Vox Sanguinis (2014) © 2014 International Society of Blood Transfusion DOI: 10.1111/vox.12159
ORIGINAL PAPER
New approach to ‘top-and-bottom’ whole blood separation using the multiunit TACSI WB system: quality of blood components A. Lotens, T. Najdovski, N. Cellier, B. Ernotte, M. Lambermont & A. Rapaille Service du Sang, Croix-Rouge de Belgique, Suarle e, Belgium
Background and Objectives TACSI whole blood system is designed to combine primary and secondary processing of six whole blood bags into plasma units, buffy coat and red blood cell concentrates. The aim of this study was to investigate the specifications and in vitro storage parameters of blood components compared with standard centrifugation and separation processing. Materials and Methods Whole blood bags, collected in CRC kits, were treated on a TACSI whole blood system. They were compared with whole blood bags collected in Composelect kits. In addition to routine quality control analyses, conservation studies were performed on red blood cell concentrates for 42 days and on plasma for 6 months. Platelets pools with five buffy coats were also created, and cellular contamination was evaluated. Results Red blood cell concentrates produced from TACSI whole blood met European quality requirements. For white blood cell count, one individual result exceeded 1 9 106 cells/unit. All plasma units fell within specifications for residual cellular contamination and storage parameters. The performances of the TACSI whole blood system allow for the preparation of low volume buffy coats with a recovery of 90% of whole blood platelets. Haemoglobin losses in TACSI BC are smaller, but this did not result in higher haemoglobin content of red cells. These BC are suitable for the production of platelet concentrates.
Received: 17 December 2013, revised 24 April 2014, accepted 25 April 2014
Conclusion From these in vitro data, red blood cell concentrates produced using TACSI whole blood are suitable for clinical use with a quality at least equivalent to the control group. Key words: blood components, separation system, storage study, TACSI WB.
Introduction Whole blood (WB) separation into components is widely adopted in order to maximize blood component quality and availability for self-sufficiency in many countries. Separation strategies are adapted to supply priorities [1]. In many European (EU) Blood Services, WB collection is driven by the red blood cell concentrate (RBCC) supply, but separation strategy is designed to maximize the use of the collected WB units for RBCC preparation and as Correspondence: Ana€ıs Lotens, Service du Sang, Croix-Rouge de Belgique, 8A Rue du Fond du Marechal, 5020 Suarlee, Belgium E-mail : [email protected]
raw material for the production of platelet concentrates (PC) and plasma units. In our centre, although we successfully validated prestorage leucocyte reduction in WB as the appropriate technique for RBCC and plasma units production [2], WB is collected in ‘top-and-bottom’ packs and separated into RBCC, buffy coat (BC) and plasma units. Leucodepletion is performed by components filtration as described for many years [3–5], widely spread throughout Europe and recently adopted by Canadian Blood Services [6]. From the beginning of the ‘top-andbottom’ systems, automation of the technique’s numerous manual steps was an important topic [3, 4]. Automated presses allowed to reduce manual handling and to obtain reproducible separations. Further developments enhanced
1
2 A. Lotens et al.
productivity and achieved shorter separation times [4, 7]. First-generation automated WB separators focused on automation of the components expression step. Recent developments focused on the integration of centrifugation and separation steps into a single instrument to produce RBCC, BC and plasma units in one step. These instruments handled a single WB unit per cycle [8], but subsequent versions could handle multiple units per run [9, 10]. Most recent evolutions focus on processing WB and BC pools in the same instrument. This study is the first published report on the use of the Terumo Automated Centrifuge & Separator Integration (TACSI) WB system. Equipment developed to perform a first run for the separation of WB collected in a specific ‘top-and-bottom’ collection set and a second run for the separation of pooled BCs into PC. However, there are published papers on the use of the TACSI PL for pooled BC platelets concentrates production [11, 12]. The quality of the components is evaluated in accordance with EU quality requirements.
