BLOOD COMPONENTS The hemostatic activity of cryopreserved platelets is mediated by phosphatidylserine-expressing platelets and platelet microparticles Lacey Johnson,1 Craig P. Coorey,1,2 and Denese C. Marks1
BACKGROUND: Cryopreservation of platelets (PLTs) at −80°C with dimethyl sulfoxide (DMSO) can extend the shelf life from 5 days to 2 years. Cryopreserved PLTs are reported to have a greater in vivo hemostatic effect than liquid-stored PLTs. As such, the aim of this study was to understand the mechanisms responsible for the hemostatic potential of cryopreserved PLTs and the contribution of the reconstitution solution to this activity. STUDY DESIGN AND METHODS: DMSO (5% final concentration) was added to buffy coat–derived PLTs, followed by prefreeze removal of DMSO and storage at −80°C. Cryopreserved PLTs (n = 8 per group) were thawed at 37°C, reconstituted with either 1 unit of thawed frozen plasma or PLT additive solution (PAS-G). In vitro assays were performed before freezing and after thawing to assess the hemostatic activity of PLTs. RESULTS: Cryopreserved PLTs expressed high levels of phosphatidylserine and contained significantly more phosphatidylserine-positive PLT microparticles than liquid-stored PLTs. This was accompanied by a significant decrease in the time to clot formation and clot strength, as measured by thromboelastography. The supernatant from cryopreserved PLTs was sufficient to reduce the phosphatidylserine-dependent clotting time and increase the thrombin generation potential. Overall, plasma-reconstituted cryopreserved PLTs were more procoagulant than those reconstituted in PAS-G. CONCLUSION: PLT cryopreservation results in the generation of phosphatidylserine-expressing PLT microparticles which contribute to the hemostatic activity. Understanding the hemostatic activity of these components may assist in extending the use of these specialized components beyond military applications.
T
o circumvent the problems associated with the short shelf life of platelets (PLTs), methods for PLT cryopreservation have been under investigation for more than 30 years.1-3 The most widely used cryopreservation protocol involves the addition of dimethyl sulfoxide (DMSO; at a final concentration of 4%-6%), followed by prefreeze removal of the DMSOcontaining supernatant and freezing as a hyperconcentrate in a low volume.4-7 After being thawed, PLTs can be reconstituted in plasma, saline, or PLT additive solutions (PASs), resulting in approximately 70% to 80% recovery.5-8 It is well established that cryopreserved PLTs exhibit an altered receptor phenotype, resulting in reduced in vitro aggregation to multiple agonists.5,6,8,9 Despite this,
ABBREVIATIONS: α-angle = clot growth; CAT = calibrated automated thrombogram; ETP = endogenous thrombin potential; K-time = speed of clot formation; MA = maximum amplitude; PAS(s) = platelet additive solution(s); R-time = time to clot formation; TEG = thromboelastography; ttPeak = time to peak. From 1Research and Development, The Australian Red Cross Blood Service; and the 2Sydney Medical School, University of Sydney, Sydney, NSW, Australia. Address correspondence to: Lacey Johnson, Research and Development, Australian Red Cross Blood Service, 17 O’Riordan Street, Alexandria, NSW 2015, Australia; e-mail: [email protected]. We acknowledge that the Australian Government fully funds the Australian Red Cross Blood Service for the provision of blood products and services to the Australian Community. This study was partly funded by the Defence Health Foundation. CPC was supported by a Summer Research Scholarship from the University of Sydney Medical School. Received for publication August 26, 2013; revision received December 11, 2013, and accepted December 12, 2013. doi: 10.1111/trf.12578 © 2014 Australian Red Cross Blood Service. Transfusion © 2014 AABB TRANSFUSION 2014;54:1917-1926. Volume 54, August 2014 TRANSFUSION
1917
JOHNSON ET AL.
cryopreserved PLTs have been shown to be more effective than liquid-preserved PLTs in restoring hemostasis and reducing nonsurgical blood loss in patients after cardiopulmonary bypass surgery.10 It has been hypothesized that the improved in vivo hemostatic function is due to cryopreserved PLTs having a procoagulant phenotype.11 PLT procoagulant activity is mediated by phosphatidylserine externalization on the outer cellular membrane, as well as through the formation of phosphatidylserine-expressing microparticles.12,13 The exposure of phosphatidylserine provides a site for the assembly of coagulation proteins, resulting in the rapid generation of thrombin and efficient blood clotting.14 The contribution of both phosphatidylserine-expressing PLTs and microparticles to global clot formation can be measured using coagulation assays sensitive to phospholipid, such as Procoag-PPL, thromboelastography (TEG), and thrombin generation assays (calibrated automated thrombogram [CAT]),15,16 which measure different aspects of coagulation. The hemostatic potential of PLTs relies not only on phosphatidylserine expression, but also additional proteins, such as clotting factors (Factor [F]VII, F IX, FX, and prothrombin), thromboxane, and tissue factor.17 These proteins are present in plasma and are also released directly from the granule stores of activated PLTs.18,19 As such, reconstituting cryopreserved PLTs in plasma or nonplasma solutions, such as PAS, will affect the available pool of procoagulant proteins. Further, the type of reconstitution solution influences PLT activation,6 potentially affecting the release of procoagulant proteins from granule stores. As such, the purpose of our study was to explore the hemostatic activity of cryopreserved PLTs and determine whether the reconstitution solution (plasma vs. PAS-G) affects this activity.
MATERIALS AND METHODS
between collection and processing. PLTs were prepared by pooling four ABO- and D-matched buffy coats with additive (SSP+, MacoPharma, Mouvaux, France). The PLTs were then separated by slow centrifugation (500 × g, 6 min) and leukoreduced (Imuguard III-S PL filter, Terumo BCT, Somerset, NJ), using an automatic blood separator (MacoPress Smart, MacoPharma), according to blood service procedures. Buffy coat–derived PLT units (n = 16) were frozen on Day 1. A sample was taken before freezing (prefreeze), by collection into the sampling pouch of the original storage bag. PLTs were frozen at −80°C as previously described,6 using DMSO at a final concentration of approximately 5%. For thawing, the cryopreserved PLT units were placed in a water bath at 37°C until they reached a temperature of 30°C (approx. 5 min). PLTs were reconstituted in either 1 unit of thawed deep frozen plasma (mean, 265 mL; frozen at −80°C and thawed to 30°C; n = 8) or 1 unit of PAS-G (225 mL; Pall Corporation, Covina, CA; n = 8), as described previously.6 Thawed and reconstituted PLTs were stored in the freezing bag (400-mL polyvinylchloride storage bag; Baxter Healthcare, Deerfield, IL) at 22°C for 6 hours without agitation, followed by agitation (Helmer, Inc., Noblesville, IN) for the remainder of storage. PLT quality was examined immediately after thawing (PT0) and again at 6 (PT6) and 24 hours (PT24) postthaw and compared to the matched prefreeze sample. PLT samples (10 mL) were removed using a sterile-docked sample pouch (Terumo Corporation, Tokyo, Japan). Before each sampling point the products were weighed and unit volumes were calculated by dividing the unit weight by the specific gravity of 1.01 for PLTs in SSP+ (prefreeze samples), 1.025 for PLTs in plasma (plasma postthaw samples), and 1.0086 for PLTs in PAS-G (PAS-G postthaw samples). All assays testing PLTs were performed immediately after sampling at each time point, while assays testing supernatant were performed in batches using samples that had been stored at −80°C.
Study design The effect of cryopreservation on the hemostatic activity of PLTs per se was performed as a paired study, with the same PLT units being tested before freezing, after thawing, and during postthaw storage. However, the comparison of the reconstitution solutions was an unpaired study design.
Preparation and cryopreservation of PLT concentrates This study had approval from the Australian Red Cross Blood Service Human Research Ethics Committee. All donations were from eligible, voluntary donors. Whole blood units (450 ± 45 mL) were collected (Day 0) into topand-bottom bags containing 63 mL of CPD (Fresenius Kabi, Bad Homburg v.d.h., Germany) and stored at 22°C 1918
TRANSFUSION Volume 54, August 2014
Laboratory analysis of PLTs and PLT microparticles The PLT count and mean PLT volume (MPV) were measured using a hematology analyzer (CellDyn Emerald, Abbott Laboratories, Abbott Park, IL). The pH was measured using a pH meter (SevenMulti, Mettler Toledo, Switzerland) at 22°C. Phosphatidylserine expression on PLTs was determined by annexin V staining and flow cytometry, as previously described.8 PLT microparticles were analyzed from the PLT concentrate without centrifugation to minimize the effects of sample manipulation.20 Beads of standard size 0.6, 1.0, and 3.0 μmol/L (Sigma-Aldrich, St Louis, MO) were used to set the gating scale for the forward light scatter parameter to allow separation of PLTs and
HEMOSTATIC ACTIVITY OF FROZEN PLTs
microparticles. A sample from the PLT component (5 μL) was diluted in annexin V binding buffer (BioLegend, San Diego, CA) and stained with CD61 (anti-human CD61-APC, Dako, Glostrup, Denmark) and annexin V (annexin V–fluorescein isothiocyanate; BioLegend) for 15 minutes at room temperature in the dark. The absolute number of PLT microparticles was assessed by flow cytometry (FACSCanto II, Becton Dickinson, Franklin Lakes, NJ) using tubes (TruCount, BD Biosciences, San Jose, CA), with 10,000 bead events collected. PLT microparticles were defined as events not more than 1.0 μmol/L in size that stained positive for both CD61 and phosphatidylserine (annexin V). PLT supernatant, containing microparticles, was collected by centrifugation at 1600 × g for 20 minutes and then 12,000 × g for 5 minutes at room temperature (22°C) and frozen in aliquots at −80°C until testing.
Coagulation analysis of PLTs and PLT microparticles For measurement of PLT clotting potential with a thromboelastogram (TEG 5000, Haemoscope Corporation, Niles, IL), the PLT concentration was adjusted to 200 × 109/L in plasma. Adjusted PLTs (1000 μL) were transferred to a kaolin-containing tube (Haemoscope Corporation) and mixed by inversion. Kaolin-activated PLTs (340 μL) were then added to a plain TEG cup (Haemoscope Corporation) containing 20 μL of calcium chloride (CaCl2; 0.2 mol/L; Haemoscope Corporation). The following TEG variables were measured at 37°C over approximately 60 minutes: R-time (time to clot initiation; min), maximum amplitude (MA; clot strength; mm), K-time (speed of clot formation; min), and α-angle (clot growth; degrees). The procoagulant activity of cryopreserved PLTs and microparticles was assessed using a procoagulant phospholipid kit (STA-Procoag-PPL kit, Diagnostica Stago Ltd, Asnieres, France) according to the manufacturer’s instructions. Briefly, a sample of the PLT concentrate or PLT supernatant (25 μL) was added to 25 μL Reagent 1 (containing human plasma depleted of procoagulant phospholipid; as a source of coagulation factors) and 100 μL Reagent 2 (containing calcium and activated FX; FXa). The combination of both FXa and phospholipid (with other required coagulation factors present in Reagent 1) results in the assembly of the prothrombinase complex and subsequent clot formation. The absence of phospholipid in the reagents makes the assay dependent on phospholipids present in the sample (either on the PLT surface or in the PLT supernatant, presumably attached to microparticles). Therefore, the higher the procoagulant phospholipid level in the sample, the faster the clotting time.21 Clot formation was analyzed using an automated coagulometer (STACompact, Diagnostica Stago Ltd) and measured in seconds.
