Clin Chem Lab Med 2015; 53(10): 1605–1611
Matthias Erber* and Geoffrey Lee
The influence of excipients commonly used in freeze drying on whole blood coagulation dynamics assessed by rotational thromboelastometry DOI 10.1515/cclm-2014-1046 Received October 24, 2014; accepted January 5, 2015; previously published online February 11, 2015
coagulation impairment due to compounds contained in lyophilized reagents. Keywords: glycine; hypertonic solutions; lyophilization; mannitol; thromboelastometry.
Abstract Background: Lyophilized reagents are used on a daily basis in coagulation diagnostics. They often contain a number of excipients in addition to the active compound. Some of these excipients may, however, influence coagulation dynamics. Methods: Besides from plasmatic coagulation bulking agents may influence platelet properties. We therefore studied the influence of a variety of bulking agents (glycine, mannitol, sucrose and trehalose) as well as a surfactant (Tween® 80) on whole blood coagulation using thromboelastometry (ROTEM®) and platelet function analysis (ROTEM® platelet). Results: Both disaccharides as well as Tween® 80 did not influence whole blood coagulation in the concentration range investigated. The addition of glycine and mannitol solutions to the ROTEM® measurement leads to an impaired clot formation as well as overall clot strength while clotting initiation remained barely influenced. Hypertonic glycine and mannitol solutions exhibit different clot formation impairment when correlated to their osmolar concentration and compared to equally osmolar NaCl-solutions. The effect of glycine was assigned to fibrin formation impairment identified with the FIBTEM assay. Platelet function analysis revealed that hypertonic glycine solutions do not alter platelet function but hypertonic mannitol and NaCl solutions do. Conclusions: While the influence observed for glycine may be due to fibrinogen precipitation, the mechanism of mannitol appears to be more complex as platelet function as well as fibrin-based clot formation are influenced. This study therefore demonstrates the necessity to check for *Corresponding author: Matthias Erber, Division of Pharmaceutics, Friedrich-Alexander University Erlangen-Nürnberg, Cauerstrasse 4, 91058 Erlangen, Germany, E-mail: [email protected] Geoffrey Lee: Division of Pharmaceutics, Friedrich-Alexander University Erlangen-Nürnberg, Erlangen, Germany
Introduction Lyophilization or freeze drying is a process which removes the solvent from a solution under vacuum after the solution was frozen. This procedure allows the drying of heat sensitive material, e.g., proteins, under gentle conditions. Many lyophilized products are embedded in so-called bulking agents or stabilizers due to the low solid content of the active ingredient or in order to protect the active ingredient during and after the lyophilization process [1]. Diverse substances may serve as bulking agents for a formulation, such as amino acids, sugars and polyols [2]. Most commonly used bulking agents are the polyol mannitol, the disaccharides sucrose and trehalose, and recently also amino acids, e.g., glycine, gain more attention [1–4]. Freeze-dried products may also contain various other excipients, which improve product properties, e.g., surfactants, binders and buffers. Many reagents used in coagulation diagnostics are protein-based which often require specific stabilization during the lyophilization process [1, 2]. As coagulation is a complex mechanism the use of unaltered whole blood provides the results closest to nature [5, 6]. Bulking agents contained in lyophilized products may, however, alter coagulation dynamics, as has been observed for several plasma substitutes [7–9], high molecular weight cryoprotectants [10] and also for mannitol solutions [11]. As most of these investigations focus on plasma coagulation after administration of the respective compound, the aim of this study is the identification of the influence of excipients commonly used in freeze-dried products on whole blood coagulation in vitro determined with rotational thromboelastometry (ROTEM®) using the tissue factor-based EXTEM® assay. Sucrose and trehalose were selected as amorphous bulking agents which
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
1606 Erber and Lee: Influence of freeze drying excipients on coagulation are widely used in freeze drying practice [1, 2]. The two disaccharides are also capable of protein stability preservation during freezing (cryoprotection) and freeze drying (lyoprotection). Mannitol is a prominent example for a crystalline bulking agent. As a crystalline alternative for mannitol as a bulking agent the amino acid glycine was selected. Glyince also shows cryo- as well as lyoprotection [12]. Tween® 80 is a commonly used surfactant as it shows good properties regarding protein stabilization during freezing and thawing [13].
Materials and methods Glycine, D-mannitol, sucrose, trehalose × 1H2O and Tween® 80 were purchased from Sigma-Aldrich GmbH (Steinheim, Germany). Citrated whole blood (10 mL; Coagulation 9 NC/10 mL; Sarstedt Monovette®) was received from the university medical center’s blood donation service and was drawn from healthy volunteers. Study protocol was approved by the Ethic Commission of the University’s Medical Center and was conducted according to the World Medical Association Declaration of Helsinki. The ROTEM® device was provided by Tem Innovations GmbH (Munich, Germany). The ROTEM® measurement uses 300 μL whole blood, 20 μL startem reagent (200 mM CaCl2 solution), and for the EXTEM® assay 20 μL of the tissue factor-based reagent are added. The FIBTEM® assay uses 20 μL of the extem® reagent and 20 μL of a solution containing cytochalasin D and 200 mM CaCl2 instead of startem. As cytochalasin D inhibits platelet activation the maximum clot firmness (MCF) observed during the FIBTEM® is therefore a measure for fibrin formation [14]. The substances were added to the measurement cup using 50 μL of differently concentrated solutions ranging from 25 mg/mL to 200 mg/mL. A lyophilized product typically contains 2%–10% (m/V) solids which results in a few milligram which are added to the measurement also depending on the added volume [2]. For mannitol additional data points were included. As Tween® 80 is not a bulking agent but a surfactant, its concentration ranged from 25 μg/mL to 200 μg/mL. A total of 50 μL isotonic saline
(0.9% NaCl solution) was added to the reference measurements to avoid effects due to different dilution of the samples. Citrated whole blood was analyzed within 4 h after blood draw and was kept in the blood sample holder of the ROTEM® device for 10 min at 37 °C prior to the measurement. The EXTEM® assay was performed according to the instructions provided by Tem Innovations GmbH. Whole blood coagulation determined with the ROTEM® device depends on several blood sample specific factors, e.g., fibrinogen level, coagulation factor level, hematocrit, platelet count [15–17] etc. ROTEM® results are therefore displayed as ratios compared to the unaltered reference value of a whole blood sample. The ratio is generated by dividing the measured value by the mean reference value (e.g., CT-ratio = CT[s]/ CTref(mean)[s]). Standard deviations also refer to the calculated ratio of the single determination. Referring to the reference measurement provides the opportunity to compare whole blood samples from different individuals. The following ROTEM® parameters were investigated (Figure 1 is a schematic graphical output generated during the ROTEM® measurement): clotting time (CT), clot formation time (CFT), amplitude 10 min after CT (A10), MCF and the maximum clotting speed (MaxV). CT is the time necessary to generate an amplitude of 2 mm, CFT is the time until an amplitude of 20 mm is reached, and the MaxV reflects the maximum value of the first deviation of the time versus amplitude profile. For all blood samples tested normal values for the parameters observed were obtained if unaltered, i.e., parameters were within the normal range [18]. All data are based on four-fold determination of each data point. Each individual substance was tested with three whole blood samples drawn from three different healthy individuals. Platelet function analysis was performed with the ROTEM® platelet module which is based on impedance aggregometry. It determines several parameters which are based on increasing resistance: A6: amplitude (representing transformed resistance) in Ω 6 min after initiation, MS: maximum increase in Ω per minute, AUC: area under the curve (AUC) in Ω/min. The ROTEM® platelet uses 150 μL whole blood which is diluted with 150 μL isotonic saline (0.9% NaCl solution). A total of 12 μL of the TRAPTEM® reagent were used as an activator. The TRAPTEM® uses thrombin activating peptide (TRAP) which leads to platelet activation by thrombin. In addition 50 μL of the differently concentrated solutions were added to the measurement. Isotonic saline served as a reference. The A6 parameter was selected for comparison. All results obtained from unaltered whole blood samples were within the reference ranges (AUC = 55–155 Ω/min;
Figure 1 Schematic ROTEM graphical output explaining the result parameters (provided by Tem Innovations GmbH).
