European Journal of Haematology 92 (367–376)
REVIEW ARTICLE
An overview of platelet indices and methods for evaluating platelet function in thrombocytopenic patients Pernille J. Vinholt1, Anne-Mette Hvas2, Mads Nybo1 1
Department of Clinical Biochemistry and Pharmacology, Odense University Hospital, Odense; 2Department of Clinical Biochemistry, Aarhus University Hospital, Aarhus, Denmark
Abstract Thrombocytopenia is associated with bleeding risk. However, in thrombocytopenic patients, platelet count does not correlate with bleeding risk and other factors are thus likely to contribute to this risk. This review presents currently available platelet-related markers available on automated haematology analysers and commonly used methods for testing platelet function. The test principles, advantages and disadvantages of each test are described. We also evaluate the current literature regarding the clinical utility of the test for prediction of bleeding in thrombocytopenia in haematological and oncological diseases. We find that several platelet-related markers are available, but information about the clinical utility in thrombocytopenia is limited. Studies support that mean platelet volume (MPV) can aid diagnosing the cause of thrombocytopenia and low MPV may be associated with bleeding in thrombocytopenia. Flow cytometry, platelet aggregometry and platelet secretion tests are used to diagnose specific platelet function defects. The flow cytometric activation marker P-selectin and surface coverage by the Cone-and-Plate[let] analyser predict bleeding in selected thrombocytopenic populations. To fully uncover the clinical utility of platelet-related tests, information about the prevalence of platelet function defects in thrombocytopenic conditions is required. Finally, knowledge of the performance in thrombocytopenic samples from patients is essential. Key words thrombocytopenia; bleeding; platelet markers; haematology; oncology Correspondence Pernille J. Vinholt, Department of Clinical Biochemistry and Pharmacology, Odense University Hospital, Sdr. Boulevard 29, DK-5000 Odense C, Denmark. Tel: +45 3053 7246; Fax: +45 6541 1911; e-mail: [email protected] Accepted for publication 29 December 2013
Platelets are essential in primary haemostasis, and it is evident that a low platelet count is a significant risk factor for bleeding (1). Thus, bleeding is a frequently occurring complication and may be the cause of death in thrombocytopenic patients (2, 3), but not all patients with thrombocytopenia experience bleeding. In a study of almost 30 000 platelet counts obtained from haematological and oncological patients, clinically significant bleeding (WHO grade ≥2) was experienced on only 25% of days where platelet count was ≤5 9 109/L (3). When platelet count was between 6 and 80 9 109/L, the risk of bleeding was 17% and did not seem to correlate with platelet count (3). Likewise, there was no clear association between platelet count and bleeding found in other studies among haematological or oncological patients (4, 5). In other thrombocytopenic conditions such as liver cirrhosis or sepsis, the relation between platelet count and bleeding has only been sparsely investigated (6, 7).
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
doi:10.1111/ejh.12262
Thus, other factors obviously contribute to the risk of bleeding in thrombocytopenia. In this regard, the haemostatic capacity of platelets depends on both number and function. The aim of this review was therefore to describe the evidence for using platelet-related markers and methods for testing platelet function in assessment of spontaneous bleeding risk in thrombocytopenic patients. Automated haematology analysers Principle
Platelet count obtained by automated analysers deviates due to differences in the applied methodology and detection algorithms (8–10). The commonly used impedance method detects cells by increase in electrical impedance when a cell passes through an aperture in the flow cytometer. Platelets
367
Platelet indices and function in thrombocytopenia
are hereby detected by cell size because the increase in impedance is proportional to cell volume (11). For optical light scatter method, cells are also analysed based on volume when cells pass through a laser beam in the flow cytometer. If a two-angled method is used, light is two-dimensional permitting additional analysis of cell granularity (11). Hence, the commonly used impedance, but also the single-angle optical light scatter method, determines platelet number by counting cells within a specific size range. To differentiate platelets from non-platelet fragments and red blood cells, the analyser generates a (log-normal) distribution curve from the initial platelet histogram (Table 1) (11–13). Inaccuracies in the platelet count can occur if non-platelet particles with the same size as platelets, for example microcytes, interfere with platelet counting obtained by methods relying on cell size. This is critical in thrombocytopenic samples as the relative effect on the platelet count of misclassified cells increases as the platelet count decreases (13). Furthermore, large platelets or platelet