T ROARN ISGFIUNSAI LO NA PR RT AI C LT EI C E The impact of whole blood processing and freezing conditions on the quality of therapeutic plasma prepared from whole blood Stefan Runkel,1 Walter E. Hitzler,1 and Peter Hellstern2

BACKGROUND: The quality of whole blood (WB)derived plasma preparations has been the subject of several studies, but there has been a lack of robust, comparative data for the different methods of processing and freezing. STUDY DESIGN AND METHODS: Six WB-derived plasma units were pooled and split (n = 16) and frozen within either 8 or 24 hours after WB collection, stored at 4°C or at room temperature (RT), and then frozen either slowly at −20°C or rapidly to below −30°C. Plasma units were tested for fibrinogen, Factor (F)V, FVII, FVIII, FXI, and von Willebrand factor (VWF), protein C (PC), protein S (PS) activity and free PS, prothrombin time, and partial thromboplastin time. RESULTS: FVIII was reduced by 9% to 19% after having been stored for 24 hours irrespective of storage temperature. Slow freezing (SF) reduced FVIII by 17% to 25% compared to rapid freezing (RF) to below −30°C. Storage temperature, but not 24-hour storage, decreased PS activity by 20% to 28%. PS activity was 8% to 17% lower in plasma units frozen slowly compared to RF. Storage and freezing had no influence on free PS. SF caused small losses of FVII and FXI activity. CONCLUSION: Twenty-four-hour hold at RT and SF both reduce FVIII levels below 70 U/dL in many plasma units. PS activity is affected substantially by storage temperature and SF, but free PS is not. With regard to plasma quality, freezing to below −30°C within 1 hour is superior to SF at −20°C.

796 TRANSFUSION Volume 55, April 2015

S

everal different whole blood (WB) processing and freezing regimes are currently available to the producers of non–pathogen-reduced, WB-derived therapeutic plasma preparations. In the United States, three types of single-donor plasma have been licensed.1 Fresh-frozen plasma (FFP) is separated from WB and frozen at not more than −18°C within 8 hours of blood collection. Plasma frozen within 24 hours after phlebotomy (PF24; known in the United States also as FP24) is prepared from WB or apheresis collection and stored at 1 to 6°C within 8 hours of collection and frozen at not more than −18°C within 24 hours of collection. Plasma frozen within 24 hours after phlebotomy held at room temperature up to 24 hours after phlebotomy (PF24RT) is prepared from WB or apheresis collection, stored at room temperature (RT; 22 ± 2°C) and frozen at not more than −18°C within 24 hours of collection. European recommendations require plasma to be separated from WB by hardspeed centrifugation between 6 and 18 hours after

ABBREVIATIONS: APTT = activated partial thromboplastin time; PC = protein C; PF24 = plasma frozen within 24 hours after phlebotomy; PF24RT = plasma frozen within 24 hours after phlebotomy held at room temperature up to 24 hours after phlebotomy; PS = protein S; PT = prothrombin time; RF = rapid freezing; RP = reference plasma; RT = room temperature; SF = slow freezing; VWF:RCo = Von Willebrand factor ristocetin cofactor activity; WB = whole blood. From the 1Transfusion Center, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany; and the 2Institute of Hemostaseology and Transfusion Medicine, Academic City Hospital, Ludwigshafen, Germany. Address reprint requests to: Peter Hellstern, MD, Academic City Hospital, Institute of Hemostaseology, 67063 Ludwigshafen, Germany; e-mail: [email protected]. This publication includes essential aspects of the dissertation by Stefan Runkel. Received for publication August 19, 2014; revision received September 16, 2014, and accepted September 17, 2014. doi: 10.1111/trf.12914 © 2014 AABB TRANSFUSION **;**:**-**. 2015;55:796–804. Volume **, ** **

TRANSFUSION

1

RUNKEL ET AL.