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
Erber and Lee: Influence of freeze drying excipients on coagulation 1607
A6 = 14–36 Ω; MS = 5–14 Ω/min) [19] with intra-whole blood sample coefficients of variation ≤ 10%. The results are also displayed as ratios compared to the mean reference value. The ROTEM® platelet results are based on three whole blood samples. Each concentration was measured in duplicates. Non-linear regression to the equation: y = y0+Aekx as well as linear regression was performed using origin 9.0 software (OriginLab, Northhampton, MA, USA). R2-values represent coefficients of determination.
Results and discussion Sucrose/trehalose Sucrose and trehalose were added to the ROTEM® cup covering the following concentrations: 25, 50, 75, 100 and 200 mg/mL. The results for sucrose and trehalose are summarized in Figure 2A and B, respectively. Coagulation initiation reflected by CT, clot stabilization reflected by CFT and A10 as well as the MCF remain unchanged upon either sucrose or trehalose addition. Both disaccharides therefore do not influence coagulation within the range studied.
Glycine Glycine revealed in a preliminary experiment a high influence on coagulation and was therefore added to the ROTEM® cup covering the following, lower concentrations: 2, 4, 8, 20, 40, 60, 80 and 100 mg/mL. The pH values of the glycine solutions ranged from 6.38 to 6.44 determined using a MP 120 pH-Meter (Mettler-Toledo, Gießen, Germany). pH of the blood samples containing 50 μL of the 100 mg/mL glycine solution were between 7.40 and 7.41 at 37 °C. Figure 2C summarizes the influence of glycine solutions on the ROTEM® parameters. Coagulation observed with the ROTEM® is uninfluenced up to a glycine concentration of 0.008 g/mL. After adding solutions with a higher glycine concentration CT remains stable, exhibiting only slight variation up to a concentration of about 0.08 g/mL but is clearly altered after adding a 0.1 g/mL solution. CFT, A10 as well as MCF start to deviate from the reference measurement at a concentration between 0.02 and 0.04 g/mL. A10 decrease depends linearly from CFT prolongation [Figure 3; A10(ratio) = –0.159 · CFT(ratio)+1.145; r2 = 0.976] which demonstrates the connection between the two parameters regarding clotting speed. It is therefore possible to predict A10 values from CFT values in this model. The applicability to clinical scenarios might be possible as well, but needs clarification. The MaxV decreases
linearly with increasing glycine concentration [Figure 3; MaxV(ratio) = –0.0067 · c(glycine)+0.985; r2 = 0.960] which demonstrates the clotting speed impairment by glycine. The initiation of the coagulation process, represented by CT, is therefore barely influenced by glycine solutions, but propagation and overall clot stability clearly are. During this study the volume replacement is 14.3% and the 0.1 g/mL glycine solution generates an osmolarity of 1332.5 mOsm. A study with glycine (c = 1200 mOsm) performed by Wilder et al. covering 5%–20% volume replacement focusing on in vitro plasma coagulation confirms the findings reported here [20]. Platelets in platelet rich plasma (PRP) can withstand a wide range of hypertonic stress showing good recovery depending on temperature [21]. The substance which is used to generate the hypertonic solution also influences platelet recovery [21]. Law showed that a substantial proportion of platelets from PRP lose their discoid shape changing to spherical or dendritic shape upon hypertonic stress [22]. The results on platelet function analysis are summarized in the next section. As the ROTEM® device is based on elasticity measurement, a possible explanation would be impaired fibrin formation or fibrin crosslinking. Although some glycine derivates show FXIIIa connected fibrin formation inhibition, glycine does not exhibit such an effect [23]. Fibrinformation and therefore elasticity increase depends on fibrinogen present in the whole blood sample [15]. Specific fibrinogen precipitation from plasma samples can be achieved with solutions containing glycine [24, 25]. Glycine solutions used for this purpose typically have concentrations > 2 M. Considering hematocrit value (HCT) of the blood samples used in this work, a glycine plasma concentration of 0.27±0.01 M results after adding 50 μL of a 100 mg/mL solution. So although glycine concentration is low it may, however, lead to partial fibrinogen precipitation and therefore result in clot destabilization.
Mannitol Mannitol was added at the following concentrations: 20, 40, 50, 60, 70, 80, 100, 120, 140, 160, 180 and 200 mg/mL. Figure 2D summarizes the results for mannitol solutions. CT remains constant over the concentration range investigated. CFT increases with increasing mannitol concentration and A10 as well as MCF decrease with increasing mannitol concentration. These findings are similar to the results obtained with glycine solutions. Besides other blood sample dependent factors, platelet count and function influence the EXTEM assay [15]. Hypertonic solutions may destabilize cells included in the coagulation process,
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
1608 Erber and Lee: Influence of freeze drying excipients on coagulation
Ratio, tsample/tref
Ratio, tsample/tref
Ratio, mmsample/mmref
B Ratio, mmsample/mmref
A
Sucrose, mg/mL
Trehalose, mg/mL
Ratio, tsample/tref
Ratio, tsample/tref
Ratio, mmsample/mmref
D Ratio, mmsample/mmref
C
Glycine, mg/mL
Mannitol, mg/mL
Ratio, tsample/tref
Ratio, mmsample/mmref
E
Tween® 80, µg/mL
Figure 2 Influence of differently concentrated excipient solutions on ROTEM® parameters. Left y-axis: CT (black squares) and CFT ratio (open squares); right y-axis: A10 (blue circles) and MCF ratio (open circles). (A) Sucrose, (B) Trehalose, (C) Glycine, (D) Mannitol, (E) Tween® 80. Results are shown as means±standard deviation, the lines represent best-fit regressions (n = 3 whole blood samples for each substance).
i.e., especially platelets [20]. The impaired platelet function may result in an overall less stable clot. The results for glycine and mannitol were therefore compared to NaCl solutions of the same osmolar concentration using both the ROTEM® thromboelastometer and the ROTEM® platelet device to identify effects due to high osmotic pressure. The osmolarity of different solutions was calculated assuming completely dissolved substances using the following equation: cosm = ϕ·x·cmol, cosm is the osmolar concentration, ϕ is the osmotic coefficient which was set
constant (ϕ = 1; ideal solution), x is the number of osmotic dissociated units per molecule substance dissolved, and cmol is the molar concentration. As CFT is the ROTEM® para meter which shows the strongest dependency, Figure 4 compares the different CFT-ratios in dependency of the solution’s osmolarity. A non-linear regression was performed on each data set using the following equation: CFT ratio = 1+0.08 · ek·c. c is the concentration of the respective substance in mOsmol/L and k is the impairment constant. Regression data are summarized in
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
Erber and Lee: Influence of freeze drying excipients on coagulation 1609 Table 1 Regression data for CFT ratio in dependency of osmolarity.