clumps may not be adequately identified due to the lack of distinction between red or white blood cells and platelets. This might result in an underestimation of the platelet count (10, 13). The advantages of these methods are that they rapidly and at low cost measure platelet count. More accurate methods have been developed but are only available on a few analysers. It includes an immunological method with application of a platelet-specific antibody or fluorescent labelling of platelets prior to counting in the flow cytometer (11). Previously, also manual counting by phase contrast microscopy was often used, but this method is time-consuming and imprecise (9). Overall, automated haematology analysers overestimate platelet count in thrombocytopenic samples compared with the immunological reference method (8, 9). The coefficient of variation (CV) increases as platelet count decreases (8, 9). In a recent survey based on United Kingdom National External Quality Assessment Scheme, CV% reached 15–43% when platelet count was below 10 9 109/L (9). Other platelet indices can be derived from the platelet distribution curve obtained from impedance or optical methods. It includes mean platelet volume (MPV), platelet distribution width (PDW) and the fraction of large platelets. The MPV describes the average platelet size reported in femtolitre (fL) and is available on most haematology analysers (11, 12, 14). The PDW is a measure of the heterogeneity in platelet size either defined as the distribution width at 20% frequency level or calculated as the standard deviation of platelet volume divided by MPV 9 100 (12). Derived platelet indices are, however, highly specific to the individual technologies, with different analysers having different reference ranges (12). As a separate feature, the fraction of large platelets can be addressed specifically: On SysmexTM analysers (Sysmex, Kobe, Japan), the parameter is called platelet large cell ratio (P-LCR) and is the percentage of platelets larger than 12 fL (11, 12). On the ADVIATM analyser (Siemens Healthcare,
368
Vinholt et al.
Erlangen, Germany), the ratio of platelets higher than 20 fL is reported (11). Factors that affect platelet counting (interference from cells or cell fragments, inadequate detection of large platelets or platelet clumps) also influence platelet indices that are calculated from the platelet distribution curve. If red blood cells are misclassified as platelets, it causes an overestimation of MPV, a higher PDW and an increase in fraction of large cells. Moreover, in severe thrombocytopenia, it may not be possible to obtain a sufficient platelet distribution curve to calculate other platelet indices than the platelet count (12). There have been special concerns about the recommended anticoagulant for platelet counting, K2 or K3 ethylenediaminetetraacetic acid (EDTA), because it affects MPV. EDTA causes an increase in MPV – from 7.9% within 30 min to 13.4% over 24 h when measured by impedance (15) – and decreases by 10% when determined by an optical method (12, 14). Time delay probably also affects other variables, for example PDW and the fraction of large cells (14). Therefore, it is recommended to process the sample within 120 min when platelet indices are determined (16). EDTA may also cause agglutination of platelets, resulting in a falsely low platelet count (pseudo-thrombocytopenia). Each laboratory should, however, have strategies to detect spurious low platelet counts (13). Reticulated platelets can be determined by flow cytometry with a fluorescent dye that selectively binds RNA. Assessment of reticulated platelets is available only for SysmexTM haematology analysers and is called immature platelet fraction (IPF). The IPF is measured in a dedicated platelet chamber and is given as a percentage of the total platelet count or as an absolute number (12). Immature platelet fraction is not affected by the above-mentioned methodological issues and is stable for 48 h (17). Finally, the ADVIATM analyser reports a mean platelet component (MPC) based on the refractive index using an optical method (11). It corresponds to platelet density, and a decrease in density suggests platelet activation (18). Clinical utility of parameters related to platelet size
Large platelets have higher platelet granule content and greater extent of secretion, membrane protein activation and platelet aggregation in vitro (19, 20). Several clinical studies have documented an association between haemostasis and platelet size as they found MPV to be an independent risk factor for thrombosis (21), and large platelets are selectively consumed during massive bleeding (22, 23). However, only few studies have investigated the relationship between MPV and spontaneous bleeding in thrombocytopenia, and no data on MPC, PDW or the fraction of large platelets have been published. Eldor et al. (24) investigated 175 haematological patients with platelet counts below 20 9 109/L. They found that MPV was better than the platelet count for identifying bleeding phenotype. Unfortunately, it was not stated whether
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Vinholt et al.