Thrombin generation was measured using a CAT (Thrombinoscope BV, Maastricht, The Netherlands). In this assay, thrombin generation occurs in the presence of both phospholipid and tissue factor, which are present in either the PLT supernatant and/or the reagents. The PRP reagent (Thrombinoscope BV) contains 1 pmol/L tissue factor and is used to assess the presence of phospholipid in the sample, whereas the MP reagent (Thrombinoscope BV) contains 4 μmol/L phospholipids and is used to assess the presence of tissue factor in the sample. MP reagent was used in the presence of corn-trypsin inhibitor (40 μg/mL; EMD Chemicals, Inc., San Diego, CA) to inhibit contact activation. PLT supernatant (80 μL), prepared as described, was transferred to a microtiter plate (Immulon 2HB, Thermo Electron Corporation, Milford, MA), containing 20 μL of either PRP reagent or MP reagent and 20 μL of thrombin calibrator reagent (Thrombinoscope BV). The thrombin calibrator was performed in parallel with each sample to correct for individual differences in color of plasma, inner filter effect, and substrate consumption. Thrombin generation was initiated by the addition of 20 μL of fluorogenic thrombin substrate and calcium chloride (Thrombinoscope BV). The activity of thrombin converts the fluorogenic substrate into a fluorescent signal, which is measured over time, allowing the production of a thrombin generation curve. The fluorescence intensity was measured every 20 seconds for 2 hours at 37°C, using a microtiter plate reader (Fluoroskan Ascent, Thermo Fisher Scientific, Vantaa, Finland) at wavelengths of 390 nm (excitation filter) and 460 nm (emission filter). Thrombinoscope software (Version 5.0.0.715, Thrombinoscope BV) was used to calculate the following variables: the lag time, representing the time until initial thrombin (10 nmol/L) had formed (min); peak height (nmol/L thrombin); time to peak (ttPeak; min), representing the time point when the peak height was reached; the endogenous thrombin potential (ETP), reflecting the area under the curve (nmol/L thrombin × min). All samples were analyzed in triplicate. The presence of tissue factor released into the supernatant was detected by Western blotting. PLT supernatants were collected as described and separated by electrophoresis in a 4% to 20% polyacrylamide gel (BioRad, Hercules, CA) and transferred to a polyvinylidene fluoride membrane (ThermoFisher Scientific, Waltham, MA). Membranes were blocked with 5% skim milk powder in phosphate-buffered saline plus 0.05% Tween 20 (PBS-T) and incubated with anti-tissue factor primary antibody (1/1000; EMD Millipore Corporation, Temecula, CA), followed by anti-mouse immunoglobulin G–horseradish peroxidase secondary antibody (Cell Signaling Technology, Beverly, MA). Enhanced chemiluminescent (ECL) reagent (Bio-Rad) was used for detection, using a digital imager (LAS4000, GE Healthcare Life Sciences, Uppsala, Volume 54, August 2014 TRANSFUSION
1919
JOHNSON ET AL.
TABLE 1. Specifications of PLTs before freezing, upon freezing, and immediately after thawing and reconstitution in plasma or PAS-G* PLT concentrate Volume (mL) PLTs (×109/unit) MPV (fL) pH Deep frozen PLTs DMSO concentration (%) Volume (mL) Thawed PLTs (PT0) DMSO content (g/unit) Volume (mL) PLTs (×109/unit) Freeze/thaw recovery (%) pH MPV (fL) Thawed PLTs (PT6) PLTs (×109/unit) pH MPV (fL) Thawed PLTs (PT24) PLTs (×109/unit) pH MPV (fL)
30% plasma/70% SSP+ 363.3 ± 12.1 357.1 ± 10.1 296.7 ± 25.5 305.5 ± 32.8 5.3 ± 0.2 5.1 ± 0.2 7.1 ± 0.0 7.1 ± 0.0 Hyperconcentrated 4.6 ± 0.2 4.7 ± 0.2 13.7 ± 4.0 15.5 ± 3.8 Plasma PAS-G 0.6 ± 0.2 0.7 ± 0.2 285.4 ± 13.7 222.5 ± 4.8† 197.1 ± 25.7 209.6 ± 24.7 66.2 ± 4.1 68.8 ± 6.3 7.3 ± 0.1 6.8 ± 0.1† 5.3 ± 0.3 5.5 ± 0.1 Plasma PAS-G 214.0 ± 24.1 250.1 ± 29.8† 7.3 ± 0.1 6.8 ± 0.1† 5.8 ± 0.2 6.1 ± 0.2† Plasma PAS-G 166.9 ± 17.5 199.8 ± 32.1† 7.3 ± 0.1 6.9 ± 0.4† 6.0 ± 0.2 6.3 ± 0.1†
* Values shown as mean ± SD; n = 8 in each group. † p < 0.05 between plasma and PAS-G, determined using unpaired two-sided t test. PT0 = immediately postthaw; PT6 = 6 hours postthaw; PT24 = 24 hours postthaw.
Sweden). Three independent experiments were performed and a representative blot is shown.
Statistical analysis Results are expressed as mean ± standard deviation (SD). Data were analyzed using computer software (GraphPad Prism, Version 5.04; GraphPad, Inc., La Jolla, CA). Paired two-sided tests were performed to compare the characteristics of the PLT products before freezing and postthaw (Table 1). The interaction effect of reconstituting PLTs in each solution and subsequent postthaw storage was assessed using a two-way repeated-measures analysis of variance (ANOVA). Where a significant interaction was observed, post hoc paired two-sided t tests were performed to assess the differences between prefreeze and postthaw samples within each reconstitution solution group (plasma or PAS-G) and post hoc unpaired two-sided t-tests were performed to assess the differences between the reconstitution solution group (plasma or PAS-G) at each time point. A p value of less than 0.05 was considered to be significant for all analyses.
RESULTS Before freezing, the PLT components met targeted specifications, which were based on published literature4 and the components were similar in the plasma and PAS-G groups (Table 1). A mean of 5% DMSO was added to each 1920
TRANSFUSION Volume 54, August 2014
PLT unit and the volume of each PLT unit at the time of freezing was similar between the groups (Table 1). After thawing, reconstitution of the PLT units in plasma or PAS-G solutions resulted in differences in the product volume, MPV, and pH (Table 1), both between solutions and compared to prefreeze variables. All PLT units maintained a pH of higher than 6.4, although the plasma units had a higher pH than PAS-G units after reconstitution (Table 1). PLT recovery was approximately 70% and was not affected by the reconstitution solution (Table 1). Interestingly, the PLT count increased between PT0 and PT6, but then decreased during subsequent storage. However, the PAS-G group maintained a higher PLT count during postthaw storage than plasma reconstituted PLTs. PLT TEG variables (R-time, K-time, angle, MA) were measured before freezing and after thawing and reconstitution in either plasma or PAS-G (Fig. 1A). The TEG variables before freezing were not significantly different between the two groups (Figs. 1B-1E; prefreeze). Freezing and thawing resulted in a significant decrease in both the R-time (Fig. 1B) and the MA (Fig. 1C), regardless of the reconstitution solution. However, the PLTs reconstituted in plasma had significantly faster clot initiation (R-time) than PLTs reconstituted in PAS-G, immediately after thawing. Similarly, the PLTs reconstituted in plasma displayed higher clot strength (MA) after thawing and storage compared to the PAS-G–reconstituted PLTs. The K-time and angle did not significantly change as a result of freezing and thawing (Figs. 1D and 1E), although the PLTs reconstituted in plasma had a significantly lower K-time than the PAS-G group. A reduction in the R-time has previously been associated with PLT activation, specifically increased phosphatidylserine expression and the presence of microparticles.16,22 As such, the expression of phosphatidylserine on PLTs and PLT microparticles was assessed. The percentage of cells expressing phosphatidylserine increased significantly after cryopreservation (Fig. 2A; p < 0.0001). Further, the PLTs reconstituted in plasma displayed a gradual decline in phosphatidylserine expression throughout storage, which remained significantly lower than the PLTs reconstituted in PAS-G. The absolute number of phosphatidylserine-expressing PLT microparticles significantly increased immediately after thawing (Fig. 2B), regardless of reconstitution solution. During 6 hours of storage, the microparticle number increased, but after 24 hours, the number of microparticles had decreased. However, the PLTs reconstituted in PAS-G had a significantly higher number of microparticles at this time point (p = 0.003). To determine whether the presence of exposed phospholipid on either PLTs or microparticles specifically contributes to coagulation, a FXa-based phospholipiddependent clotting assay (Procoag-PPL) was used. The
HEMOSTATIC ACTIVITY OF FROZEN PLTs
the PRP reagent (containing 1 pmol/L tissue factor, but no phospholipid). The data demonstrate that supernatant from thawed PLTs reconstituted in plasma had very strong thrombin generation potential (Fig. 3A). The peak height and total ETP were significantly increased after thawing, while the lag time and ttPeak thrombin generation were significantly reduced (Table 2). However, there was little change in the thrombin generation potential during postthaw storage. Interestingly, the supernatant from PLTs reconstituted in PAS-G elicited minimal thrombin generation after thawing (Fig. 3B). This was most likely due to the absence of clotting factors in the PAS-G, as cryopreserved PLTs reconstituted in PAS-G that were subsequently centrifuged and reconstituted in plasma before analysis with CAT were able to generate thrombin to a similar level to plasma reconstituted PLTs (data not shown). Thrombin generation can also be triggered by tissue factor, which is reported to be expressed on PLT microparticles.18,24 Therefore, the contribution of tissue factor to thrombin generation Fig. 1. PLT cryopreservation affects the kinetics of clot formation. (A) Representative by PLT supernatants was also assessed, TEG tracings from PLTs before freezing (black line) and immediately after thawing using the MP reagent (containing (gray line) and reconstitution in plasma or PAS-G. The TEG variables (B) R-time, (C) 4 μmol/L phospholipids, but no tissue MA, (D) K-time, and (E) α-angle were examined at the indicated time points: before factor). The results demonstrate that freezing (prefreeze), immediately postthawing (PT0), 6 hours postthawing (PT6), and supernatants obtained from cryopre24 hours postthawing (PT24). The data represent mean ± SD (error bars). *p < 0.05 served PLTs reconstituted in plasma between the plasma ( ) and PAS-G (□) groups at the indicated time point; †p < 0.05 contain sufficient tissue factor to induce compared to plasma prefreeze samples; ‡p < 0.05 compared to PAS-G prefreeze thrombin generation (Fig. 3C). Howsamples. ever, the extent of thrombin generation was less and the time to formation was longer than those obtained using the PRP reagent overall procoagulant effect of the cryopreserved PLTs (Table 2). Interestingly, the supernatants from PLTs reconwas assessed using a sample of the whole PLT concentrate stituted in PAS-G were unable to elicit thrombin genera(containing PLTs and microparticles; Fig. 2C) as well as tion using the MP reagent (Table 2), even though tissue the supernatant alone (enriched for PLT microparticles; factor could be detected in the supernatant by Western Fig. 2D). Cryopreservation significantly reduced the blotting (Fig. 3D). clotting times, with plasma reconstitution leading to faster clotting than PAS-G reconstituted PLTs at PT0 and PT6, but not at 24 hours (Figs. 2C and 2D). Postthawing, DISCUSSION the use of supernatant alone resulted in similar clotting time to the whole PLT sample, suggesting that PLT In this study, PLTs were cryopreserved in 5% DMSO, frozen microparticles are important for phospholipid-dependent at −80°C, and reconstituted in plasma or PAS-G upon clotting. thawing. The data demonstrate that cryopreserved PLTs The presence of phosphatidylserine is also known to express high amounts of phosphatidylserine and generate trigger thrombin generation.23 As such, the phospholipidhigh numbers of phosphatidylserine-expressing microparticles. Further, we suggest that microparticles present induced thrombin generation potential of microparticles in the supernatant contribute to the hemostatic function in the cryopreserved PLT supernatant was examined using Volume 54, August 2014 TRANSFUSION
1921
JOHNSON ET AL.
