Influence of Residual Platelet Count on Routine Coagulation, Factor VIII, and Factor IX Testing in Postfreeze-Thaw Samples

Transcription

1 834 Influence of Residual Platelet Count on Routine Coagulation, Factor VIII, and Factor IX Testing in Postfreeze-Thaw Samples Giuseppe Lippi, MD 1 Rossana Rossi, PhD 1 Luigi Ippolito, MD 1 Valentina Zobbi, PhD 1 Donata Azzi, PhD 1 Silvia Pipitone, MD 1 Emmanuel J. Favaloro, PhD, FFSc (RCPA) 2 Dorothy M. Adcock Funk, MD 3 1 Department of Pathology and Laboratory Medicine, Laboratory of Clinical Chemistry and Hematology, Academic Hospital of Parma, Parma, Italy 2 Department of Haematology, Institute of Clinical Pathology and Medical Research (ICPMR), Pathology West, Westmead Hospital, Westmead,NewSouthWales,Australia 3 Esoterix Inc., Englewood, Colorado Address for correspondence Giuseppe Lippi, MD, U.O. Diagnostica Ematochimica, Dipartimento di Patologia e Medicina di Laboratorio, Azienda Ospedaliero-Universitaria di Parma, Via Gramsci, 14, Parma, Italy ( glippi@ao.pr.it; ulippi@tin.it). Semin Thromb Hemost 2013;39: Abstract Keywords preanalytical variability coagulation hemostasis centrifugation freezing The use of frozen-thawed samples rather than fresh samples for specialized coagulation testing is becoming commonplace, thereby creating novel risks that may jeopardize the quality of hemostasis testing. Residual platelets (PLTs) in frozen plasma are most critical as freezinginduced activation and injury may impair routine and specialized testing after thawing. The aim of this study was to verify the impact of postcentrifugation PLT count in postfreeze-thawed samples on activated partial thromboplastin time (APTT), prothrombin time (PT), fibrinogen, factor VIII (FVIII) activity testing, and factor IX (FIX) activity testing. These parameters were herein assessed in postfreeze-thaw paired plasma samples collected from 15 healthy volunteers and subjected to 4 different centrifugation forces (i.e., 3,000, 1,500, 1,000, and 500g), using data obtained with centrifugation force of 1,500g as the gold standard, in agreement with current recommendations. Compared with reference samples, PLT counts in fresh aliquots were indistinguishable in specimens centrifuged at 1,000g,significantly lower in those centrifuged at 3,000g and significantly higher in those centrifuged at 500g.In all cases except samples centrifuged at 3,000g, the PLT count was significantly decreased in postfreezethaw compared with paired fresh specimens. In postfreeze-thaw plasma, APTT was not influenced by residual PLT count. The results of PT and fibrinogen were consistently altered in samples centrifuged at 1,000 and 500g, though the correlation with the reference measures remained clinically acceptable. Data obtained for FVIII and FIX activities revealed a positive bias in all postfreeze-thaw plasmas, achieving statistical significance in samples centrifuged at 3,000g. We conclude that alteration of centrifuge speeds away from the recommended 1,500g may influence the level of residual PLTs in sample centrifuged at lower speeds such as 500g, and therefore may make these specimens unsuitable for hemostasis testing in postfreeze-thawed plasma samples. In addition, although the changes seen in FVIII and FIX in samples centrifuged at 3,000g may reflect non-plt related effects, such changes should also be considered in this setting. published online September 10, 2013 Issue Theme Old and New Challenges in Hemophilia Management; Guest Editors, Antonio Coppola, MD, Annarita Tagliaferri, MD, and Massimo Franchini, MD. Copyright 2013 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) DOI /s ISSN