Platelet indices and function in thrombocytopenia
Table 1 Characteristics and clinical utility of platelet markers Name of test
Principle
Advantages
Disadvantages
Clinical utility
Automated flow cytometric analysis, most often impedance
Simple, rapid Low volume
Diagnostic for thrombocytopenia No correlation with bleeding risk in thrombocytopenic patients in general (3–5)
Mean platelet volume, MPV
Derived from the platelet distribution curve
Simple, rapid Low volume Widely available
Low accuracy and high CV% at low platelet count Dependent on methodology, risk of interference or lack of detection of large platelets or platelet clumps Dependent on the platelet distribution curve
Platelet distribution width, PDW
Derived from the platelet distribution curve
Simple, rapid Low volume
Fraction of large cells
Derived from the platelet distribution curve
Simple, rapid Low volume
Immature platelet fraction, IPF
Staining of RNA in platelets (flow cytometry)
Simple, rapid, low volume No interference from other cells
Mean platelet component, MPC
Refractive index obtained by optical light scatter (flow cytometry)
Simple, rapid Low volume
Limited experience
Gold standard Widely used
Time-consuming Require experienced operator, sample preparation and large blood volume High CV% High CV%
Platelet indices Platelet count
Methods for platelet function testing Light transmission Change in turbidity due to aggregometry platelet aggregation in response to agonists
Whole blood aggregometry
Platelet secretion assays
Changes in impedance due to platelet aggregation to electrodes in response to agonists (i) Lumiaggregometry: Light transmission aggregometry or whole blood aggregometry combined with bioluminescence assay for detection of dense granule release (ii) Solitary assays: Detection of released substances or granule content
Whole blood Available as a point-of-care test High sensitivity for platelet secretion defects
Dependent on the platelet distribution curve Dependent on the platelet distribution curve Limited availability
For solitary assays, sample preparation is required and no common standard exists
High predictive value for bone marrow failure as a cause of thrombocytopenia (15, 25, 27–31) Low MPV may be associated with bleeding when platelet count is
REVIEW ARTICLE
An overview of platelet indices and methods for evaluating platelet function in thrombocytopenic patients Pernille J. Vinholt1, Anne-Mette Hvas2, Mads Nybo1 1
Department of Clinical Biochemistry and Pharmacology, Odense University Hospital, Odense; 2Department of Clinical Biochemistry, Aarhus University Hospital, Aarhus, Denmark
Abstract Thrombocytopenia is associated with bleeding risk. However, in thrombocytopenic patients, platelet count does not correlate with bleeding risk and other factors are thus likely to contribute to this risk. This review presents currently available platelet-related markers available on automated haematology analysers and commonly used methods for testing platelet function. The test principles, advantages and disadvantages of each test are described. We also evaluate the current literature regarding the clinical utility of the test for prediction of bleeding in thrombocytopenia in haematological and oncological diseases. We find that several platelet-related markers are available, but information about the clinical utility in thrombocytopenia is limited. Studies support that mean platelet volume (MPV) can aid diagnosing the cause of thrombocytopenia and low MPV may be associated with bleeding in thrombocytopenia. Flow cytometry, platelet aggregometry and platelet secretion tests are used to diagnose specific platelet function defects. The flow cytometric activation marker P-selectin and surface coverage by the Cone-and-Plate[let] analyser predict bleeding in selected thrombocytopenic populations. To fully uncover the clinical utility of platelet-related tests, information about the prevalence of platelet function defects in thrombocytopenic conditions is required. Finally, knowledge of the performance in thrombocytopenic samples from patients is essential. Key words thrombocytopenia; bleeding; platelet markers; haematology; oncology Correspondence Pernille J. Vinholt, Department of Clinical Biochemistry and Pharmacology, Odense University Hospital, Sdr. Boulevard 29, DK-5000 Odense C, Denmark. Tel: +45 3053 7246; Fax: +45 6541 1911; e-mail: [email protected] Accepted for publication 29 December 2013
Platelets are essential in primary haemostasis, and it is evident that a low platelet count is a significant risk factor for bleeding (1). Thus, bleeding is a frequently occurring complication and may be the cause of death in thrombocytopenic patients (2, 3), but not all patients with thrombocytopenia experience bleeding. In a study of almost 30 000 platelet counts obtained from haematological and oncological patients, clinically significant bleeding (WHO grade ≥2) was experienced on only 25% of days where platelet count was ≤5 9 109/L (3). When platelet count was between 6 and 80 9 109/L, the risk of bleeding was 17% and did not seem to correlate with platelet count (3). Likewise, there was no clear association between platelet count and bleeding found in other studies among haematological or oncological patients (4, 5). In other thrombocytopenic conditions such as liver cirrhosis or sepsis, the relation between platelet count and bleeding has only been sparsely investigated (6, 7).