defined in this study, by dual expression of CD61 and phosphatidylserine. Our results demonstrate that there is a progressive decline in phosphatidylserine expression on the PLT surface over postthaw storage (Fig. 2A); therefore, it may be possible that phosphatidylserine is also being lost from the microparticle surface, resulting in an artifactual reduction in microparticle number. Given that the presence of PLT microparticles in cryopreserved PLT units appears to contribute to the hemostatic activity of these products, further characterization and understanding of the microparticle life span during postthaw storage would be of interest. Multiple coagulation assays were used to demonstrate that the PLT Fig. 2. Phosphatidylserine expression on PLTs and microparticles contribute to clotmicroparticles, formed as a result of ting. Flow cytometry was used to assess (A) PLT phosphatidylserine expression cryopreservation, contribute to the (annexin V–positive cells) and (B) enumerate PLT microparticles. Clotting times global coagulation potential. Both TEG were measured with Procoag-PPL assay using a sample of the (C) PLT concentrate or and the CAT are reportedly useful for (D) PLT supernatant alone after cryopreservation and reconstitution in plasma or identifying bleeding and hypercoagulaPAS-G at the indicated time-points: before freezing (prefreeze), immediately bility states in patients,27,28 and our postthawing (PT0), 6 hours postthawing (PT6), and 24 hours postthawing (PT24). results suggest that cryopreserved PLTs The data represent mean ± SD (error bars). *p < 0.05 between the plasma ( ) and are hypercoagulable, as evidenced by a PAS-G groups (□) at the indicated time point; †p < 0.05 compared to plasma reduced R-time and increased thrombin prefreeze samples; ‡p < 0.05 compared to PAS-G prefreeze samples. generation potential (ETP and thrombin peak), compared to liquid-stored PLTs (prefreeze). Although the time to clot initiation was of the cryopreserved PLT product, resulting in increased reduced after cryopreservation, the rate of clot formation phosphatidylserine-dependent clotting and thrombin was not affected (K-time and α-angle). The R-time meageneration potential. It was also found that reconstituting sures the time to initial fibrin formation, which is influPLTs in plasma, rather than PAS-G, does not affect the enced by the presence of microparticles;16 therefore, number of microparticles formed, but does result in enhanced hemostatic in vitro activity, due to the presence the increased microparticle number induced by cryoof clotting factors and other procoagulant proteins in preservation may provide an explanation for the reduced plasma. R-time. However, the angle and K-time are mainly influPhosphatidylserine, present on PLTs and PLT enced by fibrinogen levels,29 which should be equivalent microparticles, supports normal coagulation through the due to adequate plasma levels (as the PLTs are diluted in assembly of the FXa and thrombin-generating coagulafresh plasma for testing). The clot strength (MA) was also tion enzyme complexes.14 It has also been suggested that reduced in the cryopreserved PLTs, which may be an effect of the altered PLT surface receptor expression,6 as inhibiPLT microparticles are up to 100-fold more procoagulant 25 than PLTs. This study confirmed that cryopreserved PLTs tion of GPIIb/IIIa has been shown to reduce the MA.30 express high levels of phosphatidylserine and that cryoThe results of the Procoag-PPL assay also confirm that preservation results in the production of high numbers phosphatidylserine-expressing microparticles are conof phosphatidylserine-expressing PLT microparticles. tributing to clot formation, as the supernatant of Despite the high amount of PLT microparticles generated cryopreserved PLTs was sufficient to reduce clotting. This in response to cryopreservation, the numbers decreased assay is a phospholipid-dependent coagulation assay and during storage at room temperature for 24 hours. This has previously been shown to be useful for the assessment result was counterintuitive, as microparticle numbers of PLT microparticle procoagulant activity.31 have been shown to increase during extended storage (14 This study was focused primarily on the role of days) of apheresis PLTs.26 An explanation for this result phosphatidylserine-expressing microparticles. However, we did find that a sufficient amount of tissue factor was may be due to the way that the microparticles have been 1922
TRANSFUSION Volume 54, August 2014
HEMOSTATIC ACTIVITY OF FROZEN PLTs
Fig. 3. PLT microparticles from cryopreserved PLT units contribute to thrombin generation. Thrombin generation was initiated with (A and B) 1 pmol/L tissue factor (PRP reagent) and (C) 4 μmol/L phospholipids (MP reagent) using supernatants from PLT concentrates collected before freezing (PF), immediately postthawing (PT0), 6 hours postthawing (PT6), and 24 hours postthawing (PT24). (D) Western blotting for tissue factor present in the supernatant of cryopreserved PLTs reconstituted in PAS-G.
present in the supernatant of cryopreserved PLTs reconstituted in plasma to initiate thrombin generation, albeit to a lesser degree than the phosphatidylserine-induced thrombin generation. This effect was not observed in PLTs reconstituted in PAS-G, despite tissue factor being detected in the supernatant by Western blotting. It would be of interest in future studies to assess the role of tissue factor in the hemostatic function of cryopreserved PLTs, as tissue factor is known to be expressed on the surface of activated PLTs and PLT microparticles.32 Further, tissue factor provides a major stimulus for clot formation in vivo.17 The reconstitution solution appears to affect the in vitro hemostatic activity of cryopreserved PLTs. Cryopreserved PLTs have previously been reconstituted in thawed frozen plasma, saline, or PAS, and differences in recovery and activation have been reported.5-8,33 We demonstrated that PLT recovery was similar between the solutions immediately after reconstitution, while PAS-G was
able to support the maintenance of a higher PLT count during postthaw storage. However, the phosphatidylserine expression and absolute microparticle number were similar between PLT units reconstituted in plasma and PAS-G. Despite this, differences were detected in the functional assays, with shorter clotting times (measured by both TEG and Procoag-PPL) and greater thrombin generation potential observed in the PLTs reconstituted in plasma. Further, greater clot strength (MA), faster initiation (R-time; at PT0) and rate of clotting (K-time) were observed in the PLT units reconstituted in plasma, compared to PAS-G. These differences are likely due to the presence of coagulation factors in the plasma that contribute to clotting,28,34 which are absent in the units reconstituted in PAS-G. Caution does need to be taken when interpreting the role of plasma in the hemostatic capacity of cryopreserved PLTs, as the amount of plasma varies between the different time points. Specifically, the prefreeze samples contain only 30% plasma compared to Volume 54, August 2014 TRANSFUSION
1923
JOHNSON ET AL.
TABLE 2. Thrombin generation potential of PLT supernatant induced by PRP or MP reagent* PRP reagent† Parameter and time tested Lag time (min) Prefreeze PT0 PT6 PT24 p value¶ Peak (nmol/L) Prefreeze PT0 PT6 PT24 p value¶ ttPeak (min) Prefreeze PT0 PT6 PT24 p value¶ ETP (nmol/L × min) Prefreeze PT0 PT6 PT24 p value¶
Plasma
MP reagent‡ PAS-G
Plasma
PAS-G
7.97 ± 0.41 3.00 ± 0.41§ 3.84 ± 0.58§ 4.46 ± 0.64§ p < 0.0001
7.76 ± 0.93 6.73 ± 0.74§|| 6.29 ± 0.62§|| 6.19 ± 0.63§||
64.79 ± 10.98 6.68 ± 1.10§ 10.84 ± 1.62§ 15.65 ± 2.62§ NA
54.87 ± 21.11 BDL BDL BDL
36.09 ± 4.52 341.92 ± 56.74§ 280.51 ± 47.30§ 230.88 ± 35.44§ p < 0.0001
38.37 ± 13.84 6.92 ± 2.87§|| 7.74 ± 2.37§|| 8.23 ± 2.73§||
9.41 ± 3.77 332.17 ± 50.09§ 276.87 ± 57.19§ 175.34 ± 50.10§ NA
11.16 ± 8.72 BDL BDL BDL
24.91 ± 1.17 5.43 ± 1.00§ 7.09 ± 1.32§ 8.37 ± 1.26§ p < 0.0001
25.39 ± 4.10 20.55 ± 3.83§|| 14.74 ± 2.65§|| 14.54 ± 1.72§||
94.29 ± 13.20§ 8.89 ± 1.65§ 13.74 ± 2.58§ 20.60 ± 4.46§ NA
86.37 ± 27.16 BDL BDL BDL
1163.33 ± 77.76 1693.71 ± 231.75§ 1539.81 ± 182.55§ 1415.57 ± 190.96§ p < 0.0001
1184.49 ± 186.94 216.01 ± 64.18§|| 256.01 ± 87.47§|| 272.25 ± 87.52§||
BDL 6.68 ± 1.10 10.84 ± 1.62 15.65 ± 2.62 NA
BDL BDL BDL BDL
* Values shown as mean ± SD; n = 7 in each group. † PRP reagent contains 1 pmol/L tissue factor, but no phospholipid. ‡ MP reagent contains 4 μmol/L phospholipid, but no tissue factor. § p < 0.05 between prefreeze and postthaw samples, determined by paired two-sided t test. || p < 0.05 between plasma and PAS-G, determined using unpaired two-sided t test. ¶ p value for overall interaction, determined by two-way repeated-measures ANOVA. BDL = below detectable limit; NA = not applicable; PT0 = immediately postthaw; PT6 = 6 hours postthaw; PT24 = 24 hours postthaw.