STORAGE, FREEZING, AND PLASMA QUALITY

collection or from WB that has been rapidly cooled immediately after donation to maintain RT and then held at this temperature for up to 24 hours.2 In contrast to US standards, European plasma must be frozen completely to a temperature below −30°C within 1 hour (UK guidelines specify −25°C or below3). The quality of non–pathogen-reduced, WB-derived plasma types currently in use has not yet been examined sufficiently. Most comparative studies of different WB preparation methods were unpaired or analyzed aliquots from the original plasma bags, which do not reflect the influence of freezing and thawing on plasma quality in the original bags.4-12 Only one study thus far used full-sized plasma units from pooled and split plasma units for comparison between FFP and PF24RT frozen at −18°C.13 How slow freezing (SF) at −18°C and rapid freezing (RF) to less than −25 or less than −30°C as practiced in Europe impacts on coagulation factor and inhibitor activities has not yet been studied comparatively. This study compared directly the influence on plasma quality of different storage times and temperatures and the speed of freezing by preparing plasma pools from sex and ABO blood group–identical WB units, which were split into identical 230-mL plasma units.

processes. After informed consent, 96 individuals (six women, 90 men) donated 460 mL of WB into standard blood collection bags (T3902, Fresenius Hemocare, Bad Homburg, Germany) containing 74 mL of CPD. The WB units were cooled to RT within 1 hour using a blood processor (Compocool, Fresenius) and subsequently leukoreduced by inline filtration. Plasma was separated from blood cells after hard-spin centrifugation for 12 minutes at 4955 × g using an automated blood processor (Compomat G4, Fresenius). Six ABO blood group and sex-identical plasma units each were transferred to a plastic bag (LB2.2, Fresenius). Sixteen plasma pools were received from the 96 donors (eight pools blood group A, eight pools blood group O; one blood group A pool was from six female donors). After careful mixing, 230 mL of plasma was distributed from the plasma pool to each of six plastic bags (P4162, Fresenius) within 8 hours of blood collection (Fig. 1A). Six different plasma types were produced: 1. 2.

3.

MATERIALS AND METHODS Plasma preparation Plasma units were pooled and split into six identical 230-mL plasma units for the six different manufacturing

4.

Plasma frozen within 8 hours of blood collection at −20°C (FFPSF); Plasma frozen within 8 hours of blood collection, frozen within 1 hour to a temperature below −30°C (FFPRF); Plasma stored at +4°C within 8 hours of blood collection and frozen at −20°C 24 hours after collection (PF24SF); Plasma stored at +4°C within 8 hours of blood collection and frozen within 1 hour to a temperature below −30°C 24 hours after collection (PF24RF);

Fig. 1. Experimental plasma production design. (A) Plasma pools each from six ABO blood group– and sex-identical WB-derived plasma units; n = 16 in each group. (B) Two WB subunits each from 1 WB unit (n = 12). 2

TRANSFUSION

Volume **, ** **

Volume 55, April 2015 TRANSFUSION 797

RUNKEL ET AL.

5. 6.

Plasma stored at RT and frozen at −20°C 24 hours after collection (PF24RTSF); and Plasma stored at RT and frozen within 1 hour to below −30°C 24 hours after collection (PF24RTRF).

STORAGE, FREEZING, AND PLASMA QUALITY

national Reference Preparations. Calibration curves were produced for each assay run.

Statistical analysis

Two-milliliter aliquots were drawn from the plasma pools immediately after pooling, frozen in liquid nitrogen, and stored at −70°C to serve as controls (reference plasma [RP]), the assumption being that impairment of clotting factor and inhibitor activities caused by storage and freezing under these conditions would be minimal. RF was performed in a freezer from Hof (Lohra, Germany), which achieved complete freezing of plasma units to below −30°C within 14 to 45 minutes. Plasma units frozen slowly were placed on the bottom of a freezer set at −20°C. Temperature probes placed into plasma dummies showed that the time until complete freezing to −20°C under these conditions was up to 730 minutes. After freezing the slowfrozen plasma units were stored at −21 ± 0.5°C. Rapidfrozen plasma was stored at −31.7 ± 1°C. Mean storage time for both types of plasma was 111 days (minimum, 49 days; maximum, 161 days). To study the influence on quality of the time that plasma spends in the presence of blood cells, 12 further leukoreduced WB units were split into two subunits. Plasma was separated from one WB subunit 8 hours after collection, stored at RT, and frozen 24 hours after collection at −20°C (PF24RTSF 1). The second subunit was stored at RT, and the plasma was separated and frozen at −20°C 24 hours after collection (PF24RTSF 2; Fig. 1B).