Materials and methods Study design A study design with two phases was created to evaluate the TACSI WB system (Terumo BCT, Tuttlingen, Germany) for WB separation in ‘top-and-bottom’ kits (TACSI CRC kit, Terumo BCT, Japan). In the first phase (phase 0), the aim was to assess the quality of blood components obtained from 12 WB donations collected in CRC kits and separated using the centripress TACSI WB and compared with 12 WB donations collected with our standard collection packs (quadruple ‘top-and-bottom’ CE filter bag, Composelect, Fresenius Kabi, Bad Homburg, Germany). A storage study was set up including RBCC conservation at +4°C up to 42 days and fresh-frozen plasma (FFP) at -25°C up to 30 days. The second phase (phase 1) represents a confirmation study of the quality of blood components. 142 WB units were treated on TACSI WB system, and 30 random RBCC and the 30 corresponding plasma units were analysed. A storage study of plasma, which was pooled by five units was assayed with analyses on day 1 and at 6 months. Twelve PC from 5 pooled BC were realized. In addition, 100 RBCC samples were analysed by flow cytometry to determine white blood cell (WBC) count. RBCCs collected in phase 1 were transfused.
guidelines, was collected in a 63-ml citrate-phosphatedextrose (CPD) solution in either a conventional quadruple bag system with 100 ml of saline-adenine-glucose-mannitol (SAGM) preservation solution (Composelect, control group) or in a new quadruple bag system containing an additional plastic cassette and various breakaways (test group). TACSI CRC kits were used with DOCON blood mixing scales (Maco Pharma, Mouvaux, France), which had a plateau compatible with the plastic cassette of the kit. WB collected units were kept overnight at +20°C to +24°C on heat-absorbing plates (CompoCool; Fresenius Kabi) before centrifugation and separation.
Separation The Composelect bags were centrifuged in a Hettich Roto Silenta 630 RS (Hettich, Tuttlingen, Germany) at 20°C (36 9 103 g/min) and separated on T-Ace II (Terumo BCT Europe NV, Zaventem, Belgium). In this standard ‘topand-bottom’ collection bags, the SAGM is contained in a separate bag and is added to the RBCC by inversion in an additional step. The TACSI CRC WB units were weighed and paired 2 by 2 with a maximum weight difference of 5 g. Parameterization targeted a BC volume of 42 ml and an haematocrit (Hct) of 32%. Separation consisted in a sedimentation phase followed by a separation phase (17 9 103 g/min). TACSI WB system separated 6 WB bags per run and expressed the RBCC immediately during separation in the SAGM containing bag. At the end of the separation cycle, the TACSI WB separator was unloaded, inserts were disassembled, and tubings were sealed with a desktop sealing device. RBCC were hung for leucodepletion, excess air was expelled from the RBCC bag in the initial bag, tubing was sealed, and RBCC were stored in a ventilated cold room at +2 to +6°C (AFNOR validated NFX15-140). Plasma units were frozen in shock freezers (Dometic, Hosingen, Luxembourg) to -30°C or below within 60 min and stored at -25°C or lower. After 4 h resting, 5 BCs were pooled using a pooling kit (Terumo TACSI PL kit, Zaventem, Belgium) with the addition of 250 ml platelet additive solution (PAS) (SSP+, Macopharma, Mouvaux, France). BC pools were further separated into specific inserts for second separation on the same TACSI WB system (1 9 103 g/min).
Sampling and analyses Phase 0
WB collection 450 ml WB from healthy volunteer donors, meeting standard blood donor criteria in accordance with EU/national
The RBCC storage study was carried out at +4°C, and units were weighed and sampled on days 1, 14, 28 and 42. Haematological analyses, pH, glucose, potassium, ATP, 2,3 DPG and haemolysis were measured. At the © 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
Evaluation of new TACSI WB system 3
storage study end-point, bacterial contamination detection was performed with BacT-Alert system (bioMerieux, Marcy l’Etoile, France). Plasma units and BC were analysed on day 1 for haematological data. BC platelet recovery is calculated by the platelet content in BC divided by the platelet content from WB. A plasma storage study was completed with analyses on day 1 on fresh plasma and day 30 on frozen-thawed plasma for coagulation factors (V, VII, VIII, IX, X and von Willebrand factor (vW), prothrombin time (PT), D-dimers, activated partial thromboplastin time (aPTT), fibrinogen, pH and total proteins. All samples were snap-frozen and stored at a temperature below – 65°C before analyses.
Phase 1 For quality control analyses, 30 RBCC and 30 plasma units were sampled on day 1 before storage. 100 RBCC samples were analysed for WBC count using the flow cytometric method. A 6-month plasma storage study was performed in a pool-split design where 10 pools of 5 plasma units were aliquoted into 250-ml plasma units. The following analyses were then carried out: coagulation factors (II, V, VII, VIII, IX, X, vW), fibrinogen, D-dimers, PT, aPTT, albumin, proteins C and S, total proteins, antithrombin III, pH, IgG, IgA and IgM. Twelve PC produced on TACSI WB from pools of 5 BCs were analysed on day 1 for volume, haematological data and PLT recoveries. Platelet recovery is calculated by the platelet content in the final pool divided by the sum of the five platelet contents stemming from WB.