100% plasma in the samples postthawing, while the PLTs reconstituted in PAS-G contain no plasma. However, it is clear that cryopreservation results in an increased hemostatic capacity of PLTs regardless of the plasma content and that the presence of coagulation factors in plasma may provide further stimulus. Understanding the hemostatic capacity of cryopreserved PLTs may provide the knowledge to ensure the balance between hemostasis and thrombosis is maintained after transfusion. While this study, and others,9 suggests that cryopreserved PLTs may have greater hemostatic potential than liquid-stored PLTs, cryopreserved PLTs have also been shown to have reduced in vivo survival compared to fresh PLTs.5,35 The reduced survival may be due to clearance in the liver and/or spleen, as a result of the phenotypic alterations typical of cryopreserved PLTs, including increased cell surface expression of P-selectin and phosphatidylserine and loss of surface glycoproteins.35-37 Thus, the balance of hemostasis may be dependent on the clinical indication for the PLT transfusion as well as patient-specific factors. For example, during active bleeding, a procoagulant phenotype may increase the efficacy of the PLT transfusion as the cryopreservation process “primes” the PLTs, resulting in hastened clot formation. On the other hand, a 1924
TRANSFUSION Volume 54, August 2014
procoagulant PLT phenotype transfused to a nonbleeding patient, such as a prophylactic PLT transfusion, could lead to an increased risk of thrombotic complications.14 There is limited evidence to support the best clinical indication for the transfusion of cryopreserved PLTs. Cryopreserved PLTs have been used with great success in military operations since 2001, with more than 1000 units transfused to at least 333 patients.7,38 These trauma patients were likely to be actively bleeding, and a subset of them were massively transfused.7 Further, cryopreserved PLTs have been shown to effectively stem active bleeding in patients undergoing cardiothoracic surgery.10 On the other hand, cryopreserved PLTs have been transfused prophylactically,1,37 and although PLT increments were reported, it is not clear whether the PLTs were hemostatically active. Importantly, although serious transfusion reactions have not been reported in any of these studies, all were likely underpowered for safety outcomes, and events beyond the immediate transfusion period were not assessed. Another important consideration is that certain patient groups, such as those with cancer, cirrhosis, and other liver pathologies, may be more sensitive to thromboembolic complications.39-42 Therefore, prospective clinical studies are required to determine the efficacy of cryopreserved PLTs reconstituted in either plasma or PAS
HEMOSTATIC ACTIVITY OF FROZEN PLTs
in well-defined patient cohorts, if their utility is to be expanded beyond the treatment of battlefield injuries.
13. Key NS. Analysis of tissue factor positive microparticles. Thromb Res 2010;125(Suppl 1):S42-S5. 14. Keuren JF, Magdeleyns EJ, Govers-Riemslag JW, et al. Effects of storage-induced platelet microparticles on the initiation and propagation phase of blood coagulation. Br J
CONFLICT OF INTEREST
Haematol 2006;134:307-13.
The authors report no conflicts of interest or funding sources.
REFERENCES 1. Schiffer CA, Aisner J, Wiernik PH. Clinical experience with
15. Matijevic N, Wang YW, Kostousov V, et al. Decline in platelet microparticles contributes to reduced hemostatic potential of stored plasma. Thromb Res 2011;128:35-41. 16. Lawrie AS, Harrison P, Cardigan RA, et al. The characterization and impact of microparticles on haemostasis within fresh-frozen plasma. Vox Sang 2008;95:197-204.
transfusion of cryopreserved platelets. Br J Haematol 1976; 34:377-85. 2. Melaragno AJ, Carciero R, Feingold H, et al. Cryopreservation of human platelets using 6% dimethyl sulfoxide and storage at −80°. Vox Sang 1985;49:245-58. 3. Valeri CR, Ragno G, Khuri S. Freezing human platelets with 6 percent dimethyl sulfoxide with removal of the supernatant solution before freezing and storage at −80°C without postthaw processing. Transfusion 2005;45:1890-8. 4. Lelkens CC, Koning JG, De Kort B, et al. Experiences with frozen blood products in the Netherlands military. Transfus Apher Sci 2006;34:289-98. 5. Dumont LJ, Cancelas JA, Dumont DF, et al. A randomized
17. Owens AP, Mackman N. Microparticles in hemostasis and thrombosis. Circ Res 2011;108:1284-97. 18. Siddiqui FA, Desai H, Amirkhosravi A, et al. The presence and release of tissue factor from human platelets. Platelets 2002;13:247-53. 19. Blair P, Flaumenhaft R. Platelet α-granules: basic biology and clinical correlates. Blood Rev 2009;23:177-89. 20. Devine D, Serrano K, Levin E, et al. Flow cytometric detection and characterisation of platelet microparticles in platelet concentrates at expiry. Vox Sang 2012;103:124. 21. Exner T, Joseph J, Low J, et al. A new activated factor X-based clotting method with improved specificity for procoagulant phospholipid. Blood Coagul Fibrinolysis 2003;14:773-9.
controlled trial evaluating recovery and survival of 6% dimethyl sulfoxide–frozen autologous platelets in healthy volunteers. Transfusion 2013;53:128-37. 6. Johnson L, Reid S, Tan S, et al. PAS-G supports platelet reconstitution after cryopreservation in the absence of plasma. Transfusion 2013;53:2268-77. 7. Noorman F, Badloe J. −80°C frozen platelets, efficient logistics: available, compatible, safe and effective in the treatment of trauma patients with or without massive blood loss in military theatre. Transfusion 2012;52 Suppl: 33A. 8. Johnson LN, Winter KM, Reid S, et al. Cryopreservation of buffy-coat-derived platelet concentrates in dimethyl sulfoxide and platelet additive solution. Cryobiology 2011;62: 100-6. 9. Valeri CR, Macgregor H, Ragno G. Correlation between in vitro aggregation and thromboxane A2 production in fresh, liquid-preserved, and cryopreserved human platelets: effect of agonists, pH, and plasma and saline resuspension. Transfusion 2005;45:596-603. 10. Khuri S, Healey N, Macgregor H, et al. Comparison of the effects of transfusions of cryopreserved and liquidpreserved platelets on hemostasis and blood loss after cardiopulmonary bypass. J Thorac Cardiovasc Surg 1999;117: 172-83. 11. Barnard MR, Macgregor H, Ragno G, et al. Fresh, liquidpreserved, and cryopreserved platelets: adhesive surface receptors and membrane procoagulant activity. Transfusion 1999;39:880-8. 12. Freyssinet JM, Toti F. Formation of procoagulant microparticles and properties. Thromb Res 2010;125 Suppl 1:S46-8.
22. Valeri CR, Srey R, Tilahun D, et al. In vitro effects of poly-nacetyl glucosamine on the activation of platelets in platelet-rich plasma with and without red blood cells. J Trauma 2004;57:S22-5. 23. Hemker HC, Giesen P, Al Dieri R, et al. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb 2003;33:4-15. 24. Hellum M, Øvstebø R, Trøseid A, et al. Microparticleassociated tissue factor activity measured with the
25.
26.
27.
28. 29.
Zymuphen MP-TF kit and the calibrated automated thrombogram assay. Blood Coagul Fibrinolysis 2012;23: 520-6. Sinauridze EI, Kireev DA, Popenko NY, et al. Platelet microparticle membranes have 50- to 100-fold higher specific procoagulant activity than activated platelets. Thromb Haemost 2007;97:425-34. Nollet KE, Saito S, Ono T, et al. Microparticle formation in apheresis platelets is not affected by three leukoreduction filters. Transfusion 2013;53:2293-8. Ostrowski SR, Sørensen AM, Larsen CF, et al. Thrombelastography and biomarker profiles in acute coagulopathy of trauma: a prospective study. Scand J Trauma Resusc Emerg Med 2011;19:3-10. Al Dieri R, De Laat B, Hemker HC. Thrombin generation: what have we learned? Blood Rev 2012;26:197-203. Harr JN, Moore EE, Ghasabyan A, et al. Funtional fibrinogen assay indicates that fibrinogen is critical in correcting abnormal clot strength following trauma. Shock 2013;39: 45-9.
Volume 54, August 2014 TRANSFUSION
1925
JOHNSON ET AL.
30. Bowbrick VA, Mikhailidis DP, Stansby G. Value of thromboelastography in the assessment of platelet function. Clin Appl Thromb Hemost 2003;9:137-42. 31. Macey MG, Enniks N, Bevan S. Flow cytometric analysis of microparticle phenotype and their role in thrombin generation. Cytometry B Clin Cytom 2011;80B:57-63. 32. Mayne E, Funderburg NT, Sieg SF, et al. Increased platelet and microparticle activation in HIV infection: upregulation of p-selectin and tissue factor expression. J Acquir Immune Defic Syndr 2012;59:340-6. 33. Hornsey VS, McMillan L, Morrison A, et al. Freezing of buffy coat-derived, leukoreduced platelet concentrates in 6 percent dimethyl sulfoxide. Transfusion 2008;48:2508-14. 34. Berckmans RJ, Nieuwland R, Böing AN, et al. Cell-derived microparticles circulate in healthy humans and support low grade thrombin generation. Thromb Haemost 2001;85: 639-46. 35. Valeri CR, Giorgio A, Macgregor H, et al. Circulation and distribution of autotransfused fresh, liquid-preserved and cryopreserved baboon platelets. Vox Sang 2002;83:347-51. 36. Holme S, Sweeney JD, Sawyer S, et al. The expression of p-selectin during collection, processing, and storage of
1926
TRANSFUSION Volume 54, August 2014
platelet concentrates: relationship to loss of in vivo viability. Transfusion 1997;37:12-7. 37. Rumjantseva V, Hoffmeister KM. Novel and unexpected clearance mechanisms for cold platelets. Transfus Apher Sci 2010;42:63-70. 38. Badloe J, Noorman F. The Netherlands experience with frozen −80°C red cells, plasma and platelets in combat casualty care. Transfusion 2011;51 Suppl:24A. 39. Shah N, Northup P, Caldwell S. A clinical survey of bleeding, thrombosis, and blood product use in decompensated cirrhosis patients. Ann Hepatol 2012;11:686-90. 40. Falanga A, Marchetti M, Russo L. Venous thromboembolism in the hematologic malignancies. Curr Opin Oncol 2012;24:702-10. 41. Khorana A, Francis CW, Blumberg N, et al. Blood transfusions, thrombosis, and mortality in hospitalized patients with cancer. Arch Intern Med 2008;168:2377-81. 42. Van Doormaal F, Kleinjan A, Berckmans RJ, et al. Coagulation activation and microparticle-associated coagulant activity in cancer patients: an exploratory prospective study. Thromb Haemost 2012;108:160-5.