Sample sizes were estimated using computer software (StatMate 2.0 for Windows, GraphPad Software, Inc., La Jolla, CA). The estimated standard deviation for power calculations was 20 for clotting factors and inhibitors except fibrinogen. The estimated correlation coefficient was 0.9. For the paired studies, the samples size of 16 and 12 were based on the ability to detect 7 and 8% differences, respectively, between mean clotting factors, clotting inhibitors, or coagulation screening tests in the plasma pairs at a two-tailed p value of 0.05 and a power of 90%. The D’Agostino Pearson omnibus test for normality was applied to analyze the values. Because most values were not distributed normally, data are given as medians and ranges (minimum − maximum). The paired data were compared using the Wilcoxon matched pairs test. Any two-tailed probability below 0.05 was considered significant when two samples were compared. To compensate for multiple comparisons when comparing more than two samples, two-tailed p values below 0.01 were regarded as significant.

Laboratory tests

Table 1 shows the coagulation factor and inhibitor levels, PT ratios, and APTT in the three plasma types stored under different conditions before freezing and frozen slowly at −20°C (left part of Table 1) or rapidly to below −30°C (right part of Table 1). Storage temperature before freezing (PF24 vs. PF24RT) had no observable influence on FVIII activity. Markedly and significantly lower plasma levels of FVIII were observed in PF24 and PF24RT compared to FFP in both freezing regimes. In PF24 and PF24RT frozen slowly to −20°C (SF) FVIII was 19% and 18% lower, respectively, than in FFPSF (p < 0.0001). PF24 and PF24RT frozen rapidly to below −30°C (RF) had 11 and 9% lower FVIII activities, respectively, than FFPRF (p < 0.0001). Storage at RT reduced PS activity but not free PS strongly when compared with storage at 4°C (PF24SF vs. PF24RTSF, 27%, p < 0.0001; PF24RF vs. PF24RTRF, 20%, p < 0.0001), but PS activity was not significantly influenced by storage time. A small but significantly lower FVII activity was seen in PF24RF compared to FFPRF (7%, p = 0.0001). Neither storage time nor storage temperature had any significant impact on the levels of FV, FXI, fibrinogen, VWF:RCo, PC, and free PS.

Two technical assistants thawed the plasma units and RP aliquots at 37°C and performed analysis within 4 hours of thawing. Factor (F)VIII (FVIII) and protein S (PS) were measured within 2 hours after thawing. All assays were run on analyzers (ACL TOP, Instrumentation Laboratory, Kirchheim, Germany), using reagents (HemosIL, Instrumentation Laboratory) from the same company. Single lots of all reagents were used. The prothrombin time (PT) ratios and the activated partial thromboplastin time (APTT) were determined using reagents (RecombiPlasTin and SynthASil, Instrumentation Laboratory), respectively. Fibrinogen was measured using the Clauss method. FV, FVII, FVIII, and FXI were measured using one-stage clotting assays and factordeficient plasmas. Protein C (PC) was assessed by chromogenic substrate assay, and PS activity, by a two-stage clotting assay using PS-deficient plasma. A latex immunoassay was used to determine free PS. Von Willebrand factor (VWF) ristocetin cofactor activity (VWF:RCo) was quantified turbidimetrically. All quantitative assays were calibrated with calibration plasma (HemosIL, Instrumentation Laboratory), which had been carefully calibrated against current Inter798 TRANSFUSION Volume 55, April 2015

RESULTS Effect of prefreezing storage conditions on coagulation factors, inhibitors, and coagulation screening tests

Volume **, ** **

TRANSFUSION

3

4

TRANSFUSION

Volume **, ** ** 92 (76-108)

79 (54-94)

99 (89-107)

35.8 (32.2-40.6)

1.05 (0.98-1.16)

82 (56-150)

2.25 (1.7-3.3)

87 (76-100)

57 (33-79)

71 (57-90)

B: PF24SF 69 (58-91)

84 (77-108)

57 (40-76)

97 (84-122)

35.8 (31.6-40.6)

1.06 (0.89-1.16)

91 (32-170)

2.26 (1.74-2.95)

85 (69-120)

58 (34-103)

74 (53-88)

C: PF24RTSF 68 (54-97)

% diff p value 9 NS 2 NS 18 0.002 1 NS 0 NS 3 NS 3 NS 4

The impact of whole blood processing and freezing conditions on the quality of therapeutic plasma prepared from whole blood.

The quality of whole blood (WB)-derived plasma preparations has been the subject of several studies, but there has been a lack of robust, comparative ...
145KB Sizes 1 Downloads 5 Views