(Dade Behring) on the Sysmex device. FVIII and FIX were quantified using a one-stage clotting time with Actin-FS (Dade Behring) on a CA-1500 automate (Sysmex Ltd). FII, FV, FVII and FX concentrations were measured using a one-stage clotting time with Innovin (Dade Behring) on a CA-7000 automate (Sysmex Ltd). The vW factor was measured with an immunological latex test using a vW Factor Antigen Kit (Instrumentation Laboratory, Zaventem, Belgium). D-dimers were tested with an immunological latex test using a D-dimer HS 500 Kit (Instrumentation Laboratory). Albumin was analysed by capillary electrophoresis (Capillarys, SEBIA, Evry, France) with a Protein 6 Kit. The total protein level was measured using a spectrophotometric method (Biuret) on a DXC 800 automate (Beckman Coulter, Fullerton, CA, USA). Protein C and antithrombin III levels were tested in a chromogenic assay, Berichrom Protein C and Berichrom Antithrombin (Dade Behring), respectively, on a CA-7000 Automate (Sysmex UK Ltd). Protein S concentration was evaluated by an in-house ELISA method, using Dalko antibodies (DakoCytomation, Cambridgeshire, UK). IgG, IgA and IgM were analysed by immunoturbidimetry (Cobas 8000, module 502; Roche).
Clinical use One hundred and thirty RBCC units were issued to one hospital for transfusion, followed by a routine haemovigilance programme.
Statistical analysis Assays Haematological analyses were performed on a haematological analyser (XE-2100D; Sysmex, Netherlands). Free Hb was measured with a modified Harboe method using Evolis automate [13] (Biorad, France). Leucocytes in the RBCC and plasma units were counted using a cytometric method (FACSCalibur; Becton Dickinson, Palo Alto, CA, USA) with a BD leucocount kit (Becton Dickinson). pH was measured using a pH meter (Inolab pH 720, WTW) at 22°C. Supernatant was assayed for extracellular K+ and glucose using a blood gas analyser (RapidLab). Haemolysis was calculated applying the following formula: haemolysis = (100% - Hct% x free Hb)/total Hb. After perchloric acid extraction, 2,3 DPG was analysed using the 2,3 DPG kit from Roche (Mannheim, Germany) and ATP was analysed with the glucose/hexokinase reaction [14]. PT and aPTT were tested on a CA-7000 automated coagulation analyser (Sysmex Ltd, Milton Keynes, UK) using a recombinant reagent Actin-FS (Dade Behring, Marburg, Germany). The fibrinogen concentration was measured by Clauss’s method with thrombin reagent © 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
Results are expressed as mean and standard deviation (SD) except otherwise stated. The values were analysed using the Anderson-Darling test for normal distribution. For results with a normal distribution, a Student’s t-test was used to compare said results, test group versus control group. When results from the same group were compared, we used the paired t-test. We considered P values 90% in table 2). Mean Hb content was not statistically different between the 2 groups. Results indicate that the leucodepletion filter in TACSI CRC kit is probably less efficient than in the control group but nevertheless produces RBCC compliant
with EU requirements for residual WBC counts and suitable for transfusion. Haemolysis results were similar than previous results published by Gulliksson with similar blood bags (Terumo and Fresenius T&B 4ple bags overnight storage) [16] indicating that the TACSI WB procedure does not induce RBC damage, which further develops during storage. From the start of the storage study, the glucose concentration of the control group was slightly higher than in the test group and remained statistically different until day 42. This difference is explained with the significant different volume results. In both groups, residual glucose was around 50% of initial concentration at the end of storage and was available in sufficient quantity to sustain cell metabolism. Potassium concentrations in both groups were 65) at the end of storage, and results were similar to those reported by Johnson et al. [10] and Thomas and colleagues [17]. Residual cellular contaminations of plasma units from the test group fell within specifications and were significantly lower than in the control group. Storage study results confirm that plasma units from test and control groups behave in a similar manner and are in line with the literature [2, 21]. All studied results fell within wellaccepted limits [21] indicating that plasma units produced by TACSI WB are of acceptable quality and show very low residual cellular contamination. As regards BC, the separation processes for control and test group were not comparable, and the physical and haematological parameters of the BC for both groups were intentionally modified to optimize RBC recovery in RBCC and to preserve platelet yield in PC. The performances of the TACSI WB allowed for smaller losses of Hb, PC preparation with pools of 5 BC and PLT recovery higher than 90%. BC from the TACSI WB group showed mean volume and mean PLT content similar to data published by Tardivel et al. [22] Concerning the PLT components, additional studies must be performed to assess in vitro functional properties for PLT components to confirm their suitability for transfusion. In addition to logistic impacts, the introduction of TACSI WB in routine collection operations required accurate training of collection teams, adapted blood mixing © 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