Glycine, mg/mL
Ratio, mmsample/mmref
MaxV ratio, vsample/vref
Substance
CFT-Ratio, tsample/tref
Figure 3 Glycine influence on clotting speed. Left y-axis/lower x-axis: Correlation between A10 ratio and CFT ratio (black squares); right y-axis/upper x-axis: clot formation speed (MaxV) in dependency of glycine concentration (open circles). Results are shown as means±standard deviation (n = 3 whole blood samples).
CFT-Ratio, tsample/tref
Table 1. CFT ratios after the addition of differently concentrated NaCl solutions were slightly influenced. NaCl solutions with an osmolarity up to 1100 mOsm do therefore not influence clot formation. Comparing glycine and mannitol CFT ratios at the same osmolarity shows the higher influence of glycine on clot formation. As the regression represents a first order – like CFT – ratio/osmolarity correlation, the values for the impairment constant reflect the magnitude of clot formation impairment by the different substances. Different substances at definite osmolarities therefore exhibit different tendency to alter clotting and clot formation in whole blood samples, i.e., clot formation impairment is dependent on the solutions osmolarity but also depends on the substance dissolved. The absence of clotting impairment after NaCl addition demonstrates that
NaCl Mannitol Glycine
k
R2
χ2
0.97 2.11 2.69
0.751 0.947 0.959
0.442 0.406 0.652
osmolarity connected stresses are not alone responsible for the observed CFT prolongation. The ROTEM® platelet results are summarized in Table 2. The influence of glycine on platelet function is minimal even at high concentrations, yet it shows high deviations. Mannitol as well as NaCl solutions clearly impair platelet function with increasing solute concentration. This confirms the assumption that the origin of the influence of mannitol and glycine on whole blood coagulation is different. NaCl and mannitol solutions have a similar effect on platelet function at the same osmolar concentration, this effect is therefore attributed to osmotic pressure. As glycine does not exhibit a similar influence it is assumed that glycine may be rapidly transported into platelets thus reducing osmotic pressure. This transport could be due to passive diffusion [26] or by active transport as has been reported for L-arginine in platelets [27] or glycine into other cells [28, 29]. During the FIBTEM® assay (Figure 5) both mannitol and glycine showed an impaired clot stability, i.e., decreased MCF compared to the reference. The effect on fibrin-based clot formation was stronger for glycine decreasing the MCF at a concentration of 100 mg/mL to about 50% of the initial MCF. Addition of a mannitol solution of 200 mg/mL decreased the MCF to about 80% of the initial MCF. This experiment confirms the assumption made earlier regarding clot destabilization by glycine due to fibrin formation impairment. Considering the ROTEM® platelet results the destabilizing effect of glycine on clot formation is majorly attributed to the impairment of fibrin formation. Within the range observed mannitol Table 2 ROTEM® platelet results for glycine, mannitol and NaCl solutions (n = 3 whole blood samples). c, osmol/L
Osmolarity, osmol/L
Figure 4 CFT ratio in dependency of the solution’s osmolarity and the substance dissolved. Results are shown as means±standard deviation.
0.274 0.547 0.658 0.767 0.85 0.986 1.096
Glycine
Mannitol
NaCl
A6 ratio
sdv
A6 ratio
sdv
A6 ratio
sdv
0.771 0.789 1.046 0.954 0.899 0.936 0.954
0.049 0.130 0.078 0.052 0.026 0.026 0.156
0.913 0.893 0.892 0.885 0.699 0.655 0.454
0.062 0.071 0.064 0.070 0.100 0.089 0.063
0.714 0.661 0.699 0.608 0.620 0.578 0.476
0.074 0.074 0.026 0.009 0.027 0.017 0.054
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
FIBTEM MCF ratio, mmsample/mmref
1610 Erber and Lee: Influence of freeze drying excipients on coagulation
Concentration, mg/mL
Figure 5 FIBTEM MCF dependency of glycine and mannitol concentration. Results are shown as means±standard deviation, the dashed line represents a best-fit regression (n = 3 whole blood samples).
only slightly decreases FIBTEM® MCF but clearly affects platelet function. The influence of mannitol on clot destabilization therefore appears to be more complex, i.e., a combination of at least two mechanisms, as platelet function as well as fibrin formation are effected.
Tween® 80 As Tween® 80 is a surfactant which is typically used at low concentrations [approx. 0.01% (m/V) [2]], solutions with the following concentrations were added to the whole blood samples: 25, 50, 75, 100, 150 and 200 μg/mL. Within the range observed Tween® 80 did not influence coagulation kinetics determined with the ROTEM® device (Figure 2E). Some values deviate from the reference value which is negligible due to larger variation. When used in concentrations typical to preserve protein activity during freezing Tween® 80 is, therefore, uncritical regarding coagulation. The precise influence of Tween® associated surface tension reduction needs further investigation. As the reduced surface tension may lead to reduced friction at all surfaces, as was reported for Pluronic® F-68 [30], the influence of blood surface tension on the ROTEM® measurement should be subjected to a future experiment.