2 Influence of Residual PLT Count in Postfreeze-Thaw Samples Lippi et al. 835 Specialized coagulation testing is undergoing a substantial revolution, driven by a convergence of causalities. The evolving scenario of laboratory diagnostics comprises small local laboratories being increasing consolidated into larger facilities according to a typical hub-and-spoke model to save in personnel costs, instrumentation rentals, and reagents. 1 Accordingly, samples dedicated to esoteric testing are increasingly shipped in a refrigerated or frozen format to reference or core laboratories. The deleterious effects of ongoing economic crises in most geographies are generating enormous pressures on health-care systems, 2 which are now subjected to a necessary process of prioritizing expenditures and revising healthcare pathways. In this emerging scenario, several clinical and specialized laboratories are now forced to reduce costs and develop different, more economically efficient patterns of activities, for example, where continuous daily testing is increasingly replaced by weekly or even monthly batch testing to save on reagents, calibrations, and frequency of quality control assessments. 1 The combination of these latter aspects has generated novel problems and increased opportunity for adverse errors throughout the total testing process. As regards coagulation testing, in particular, processing of frozen-thawed (rather than fresh) samples for specialized hemostasis testing is becoming increasingly commonplace, and here, novel problems emerge. Plasma membranes of blood cells, including red blood cells (RBCs), white blood cells (WBCs), and platelets (PLTs), undergo permeability changes and irreversible injury during freezing. 3,4 Lipid phase transitions during freezing of blood result in membranes becoming transiently leaky and impair the structural organization of membrane components, thus generating coalescence of lipid rafts. These events are largely irreversible on rewarming and typically cause breakdown of membrane elements and release of a variety of intracellular substances. 5 Another problem specific for PLTs is the well-known cold-induced activation, which ultimately mimics thrombin-induced activation. 6 According to this evidence, the residual PLTs (perhaps related to numerical count) in frozen plasma specimens become critical, as freezing-induced activation and injury may impair both routine and specialized hemostasis testing in thawed samples, as reported by current Clinical Laboratory Standards Institute (CLSI) guidelines. 7 These guidelines state that although routine clotting tests assessed on fresh plasma are basically unbiased by a PLT count of at least up to /L, a PLT-depleted sample (i.e., PLT count < /L) should be obtained when specimens are frozen for delayed testing. This phenomenon is also well known in the area of lupus anticoagulant (LA) testing, where residual (frozen-thawed) PLTs may lead to quenching of LA activity, and thus false LA negative finding However, the effect of residual PLTs on other specialized hemostasis tests remains underexplored. Accordingly, the aim of this study was to verify the impact of postcentrifugation residual PLT counts in frozen-thawed samples on activated partial thromboplastin time (APTT), prothrombin time (PT) fibrinogen, factor VIII (FVIII) testing, and factor IX (FIX) testing. Patients and Methods The sample population consisted of 15 healthy volunteers (mean age, 44 years; range, years; 2 men and 13 women), enrolled among laboratory staff of an academic affiliated institution. Blood was collected early in the morning (i.e., between 8:00 and 8:30 AM) by direct venipuncture with 21-G straight needle. The tourniquet was applied to locate a suitable vein of the forearm and immediately released after insertion of the first blood tube, to prevent venous stasis. 