BACKGROUND: Cryopreservation of platelets (PLTs) at −80°C with dimethyl sulfoxide (DMSO) can extend the shelf life from 5 days to 2 years. Cryopreserved PLTs are reported to have a greater in vivo hemostatic effect than liquid-stored PLTs. As such, the aim of this study was to understand the mechanisms responsible for the hemostatic potential of cryopreserved PLTs and the contribution of the reconstitution solution to this activity. STUDY DESIGN AND METHODS: DMSO (5% final concentration) was added to buffy coat–derived PLTs, followed by prefreeze removal of DMSO and storage at −80°C. Cryopreserved PLTs (n = 8 per group) were thawed at 37°C, reconstituted with either 1 unit of thawed frozen plasma or PLT additive solution (PAS-G). In vitro assays were performed before freezing and after thawing to assess the hemostatic activity of PLTs. RESULTS: Cryopreserved PLTs expressed high levels of phosphatidylserine and contained significantly more phosphatidylserine-positive PLT microparticles than liquid-stored PLTs. This was accompanied by a significant decrease in the time to clot formation and clot strength, as measured by thromboelastography. The supernatant from cryopreserved PLTs was sufficient to reduce the phosphatidylserine-dependent clotting time and increase the thrombin generation potential. Overall, plasma-reconstituted cryopreserved PLTs were more procoagulant than those reconstituted in PAS-G. CONCLUSION: PLT cryopreservation results in the generation of phosphatidylserine-expressing PLT microparticles which contribute to the hemostatic activity. Understanding the hemostatic activity of these components may assist in extending the use of these specialized components beyond military applications.
T
o circumvent the problems associated with the short shelf life of platelets (PLTs), methods for PLT cryopreservation have been under investigation for more than 30 years.1-3 The most widely used cryopreservation protocol involves the addition of dimethyl sulfoxide (DMSO; at a final concentration of 4%-6%), followed by prefreeze removal of the DMSOcontaining supernatant and freezing as a hyperconcentrate in a low volume.4-7 After being thawed, PLTs can be reconstituted in plasma, saline, or PLT additive solutions (PASs), resulting in approximately 70% to 80% recovery.5-8 It is well established that cryopreserved PLTs exhibit an altered receptor phenotype, resulting in reduced in vitro aggregation to multiple agonists.5,6,8,9 Despite this,
ABBREVIATIONS: α-angle = clot growth; CAT = calibrated automated thrombogram; ETP = endogenous thrombin potential; K-time = speed of clot formation; MA = maximum amplitude; PAS(s) = platelet additive solution(s); R-time = time to clot formation; TEG = thromboelastography; ttPeak = time to peak. From 1Research and Development, The Australian Red Cross Blood Service; and the 2Sydney Medical School, University of Sydney, Sydney, NSW, Australia. Address correspondence to: Lacey Johnson, Research and Development, Australian Red Cross Blood Service, 17 O’Riordan Street, Alexandria, NSW 2015, Australia; e-mail: [email protected]. We acknowledge that the Australian Government fully funds the Australian Red Cross Blood Service for the provision of blood products and services to the Australian Community. This study was partly funded by the Defence Health Foundation. CPC was supported by a Summer Research Scholarship from the University of Sydney Medical School. Received for publication August 26, 2013; revision received December 11, 2013, and accepted December 12, 2013. doi: 10.1111/trf.12578 © 2014 Australian Red Cross Blood Service. Transfusion © 2014 AABB TRANSFUSION 2014;54:1917-1926. Volume 54, August 2014 TRANSFUSION
1917
JOHNSON ET AL.
cryopreserved PLTs have been shown to be more effective than liquid-preserved PLTs in restoring hemostasis and reducing nonsurgical blood loss in patients after cardiopulmonary bypass surgery.10 It has been hypothesized that the improved in vivo hemostatic function is due to cryopreserved PLTs having a procoagulant phenotype.11 PLT procoagulant activity is mediated by phosphatidylserine externalization on the outer cellular membrane, as well as through the formation of phosphatidylserine-expressing microparticles.12,13 The exposure of phosphatidylserine provides a site for the assembly of coagulation proteins, resulting in the rapid generation of thrombin and efficient blood clotting.14 The contribution of both phosphatidylserine-expressing PLTs and microparticles to global clot formation can be measured using coagulation assays sensitive to phospholipid, such as Procoag-PPL, thromboelastography (TEG), and thrombin generation assays (calibrated automated thrombogram [CAT]),15,16 which measure different aspects of coagulation. The hemostatic potential of PLTs relies not only on phosphatidylserine expression, but also additional proteins, such as clotting factors (Factor [F]VII, F IX, FX, and prothrombin), thromboxane, and tissue factor.17 These proteins are present in plasma and are also released directly from the granule stores of activated PLTs.18,19 As such, reconstituting cryopreserved PLTs in plasma or nonplasma solutions, such as PAS, will affect the available pool of procoagulant proteins. Further, the type of reconstitution solution influences PLT activation,6 potentially affecting the release of procoagulant proteins from granule stores. As such, the purpose of our study was to explore the hemostatic activity of cryopreserved PLTs and determine whether the reconstitution solution (plasma vs. PAS-G) affects this activity.
MATERIALS AND METHODS
between collection and processing. PLTs were prepared by pooling four ABO- and D-matched buffy coats with additive (SSP+, MacoPharma, Mouvaux, France). The PLTs were then separated by slow centrifugation (500 × g, 6 min) and leukoreduced (Imuguard III-S PL filter, Terumo BCT, Somerset, NJ), using an automatic blood separator (MacoPress Smart, MacoPharma), according to blood service procedures. Buffy coat–derived PLT units (n = 16) were frozen on Day 1. A sample was taken before freezing (prefreeze), by collection into the sampling pouch of the original storage bag. PLTs were frozen at −80°C as previously described,6 using DMSO at a final concentration of approximately 5%. For thawing, the cryopreserved PLT units were placed in a water bath at 37°C until they reached a temperature of 30°C (approx. 5 min). PLTs were reconstituted in either 1 unit of thawed deep frozen plasma (mean, 265 mL; frozen at −80°C and thawed to 30°C; n = 8) or 1 unit of PAS-G (225 mL; Pall Corporation, Covina, CA; n = 8), as described previously.6 Thawed and reconstituted PLTs were stored in the freezing bag (400-mL polyvinylchloride storage bag; Baxter Healthcare, Deerfield, IL) at 22°C for 6 hours without agitation, followed by agitation (Helmer, Inc., Noblesville, IN) for the remainder of storage. PLT quality was examined immediately after thawing (PT0) and again at 6 (PT6) and 24 hours (PT24) postthaw and compared to the matched prefreeze sample. PLT samples (10 mL) were removed using a sterile-docked sample pouch (Terumo Corporation, Tokyo, Japan). Before each sampling point the products were weighed and unit volumes were calculated by dividing the unit weight by the specific gravity of 1.01 for PLTs in SSP+ (prefreeze samples), 1.025 for PLTs in plasma (plasma postthaw samples), and 1.0086 for PLTs in PAS-G (PAS-G postthaw samples). All assays testing PLTs were performed immediately after sampling at each time point, while assays testing supernatant were performed in batches using samples that had been stored at −80°C.
Study design The effect of cryopreservation on the hemostatic activity of PLTs per se was performed as a paired study, with the same PLT units being tested before freezing, after thawing, and during postthaw storage. However, the comparison of the reconstitution solutions was an unpaired study design.
Preparation and cryopreservation of PLT concentrates This study had approval from the Australian Red Cross Blood Service Human Research Ethics Committee. All donations were from eligible, voluntary donors. Whole blood units (450 ± 45 mL) were collected (Day 0) into topand-bottom bags containing 63 mL of CPD (Fresenius Kabi, Bad Homburg v.d.h., Germany) and stored at 22°C 1918
TRANSFUSION Volume 54, August 2014
Laboratory analysis of PLTs and PLT microparticles The PLT count and mean PLT volume (MPV) were measured using a hematology analyzer (CellDyn Emerald, Abbott Laboratories, Abbott Park, IL). The pH was measured using a pH meter (SevenMulti, Mettler Toledo, Switzerland) at 22°C. Phosphatidylserine expression on PLTs was determined by annexin V staining and flow cytometry, as previously described.8 PLT microparticles were analyzed from the PLT concentrate without centrifugation to minimize the effects of sample manipulation.20 Beads of standard size 0.6, 1.0, and 3.0 μmol/L (Sigma-Aldrich, St Louis, MO) were used to set the gating scale for the forward light scatter parameter to allow separation of PLTs and
HEMOSTATIC ACTIVITY OF FROZEN PLTs
microparticles. A sample from the PLT component (5 μL) was diluted in annexin V binding buffer (BioLegend, San Diego, CA) and stained with CD61 (anti-human CD61-APC, Dako, Glostrup, Denmark) and annexin V (annexin V–fluorescein isothiocyanate; BioLegend) for 15 minutes at room temperature in the dark. The absolute number of PLT microparticles was assessed by flow cytometry (FACSCanto II, Becton Dickinson, Franklin Lakes, NJ) using tubes (TruCount, BD Biosciences, San Jose, CA), with 10,000 bead events collected. PLT microparticles were defined as events not more than 1.0 μmol/L in size that stained positive for both CD61 and phosphatidylserine (annexin V). PLT supernatant, containing microparticles, was collected by centrifugation at 1600 × g for 20 minutes and then 12,000 × g for 5 minutes at room temperature (22°C) and frozen in aliquots at −80°C until testing.
Coagulation analysis of PLTs and PLT microparticles For measurement of PLT clotting potential with a thromboelastogram (TEG 5000, Haemoscope Corporation, Niles, IL), the PLT concentration was adjusted to 200 × 109/L in plasma. Adjusted PLTs (1000 μL) were transferred to a kaolin-containing tube (Haemoscope Corporation) and mixed by inversion. Kaolin-activated PLTs (340 μL) were then added to a plain TEG cup (Haemoscope Corporation) containing 20 μL of calcium chloride (CaCl2; 0.2 mol/L; Haemoscope Corporation). The following TEG variables were measured at 37°C over approximately 60 minutes: R-time (time to clot initiation; min), maximum amplitude (MA; clot strength; mm), K-time (speed of clot formation; min), and α-angle (clot growth; degrees). The procoagulant activity of cryopreserved PLTs and microparticles was assessed using a procoagulant phospholipid kit (STA-Procoag-PPL kit, Diagnostica Stago Ltd, Asnieres, France) according to the manufacturer’s instructions. Briefly, a sample of the PLT concentrate or PLT supernatant (25 μL) was added to 25 μL Reagent 1 (containing human plasma depleted of procoagulant phospholipid; as a source of coagulation factors) and 100 μL Reagent 2 (containing calcium and activated FX; FXa). The combination of both FXa and phospholipid (with other required coagulation factors present in Reagent 1) results in the assembly of the prothrombinase complex and subsequent clot formation. The absence of phospholipid in the reagents makes the assay dependent on phospholipids present in the sample (either on the PLT surface or in the PLT supernatant, presumably attached to microparticles). Therefore, the higher the procoagulant phospholipid level in the sample, the faster the clotting time.21 Clot formation was analyzed using an automated coagulometer (STACompact, Diagnostica Stago Ltd) and measured in seconds.