Conclusions In summary, both disaccharides investigated in this study neither influence CT nor the clot firmness parameters and are therefore uncritical regarding whole blood coagulation diagnostics. Tween® 80 used in a concentration typical
for surfactants in freeze drying did not also influence the ROTEM® measurement. Both crystalline bulking agents, glycine and mannitol, did, however, alter the coagulation profile obtained during ROTEM® analysis. There was no influence on clotting initiation by mannitol and the influence on CT was small for glycine. These findings confirm previous studies on plasmatic coagulation. Observing coagulation beyond CT, both clotting speed and clot strength are impaired by glycine and mannitol solutions. The effect observed for both solutions is, however, not only due to increased osmolarity as the comparison to NaCl solutions of equal osmolar concentration demonstrates. The ROTEM® platelet results show that coagulation impairment by glycine is not due to platelet function impairment which could be a result from glycine transport, either active or passive, into platelets. Mannitol and NaCl solutions alter platelet function in a similar way due to osmotic pressure. As hypertonic NaCl solutions do not influence coagulation, the effect of mannitol could result from both fibrin formation impairment which was confirmed by a decreased MCF during the FIBTEM® experiment and platelet function impairment. Based on the FIBTEM® results the effect of glycine on coagulation was attributed to fibrin formation impairment. Although glycine concentration was low compared to studies focusing on fibrinogen precipitation the effect of glycine on whole blood coagulation may, however, be due to precipitation dependent decreased fibrinogen concentration. The results reported in this study were generated by in vitro analysis using high concentrations of the respective solute. The results may therefore be of limited applicability in clinical practice. The thresholds reported, however, need to be considered when: 1) diluting a whole blood sample with solutions containing the respective compounds; 2) applying a highly concentrated mannitol solution, e.g., in order to reduce intracranial pressure; and when 3) dissolving lyophilized reagents directly with blood samples. Acknowledgments: The university medical center’s blood donation service as well as the volunteers are gratefully acknowledged for the preparation of the whole blood samples and their participation, respectively. Gaby Lange and Ilona Muslaeva are gratefully acknowledged for performing the ROTEM® platelet analysis and their expert technical assistance. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission. Financial support: None declared. Employment or leadership: None declared. Honorarium: None declared.
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
Erber and Lee: Influence of freeze drying excipients on coagulation 1611
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.
References 1. Pikal MJ. Freeze drying. In: Swarbrick J, editor. Encyclopedia of pharmaceutical technology, 3rd ed. London: Informa Healthcare, 2002:1299–326. 2. Wang W. Lyophilization and development of solid protein pharmaceuticals. [review] Int J Pharm 2000;203:1–60. 3. Tang XC, Nail SL, Pikal MJ. Freeze-drying process design by manometric temperature measurement: design of a smart freeze-dryer. Pharm Res 2005;22:685–700. 4. Wang DQ, Hey JM, Nail SL. Effect of collapse on the stability of freeze-dried recombinant factor VIII and alpha-amylase. J Pharm Sci 2004;93:1253–63. 5. Rand MD, Lock J, Van’t Veer C, Gaffney D, Mann K. Blood clotting in minimally altered whole blood. Blood 1996;88: 3432–45. 6. Roche AM, James MF, Grocott MP, Mythen MG. Citrated bood does not reliably reflect fresh whole blood coagulability in trials of in vitro hemodilution. Anesth Analg 2003;96:58–61. 7. Mardel S, Saunders F, Allen H, Menezes G, Edwards C, Ollerenshaw L, et al. Reduced quality of clot formation with gelatinbased plasma substitutes. Br J Anaest 1998;80:204–7. 8. Petroianu GA, Liu J, Maleck WH, Mattinger C, Bergler WF. The effect of in vitro hemodilution with gelatin, dextran, hydroxyethyl starch, or Ringer’s solution on thrombelastograph®. Anest Anal 2000;90:795–800. 9. Niemi TT, Kuitunen AH. Artificial colloids impair haemostasis. An in vitro study using thromboelastometry coagulation analysis. Acta Anaesthesiol Scand 2005;49:373–8. 10. Bakaltcheva I. Effects of high-molecular-weight cryoprotectants on platelets and the coagulation system. Cryobiology 2000;40:283–93. 11. Luostarinen T, Niiya T, Schramko A, Rosenberg P, Niemi T. Comparison of hypertonic saline and mannitol on whole blood coagulation in vitro assessed by thromboelastometry. Neurocrit Care 2011;14:238–43. 12. Carpenter JF, Hand SC, Crowe LM, Crowe JH. Cryoprotection of phosphofructokinase with organic solutes: characterization of enhanced protection in the presence of divalent cations. Arch Biochem Biophys 1986;250:505–12. 13. Chang BS, Kendrick BS, Carpenter JF. Surface-induced denaturation of proteins during freezing and its inhibition by surfactants. J Pharm Sci 1996;85:1325–30.
14. Casella JF, Flanagan MD, Lin S. Cytochalasin D inhibits actin polymerization and induces depolymerization of actin filaments formed during platelet shape change. Nature 1981;293:302–5. 15. Theusinger OM, Schroder CM, Eismon J, Emmert MY, Seifert B, Spahn DR, et al. The influence of laboratory coagulation tests and clotting factor levels on Rotation Thromboelastometry (ROTEM(R)) during major surgery with hemorrhage. Anesth Analg 2013;117:314–21. 16. Ogawa S, Szlam F, Chen EP, Nishimura T, Kim H, Roback JD, et al. A comparative evaluation of rotation thromboelastometry and standard coagulation tests in hemodilution-induced coagulation changes after cardiac surgery. Transfusion 2012;52:14–22. 17. Ogawa S, Szlam F, Bolliger D, Nishimura T, Chen EP, Tanaka KA. The impact of hematocrit on fibrin clot formation assessed by rotational thromboelastometry. Anesth Analg 2012;115:16–21. 18. Lang T, Bauters A, Braun SL, Pötzsch B, von Pape K-W, Kolde H-J, et al. Multi-centre investigation on reference ranges for ROTEM thromboelastometry. Blood Coagul Fibrinolysis 2005;16: 301–10. 19. Tem-Innovations-GmbH. Instructions for use – trap-tem® ref 600-13. Reference ranges 2014. p 9. 20. Wilder DM. Hypertonic resuscitation and blood coagulation in vitro comparison of several hypertonic solutions for their action on platelets and plasma coagulation. Thromb Res 2002;107:255–61. 21. Armitage WJ, Parmar N, Hunt CJ. The effects of osmotic stress on human platelets. J Cell Physiol 1985;123:241–8. 22. Law P. The tolerance of human platelets to osmotic stress. Exp Hematol 1983;11:351–7. 23. Lorand L, Konishi K, Jacobsen A. Transpeptidation mechanism in blood clotting. Nature 1962;194:1148–9. 24. Kazal LA, Amsel S, Miller OP, Tocantins LM. The preparation and some properties of fibrinogen precipitated from human plasma by glycine. Exp Biol Med 1963;113:989–94. 25. Mosesson MW, Sherry S. The preparation and properties of human fibrinogen of relatively high solubility. Biochemistry 1966;5:2829–35. 26. Boullin DJ, Green AR. Mechanisms by which human blood platelets accumulate glycine, gaba and amino acid precursors of putative neurotransmitters. Br J Pharmacol 1972;45:83–94. 27. Vasta V, Meacci E, Farnararo M, Bruni P. Identification of a specific transport system for l-arginine in human platelets. Biochem Biophys Res Comm 1995;206:878–84. 28. Ellory JC, Jones SE, Young JD. Glycine transport in human erythrocytes. J Physiol 1981;320:403–22. 29. Bannai S, Tateishi N. Role of membrane transport in metabolism and function of glutathione in mammals. J Membr Biol 1986;89:1–8. 30. Hymes AC, Safavian MH, Gunther T, Forsman J. The influence of an industrial surfactant pluronic F-68, in the treatment of hemorrhagic shock. J Surg Res 1971;11:191–7.