13 In each subject, blood was drawn in 9 consecutive, 2.7-mL primary blood tubes containing mmol/l buffered sodium citrate (BD Vacutainer plastic whole blood tube, reference number ; Becton Dickinson, Franklin Lakes, NJ), according to a specific sequence. From subject 1 to subject 8, the sequence was discard tube, 3,000-F, 3,000-S, 1,500-F, 1,500-S, 1,000-F, 1,000-S, 500-F, and 500-S, whereas from subject 9 to subject 15, the sequence was reversed (i.e., from 500-S 3,000-F) except for the discard tube, which was always collected first. Blood tubes were mixed immediately after collection by six times gentle inversion and then centrifuged, with brake on, as follows: The two blood tubes labeled 3,000 were centrifuged at 3,000g for 15 minute at room temperature; the two blood tubes labeled 1,500 were centrifuged at 1,500g for 15 minute at room temperature (i.e., according to CLSI reference procedure) 7 ; the two blood tubes labeled 1,000 were centrifuged at 1,000g for 15 minute at room temperature; and the two blood tubes labeled 500 were centrifuged at 500g for 15 minute at room temperature. After centrifugation, plasma was immediately separated. Aliquots labeled as S (i.e., stored ) were frozen for 1 week at 80 C, whereas aliquots labeled as F (i.e., fresh ) were immediately tested. For F samples, PLT count was assessed in all aliquots, whereas APTT, PT, fibrinogen, FVIII, and FIX were assessed in the reference aliquot 1,500-F. For samples S, which were thawed at room temperature after 1 week of storage at 80 C, PLT count, APTT, PT, fibrinogen, FVIII, and FIX were tested in all aliquots. The WBC, RBC, and PLT counts were assessed with Advia 2120 (Siemens Healthcare Diagnostics, Tarrytown, NY). In this analyzer, PLTs are enumerated and sized by a dual-angle optical method and the linearity of PLT count is comprised between 5 and 3, /L. 14 Coagulation testing was performed on ACL TOP 700 (Instrumentation Laboratory, Bedford, MA). The routine clotting assays, thus including PT, APTT, and fibrinogen, were performed using RecombiPlasTin, SynthASil, and FibrinogenC-XL (Instrumentation Laboratory). The were assessed with SynthASil/Hemosil plasma-deficient FVIII (Instrumentation Laboratory) and Electrachrome Factor VIII (Instrumentation Laboratory). The FIX activity based on a one-stage clotting assay (FIX:C) was assessed with SynthASil/ Hemosil Plasma-deficient FIX (Instrumentation Laboratory). Results of all tests were reported as median and interquartile range (IQR). The significance of difference and the bias against the reference measures (i.e., those obtained on 1,500-F samples for PLTs and 1,500-S samples for coagulation testing, as for current CLSI recommendations) 7 were assessed

3 836 Influence of Residual PLT Count in Postfreeze-Thaw Samples Lippi et al. Table 1 Summary of main results for cell-based tests 1,500-F 3,000-F 1,000-F 500-F 3,000-S 1,500-S 1,000-S 500-S Value Value p Value p Value p Value p Value p Value p Value p 9(7 12) 4 (3 5) < (5 12) (18 36) < (3 6) < (4 5) < (4 9) (8 17) 0.04 PLT count ( 10 9 /L) 0.23 NT NT NT NT ( ) ( ) 0.01 ( ) 0.01 ( ) WBC count ( 10 9 /L) 0.09 NT NT NT NT RBC count ( /L) Abbreviations: F, fresh; NT, not tested; PLT, platelet; RBC, red blood cell; WBC, white blood cell; S, stored. Note: Results of platelet white blood cell, red blood cell counts on fresh, and postfreeze-thaw specimens. Results are shown as median and interquartile range, whereas the significance of difference is calculated against the 1500-F specimen, as for current Clinical Laboratory Standards Institute (CLSI) recommendations. 