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
doi:10.1111/ejh.12262
Thus, other factors obviously contribute to the risk of bleeding in thrombocytopenia. In this regard, the haemostatic capacity of platelets depends on both number and function. The aim of this review was therefore to describe the evidence for using platelet-related markers and methods for testing platelet function in assessment of spontaneous bleeding risk in thrombocytopenic patients. Automated haematology analysers Principle
Platelet count obtained by automated analysers deviates due to differences in the applied methodology and detection algorithms (8–10). The commonly used impedance method detects cells by increase in electrical impedance when a cell passes through an aperture in the flow cytometer. Platelets
367
Platelet indices and function in thrombocytopenia
are hereby detected by cell size because the increase in impedance is proportional to cell volume (11). For optical light scatter method, cells are also analysed based on volume when cells pass through a laser beam in the flow cytometer. If a two-angled method is used, light is two-dimensional permitting additional analysis of cell granularity (11). Hence, the commonly used impedance, but also the single-angle optical light scatter method, determines platelet number by counting cells within a specific size range. To differentiate platelets from non-platelet fragments and red blood cells, the analyser generates a (log-normal) distribution curve from the initial platelet histogram (Table 1) (11–13). Inaccuracies in the platelet count can occur if non-platelet particles with the same size as platelets, for example microcytes, interfere with platelet counting obtained by methods relying on cell size. This is critical in thrombocytopenic samples as the relative effect on the platelet count of misclassified cells increases as the platelet count decreases (13). Furthermore, large platelets or platelet clumps may not be adequately identified due to the lack of distinction between red or white blood cells and platelets. This might result in an underestimation of the platelet count (10, 13). The advantages of these methods are that they rapidly and at low cost measure platelet count. More accurate methods have been developed but are only available on a few analysers. It includes an immunological method with application of a platelet-specific antibody or fluorescent labelling of platelets prior to counting in the flow cytometer (11). Previously, also manual counting by phase contrast microscopy was often used, but this method is time-consuming and imprecise (9). Overall, automated haematology analysers overestimate platelet count in thrombocytopenic samples compared with the immunological reference method (8, 9). The coefficient of variation (CV) increases as platelet count decreases (8, 9). In a recent survey based on United Kingdom National External Quality Assessment Scheme, CV% reached 15–43% when platelet count was below 10 9 109/L (9). Other platelet indices can be derived from the platelet distribution curve obtained from impedance or optical methods. It includes mean platelet volume (MPV), platelet distribution width (PDW) and the fraction of large platelets. The MPV describes the average platelet size reported in femtolitre (fL) and is available on most haematology analysers (11, 12, 14). The PDW is a measure of the heterogeneity in platelet size either defined as the distribution width at 20% frequency level or calculated as the standard deviation of platelet volume divided by MPV 9 100 (12). Derived platelet indices are, however, highly specific to the individual technologies, with different analysers having different reference ranges (12). As a separate feature, the fraction of large platelets can be addressed specifically: On SysmexTM analysers (Sysmex, Kobe, Japan), the parameter is called platelet large cell ratio (P-LCR) and is the percentage of platelets larger than 12 fL (11, 12). On the ADVIATM analyser (Siemens Healthcare,
368
Vinholt et al.