Evaluation of new TACSI WB system 7
scales with a plateau that can accommodate the plastic cassette from the CRC kit. This cassette makes whole blood donations more bulky with significant impact on WB packaging for transportation to processing centre. TACSI WB is a new technology with a high level of automation of processing steps that required a redesign of the WB processing facility. The process flow was simplified, and the footprint of the processing facility was significantly reduced. Intensive training of the processing team was necessary before working with this system, in particular for the mounting of the CRC collection kit in the specific inserts. This step is an important manual step to avoid centrifugation and separation problems during processing. In addition, the insert mounting time is critical for capacity and productivity. In this study with a limited number of processed donations, reject levels imputable to TACSI WB were acceptable (21%) but could be improved compared with our routine operation (04%, data not shown). These operational aspects must be further studied in a broader feasibility study. TASCI WB offers the possibility to optimize RBC yields and maintains acceptable PLT recoveries in small volume BC. Average plasma recovery is 11 ml less than in our standard process and needs further optimization of the
parameters. Transfused RBCCs were monitored using a routine haemovigilance programme, and no transfusion reactions were reported to date. The introduction of TACSI WB in blood services will cause organizational changes and a process flow redesign. A study is planned with 1500 WB units and will specially focus on process flow and productivity of the TACSI WB system.
Acknowledgements The authors thank the Cliniques Universitaires de Bruxelles, H^opital Erasme for the transfusion of TACSI WB blood components and for haemovigilance follow-up and reporting. The authors also thank the Coagulation and Haemostasis Laboratory of the Cliniques Universitaires Saint-Luc UCL Bruxelles and the Clinical Biology Laboratory in Centre Hospitalier Regional de Namur for the analyses.
Conflict of interest The study was sponsored by TerumoBCT. The authors declare no other conflict of interest related to the paper’s subject.
References 1 Heaton L, Rebulla P, Pappalettera M, et al.: A Comparative Analysis of Different Methods for Routine Blood Component Preparation. Transfus Med Rev 1997; 11:116–129 2 Rapaille A, Moore G, Siquet J, et al.: Prestorage leukocyte reduction with in-line filtration of whole blood: evaluation of red cells and plasma storage. Vox Sang 1997; 73:28–35 3 Pietersz RNI, Dekker WJA, Reesink HW: Comparison of a conventional quadruple-bag system with a ‘Topand-Bottom’ system for blood processing. Vox Sang 1990; 59:205–208 4 Claes F, Eriksson L, Hedlund K, et al.: The bottom and top system: a new technique for blood component preparation and storage. Vox Sang 1988; 55:211–217 5 Pietersz RNI, Reesink HW, de Korte D, et al.: Storage of leukocyte-poor red cell concentrates: filtration in a closed system using connection device. Vox Sang 1989; 57:29–36 6 Sheffield WP, Bhakta V, Jenkins C, et al.: Conversion to the buffy coat
7
8
9
10
11
method and quality of frozen plasma derived from whole blood donations in Canada. Transfusion 2010; 50:1043–1049 Van der Meer PF, Pietersz RNI, Hinloopen B, et al.: Automated separation of whole blood in Top and Bottom bags into components using the compomat G4. Vox Sang 1999; 76:90–99 Thomas S, Beard M, Garwood M, et al.: Blood components produced from whole blood using the Atreus processing system. Transfusion 2008; 48:2515–2524 Lagerberg JW, Salado-Jimena JA, L€ of H, et al.: Evaluation of the quality of blood components obtained after automated separation of whole blood by a new multiunit processor. Transfusion 2013; 53:1798–1807 Johnson L, Winter K, Kwok M, et al.: Evaluation of the quality of blood components prepared using the Reveos automated blood processing system. Vox Sang 2013; 105:225–235 Sandgren P, Hild M, Sj€ ofin A, et al.: Storage of Buffy-Coat-derived platelets
© 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
12
13
14
15
16
in additive solutions; in vitro effects on platelets prepared by the novel TACSI system and stored in plastic containers with different gas permeability. Vox Sang 2010; 99:341–347 Plaza EM, Cespedes P, Fernandez H, et al.: Quality assessment of buffycoat-derived leucopleted platelet concentrates in PAS-plasma, prepared by the OrbiSac or TACSI automated system. Vox Sang 2013; 106:38–44 Harboe M: A method for determination of hemoglobin in plasma by nearultraviolet spectrophotometry. Scand J Clin Lab Invest 1959; 11:66–70 de Korte D, Kleine M, Korsten HG, et al.: Prolonged maintenance of 2,3 diphosphoglycerate acid and adenosine triphosphate in red blood cells during storage. Transfusion 2008; 48:1081– 1089 Guide to the Preparation, Use and Quality Assurance of Blood Components, 16th edn. Strasbourg, France, EDQM, 2011. Gulliksson H, van der Peer PF: Storage of whole blood overnight in different