Thrombin generation was measured using a CAT (Thrombinoscope BV, Maastricht, The Netherlands). In this assay, thrombin generation occurs in the presence of both phospholipid and tissue factor, which are present in either the PLT supernatant and/or the reagents. The PRP reagent (Thrombinoscope BV) contains 1 pmol/L tissue factor and is used to assess the presence of phospholipid in the sample, whereas the MP reagent (Thrombinoscope BV) contains 4 μmol/L phospholipids and is used to assess the presence of tissue factor in the sample. MP reagent was used in the presence of corn-trypsin inhibitor (40 μg/mL; EMD Chemicals, Inc., San Diego, CA) to inhibit contact activation. PLT supernatant (80 μL), prepared as described, was transferred to a microtiter plate (Immulon 2HB, Thermo Electron Corporation, Milford, MA), containing 20 μL of either PRP reagent or MP reagent and 20 μL of thrombin calibrator reagent (Thrombinoscope BV). The thrombin calibrator was performed in parallel with each sample to correct for individual differences in color of plasma, inner filter effect, and substrate consumption. Thrombin generation was initiated by the addition of 20 μL of fluorogenic thrombin substrate and calcium chloride (Thrombinoscope BV). The activity of thrombin converts the fluorogenic substrate into a fluorescent signal, which is measured over time, allowing the production of a thrombin generation curve. The fluorescence intensity was measured every 20 seconds for 2 hours at 37°C, using a microtiter plate reader (Fluoroskan Ascent, Thermo Fisher Scientific, Vantaa, Finland) at wavelengths of 390 nm (excitation filter) and 460 nm (emission filter). Thrombinoscope software (Version 5.0.0.715, Thrombinoscope BV) was used to calculate the following variables: the lag time, representing the time until initial thrombin (10 nmol/L) had formed (min); peak height (nmol/L thrombin); time to peak (ttPeak; min), representing the time point when the peak height was reached; the endogenous thrombin potential (ETP), reflecting the area under the curve (nmol/L thrombin × min). All samples were analyzed in triplicate. The presence of tissue factor released into the supernatant was detected by Western blotting. PLT supernatants were collected as described and separated by electrophoresis in a 4% to 20% polyacrylamide gel (BioRad, Hercules, CA) and transferred to a polyvinylidene fluoride membrane (ThermoFisher Scientific, Waltham, MA). Membranes were blocked with 5% skim milk powder in phosphate-buffered saline plus 0.05% Tween 20 (PBS-T) and incubated with anti-tissue factor primary antibody (1/1000; EMD Millipore Corporation, Temecula, CA), followed by anti-mouse immunoglobulin G–horseradish peroxidase secondary antibody (Cell Signaling Technology, Beverly, MA). Enhanced chemiluminescent (ECL) reagent (Bio-Rad) was used for detection, using a digital imager (LAS4000, GE Healthcare Life Sciences, Uppsala, Volume 54, August 2014 TRANSFUSION
1919
JOHNSON ET AL.
TABLE 1. Specifications of PLTs before freezing, upon freezing, and immediately after thawing and reconstitution in plasma or PAS-G* PLT concentrate Volume (mL) PLTs (×109/unit) MPV (fL) pH Deep frozen PLTs DMSO concentration (%) Volume (mL) Thawed PLTs (PT0) DMSO content (g/unit) Volume (mL) PLTs (×109/unit) Freeze/thaw recovery (%) pH MPV (fL) Thawed PLTs (PT6) PLTs (×109/unit) pH MPV (fL) Thawed PLTs (PT24) PLTs (×109/unit) pH MPV (fL)
30% plasma/70% SSP+ 363.3 ± 12.1 357.1 ± 10.1 296.7 ± 25.5 305.5 ± 32.8 5.3 ± 0.2 5.1 ± 0.2 7.1 ± 0.0 7.1 ± 0.0 Hyperconcentrated 4.6 ± 0.2 4.7 ± 0.2 13.7 ± 4.0 15.5 ± 3.8 Plasma PAS-G 0.6 ± 0.2 0.7 ± 0.2 285.4 ± 13.7 222.5 ± 4.8† 197.1 ± 25.7 209.6 ± 24.7 66.2 ± 4.1 68.8 ± 6.3 7.3 ± 0.1 6.8 ± 0.1† 5.3 ± 0.3 5.5 ± 0.1 Plasma PAS-G 214.0 ± 24.1 250.1 ± 29.8† 7.3 ± 0.1 6.8 ± 0.1† 5.8 ± 0.2 6.1 ± 0.2† Plasma PAS-G 166.9 ± 17.5 199.8 ± 32.1† 7.3 ± 0.1 6.9 ± 0.4† 6.0 ± 0.2 6.3 ± 0.1†
* Values shown as mean ± SD; n = 8 in each group. † p < 0.05 between plasma and PAS-G, determined using unpaired two-sided t test. PT0 = immediately postthaw; PT6 = 6 hours postthaw; PT24 = 24 hours postthaw.
Sweden). Three independent experiments were performed and a representative blot is shown.
Statistical analysis Results are expressed as mean ± standard deviation (SD). Data were analyzed using computer software (GraphPad Prism, Version 5.04; GraphPad, Inc., La Jolla, CA). Paired two-sided tests were performed to compare the characteristics of the PLT products before freezing and postthaw (Table 1). The interaction effect of reconstituting PLTs in each solution and subsequent postthaw storage was assessed using a two-way repeated-measures analysis of variance (ANOVA). Where a significant interaction was observed, post hoc paired two-sided t tests were performed to assess the differences between prefreeze and postthaw samples within each reconstitution solution group (plasma or PAS-G) and post hoc unpaired two-sided t-tests were performed to assess the differences between the reconstitution solution group (plasma or PAS-G) at each time point. A p value of less than 0.05 was considered to be significant for all analyses.
RESULTS Before freezing, the PLT components met targeted specifications, which were based on published literature4 and the components were similar in the plasma and PAS-G groups (Table 1). A mean of 5% DMSO was added to each 1920
TRANSFUSION Volume 54, August 2014
PLT unit and the volume of each PLT unit at the time of freezing was similar between the groups (Table 1). After thawing, reconstitution of the PLT units in plasma or PAS-G solutions resulted in differences in the product volume, MPV, and pH (Table 1), both between solutions and compared to prefreeze variables. All PLT units maintained a pH of higher than 6.4, although the plasma units had a higher pH than PAS-G units after reconstitution (Table 1). PLT recovery was approximately 70% and was not affected by the reconstitution solution (Table 1). Interestingly, the PLT count increased between PT0 and PT6, but then decreased during subsequent storage. However, the PAS-G group maintained a higher PLT count during postthaw storage than plasma reconstituted PLTs. PLT TEG variables (R-time, K-time, angle, MA) were measured before freezing and after thawing and reconstitution in either plasma or PAS-G (Fig. 1A). The TEG variables before freezing were not significantly different between the two groups (Figs. 1B-1E; prefreeze). Freezing and thawing resulted in a significant decrease in both the R-time (Fig. 1B) and the MA (Fig. 1C), regardless of the reconstitution solution. However, the PLTs reconstituted in plasma had significantly faster clot initiation (R-time) than PLTs reconstituted in PAS-G, immediately after thawing. Similarly, the PLTs reconstituted in plasma displayed higher clot strength (MA) after thawing and storage compared to the PAS-G–reconstituted PLTs. The K-time and angle did not significantly change as a result of freezing and thawing (Figs. 1D and 1E), although the PLTs reconstituted in plasma had a significantly lower K-time than the PAS-G group. A reduction in the R-time has previously been associated with PLT activation, specifically increased phosphatidylserine expression and the presence of microparticles.16,22 As such, the expression of phosphatidylserine on PLTs and PLT microparticles was assessed. The percentage of cells expressing phosphatidylserine increased significantly after cryopreservation (Fig. 2A; p < 0.0001). Further, the PLTs reconstituted in plasma displayed a gradual decline in phosphatidylserine expression throughout storage, which remained significantly lower than the PLTs reconstituted in PAS-G. The absolute number of phosphatidylserine-expressing PLT microparticles significantly increased immediately after thawing (Fig. 2B), regardless of reconstitution solution. During 6 hours of storage, the microparticle number increased, but after 24 hours, the number of microparticles had decreased. However, the PLTs reconstituted in PAS-G had a significantly higher number of microparticles at this time point (p = 0.003). To determine whether the presence of exposed phospholipid on either PLTs or microparticles specifically contributes to coagulation, a FXa-based phospholipiddependent clotting assay (Procoag-PPL) was used. The
HEMOSTATIC ACTIVITY OF FROZEN PLTs
the PRP reagent (containing 1 pmol/L tissue factor, but no phospholipid). The data demonstrate that supernatant from thawed PLTs reconstituted in plasma had very strong thrombin generation potential (Fig. 3A). The peak height and total ETP were significantly increased after thawing, while the lag time and ttPeak thrombin generation were significantly reduced (Table 2). However, there was little change in the thrombin generation potential during postthaw storage. Interestingly, the supernatant from PLTs reconstituted in PAS-G elicited minimal thrombin generation after thawing (Fig. 3B). This was most likely due to the absence of clotting factors in the PAS-G, as cryopreserved PLTs reconstituted in PAS-G that were subsequently centrifuged and reconstituted in plasma before analysis with CAT were able to generate thrombin to a similar level to plasma reconstituted PLTs (data not shown). Thrombin generation can also be triggered by tissue factor, which is reported to be expressed on PLT microparticles.18,24 Therefore, the contribution of tissue factor to thrombin generation Fig. 1. PLT cryopreservation affects the kinetics of clot formation. (A) Representative by PLT supernatants was also assessed, TEG tracings from PLTs before freezing (black line) and immediately after thawing using the MP reagent (containing (gray line) and reconstitution in plasma or PAS-G. The TEG variables (B) R-time, (C) 4 μmol/L phospholipids, but no tissue MA, (D) K-time, and (E) α-angle were examined at the indicated time points: before factor). The results demonstrate that freezing (prefreeze), immediately postthawing (PT0), 6 hours postthawing (PT6), and supernatants obtained from cryopre24 hours postthawing (PT24). The data represent mean ± SD (error bars). *p < 0.05 served PLTs reconstituted in plasma between the plasma ( ) and PAS-G (□) groups at the indicated time point; †p < 0.05 contain sufficient tissue factor to induce compared to plasma prefreeze samples; ‡p < 0.05 compared to PAS-G prefreeze thrombin generation (Fig. 3C). Howsamples. ever, the extent of thrombin generation was less and the time to formation was longer than those obtained using the PRP reagent overall procoagulant effect of the cryopreserved PLTs (Table 2). Interestingly, the supernatants from PLTs reconwas assessed using a sample of the whole PLT concentrate stituted in PAS-G were unable to elicit thrombin genera(containing PLTs and microparticles; Fig. 2C) as well as tion using the MP reagent (Table 2), even though tissue the supernatant alone (enriched for PLT microparticles; factor could be detected in the supernatant by Western Fig. 2D). Cryopreservation significantly reduced the blotting (Fig. 3D). clotting times, with plasma reconstitution leading to faster clotting than PAS-G reconstituted PLTs at PT0 and PT6, but not at 24 hours (Figs. 2C and 2D). Postthawing, DISCUSSION the use of supernatant alone resulted in similar clotting time to the whole PLT sample, suggesting that PLT In this study, PLTs were cryopreserved in 5% DMSO, frozen microparticles are important for phospholipid-dependent at −80°C, and reconstituted in plasma or PAS-G upon clotting. thawing. The data demonstrate that cryopreserved PLTs The presence of phosphatidylserine is also known to express high amounts of phosphatidylserine and generate trigger thrombin generation.23 As such, the phospholipidhigh numbers of phosphatidylserine-expressing microparticles. Further, we suggest that microparticles present induced thrombin generation potential of microparticles in the supernatant contribute to the hemostatic function in the cryopreserved PLT supernatant was examined using Volume 54, August 2014 TRANSFUSION
1921
JOHNSON ET AL.