Erlangen, Germany), the ratio of platelets higher than 20 fL is reported (11). Factors that affect platelet counting (interference from cells or cell fragments, inadequate detection of large platelets or platelet clumps) also influence platelet indices that are calculated from the platelet distribution curve. If red blood cells are misclassified as platelets, it causes an overestimation of MPV, a higher PDW and an increase in fraction of large cells. Moreover, in severe thrombocytopenia, it may not be possible to obtain a sufficient platelet distribution curve to calculate other platelet indices than the platelet count (12). There have been special concerns about the recommended anticoagulant for platelet counting, K2 or K3 ethylenediaminetetraacetic acid (EDTA), because it affects MPV. EDTA causes an increase in MPV – from 7.9% within 30 min to 13.4% over 24 h when measured by impedance (15) – and decreases by 10% when determined by an optical method (12, 14). Time delay probably also affects other variables, for example PDW and the fraction of large cells (14). Therefore, it is recommended to process the sample within 120 min when platelet indices are determined (16). EDTA may also cause agglutination of platelets, resulting in a falsely low platelet count (pseudo-thrombocytopenia). Each laboratory should, however, have strategies to detect spurious low platelet counts (13). Reticulated platelets can be determined by flow cytometry with a fluorescent dye that selectively binds RNA. Assessment of reticulated platelets is available only for SysmexTM haematology analysers and is called immature platelet fraction (IPF). The IPF is measured in a dedicated platelet chamber and is given as a percentage of the total platelet count or as an absolute number (12). Immature platelet fraction is not affected by the above-mentioned methodological issues and is stable for 48 h (17). Finally, the ADVIATM analyser reports a mean platelet component (MPC) based on the refractive index using an optical method (11). It corresponds to platelet density, and a decrease in density suggests platelet activation (18). Clinical utility of parameters related to platelet size
Large platelets have higher platelet granule content and greater extent of secretion, membrane protein activation and platelet aggregation in vitro (19, 20). Several clinical studies have documented an association between haemostasis and platelet size as they found MPV to be an independent risk factor for thrombosis (21), and large platelets are selectively consumed during massive bleeding (22, 23). However, only few studies have investigated the relationship between MPV and spontaneous bleeding in thrombocytopenia, and no data on MPC, PDW or the fraction of large platelets have been published. Eldor et al. (24) investigated 175 haematological patients with platelet counts below 20 9 109/L. They found that MPV was better than the platelet count for identifying bleeding phenotype. Unfortunately, it was not stated whether
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Vinholt et al.
Platelet indices and function in thrombocytopenia
Table 1 Characteristics and clinical utility of platelet markers Name of test
Principle
Advantages
Disadvantages
Clinical utility
Automated flow cytometric analysis, most often impedance
Simple, rapid Low volume
Diagnostic for thrombocytopenia No correlation with bleeding risk in thrombocytopenic patients in general (3–5)
Mean platelet volume, MPV
Derived from the platelet distribution curve
Simple, rapid Low volume Widely available
Low accuracy and high CV% at low platelet count Dependent on methodology, risk of interference or lack of detection of large platelets or platelet clumps Dependent on the platelet distribution curve
Platelet distribution width, PDW
Derived from the platelet distribution curve
Simple, rapid Low volume
Fraction of large cells
Derived from the platelet distribution curve
Simple, rapid Low volume
Immature platelet fraction, IPF
Staining of RNA in platelets (flow cytometry)
Simple, rapid, low volume No interference from other cells
Mean platelet component, MPC
Refractive index obtained by optical light scatter (flow cytometry)
Simple, rapid Low volume
Limited experience
Gold standard Widely used
Time-consuming Require experienced operator, sample preparation and large blood volume High CV% High CV%
Platelet indices Platelet count
Methods for platelet function testing Light transmission Change in turbidity due to aggregometry platelet aggregation in response to agonists
Whole blood aggregometry
Platelet secretion assays
Changes in impedance due to platelet aggregation to electrodes in response to agonists (i) Lumiaggregometry: Light transmission aggregometry or whole blood aggregometry combined with bioluminescence assay for detection of dense granule release (ii) Solitary assays: Detection of released substances or granule content
Whole blood Available as a point-of-care test High sensitivity for platelet secretion defects
Dependent on the platelet distribution curve Dependent on the platelet distribution curve Limited availability
For solitary assays, sample preparation is required and no common standard exists
High predictive value for bone marrow failure as a cause of thrombocytopenia (15, 25, 27–31) Low MPV may be associated with bleeding when platelet count is