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
Matthias Erber* and Geoffrey Lee
The influence of excipients commonly used in freeze drying on whole blood coagulation dynamics assessed by rotational thromboelastometry DOI 10.1515/cclm-2014-1046 Received October 24, 2014; accepted January 5, 2015; previously published online February 11, 2015
coagulation impairment due to compounds contained in lyophilized reagents. Keywords: glycine; hypertonic solutions; lyophilization; mannitol; thromboelastometry.
Abstract Background: Lyophilized reagents are used on a daily basis in coagulation diagnostics. They often contain a number of excipients in addition to the active compound. Some of these excipients may, however, influence coagulation dynamics. Methods: Besides from plasmatic coagulation bulking agents may influence platelet properties. We therefore studied the influence of a variety of bulking agents (glycine, mannitol, sucrose and trehalose) as well as a surfactant (Tween® 80) on whole blood coagulation using thromboelastometry (ROTEM®) and platelet function analysis (ROTEM® platelet). Results: Both disaccharides as well as Tween® 80 did not influence whole blood coagulation in the concentration range investigated. The addition of glycine and mannitol solutions to the ROTEM® measurement leads to an impaired clot formation as well as overall clot strength while clotting initiation remained barely influenced. Hypertonic glycine and mannitol solutions exhibit different clot formation impairment when correlated to their osmolar concentration and compared to equally osmolar NaCl-solutions. The effect of glycine was assigned to fibrin formation impairment identified with the FIBTEM assay. Platelet function analysis revealed that hypertonic glycine solutions do not alter platelet function but hypertonic mannitol and NaCl solutions do. Conclusions: While the influence observed for glycine may be due to fibrinogen precipitation, the mechanism of mannitol appears to be more complex as platelet function as well as fibrin-based clot formation are influenced. This study therefore demonstrates the necessity to check for *Corresponding author: Matthias Erber, Division of Pharmaceutics, Friedrich-Alexander University Erlangen-Nürnberg, Cauerstrasse 4, 91058 Erlangen, Germany, E-mail: [email protected] Geoffrey Lee: Division of Pharmaceutics, Friedrich-Alexander University Erlangen-Nürnberg, Erlangen, Germany
Introduction Lyophilization or freeze drying is a process which removes the solvent from a solution under vacuum after the solution was frozen. This procedure allows the drying of heat sensitive material, e.g., proteins, under gentle conditions. Many lyophilized products are embedded in so-called bulking agents or stabilizers due to the low solid content of the active ingredient or in order to protect the active ingredient during and after the lyophilization process [1]. Diverse substances may serve as bulking agents for a formulation, such as amino acids, sugars and polyols [2]. Most commonly used bulking agents are the polyol mannitol, the disaccharides sucrose and trehalose, and recently also amino acids, e.g., glycine, gain more attention [1–4]. Freeze-dried products may also contain various other excipients, which improve product properties, e.g., surfactants, binders and buffers. Many reagents used in coagulation diagnostics are protein-based which often require specific stabilization during the lyophilization process [1, 2]. As coagulation is a complex mechanism the use of unaltered whole blood provides the results closest to nature [5, 6]. Bulking agents contained in lyophilized products may, however, alter coagulation dynamics, as has been observed for several plasma substitutes [7–9], high molecular weight cryoprotectants [10] and also for mannitol solutions [11]. As most of these investigations focus on plasma coagulation after administration of the respective compound, the aim of this study is the identification of the influence of excipients commonly used in freeze-dried products on whole blood coagulation in vitro determined with rotational thromboelastometry (ROTEM®) using the tissue factor-based EXTEM® assay. Sucrose and trehalose were selected as amorphous bulking agents which
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
1606 Erber and Lee: Influence of freeze drying excipients on coagulation are widely used in freeze drying practice [1, 2]. The two disaccharides are also capable of protein stability preservation during freezing (cryoprotection) and freeze drying (lyoprotection). Mannitol is a prominent example for a crystalline bulking agent. As a crystalline alternative for mannitol as a bulking agent the amino acid glycine was selected. Glyince also shows cryo- as well as lyoprotection [12]. Tween® 80 is a commonly used surfactant as it shows good properties regarding protein stabilization during freezing and thawing [13].
Materials and methods Glycine, D-mannitol, sucrose, trehalose × 1H2O and Tween® 80 were purchased from Sigma-Aldrich GmbH (Steinheim, Germany). Citrated whole blood (10 mL; Coagulation 9 NC/10 mL; Sarstedt Monovette®) was received from the university medical center’s blood donation service and was drawn from healthy volunteers. Study protocol was approved by the Ethic Commission of the University’s Medical Center and was conducted according to the World Medical Association Declaration of Helsinki. The ROTEM® device was provided by Tem Innovations GmbH (Munich, Germany). The ROTEM® measurement uses 300 μL whole blood, 20 μL startem reagent (200 mM CaCl2 solution), and for the EXTEM® assay 20 μL of the tissue factor-based reagent are added. The FIBTEM® assay uses 20 μL of the extem® reagent and 20 μL of a solution containing cytochalasin D and 200 mM CaCl2 instead of startem. As cytochalasin D inhibits platelet activation the maximum clot firmness (MCF) observed during the FIBTEM® is therefore a measure for fibrin formation [14]. The substances were added to the measurement cup using 50 μL of differently concentrated solutions ranging from 25 mg/mL to 200 mg/mL. A lyophilized product typically contains 2%–10% (m/V) solids which results in a few milligram which are added to the measurement also depending on the added volume [2]. For mannitol additional data points were included. As Tween® 80 is not a bulking agent but a surfactant, its concentration ranged from 25 μg/mL to 200 μg/mL. A total of 50 μL isotonic saline
(0.9% NaCl solution) was added to the reference measurements to avoid effects due to different dilution of the samples. Citrated whole blood was analyzed within 4 h after blood draw and was kept in the blood sample holder of the ROTEM® device for 10 min at 37 °C prior to the measurement. The EXTEM® assay was performed according to the instructions provided by Tem Innovations GmbH. Whole blood coagulation determined with the ROTEM® device depends on several blood sample specific factors, e.g., fibrinogen level, coagulation factor level, hematocrit, platelet count [15–17] etc. ROTEM® results are therefore displayed as ratios compared to the unaltered reference value of a whole blood sample. The ratio is generated by dividing the measured value by the mean reference value (e.g., CT-ratio = CT[s]/ CTref(mean)[s]). Standard deviations also refer to the calculated ratio of the single determination. Referring to the reference measurement provides the opportunity to compare whole blood samples from different individuals. The following ROTEM® parameters were investigated (Figure 1 is a schematic graphical output generated during the ROTEM® measurement): clotting time (CT), clot formation time (CFT), amplitude 10 min after CT (A10), MCF and the maximum clotting speed (MaxV). CT is the time necessary to generate an amplitude of 2 mm, CFT is the time until an amplitude of 20 mm is reached, and the MaxV reflects the maximum value of the first deviation of the time versus amplitude profile. For all blood samples tested normal values for the parameters observed were obtained if unaltered, i.e., parameters were within the normal range [18]. All data are based on four-fold determination of each data point. Each individual substance was tested with three whole blood samples drawn from three different healthy individuals. Platelet function analysis was performed with the ROTEM® platelet module which is based on impedance aggregometry. It determines several parameters which are based on increasing resistance: A6: amplitude (representing transformed resistance) in Ω 6 min after initiation, MS: maximum increase in Ω per minute, AUC: area under the curve (AUC) in Ω/min. The ROTEM® platelet uses 150 μL whole blood which is diluted with 150 μL isotonic saline (0.9% NaCl solution). A total of 12 μL of the TRAPTEM® reagent were used as an activator. The TRAPTEM® uses thrombin activating peptide (TRAP) which leads to platelet activation by thrombin. In addition 50 μL of the differently concentrated solutions were added to the measurement. Isotonic saline served as a reference. The A6 parameter was selected for comparison. All results obtained from unaltered whole blood samples were within the reference ranges (AUC = 55–155 Ω/min;
Figure 1 Schematic ROTEM graphical output explaining the result parameters (provided by Tem Innovations GmbH).