7 with paired Mann Whitney U test, Spearman correlation, and Bland and Altman plots, using Analyze-it for Microsoft Excel (Analyze-it Software Ltd, Leeds, United Kingdom). The statistical significance was fixed at p < Evaluation of the bias against results of the reference 1,500-F samples may be misleading, as the measurements on fresh and postfreezethaw plasma samples have been performed in two separate analytical sessions (R. A. Marlar, J. N. Gausman, J. Engel. Unpublished data on Validation of hemostasis and coagulation assays: recommendations and guidelines ; 2013). All volunteers gave explicit informed consent for participation to this study, which was performed in accordance with the Declaration of Helsinki and under the terms of all relevant local legislations. Results The results of hematological testing (cellular counts) on the fresh and frozen specimens are shown in Table 1.Thevalues of both WBC and RBC count on the different specimens did not significantly differ from those of the reference 1,500-F sample (i.e., those centrifuged according to CLSI recommendations). 7 The PLT counts were indistinguishable in 1,000-F samples, significantly lower in 3,000-F samples and significantly higher in 500-F aliquots ( Fig. 1). In all cases except samples centrifuged at 3,000g, the PLT count significantly decreased in postfreeze-thaw samples compared with the paired fresh specimens, consistent with the hypothesis of effective breakdown of these elements during the freezing cycle ( Table 1 and Fig. 2). The results of coagulation testing in postfreeze-thaw samples are shown in Table 2. As compared with the reference 1,500-S sample, significant bias was observed for FVIII:C, FVIII:A, and FIX: C in 3,000-S aliquots, and PT, fibrinogen, and FIX:C in both 1,000- S and 500-S aliquots. This bias evaluation was performed on stored samples, as the measurements on fresh and postfreezethaw plasma samples have been performed in two separate analytical sessions and thus bias assessment against reference 1,500-F samples may be misleading because of the high interassay variability that characterizes virtually all clotting assays (R. A. Marlar, J. N. Gausman, J. Engel. Unpublished data on Validation of hemostasis and coagulation assays: recommendations and guidelines ; 2013). However, as shown in Table 3, we have also assessed the Spearman correlation between fresh and stored samples because this statistical approach is less influenced by interassay bias. Highly significant correlations were found for APTT, PT, fibrinogen, and FIX:C. As regards FVIII, the correlations observed for FVIII:C in both aliquots 1,500-S and 1,000-S were also highly statistically significant but not so for aliquots 3,000-S and 500-S (p value of 0.04 and 0.05). Significant correlations were also found for FVIII:A, although the correlation coefficients were always lower than those found for APTT, PT, fibrinogen, and FIX:C. Discussion Owing to considerable advances in analytical quality, 15 the most vulnerable part of hemostasis testing is now most

4 Influence of Residual PLT Count in Postfreeze-Thaw Samples Lippi et al. 837 Platelet co ount ( x 10 9 /L) p = g 1,000g 1,500g 3,000g Centrifuga on force Fig. 1 Platelet (PLT) counts in differently processed samples. PLT count (median and interquartile range) in fresh plasma specimens after centrifugation for 15 minutes at 500, 1,000, 1,500, and 3,000g. The dotted line designates the upper threshold according to the current Clinical Laboratory Standards Institute recommendations. clearly represented by the preanalytical phase, which embraces all those activities concerning collection, 7,16,17 processing, transportation, and storage 7,18 of diagnostics specimens. Although there is evidence that routine clotting tests (i.e., APTT, PT, fibrinogen, and thrombin time) performed on fresh plasma samples are not affected by PLT counts of up to /L, 19 a PLT-depleted sample should be obtained if the specimen is to be frozen for subsequent testing. Such a recommendation, whereby primary blood tubes should be centrifuged in a manner to ensure PLT depletion, has been provided by two separate expert groups, CLSI 7 and the Haemostasis and Thrombosis Task Force of the British Committee for Standards in Haematology (BCSH). 