanworth SJ, Hyde C, Heddle N, Rebulla P, Brunskill S, Murphy MF. Prophylactic platelet transfusion for haemorrhage after chemotherapy and stem cell transplantation. Cochrane Database Syst Rev 2004; CD004269.
  • 11
    Michelson A, Linden MD, Barnard MR, Furman MI, Frelinger AL. Flow cytometry. In: Michelson A, ed. Platelets, 2nd edn. San Diego, CA: Elsevier/Academic Press, 2007: 54563.
  • 12
    Gurbel PA, Becker RC, Mann KG, Steinhubl SR, Michelson AD. Platelet function monitoring in patients with coronary artery disease. J Am Coll Cardiol 2007; 50: 182234.
  • 13
    Michelson AD. Antiplatelet therapies for the treatment of cardiovascular disease. Nat Rev Drug Discov 2010; 9: 15469.
  • 14
    Panzer S, Rieger M, Vormittag R, Eichelberger B, Dunkler D, Pabinger I. Platelet function to estimate the bleeding risk in autoimmune thrombocytopenia. Eur J Clin Invest 2007; 37: 81419.
  • 15
    Michelson AD. Evaluation of platelet function by flow cytometry. Pathophysiol Haemost Thromb 2006; 35: 6782.
  • 16
    Page LK, Psaila B, Provan D, Michael Hamilton J, Jenkins JM, Elish AS, Lesser ML, Bussel JB. The immune thrombocytopenic purpura (ITP) bleeding score: assessment of bleeding in patients with ITP. Br J Haematol 2007; 138: 2458.
  • 17
    Kestin AS, Valeri CR, Khuri SF, Loscalzo J, Ellis PA, MacGregor H, Birjiniuk V, Ouimet H, Pasche B, Nelson MJ. The platelet function defect of cardiopulmonary bypass. Blood 1993; 82: 10717.
  • 18
    Ruisi MM, Psaila B, Ward MJ, Villarica G, Bussel JB. Stability of measurement of the immature platelet fraction. Am J Hematol 2010; 85: 6224.
  • 19
    Shattil SJ, Hoxie JA, Cunningham M, Brass LF. Changes in the platelet membrane glycoprotein IIb.IIIa complex during platelet activation. J Biol Chem 1985; 260: 1110714.
  • 20
    Bolton-Maggs P. Severe bleeding in idiopathic thrombocytopenic purpura. J Pediatr Hematol Oncol 2003; 25(Suppl. 1): S4751.
  • 21
    Eldor A, Avitzour M, Or R, Hanna R, Penchas S. Prediction of haemorrhagic diathesis in thrombocytopenia by mean platelet volume. Br Med J (Clin Res Ed) 1982; 285: 397400.
  • 22
    Kenet G, Lubetsky A, Shenkman B, Tamarin I, Dardik R, Rechavi G, Barzilai A, Martinowitz U, Savion N, Varon D. Cone and platelet analyser (CPA): a new test for the prediction of bleeding among thrombocytopenic patients. Br J Haematol 1998; 101: 2559.
  • 23
    Chang M, Nakagawa PA, Williams SA, Schwartz MR, Imfeld KL, Buzby JS, Nugent DJ. Immune thrombocytopenic purpura (ITP) plasma and purified ITP monoclonal autoantibodies inhibit megakaryocytopoiesis in vitro. Blood 2003; 102: 88795.
  • 24
    McMillan R, Wang L, Tomer A, Nichol J, Pistillo J. Suppression of in vitro megakaryocyte production by antiplatelet autoantibodies from adult patients with chronic ITP. Blood 2004; 103: 13649.
  • 25
    Niessner H, Clemetson KJ, Panzer S, Mueller-Eckhardt C, Santoso S, Bettelheim P. Acquired thrombasthenia due to GPIIb/IIIa-specific platelet autoantibodies. Blood 1986; 68: 5716.
  • 26
    Han Y, Nurden A, Combrie R, Pasquet JM. Redistribution of glycoprotein Ib within platelets in response to protease-activated receptors 1 and 4: roles of cytoskeleton and calcium. J Thromb Haemost 2003; 1: 220615.
  • 27
    Michelson AD, Ellis PA, Barnard MR, Matic GB, Viles AF, Kestin AS. Downregulation of the platelet surface glycoprotein Ib–IX complex in whole blood stimulated by thrombin, adenosine diphosphate, or an in vivo wound. Blood 1991; 77: 7709.
  • 28
    LaRosa CA, Rohrer MJ, Benoit SE, Barnard MR, Michelson AD. Neutrophil cathepsin G modulates the platelet surface expression of the glycoprotein (GP) Ib–IX complex by proteolysis of the von Willebrand factor binding site on GPIb alpha and by a cytoskeletal-mediated redistribution of the remainder of the complex. Blood 1994; 84: 15868.
  • 29
    Stanworth SJ, Dyer C, Choo L, Bakrania L, Copplestone A, Llewelyn C, Norfolk D, Powter G, Littlewood T, Wood EM, Murphy MF. Do all patients with hematologic malignancies and severe thrombocytopenia need prophylactic platelet transfusions? Background, rationale, and design of a clinical trial (trial of platelet prophylaxis) to assess the effectiveness of prophylactic platelet transfusions. Transfus Med Rev 2010; 24: 16371.