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
Erber and Lee: Influence of freeze drying excipients on coagulation 1607
A6 = 14–36 Ω; MS = 5–14 Ω/min) [19] with intra-whole blood sample coefficients of variation ≤ 10%. The results are also displayed as ratios compared to the mean reference value. The ROTEM® platelet results are based on three whole blood samples. Each concentration was measured in duplicates. Non-linear regression to the equation: y = y0+Aekx as well as linear regression was performed using origin 9.0 software (OriginLab, Northhampton, MA, USA). R2-values represent coefficients of determination.
Results and discussion Sucrose/trehalose Sucrose and trehalose were added to the ROTEM® cup covering the following concentrations: 25, 50, 75, 100 and 200 mg/mL. The results for sucrose and trehalose are summarized in Figure 2A and B, respectively. Coagulation initiation reflected by CT, clot stabilization reflected by CFT and A10 as well as the MCF remain unchanged upon either sucrose or trehalose addition. Both disaccharides therefore do not influence coagulation within the range studied.
Glycine Glycine revealed in a preliminary experiment a high influence on coagulation and was therefore added to the ROTEM® cup covering the following, lower concentrations: 2, 4, 8, 20, 40, 60, 80 and 100 mg/mL. The pH values of the glycine solutions ranged from 6.38 to 6.44 determined using a MP 120 pH-Meter (Mettler-Toledo, Gießen, Germany). pH of the blood samples containing 50 μL of the 100 mg/mL glycine solution were between 7.40 and 7.41 at 37 °C. Figure 2C summarizes the influence of glycine solutions on the ROTEM® parameters. Coagulation observed with the ROTEM® is uninfluenced up to a glycine concentration of 0.008 g/mL. After adding solutions with a higher glycine concentration CT remains stable, exhibiting only slight variation up to a concentration of about 0.08 g/mL but is clearly altered after adding a 0.1 g/mL solution. CFT, A10 as well as MCF start to deviate from the reference measurement at a concentration between 0.02 and 0.04 g/mL. A10 decrease depends linearly from CFT prolongation [Figure 3; A10(ratio) = –0.159 · CFT(ratio)+1.145; r2 = 0.976] which demonstrates the connection between the two parameters regarding clotting speed. It is therefore possible to predict A10 values from CFT values in this model. The applicability to clinical scenarios might be possible as well, but needs clarification. The MaxV decreases
linearly with increasing glycine concentration [Figure 3; MaxV(ratio) = –0.0067 · c(glycine)+0.985; r2 = 0.960] which demonstrates the clotting speed impairment by glycine. The initiation of the coagulation process, represented by CT, is therefore barely influenced by glycine solutions, but propagation and overall clot stability clearly are. During this study the volume replacement is 14.3% and the 0.1 g/mL glycine solution generates an osmolarity of 1332.5 mOsm. A study with glycine (c = 1200 mOsm) performed by Wilder et al. covering 5%–20% volume replacement focusing on in vitro plasma coagulation confirms the findings reported here [20]. Platelets in platelet rich plasma (PRP) can withstand a wide range of hypertonic stress showing good recovery depending on temperature [21]. The substance which is used to generate the hypertonic solution also influences platelet recovery [21]. Law showed that a substantial proportion of platelets from PRP lose their discoid shape changing to spherical or dendritic shape upon hypertonic stress [22]. The results on platelet function analysis are summarized in the next section. As the ROTEM® device is based on elasticity measurement, a possible explanation would be impaired fibrin formation or fibrin crosslinking. Although some glycine derivates show FXIIIa connected fibrin formation inhibition, glycine does not exhibit such an effect [23]. Fibrinformation and therefore elasticity increase depends on fibrinogen present in the whole blood sample [15]. Specific fibrinogen precipitation from plasma samples can be achieved with solutions containing glycine [24, 25]. Glycine solutions used for this purpose typically have concentrations > 2 M. Considering hematocrit value (HCT) of the blood samples used in this work, a glycine plasma concentration of 0.27±0.01 M results after adding 50 μL of a 100 mg/mL solution. So although glycine concentration is low it may, however, lead to partial fibrinogen precipitation and therefore result in clot destabilization.
Mannitol Mannitol was added at the following concentrations: 20, 40, 50, 60, 70, 80, 100, 120, 140, 160, 180 and 200 mg/mL. Figure 2D summarizes the results for mannitol solutions. CT remains constant over the concentration range investigated. CFT increases with increasing mannitol concentration and A10 as well as MCF decrease with increasing mannitol concentration. These findings are similar to the results obtained with glycine solutions. Besides other blood sample dependent factors, platelet count and function influence the EXTEM assay [15]. Hypertonic solutions may destabilize cells included in the coagulation process,
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
1608 Erber and Lee: Influence of freeze drying excipients on coagulation
Ratio, tsample/tref
Ratio, tsample/tref
Ratio, mmsample/mmref
B Ratio, mmsample/mmref
A
Sucrose, mg/mL
Trehalose, mg/mL
Ratio, tsample/tref
Ratio, tsample/tref
Ratio, mmsample/mmref
D Ratio, mmsample/mmref
C
Glycine, mg/mL
Mannitol, mg/mL
Ratio, tsample/tref
Ratio, mmsample/mmref
E
Tween® 80, µg/mL
Figure 2 Influence of differently concentrated excipient solutions on ROTEM® parameters. Left y-axis: CT (black squares) and CFT ratio (open squares); right y-axis: A10 (blue circles) and MCF ratio (open circles). (A) Sucrose, (B) Trehalose, (C) Glycine, (D) Mannitol, (E) Tween® 80. Results are shown as means±standard deviation, the lines represent best-fit regressions (n = 3 whole blood samples for each substance).
i.e., especially platelets [20]. The impaired platelet function may result in an overall less stable clot. The results for glycine and mannitol were therefore compared to NaCl solutions of the same osmolar concentration using both the ROTEM® thromboelastometer and the ROTEM® platelet device to identify effects due to high osmotic pressure. The osmolarity of different solutions was calculated assuming completely dissolved substances using the following equation: cosm = ϕ·x·cmol, cosm is the osmolar concentration, ϕ is the osmotic coefficient which was set
constant (ϕ = 1; ideal solution), x is the number of osmotic dissociated units per molecule substance dissolved, and cmol is the molar concentration. As CFT is the ROTEM® para meter which shows the strongest dependency, Figure 4 compares the different CFT-ratios in dependency of the solution’s osmolarity. A non-linear regression was performed on each data set using the following equation: CFT ratio = 1+0.08 · ek·c. c is the concentration of the respective substance in mOsmol/L and k is the impairment constant. Regression data are summarized in
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
Erber and Lee: Influence of freeze drying excipients on coagulation 1609 Table 1 Regression data for CFT ratio in dependency of osmolarity.