20 Because of emerging economic constraints and organizational issues, most specialized hemostasis tests are now performed on postfreeze-thaw samples and rarely on fresh centrifuged plasma. It is therefore clear that an appropriate plasma matrix, possessing a very low PLT count, may be necessary to generate unbiased results of both first and second line coagulation tests in postfreeze-thaw samples. According to the results of this study, the median PLT count ( /L; IQR, /L) obtained in fresh samples centrifuged as for current recommendations (i.e., at 1,500g per 10 minute) was nearly equivalent to the threshold of CLSI guidelines, which state that samples should be centrifuged to consistently produce PLT poor plasma, with a PLT count < /L. 7 The percentage of samples exceeding this limit was 47% (7/15) in 1,500-F and 1,000-F aliquots, 93% (14/15) in 500-F samples and 0% (0/15) in 3,000-F specimens ( Fig. 1). The results in the 3,000-F aliquots are in agreement with those earlier published by Sultan, who showed that 5-minute centrifugation at 3,000g generates a residual PLT count < /L in fresh plasma. 21 It is thereby reasonable to consider whether a greater centrifugal force, perhaps for shorter time, 22 may be recommended to achieve the target of < /L PLT in all samples, provided that the integrity of specimens remains unaffected. As regards hemostasis testing in postfreeze-thaw plasma, APTT seems overall the most robust assay, regardless of the residual PLT count, and this is somewhat surprising considering that this test is strongly influenced by several preanalytical variables. 23,24 The PT results were consistently shortened in samples centrifuged at 1,000 and 500g,although results were still significantly correlated with the reference measures. Like PT, the results of fibrinogen were statistically altered in specimens centrifuged at 1,000 and 500g,although the correlation with the reference measures remains clinically acceptable. The data obtained for FVIII revealed a (positive) bias in all postfreeze-thaw plasmas (except FVIII:A in samples centrifuged at 1,000g), achieving statistical significance for both assays in samples centrifuged at 3,000g. Nearly identical results were found for FIX:C, with bias reaching statistical significance in all aliquots. Although these findings warrant confirmation in independent studies, including samples from patients with hemophilia, the influence of preanalytical changes caused by sample processing, including centrifugation speed, could differentially influence test findings during investigation of hemophilia patients in different centers p = /L) Platelet count ( x 10 P p = ,000-F 3,000-S 1,500-F 1,500-S 1,000-F 1,000-S 500-F 500-S Aliquots Fig. 2 Comparative platelet (PLT) counts in fresh versus frozen samples. PLT count (median and interquartile range) in fresh (-F) and postfreezethaw (-S) plasma specimens after centrifugation for 15 minute at 500, 1,000, 1,500, and 3,000g.

5 838 Influence of Residual PLT Count in Postfreeze-Thaw Samples Lippi et al. Table 2 Summary of main results for hemostasis tests 1,500-S 3,000-S p 1,000-S p 500-S p Value Value Value Value APTT (s) Values 30.6 ( ) 30.0 ( ) ( ) ( ) 0.39 Bias 0.4 ( 1.2 to 0.5) 0.1 ( 0.8 to 0.7) 0.3 ( 1.1 to 0.5) PT (s) Values 10.7 ( ) 10.5 ( ) ( ) ( ) < 0.01 Bias 0.2 ( 0.4 to 0.0) 0.4 ( 0.6 to 0.1) 0.4 ( 0.6 to 0.2) FBG (g/l) Values 277 ( ) 269 ( ) ( ) ( ) < 0.01 Bias 12 ( 9.5 to 33) 23 (8 38) 27 (15 40) FVIII:C (U/dL) Values 65 (61 85) 76 (65 98) (61 90) (61 100) 0.08 Bias 14 (1 27) 5 ( 2 to13) 9( 1 to 19) FVIII:A (U/dL) Values 60 (52 76) 70 (58 84) (50 76) (44 84) 0.43 Bias 11 (1 22) 0.7 ( 8 to9) 4( 6 to 14) FIX:C (U/dL) Values 82 (70 110) 99 (86 115) (85 115) < (83 121) < 0.01 Bias 10 (5 15) 10 (5 14) 12 (8 to 16) Abbreviations: APTT, activated partial thromboplastin time; CLSI, Clinical Laboratory Standards Institute; F, fresh; FBG, fibrinogen; FVIII:A, factor VIII activity based on a one-stage chromogenic activity; FVIII:C, factor VIII activity based on a one-stage clotting assay; FIX:C, factor IX activity based on a one-stage clotting assay; PT, prothrombin time; S, stored. Note: Results of activated partial thromboplastin time, prothrombin time, fibrinogen, factor VIII activity based on a one-stage clotting assay (FVIII:C) or chromogenic activity (FVIII:A) and factor IX activity based on a one-stage clotting assay (FIX:C) (median and interquartile range) on the different fresh, and postfreeze-thaw specimens. The significance of difference is calculated against the 1500-S specimen, as for current Clinical Laboratory Standards Institute (CLSI) recommendations. 7 Accurate performance of hemostasis testing and avoidance of potential sources of interference is pivotal for diagnosis and management of coagulation disorders, including hemophilia. 25 It has been previously established that FVIII from von Willebrand factor/fviii complex effectively binds to PLTs when these become activated, 26 and also that PLTs contain releasable amount of factor IX. 27 It is therefore feasible that breakdown of PLT is associated with release of modest but significant amounts of FVIII and FIX bound to their surface along with phospholipids and a variety of other intracellular substances that would finally interfere with coagulation testing. This may therefore explain some of our findings, and Table 3 Summary data for Spearman correlations 3,000-S 1,500-S 1,000-S 500-S APTT 0.82 (< 0.01) 0.94 (< 0.01) 0.90 (< 0.01) 0.80 (< 0.01) PT 0.92 (< 0.01) 0.90 (< 0.01) 0.97 (< 0.01) 0.78 (< 0.01) FBG 0.86 (< 0.01) 0.94 (< 0.01) 0.84 (< 0.01) 0.89 (< 0.01) FVIII:C 0.53 (0.04) 0.67 (0.01) 0.66 (0.01) 0.52 (0.05) FVIII:A 0.68 (0.01) 0.65 (0.01) 0.77 (< 0.01) 0.64 (0.01) FIX:C 0.99 (< 0.01) 0.97 (< 0.01) 0.97 (< 0.01) 0.97 (0.01) Abbreviations: APTT, activated partial thromboplastin time; FBG, fibrinogen; FIX:C, factor IX activity based on a one-stage clotting assay; FVIII:C, factor VIII activity based on a one-stage clotting assay; FVIII:A, factor VIII activity based on a one-stage chromogenic activity; PT, prothrombin time; S, stored. Note: Spearman correlations of activated partial thromboplastin time, prothrombin time, fibrinogen, factor VIII activity based on a one-stage clotting assay (FVIII:C) or chromogenic activity (FVIII:A) and factor IX activity based on a one-stage clotting assay (FIX:C) measured on the reference 1500-Fand postfreeze-thaw specimens. Values in parentheses represent statistical p values.

6 Influence of Residual PLT Count in Postfreeze-Thaw Samples Lippi et al. 839 suggest that centrifugation of samples below the recommended 1,500g, and for example, at 500g, may make such samples unsuitable for routine and specialized hemostasis testing postfreezing and thawing. Conversely, the significant increase of FVIII and FIX activities observed in plasma samples after centrifugation at 3,000g may provide evidence that this centrifugation force might be excessive, leading to spurious increase of these clotting factors, potentially because of other activation events during sample centrifugation. References 1 Lippi G, Mattiuzzi C. Testing volume is not synonymous of cost, value and efficacy in laboratory diagnostics. Clin Chem Lab Med 2013;51(2): Karanikolos M, Mladovsky P, Cylus J, et al. Financial crisis, austerity, and health in Europe. Lancet 