8 A. Lotens et al.
blood bags preceding preparation of blood components: in vitro effects on red blood cells. Blood Transfus 2009; 7:210–215 17 Thomas S, Hancock V, Cardigan R: Repeated short-term warming of red blood cell concentrates has minimal effect on their quality. Vox Sang 2012; 103:113–121 18 Van der Meer PF, Cancelas JA, Cardigan R, et al.: Evaluation of overnight hold of whole blood at room temperature before component processing:
effect of red blood cell (RBC) additive solutions on in vitro RBC measures. Transfusion 2011; 51(Suppl):15s–24s 19 Kurup PA, Arun P, Gayathri NS, et al.: Modified formulation of CPDA for storage of whole blood, and of SAGM for storage of red blood cells, to maintain the concentration of 2,3-diphosphoglycerate. Vox Sang 2003; 85: 253–261 20 Van Der Meer PF, Pietersz RN: Overnight storage of whole blood: a comparison of two designs of butane-
1,4-Diol cooling plates. Transfusion 2007; 47:2038–2043 21 Moog R, Reichenberg S, Hoburg A, et al.: Quality of methylene blue-treated fresh-frozen plasma stored up to 27 months. Transfusion 2010; 50:516– 518 22 Tardivel R, Bois S, Vignoli C, et al.: Automatisation de la preparation des produits sanguins labiles. Transfus Clin Biol 2009; 16:175–178
© 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
ORIGINAL PAPER
New approach to ‘top-and-bottom’ whole blood separation using the multiunit TACSI WB system: quality of blood components A. Lotens, T. Najdovski, N. Cellier, B. Ernotte, M. Lambermont & A. Rapaille Service du Sang, Croix-Rouge de Belgique, Suarle e, Belgium
Background and Objectives TACSI whole blood system is designed to combine primary and secondary processing of six whole blood bags into plasma units, buffy coat and red blood cell concentrates. The aim of this study was to investigate the specifications and in vitro storage parameters of blood components compared with standard centrifugation and separation processing. Materials and Methods Whole blood bags, collected in CRC kits, were treated on a TACSI whole blood system. They were compared with whole blood bags collected in Composelect kits. In addition to routine quality control analyses, conservation studies were performed on red blood cell concentrates for 42 days and on plasma for 6 months. Platelets pools with five buffy coats were also created, and cellular contamination was evaluated. Results Red blood cell concentrates produced from TACSI whole blood met European quality requirements. For white blood cell count, one individual result exceeded 1 9 106 cells/unit. All plasma units fell within specifications for residual cellular contamination and storage parameters. The performances of the TACSI whole blood system allow for the preparation of low volume buffy coats with a recovery of 90% of whole blood platelets. Haemoglobin losses in TACSI BC are smaller, but this did not result in higher haemoglobin content of red cells. These BC are suitable for the production of platelet concentrates.
Received: 17 December 2013, revised 24 April 2014, accepted 25 April 2014
Conclusion From these in vitro data, red blood cell concentrates produced using TACSI whole blood are suitable for clinical use with a quality at least equivalent to the control group. Key words: blood components, separation system, storage study, TACSI WB.
Introduction Whole blood (WB) separation into components is widely adopted in order to maximize blood component quality and availability for self-sufficiency in many countries. Separation strategies are adapted to supply priorities [1]. In many European (EU) Blood Services, WB collection is driven by the red blood cell concentrate (RBCC) supply, but separation strategy is designed to maximize the use of the collected WB units for RBCC preparation and as Correspondence: Ana€ıs Lotens, Service du Sang, Croix-Rouge de Belgique, 8A Rue du Fond du Marechal, 5020 Suarlee, Belgium E-mail : [email protected]
raw material for the production of platelet concentrates (PC) and plasma units. In our centre, although we successfully validated prestorage leucocyte reduction in WB as the appropriate technique for RBCC and plasma units production [2], WB is collected in ‘top-and-bottom’ packs and separated into RBCC, buffy coat (BC) and plasma units. Leucodepletion is performed by components filtration as described for many years [3–5], widely spread throughout Europe and recently adopted by Canadian Blood Services [6]. From the beginning of the ‘top-andbottom’ systems, automation of the technique’s numerous manual steps was an important topic [3, 4]. Automated presses allowed to reduce manual handling and to obtain reproducible separations. Further developments enhanced