defined in this study, by dual expression of CD61 and phosphatidylserine. Our results demonstrate that there is a progressive decline in phosphatidylserine expression on the PLT surface over postthaw storage (Fig. 2A); therefore, it may be possible that phosphatidylserine is also being lost from the microparticle surface, resulting in an artifactual reduction in microparticle number. Given that the presence of PLT microparticles in cryopreserved PLT units appears to contribute to the hemostatic activity of these products, further characterization and understanding of the microparticle life span during postthaw storage would be of interest. Multiple coagulation assays were used to demonstrate that the PLT Fig. 2. Phosphatidylserine expression on PLTs and microparticles contribute to clotmicroparticles, formed as a result of ting. Flow cytometry was used to assess (A) PLT phosphatidylserine expression cryopreservation, contribute to the (annexin V–positive cells) and (B) enumerate PLT microparticles. Clotting times global coagulation potential. Both TEG were measured with Procoag-PPL assay using a sample of the (C) PLT concentrate or and the CAT are reportedly useful for (D) PLT supernatant alone after cryopreservation and reconstitution in plasma or identifying bleeding and hypercoagulaPAS-G at the indicated time-points: before freezing (prefreeze), immediately bility states in patients,27,28 and our postthawing (PT0), 6 hours postthawing (PT6), and 24 hours postthawing (PT24). results suggest that cryopreserved PLTs The data represent mean ± SD (error bars). *p < 0.05 between the plasma ( ) and are hypercoagulable, as evidenced by a PAS-G groups (□) at the indicated time point; †p < 0.05 compared to plasma reduced R-time and increased thrombin prefreeze samples; ‡p < 0.05 compared to PAS-G prefreeze samples. generation potential (ETP and thrombin peak), compared to liquid-stored PLTs (prefreeze). Although the time to clot initiation was of the cryopreserved PLT product, resulting in increased reduced after cryopreservation, the rate of clot formation phosphatidylserine-dependent clotting and thrombin was not affected (K-time and α-angle). The R-time meageneration potential. It was also found that reconstituting sures the time to initial fibrin formation, which is influPLTs in plasma, rather than PAS-G, does not affect the enced by the presence of microparticles;16 therefore, number of microparticles formed, but does result in enhanced hemostatic in vitro activity, due to the presence the increased microparticle number induced by cryoof clotting factors and other procoagulant proteins in preservation may provide an explanation for the reduced plasma. R-time. However, the angle and K-time are mainly influPhosphatidylserine, present on PLTs and PLT enced by fibrinogen levels,29 which should be equivalent microparticles, supports normal coagulation through the due to adequate plasma levels (as the PLTs are diluted in assembly of the FXa and thrombin-generating coagulafresh plasma for testing). The clot strength (MA) was also tion enzyme complexes.14 It has also been suggested that reduced in the cryopreserved PLTs, which may be an effect of the altered PLT surface receptor expression,6 as inhibiPLT microparticles are up to 100-fold more procoagulant 25 than PLTs. This study confirmed that cryopreserved PLTs tion of GPIIb/IIIa has been shown to reduce the MA.30 express high levels of phosphatidylserine and that cryoThe results of the Procoag-PPL assay also confirm that preservation results in the production of high numbers phosphatidylserine-expressing microparticles are conof phosphatidylserine-expressing PLT microparticles. tributing to clot formation, as the supernatant of Despite the high amount of PLT microparticles generated cryopreserved PLTs was sufficient to reduce clotting. This in response to cryopreservation, the numbers decreased assay is a phospholipid-dependent coagulation assay and during storage at room temperature for 24 hours. This has previously been shown to be useful for the assessment result was counterintuitive, as microparticle numbers of PLT microparticle procoagulant activity.31 have been shown to increase during extended storage (14 This study was focused primarily on the role of days) of apheresis PLTs.26 An explanation for this result phosphatidylserine-expressing microparticles. However, we did find that a sufficient amount of tissue factor was may be due to the way that the microparticles have been 1922
TRANSFUSION Volume 54, August 2014
HEMOSTATIC ACTIVITY OF FROZEN PLTs
Fig. 3. PLT microparticles from cryopreserved PLT units contribute to thrombin generation. Thrombin generation was initiated with (A and B) 1 pmol/L tissue factor (PRP reagent) and (C) 4 μmol/L phospholipids (MP reagent) using supernatants from PLT concentrates collected before freezing (PF), immediately postthawing (PT0), 6 hours postthawing (PT6), and 24 hours postthawing (PT24). (D) Western blotting for tissue factor present in the supernatant of cryopreserved PLTs reconstituted in PAS-G.
present in the supernatant of cryopreserved PLTs reconstituted in plasma to initiate thrombin generation, albeit to a lesser degree than the phosphatidylserine-induced thrombin generation. This effect was not observed in PLTs reconstituted in PAS-G, despite tissue factor being detected in the supernatant by Western blotting. It would be of interest in future studies to assess the role of tissue factor in the hemostatic function of cryopreserved PLTs, as tissue factor is known to be expressed on the surface of activated PLTs and PLT microparticles.32 Further, tissue factor provides a major stimulus for clot formation in vivo.17 The reconstitution solution appears to affect the in vitro hemostatic activity of cryopreserved PLTs. Cryopreserved PLTs have previously been reconstituted in thawed frozen plasma, saline, or PAS, and differences in recovery and activation have been reported.5-8,33 We demonstrated that PLT recovery was similar between the solutions immediately after reconstitution, while PAS-G was
able to support the maintenance of a higher PLT count during postthaw storage. However, the phosphatidylserine expression and absolute microparticle number were similar between PLT units reconstituted in plasma and PAS-G. Despite this, differences were detected in the functional assays, with shorter clotting times (measured by both TEG and Procoag-PPL) and greater thrombin generation potential observed in the PLTs reconstituted in plasma. Further, greater clot strength (MA), faster initiation (R-time; at PT0) and rate of clotting (K-time) were observed in the PLT units reconstituted in plasma, compared to PAS-G. These differences are likely due to the presence of coagulation factors in the plasma that contribute to clotting,28,34 which are absent in the units reconstituted in PAS-G. Caution does need to be taken when interpreting the role of plasma in the hemostatic capacity of cryopreserved PLTs, as the amount of plasma varies between the different time points. Specifically, the prefreeze samples contain only 30% plasma compared to Volume 54, August 2014 TRANSFUSION
1923
JOHNSON ET AL.
TABLE 2. Thrombin generation potential of PLT supernatant induced by PRP or MP reagent* PRP reagent† Parameter and time tested Lag time (min) Prefreeze PT0 PT6 PT24 p value¶ Peak (nmol/L) Prefreeze PT0 PT6 PT24 p value¶ ttPeak (min) Prefreeze PT0 PT6 PT24 p value¶ ETP (nmol/L × min) Prefreeze PT0 PT6 PT24 p value¶
Plasma
MP reagent‡ PAS-G
Plasma
PAS-G
7.97 ± 0.41 3.00 ± 0.41§ 3.84 ± 0.58§ 4.46 ± 0.64§ p < 0.0001
7.76 ± 0.93 6.73 ± 0.74§|| 6.29 ± 0.62§|| 6.19 ± 0.63§||
64.79 ± 10.98 6.68 ± 1.10§ 10.84 ± 1.62§ 15.65 ± 2.62§ NA
54.87 ± 21.11 BDL BDL BDL
36.09 ± 4.52 341.92 ± 56.74§ 280.51 ± 47.30§ 230.88 ± 35.44§ p < 0.0001
38.37 ± 13.84 6.92 ± 2.87§|| 7.74 ± 2.37§|| 8.23 ± 2.73§||
9.41 ± 3.77 332.17 ± 50.09§ 276.87 ± 57.19§ 175.34 ± 50.10§ NA
11.16 ± 8.72 BDL BDL BDL
24.91 ± 1.17 5.43 ± 1.00§ 7.09 ± 1.32§ 8.37 ± 1.26§ p < 0.0001
25.39 ± 4.10 20.55 ± 3.83§|| 14.74 ± 2.65§|| 14.54 ± 1.72§||
94.29 ± 13.20§ 8.89 ± 1.65§ 13.74 ± 2.58§ 20.60 ± 4.46§ NA
86.37 ± 27.16 BDL BDL BDL
1163.33 ± 77.76 1693.71 ± 231.75§ 1539.81 ± 182.55§ 1415.57 ± 190.96§ p < 0.0001
1184.49 ± 186.94 216.01 ± 64.18§|| 256.01 ± 87.47§|| 272.25 ± 87.52§||
BDL 6.68 ± 1.10 10.84 ± 1.62 15.65 ± 2.62 NA
BDL BDL BDL BDL
* Values shown as mean ± SD; n = 7 in each group. † PRP reagent contains 1 pmol/L tissue factor, but no phospholipid. ‡ MP reagent contains 4 μmol/L phospholipid, but no tissue factor. § p < 0.05 between prefreeze and postthaw samples, determined by paired two-sided t test. || p < 0.05 between plasma and PAS-G, determined using unpaired two-sided t test. ¶ p value for overall interaction, determined by two-way repeated-measures ANOVA. BDL = below detectable limit; NA = not applicable; PT0 = immediately postthaw; PT6 = 6 hours postthaw; PT24 = 24 hours postthaw.