Glycine, mg/mL
Ratio, mmsample/mmref
MaxV ratio, vsample/vref
Substance
CFT-Ratio, tsample/tref
Figure 3 Glycine influence on clotting speed. Left y-axis/lower x-axis: Correlation between A10 ratio and CFT ratio (black squares); right y-axis/upper x-axis: clot formation speed (MaxV) in dependency of glycine concentration (open circles). Results are shown as means±standard deviation (n = 3 whole blood samples).
CFT-Ratio, tsample/tref
Table 1. CFT ratios after the addition of differently concentrated NaCl solutions were slightly influenced. NaCl solutions with an osmolarity up to 1100 mOsm do therefore not influence clot formation. Comparing glycine and mannitol CFT ratios at the same osmolarity shows the higher influence of glycine on clot formation. As the regression represents a first order – like CFT – ratio/osmolarity correlation, the values for the impairment constant reflect the magnitude of clot formation impairment by the different substances. Different substances at definite osmolarities therefore exhibit different tendency to alter clotting and clot formation in whole blood samples, i.e., clot formation impairment is dependent on the solutions osmolarity but also depends on the substance dissolved. The absence of clotting impairment after NaCl addition demonstrates that
NaCl Mannitol Glycine
k
R2
χ2
0.97 2.11 2.69
0.751 0.947 0.959
0.442 0.406 0.652
osmolarity connected stresses are not alone responsible for the observed CFT prolongation. The ROTEM® platelet results are summarized in Table 2. The influence of glycine on platelet function is minimal even at high concentrations, yet it shows high deviations. Mannitol as well as NaCl solutions clearly impair platelet function with increasing solute concentration. This confirms the assumption that the origin of the influence of mannitol and glycine on whole blood coagulation is different. NaCl and mannitol solutions have a similar effect on platelet function at the same osmolar concentration, this effect is therefore attributed to osmotic pressure. As glycine does not exhibit a similar influence it is assumed that glycine may be rapidly transported into platelets thus reducing osmotic pressure. This transport could be due to passive diffusion [26] or by active transport as has been reported for L-arginine in platelets [27] or glycine into other cells [28, 29]. During the FIBTEM® assay (Figure 5) both mannitol and glycine showed an impaired clot stability, i.e., decreased MCF compared to the reference. The effect on fibrin-based clot formation was stronger for glycine decreasing the MCF at a concentration of 100 mg/mL to about 50% of the initial MCF. Addition of a mannitol solution of 200 mg/mL decreased the MCF to about 80% of the initial MCF. This experiment confirms the assumption made earlier regarding clot destabilization by glycine due to fibrin formation impairment. Considering the ROTEM® platelet results the destabilizing effect of glycine on clot formation is majorly attributed to the impairment of fibrin formation. Within the range observed mannitol Table 2 ROTEM® platelet results for glycine, mannitol and NaCl solutions (n = 3 whole blood samples). c, osmol/L
Osmolarity, osmol/L
Figure 4 CFT ratio in dependency of the solution’s osmolarity and the substance dissolved. Results are shown as means±standard deviation.
0.274 0.547 0.658 0.767 0.85 0.986 1.096
Glycine
Mannitol
NaCl
A6 ratio
sdv
A6 ratio
sdv
A6 ratio
sdv
0.771 0.789 1.046 0.954 0.899 0.936 0.954
0.049 0.130 0.078 0.052 0.026 0.026 0.156
0.913 0.893 0.892 0.885 0.699 0.655 0.454
0.062 0.071 0.064 0.070 0.100 0.089 0.063
0.714 0.661 0.699 0.608 0.620 0.578 0.476
0.074 0.074 0.026 0.009 0.027 0.017 0.054
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
FIBTEM MCF ratio, mmsample/mmref
1610 Erber and Lee: Influence of freeze drying excipients on coagulation
Concentration, mg/mL
Figure 5 FIBTEM MCF dependency of glycine and mannitol concentration. Results are shown as means±standard deviation, the dashed line represents a best-fit regression (n = 3 whole blood samples).
only slightly decreases FIBTEM® MCF but clearly affects platelet function. The influence of mannitol on clot destabilization therefore appears to be more complex, i.e., a combination of at least two mechanisms, as platelet function as well as fibrin formation are effected.
Tween® 80 As Tween® 80 is a surfactant which is typically used at low concentrations [approx. 0.01% (m/V) [2]], solutions with the following concentrations were added to the whole blood samples: 25, 50, 75, 100, 150 and 200 μg/mL. Within the range observed Tween® 80 did not influence coagulation kinetics determined with the ROTEM® device (Figure 2E). Some values deviate from the reference value which is negligible due to larger variation. When used in concentrations typical to preserve protein activity during freezing Tween® 80 is, therefore, uncritical regarding coagulation. The precise influence of Tween® associated surface tension reduction needs further investigation. As the reduced surface tension may lead to reduced friction at all surfaces, as was reported for Pluronic® F-68 [30], the influence of blood surface tension on the ROTEM® measurement should be subjected to a future experiment.
Conclusions In summary, both disaccharides investigated in this study neither influence CT nor the clot firmness parameters and are therefore uncritical regarding whole blood coagulation diagnostics. Tween® 80 used in a concentration typical
for surfactants in freeze drying did not also influence the ROTEM® measurement. Both crystalline bulking agents, glycine and mannitol, did, however, alter the coagulation profile obtained during ROTEM® analysis. There was no influence on clotting initiation by mannitol and the influence on CT was small for glycine. These findings confirm previous studies on plasmatic coagulation. Observing coagulation beyond CT, both clotting speed and clot strength are impaired by glycine and mannitol solutions. The effect observed for both solutions is, however, not only due to increased osmolarity as the comparison to NaCl solutions of equal osmolar concentration demonstrates. The ROTEM® platelet results show that coagulation impairment by glycine is not due to platelet function impairment which could be a result from glycine transport, either active or passive, into platelets. Mannitol and NaCl solutions alter platelet function in a similar way due to osmotic pressure. As hypertonic NaCl solutions do not influence coagulation, the effect of mannitol could result from both fibrin formation impairment which was confirmed by a decreased MCF during the FIBTEM® experiment and platelet function impairment. Based on the FIBTEM® results the effect of glycine on coagulation was attributed to fibrin formation impairment. Although glycine concentration was low compared to studies focusing on fibrinogen precipitation the effect of glycine on whole blood coagulation may, however, be due to precipitation dependent decreased fibrinogen concentration. The results reported in this study were generated by in vitro analysis using high concentrations of the respective solute. The results may therefore be of limited applicability in clinical practice. The thresholds reported, however, need to be considered when: 1) diluting a whole blood sample with solutions containing the respective compounds; 2) applying a highly concentrated mannitol solution, e.g., in order to reduce intracranial pressure; and when 3) dissolving lyophilized reagents directly with blood samples. Acknowledgments: The university medical center’s blood donation service as well as the volunteers are gratefully acknowledged for the preparation of the whole blood samples and their participation, respectively. Gaby Lange and Ilona Muslaeva are gratefully acknowledged for performing the ROTEM® platelet analysis and their expert technical assistance. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission. Financial support: None declared. Employment or leadership: None declared. Honorarium: None declared.