2013;381(9874): Lippi G. Interference studies: focus on blood cell lysates preparation and testing. Clin Lab 2012;58(3 4): Lippi G, Musa R, Avanzini P, Aloe R, Pipitone S, Sandei F. Influence of in vitro hemolysis on hematological testing on Advia Int J Lab Hematol 2012;34(2): Lippi G, Salvagno GL, Montagnana M, Guidi GC. Reliability of the thrombin-generation assay in frozen-thawed platelet-rich plasma. Clin Chem 2006;52(9): Stoll C, Wolkers WF. Membrane stability during biopreservation of blood cells. Transfus Med Hemother 2011;38(2): Adcock DM, Hoefner DM, Kottke-Marchant K, Marlar RA, Szamosi DI, Warunek DJ. Collection, Transport, and Processing of Blood Specimens for Testing Plasma-Based Coagulation Assays and Molecular Hemostasis Assays: Approved Guideline-Fifth Edition. Clinical Laboratory Standards Institute. Wayne, PA: CLSI document H21 A5; Froom P, Barak M. Testing for lupus anticoagulants fresh or frozen? Clin Chem Lab Med 2012;50(9): Pradella P, Azzarini G, Santarossa L, et al. Cooperation experience in a multicentre study to define the upper limits in a normal population for the diagnostic assessment of the functional lupus anticoagulant assays. Clin Chem Lab Med 2013;51(2): Favaloro EJ. Trials and tribulations in lupus anticoagulant testing. Clin Chem Lab Med 2013;51(2): Kershaw G, Suresh S, Orellana D, Nguy YM. Laboratory identification of lupus anticoagulants. Semin Thromb Hemost 2012;38(4): Lakos G. Interference in antiphospholipid antibody assays. Semin Thromb Hemost 2012;38(4): Lippi G, Salvagno GL, Montagnana M, Guidi GC. Short-term venous stasis influences routine coagulation testing. Blood Coagul Fibrinolysis 2005;16(6): Harris N, Kunicka J, Kratz A. The ADVIA 2120 hematology system: flow cytometry-based analysis of blood and body fluids in the routine hematology laboratory. Lab Hematol 2005;11(1): Plebani M, Favaloro EJ, Lippi G. Patient safety and quality in laboratory and hemostasis testing: a renewed loop? Semin Thromb Hemost 2012;38(6): Lippi G, Franchini M, Montagnana M, Salvagno GL, Poli G, Guidi GC. Quality and reliability of routine coagulation testing: can we trust that sample? Blood Coagul Fibrinolysis 2006;17(7): Lippi G, Salvagno GL, Montagnana M, Lima-Oliveira G, Guidi GC, Favaloro EJ. Quality standards for sample collection in coagulation testing. Semin Thromb Hemost 2012;38(6): Adcock Funk DM, Lippi G, Favaloro EJ. Quality standards for sample processing, transportation, and storage in hemostasis testing. Semin Thromb Hemost 2012;38(6): Carroll WE, Wollitzer AO, Harris L, Ling MC, Whitaker WL, Jackson RD. The significance of platelet counts in coagulation studies. J Med 2001;32(1-2): Mackie I, Cooper P, Lawrie A, Kitchen S, Gray E, Laffan M; British Committee for Standards in Haematology. Guidelines on the laboratory aspects of assays used in haemostasis and thrombosis. Int J Lab Hematol 2013;35(1): Sultan A. Five-minute preparation of platelet-poor plasma for routine coagulation testing. East Mediterr Health J 2010;16(2): Lippi G, Salvagno GL, Montagnana M, Manzato F, Guidi GC. Influence of the centrifuge time of primary plasma tubes on routine coagulation testing. Blood Coagul Fibrinolysis 2007;18(5): Lippi G, Favaloro EJ. Activated partial thromboplastin time: new tricks for an old dogma. Semin Thromb Hemost 2008;34(7): Lippi G, Salvagno GL, Ippolito L, Franchini M, Favaloro EJ. Shortened activated partial thromboplastin time: causes and management. Blood Coagul Fibrinolysis 2010;21(5): Carcao MD. The diagnosis and management of congenital hemophilia. Semin Thromb Hemost 2012;38(7): Franchini M, Lippi G. Recombinant factor VIII concentrates. Semin Thromb Hemost 2010;36(5): Romp KG, Monroe DM, Hoffman M. Platelets contain releasable coagulation factor IX antigen. Blood Coagul Fibrinolysis 1993;4(6):