1
2 A. Lotens et al.
productivity and achieved shorter separation times [4, 7]. First-generation automated WB separators focused on automation of the components expression step. Recent developments focused on the integration of centrifugation and separation steps into a single instrument to produce RBCC, BC and plasma units in one step. These instruments handled a single WB unit per cycle [8], but subsequent versions could handle multiple units per run [9, 10]. Most recent evolutions focus on processing WB and BC pools in the same instrument. This study is the first published report on the use of the Terumo Automated Centrifuge & Separator Integration (TACSI) WB system. Equipment developed to perform a first run for the separation of WB collected in a specific ‘top-and-bottom’ collection set and a second run for the separation of pooled BCs into PC. However, there are published papers on the use of the TACSI PL for pooled BC platelets concentrates production [11, 12]. The quality of the components is evaluated in accordance with EU quality requirements.
Materials and methods Study design A study design with two phases was created to evaluate the TACSI WB system (Terumo BCT, Tuttlingen, Germany) for WB separation in ‘top-and-bottom’ kits (TACSI CRC kit, Terumo BCT, Japan). In the first phase (phase 0), the aim was to assess the quality of blood components obtained from 12 WB donations collected in CRC kits and separated using the centripress TACSI WB and compared with 12 WB donations collected with our standard collection packs (quadruple ‘top-and-bottom’ CE filter bag, Composelect, Fresenius Kabi, Bad Homburg, Germany). A storage study was set up including RBCC conservation at +4°C up to 42 days and fresh-frozen plasma (FFP) at -25°C up to 30 days. The second phase (phase 1) represents a confirmation study of the quality of blood components. 142 WB units were treated on TACSI WB system, and 30 random RBCC and the 30 corresponding plasma units were analysed. A storage study of plasma, which was pooled by five units was assayed with analyses on day 1 and at 6 months. Twelve PC from 5 pooled BC were realized. In addition, 100 RBCC samples were analysed by flow cytometry to determine white blood cell (WBC) count. RBCCs collected in phase 1 were transfused.
guidelines, was collected in a 63-ml citrate-phosphatedextrose (CPD) solution in either a conventional quadruple bag system with 100 ml of saline-adenine-glucose-mannitol (SAGM) preservation solution (Composelect, control group) or in a new quadruple bag system containing an additional plastic cassette and various breakaways (test group). TACSI CRC kits were used with DOCON blood mixing scales (Maco Pharma, Mouvaux, France), which had a plateau compatible with the plastic cassette of the kit. WB collected units were kept overnight at +20°C to +24°C on heat-absorbing plates (CompoCool; Fresenius Kabi) before centrifugation and separation.
Separation The Composelect bags were centrifuged in a Hettich Roto Silenta 630 RS (Hettich, Tuttlingen, Germany) at 20°C (36 9 103 g/min) and separated on T-Ace II (Terumo BCT Europe NV, Zaventem, Belgium). In this standard ‘topand-bottom’ collection bags, the SAGM is contained in a separate bag and is added to the RBCC by inversion in an additional step. The TACSI CRC WB units were weighed and paired 2 by 2 with a maximum weight difference of 5 g. Parameterization targeted a BC volume of 42 ml and an haematocrit (Hct) of 32%. Separation consisted in a sedimentation phase followed by a separation phase (17 9 103 g/min). TACSI WB system separated 6 WB bags per run and expressed the RBCC immediately during separation in the SAGM containing bag. At the end of the separation cycle, the TACSI WB separator was unloaded, inserts were disassembled, and tubings were sealed with a desktop sealing device. RBCC were hung for leucodepletion, excess air was expelled from the RBCC bag in the initial bag, tubing was sealed, and RBCC were stored in a ventilated cold room at +2 to +6°C (AFNOR validated NFX15-140). Plasma units were frozen in shock freezers (Dometic, Hosingen, Luxembourg) to -30°C or below within 60 min and stored at -25°C or lower. After 4 h resting, 5 BCs were pooled using a pooling kit (Terumo TACSI PL kit, Zaventem, Belgium) with the addition of 250 ml platelet additive solution (PAS) (SSP+, Macopharma, Mouvaux, France). BC pools were further separated into specific inserts for second separation on the same TACSI WB system (1 9 103 g/min).