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
Erber and Lee: Influence of freeze drying excipients on coagulation 1611
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.
References 1. Pikal MJ. Freeze drying. In: Swarbrick J, editor. Encyclopedia of pharmaceutical technology, 3rd ed. London: Informa Healthcare, 2002:1299–326. 2. Wang W. Lyophilization and development of solid protein pharmaceuticals. [review] Int J Pharm 2000;203:1–60. 3. Tang XC, Nail SL, Pikal MJ. Freeze-drying process design by manometric temperature measurement: design of a smart freeze-dryer. Pharm Res 2005;22:685–700. 4. Wang DQ, Hey JM, Nail SL. Effect of collapse on the stability of freeze-dried recombinant factor VIII and alpha-amylase. J Pharm Sci 2004;93:1253–63. 5. Rand MD, Lock J, Van’t Veer C, Gaffney D, Mann K. Blood clotting in minimally altered whole blood. Blood 1996;88: 3432–45. 6. Roche AM, James MF, Grocott MP, Mythen MG. Citrated bood does not reliably reflect fresh whole blood coagulability in trials of in vitro hemodilution. Anesth Analg 2003;96:58–61. 7. Mardel S, Saunders F, Allen H, Menezes G, Edwards C, Ollerenshaw L, et al. Reduced quality of clot formation with gelatinbased plasma substitutes. Br J Anaest 1998;80:204–7. 8. Petroianu GA, Liu J, Maleck WH, Mattinger C, Bergler WF. The effect of in vitro hemodilution with gelatin, dextran, hydroxyethyl starch, or Ringer’s solution on thrombelastograph®. Anest Anal 2000;90:795–800. 9. Niemi TT, Kuitunen AH. Artificial colloids impair haemostasis. An in vitro study using thromboelastometry coagulation analysis. Acta Anaesthesiol Scand 2005;49:373–8. 10. Bakaltcheva I. Effects of high-molecular-weight cryoprotectants on platelets and the coagulation system. Cryobiology 2000;40:283–93. 11. Luostarinen T, Niiya T, Schramko A, Rosenberg P, Niemi T. Comparison of hypertonic saline and mannitol on whole blood coagulation in vitro assessed by thromboelastometry. Neurocrit Care 2011;14:238–43. 12. Carpenter JF, Hand SC, Crowe LM, Crowe JH. Cryoprotection of phosphofructokinase with organic solutes: characterization of enhanced protection in the presence of divalent cations. Arch Biochem Biophys 1986;250:505–12. 13. Chang BS, Kendrick BS, Carpenter JF. Surface-induced denaturation of proteins during freezing and its inhibition by surfactants. J Pharm Sci 1996;85:1325–30.
14. Casella JF, Flanagan MD, Lin S. Cytochalasin D inhibits actin polymerization and induces depolymerization of actin filaments formed during platelet shape change. Nature 1981;293:302–5. 15. Theusinger OM, Schroder CM, Eismon J, Emmert MY, Seifert B, Spahn DR, et al. The influence of laboratory coagulation tests and clotting factor levels on Rotation Thromboelastometry (ROTEM(R)) during major surgery with hemorrhage. Anesth Analg 2013;117:314–21. 16. Ogawa S, Szlam F, Chen EP, Nishimura T, Kim H, Roback JD, et al. A comparative evaluation of rotation thromboelastometry and standard coagulation tests in hemodilution-induced coagulation changes after cardiac surgery. Transfusion 2012;52:14–22. 17. Ogawa S, Szlam F, Bolliger D, Nishimura T, Chen EP, Tanaka KA. The impact of hematocrit on fibrin clot formation assessed by rotational thromboelastometry. Anesth Analg 2012;115:16–21. 18. Lang T, Bauters A, Braun SL, Pötzsch B, von Pape K-W, Kolde H-J, et al. Multi-centre investigation on reference ranges for ROTEM thromboelastometry. Blood Coagul Fibrinolysis 2005;16: 301–10. 19. Tem-Innovations-GmbH. Instructions for use – trap-tem® ref 600-13. Reference ranges 2014. p 9. 20. Wilder DM. Hypertonic resuscitation and blood coagulation in vitro comparison of several hypertonic solutions for their action on platelets and plasma coagulation. Thromb Res 2002;107:255–61. 21. Armitage WJ, Parmar N, Hunt CJ. The effects of osmotic stress on human platelets. J Cell Physiol 1985;123:241–8. 22. Law P. The tolerance of human platelets to osmotic stress. Exp Hematol 1983;11:351–7. 23. Lorand L, Konishi K, Jacobsen A. Transpeptidation mechanism in blood clotting. Nature 1962;194:1148–9. 24. Kazal LA, Amsel S, Miller OP, Tocantins LM. The preparation and some properties of fibrinogen precipitated from human plasma by glycine. Exp Biol Med 1963;113:989–94. 25. Mosesson MW, Sherry S. The preparation and properties of human fibrinogen of relatively high solubility. Biochemistry 1966;5:2829–35. 26. Boullin DJ, Green AR. Mechanisms by which human blood platelets accumulate glycine, gaba and amino acid precursors of putative neurotransmitters. Br J Pharmacol 1972;45:83–94. 27. Vasta V, Meacci E, Farnararo M, Bruni P. Identification of a specific transport system for l-arginine in human platelets. Biochem Biophys Res Comm 1995;206:878–84. 28. Ellory JC, Jones SE, Young JD. Glycine transport in human erythrocytes. J Physiol 1981;320:403–22. 29. Bannai S, Tateishi N. Role of membrane transport in metabolism and function of glutathione in mammals. J Membr Biol 1986;89:1–8. 30. Hymes AC, Safavian MH, Gunther T, Forsman J. The influence of an industrial surfactant pluronic F-68, in the treatment of hemorrhagic shock. J Surg Res 1971;11:191–7.
Brought to you by | Stockholms Universitet Authenticated Download Date | 9/5/15 7:22 AM