100% plasma in the samples postthawing, while the PLTs reconstituted in PAS-G contain no plasma. However, it is clear that cryopreservation results in an increased hemostatic capacity of PLTs regardless of the plasma content and that the presence of coagulation factors in plasma may provide further stimulus. Understanding the hemostatic capacity of cryopreserved PLTs may provide the knowledge to ensure the balance between hemostasis and thrombosis is maintained after transfusion. While this study, and others,9 suggests that cryopreserved PLTs may have greater hemostatic potential than liquid-stored PLTs, cryopreserved PLTs have also been shown to have reduced in vivo survival compared to fresh PLTs.5,35 The reduced survival may be due to clearance in the liver and/or spleen, as a result of the phenotypic alterations typical of cryopreserved PLTs, including increased cell surface expression of P-selectin and phosphatidylserine and loss of surface glycoproteins.35-37 Thus, the balance of hemostasis may be dependent on the clinical indication for the PLT transfusion as well as patient-specific factors. For example, during active bleeding, a procoagulant phenotype may increase the efficacy of the PLT transfusion as the cryopreservation process “primes” the PLTs, resulting in hastened clot formation. On the other hand, a 1924
TRANSFUSION Volume 54, August 2014
procoagulant PLT phenotype transfused to a nonbleeding patient, such as a prophylactic PLT transfusion, could lead to an increased risk of thrombotic complications.14 There is limited evidence to support the best clinical indication for the transfusion of cryopreserved PLTs. Cryopreserved PLTs have been used with great success in military operations since 2001, with more than 1000 units transfused to at least 333 patients.7,38 These trauma patients were likely to be actively bleeding, and a subset of them were massively transfused.7 Further, cryopreserved PLTs have been shown to effectively stem active bleeding in patients undergoing cardiothoracic surgery.10 On the other hand, cryopreserved PLTs have been transfused prophylactically,1,37 and although PLT increments were reported, it is not clear whether the PLTs were hemostatically active. Importantly, although serious transfusion reactions have not been reported in any of these studies, all were likely underpowered for safety outcomes, and events beyond the immediate transfusion period were not assessed. Another important consideration is that certain patient groups, such as those with cancer, cirrhosis, and other liver pathologies, may be more sensitive to thromboembolic complications.39-42 Therefore, prospective clinical studies are required to determine the efficacy of cryopreserved PLTs reconstituted in either plasma or PAS
HEMOSTATIC ACTIVITY OF FROZEN PLTs
in well-defined patient cohorts, if their utility is to be expanded beyond the treatment of battlefield injuries.
13. Key NS. Analysis of tissue factor positive microparticles. Thromb Res 2010;125(Suppl 1):S42-S5. 14. Keuren JF, Magdeleyns EJ, Govers-Riemslag JW, et al. Effects of storage-induced platelet microparticles on the initiation and propagation phase of blood coagulation. Br J
CONFLICT OF INTEREST
Haematol 2006;134:307-13.
The authors report no conflicts of interest or funding sources.
REFERENCES 1. Schiffer CA, Aisner J, Wiernik PH. Clinical experience with
15. Matijevic N, Wang YW, Kostousov V, et al. Decline in platelet microparticles contributes to reduced hemostatic potential of stored plasma. Thromb Res 2011;128:35-41. 16. Lawrie AS, Harrison P, Cardigan RA, et al. The characterization and impact of microparticles on haemostasis within fresh-frozen plasma. Vox Sang 2008;95:197-204.
transfusion of cryopreserved platelets. Br J Haematol 1976; 34:377-85. 2. Melaragno AJ, Carciero R, Feingold H, et al. Cryopreservation of human platelets using 6% dimethyl sulfoxide and storage at −80°. Vox Sang 1985;49:245-58. 3. Valeri CR, Ragno G, Khuri S. Freezing human platelets with 6 percent dimethyl sulfoxide with removal of the supernatant solution before freezing and storage at −80°C without postthaw processing. Transfusion 2005;45:1890-8. 4. Lelkens CC, Koning JG, De Kort B, et al. Experiences with frozen blood products in the Netherlands military. Transfus Apher Sci 2006;34:289-98. 5. Dumont LJ, Cancelas JA, Dumont DF, et al. A randomized
17. Owens AP, Mackman N. Microparticles in hemostasis and thrombosis. Circ Res 2011;108:1284-97. 18. Siddiqui FA, Desai H, Amirkhosravi A, et al. The presence and release of tissue factor from human platelets. Platelets 2002;13:247-53. 19. Blair P, Flaumenhaft R. Platelet α-granules: basic biology and clinical correlates. Blood Rev 2009;23:177-89. 20. Devine D, Serrano K, Levin E, et al. Flow cytometric detection and characterisation of platelet microparticles in platelet concentrates at expiry. Vox Sang 2012;103:124. 21. Exner T, Joseph J, Low J, et al. A new activated factor X-based clotting method with improved specificity for procoagulant phospholipid. Blood Coagul Fibrinolysis 2003;14:773-9.
controlled trial evaluating recovery and survival of 6% dimethyl sulfoxide–frozen autologous platelets in healthy volunteers. Transfusion 2013;53:128-37. 6. Johnson L, Reid S, Tan S, et al. PAS-G supports platelet reconstitution after cryopreservation in the absence of plasma. Transfusion 2013;53:2268-77. 7. Noorman F, Badloe J. −80°C frozen platelets, efficient logistics: available, compatible, safe and effective in the treatment of trauma patients with or without massive blood loss in military theatre. Transfusion 2012;52 Suppl: 33A. 8. Johnson LN, Winter KM, Reid S, et al. Cryopreservation of buffy-coat-derived platelet concentrates in dimethyl sulfoxide and platelet additive solution. Cryobiology 2011;62: 100-6. 9. Valeri CR, Macgregor H, Ragno G. Correlation between in vitro aggregation and thromboxane A2 production in fresh, liquid-preserved, and cryopreserved human platelets: effect of agonists, pH, and plasma and saline resuspension. Transfusion 2005;45:596-603. 10. Khuri S, Healey N, Macgregor H, et al. Comparison of the effects of transfusions of cryopreserved and liquidpreserved platelets on hemostasis and blood loss after cardiopulmonary bypass. J Thorac Cardiovasc Surg 1999;117: 172-83. 11. Barnard MR, Macgregor H, Ragno G, et al. Fresh, liquidpreserved, and cryopreserved platelets: adhesive surface receptors and membrane procoagulant activity. Transfusion 1999;39:880-8. 12. Freyssinet JM, Toti F. Formation of procoagulant microparticles and properties. Thromb Res 2010;125 Suppl 1:S46-8.
22. Valeri CR, Srey R, Tilahun D, et al. In vitro effects of poly-nacetyl glucosamine on the activation of platelets in platelet-rich plasma with and without red blood cells. J Trauma 2004;57:S22-5. 23. Hemker HC, Giesen P, Al Dieri R, et al. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol Haemost Thromb 2003;33:4-15. 24. Hellum M, Øvstebø R, Trøseid A, et al. Microparticleassociated tissue factor activity measured with the
25.
26.
27.
28. 29.
Zymuphen MP-TF kit and the calibrated automated thrombogram assay. Blood Coagul Fibrinolysis 2012;23: 520-6. Sinauridze EI, Kireev DA, Popenko NY, et al. Platelet microparticle membranes have 50- to 100-fold higher specific procoagulant activity than activated platelets. Thromb Haemost 2007;97:425-34. Nollet KE, Saito S, Ono T, et al. Microparticle formation in apheresis platelets is not affected by three leukoreduction filters. Transfusion 2013;53:2293-8. Ostrowski SR, Sørensen AM, Larsen CF, et al. Thrombelastography and biomarker profiles in acute coagulopathy of trauma: a prospective study. Scand J Trauma Resusc Emerg Med 2011;19:3-10. Al Dieri R, De Laat B, Hemker HC. Thrombin generation: what have we learned? Blood Rev 2012;26:197-203. Harr JN, Moore EE, Ghasabyan A, et al. Funtional fibrinogen assay indicates that fibrinogen is critical in correcting abnormal clot strength following trauma. Shock 2013;39: 45-9.
Volume 54, August 2014 TRANSFUSION
1925
JOHNSON ET AL.
30. Bowbrick VA, Mikhailidis DP, Stansby G. Value of thromboelastography in the assessment of platelet function. Clin Appl Thromb Hemost 2003;9:137-42. 31. Macey MG, Enniks N, Bevan S. Flow cytometric analysis of microparticle phenotype and their role in thrombin generation. Cytometry B Clin Cytom 2011;80B:57-63. 32. Mayne E, Funderburg NT, Sieg SF, et al. Increased platelet and microparticle activation in HIV infection: upregulation of p-selectin and tissue factor expression. J Acquir Immune Defic Syndr 2012;59:340-6. 33. Hornsey VS, McMillan L, Morrison A, et al. Freezing of buffy coat-derived, leukoreduced platelet concentrates in 6 percent dimethyl sulfoxide. Transfusion 2008;48:2508-14. 34. Berckmans RJ, Nieuwland R, Böing AN, et al. Cell-derived microparticles circulate in healthy humans and support low grade thrombin generation. Thromb Haemost 2001;85: 639-46. 35. Valeri CR, Giorgio A, Macgregor H, et al. Circulation and distribution of autotransfused fresh, liquid-preserved and cryopreserved baboon platelets. Vox Sang 2002;83:347-51. 36. Holme S, Sweeney JD, Sawyer S, et al. The expression of p-selectin during collection, processing, and storage of
1926
TRANSFUSION Volume 54, August 2014
platelet concentrates: relationship to loss of in vivo viability. Transfusion 1997;37:12-7. 37. Rumjantseva V, Hoffmeister KM. Novel and unexpected clearance mechanisms for cold platelets. Transfus Apher Sci 2010;42:63-70. 38. Badloe J, Noorman F. The Netherlands experience with frozen −80°C red cells, plasma and platelets in combat casualty care. Transfusion 2011;51 Suppl:24A. 39. Shah N, Northup P, Caldwell S. A clinical survey of bleeding, thrombosis, and blood product use in decompensated cirrhosis patients. Ann Hepatol 2012;11:686-90. 40. Falanga A, Marchetti M, Russo L. Venous thromboembolism in the hematologic malignancies. Curr Opin Oncol 2012;24:702-10. 41. Khorana A, Francis CW, Blumberg N, et al. Blood transfusions, thrombosis, and mortality in hospitalized patients with cancer. Arch Intern Med 2008;168:2377-81. 42. Van Doormaal F, Kleinjan A, Berckmans RJ, et al. Coagulation activation and microparticle-associated coagulant activity in cancer patients: an exploratory prospective study. Thromb Haemost 2012;108:160-5.