Sampling and analyses Phase 0
WB collection 450 ml WB from healthy volunteer donors, meeting standard blood donor criteria in accordance with EU/national
The RBCC storage study was carried out at +4°C, and units were weighed and sampled on days 1, 14, 28 and 42. Haematological analyses, pH, glucose, potassium, ATP, 2,3 DPG and haemolysis were measured. At the © 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
Evaluation of new TACSI WB system 3
storage study end-point, bacterial contamination detection was performed with BacT-Alert system (bioMerieux, Marcy l’Etoile, France). Plasma units and BC were analysed on day 1 for haematological data. BC platelet recovery is calculated by the platelet content in BC divided by the platelet content from WB. A plasma storage study was completed with analyses on day 1 on fresh plasma and day 30 on frozen-thawed plasma for coagulation factors (V, VII, VIII, IX, X and von Willebrand factor (vW), prothrombin time (PT), D-dimers, activated partial thromboplastin time (aPTT), fibrinogen, pH and total proteins. All samples were snap-frozen and stored at a temperature below – 65°C before analyses.
Phase 1 For quality control analyses, 30 RBCC and 30 plasma units were sampled on day 1 before storage. 100 RBCC samples were analysed for WBC count using the flow cytometric method. A 6-month plasma storage study was performed in a pool-split design where 10 pools of 5 plasma units were aliquoted into 250-ml plasma units. The following analyses were then carried out: coagulation factors (II, V, VII, VIII, IX, X, vW), fibrinogen, D-dimers, PT, aPTT, albumin, proteins C and S, total proteins, antithrombin III, pH, IgG, IgA and IgM. Twelve PC produced on TACSI WB from pools of 5 BCs were analysed on day 1 for volume, haematological data and PLT recoveries. Platelet recovery is calculated by the platelet content in the final pool divided by the sum of the five platelet contents stemming from WB.
(Dade Behring) on the Sysmex device. FVIII and FIX were quantified using a one-stage clotting time with Actin-FS (Dade Behring) on a CA-1500 automate (Sysmex Ltd). FII, FV, FVII and FX concentrations were measured using a one-stage clotting time with Innovin (Dade Behring) on a CA-7000 automate (Sysmex Ltd). The vW factor was measured with an immunological latex test using a vW Factor Antigen Kit (Instrumentation Laboratory, Zaventem, Belgium). D-dimers were tested with an immunological latex test using a D-dimer HS 500 Kit (Instrumentation Laboratory). Albumin was analysed by capillary electrophoresis (Capillarys, SEBIA, Evry, France) with a Protein 6 Kit. The total protein level was measured using a spectrophotometric method (Biuret) on a DXC 800 automate (Beckman Coulter, Fullerton, CA, USA). Protein C and antithrombin III levels were tested in a chromogenic assay, Berichrom Protein C and Berichrom Antithrombin (Dade Behring), respectively, on a CA-7000 Automate (Sysmex UK Ltd). Protein S concentration was evaluated by an in-house ELISA method, using Dalko antibodies (DakoCytomation, Cambridgeshire, UK). IgG, IgA and IgM were analysed by immunoturbidimetry (Cobas 8000, module 502; Roche).
Clinical use One hundred and thirty RBCC units were issued to one hospital for transfusion, followed by a routine haemovigilance programme.
Statistical analysis Assays Haematological analyses were performed on a haematological analyser (XE-2100D; Sysmex, Netherlands). Free Hb was measured with a modified Harboe method using Evolis automate [13] (Biorad, France). Leucocytes in the RBCC and plasma units were counted using a cytometric method (FACSCalibur; Becton Dickinson, Palo Alto, CA, USA) with a BD leucocount kit (Becton Dickinson). pH was measured using a pH meter (Inolab pH 720, WTW) at 22°C. Supernatant was assayed for extracellular K+ and glucose using a blood gas analyser (RapidLab). Haemolysis was calculated applying the following formula: haemolysis = (100% - Hct% x free Hb)/total Hb. After perchloric acid extraction, 2,3 DPG was analysed using the 2,3 DPG kit from Roche (Mannheim, Germany) and ATP was analysed with the glucose/hexokinase reaction [14]. PT and aPTT were tested on a CA-7000 automated coagulation analyser (Sysmex Ltd, Milton Keynes, UK) using a recombinant reagent Actin-FS (Dade Behring, Marburg, Germany). The fibrinogen concentration was measured by Clauss’s method with thrombin reagent © 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
Results are expressed as mean and standard deviation (SD) except otherwise stated. The values were analysed using the Anderson-Darling test for normal distribution. For results with a normal distribution, a Student’s t-test was used to compare said results, test group versus control group. When results from the same group were compared, we used the paired t-test. We considered P values 90% in table 2). Mean Hb content was not statistically different between the 2 groups. Results indicate that the leucodepletion filter in TACSI CRC kit is probably less efficient than in the control group but nevertheless produces RBCC compliant
with EU requirements for residual WBC counts and suitable for transfusion. Haemolysis results were similar than previous results published by Gulliksson with similar blood bags (Terumo and Fresenius T&B 4ple bags overnight storage) [16] indicating that the TACSI WB procedure does not induce RBC damage, which further develops during storage. From the start of the storage study, the glucose concentration of the control group was slightly higher than in the test group and remained statistically different until day 42. This difference is explained with the significant different volume results. In both groups, residual glucose was around 50% of initial concentration at the end of storage and was available in sufficient quantity to sustain cell metabolism. Potassium concentrations in both groups were 65) at the end of storage, and results were similar to those reported by Johnson et al. [10] and Thomas and colleagues [17]. Residual cellular contaminations of plasma units from the test group fell within specifications and were significantly lower than in the control group. Storage study results confirm that plasma units from test and control groups behave in a similar manner and are in line with the literature [2, 21]. All studied results fell within wellaccepted limits [21] indicating that plasma units produced by TACSI WB are of acceptable quality and show very low residual cellular contamination. As regards BC, the separation processes for control and test group were not comparable, and the physical and haematological parameters of the BC for both groups were intentionally modified to optimize RBC recovery in RBCC and to preserve platelet yield in PC. The performances of the TACSI WB allowed for smaller losses of Hb, PC preparation with pools of 5 BC and PLT recovery higher than 90%. BC from the TACSI WB group showed mean volume and mean PLT content similar to data published by Tardivel et al. [22] Concerning the PLT components, additional studies must be performed to assess in vitro functional properties for PLT components to confirm their suitability for transfusion. In addition to logistic impacts, the introduction of TACSI WB in routine collection operations required accurate training of collection teams, adapted blood mixing © 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
Evaluation of new TACSI WB system 7
scales with a plateau that can accommodate the plastic cassette from the CRC kit. This cassette makes whole blood donations more bulky with significant impact on WB packaging for transportation to processing centre. TACSI WB is a new technology with a high level of automation of processing steps that required a redesign of the WB processing facility. The process flow was simplified, and the footprint of the processing facility was significantly reduced. Intensive training of the processing team was necessary before working with this system, in particular for the mounting of the CRC collection kit in the specific inserts. This step is an important manual step to avoid centrifugation and separation problems during processing. In addition, the insert mounting time is critical for capacity and productivity. In this study with a limited number of processed donations, reject levels imputable to TACSI WB were acceptable (21%) but could be improved compared with our routine operation (04%, data not shown). These operational aspects must be further studied in a broader feasibility study. TASCI WB offers the possibility to optimize RBC yields and maintains acceptable PLT recoveries in small volume BC. Average plasma recovery is 11 ml less than in our standard process and needs further optimization of the
parameters. Transfused RBCCs were monitored using a routine haemovigilance programme, and no transfusion reactions were reported to date. The introduction of TACSI WB in blood services will cause organizational changes and a process flow redesign. A study is planned with 1500 WB units and will specially focus on process flow and productivity of the TACSI WB system.
Acknowledgements The authors thank the Cliniques Universitaires de Bruxelles, H^opital Erasme for the transfusion of TACSI WB blood components and for haemovigilance follow-up and reporting. The authors also thank the Coagulation and Haemostasis Laboratory of the Cliniques Universitaires Saint-Luc UCL Bruxelles and the Clinical Biology Laboratory in Centre Hospitalier Regional de Namur for the analyses.
Conflict of interest The study was sponsored by TerumoBCT. The authors declare no other conflict of interest related to the paper’s subject.
References 1 Heaton L, Rebulla P, Pappalettera M, et al.: A Comparative Analysis of Different Methods for Routine Blood Component Preparation. Transfus Med Rev 1997; 11:116–129 2 Rapaille A, Moore G, Siquet J, et al.: Prestorage leukocyte reduction with in-line filtration of whole blood: evaluation of red cells and plasma storage. Vox Sang 1997; 73:28–35 3 Pietersz RNI, Dekker WJA, Reesink HW: Comparison of a conventional quadruple-bag system with a ‘Topand-Bottom’ system for blood processing. Vox Sang 1990; 59:205–208 4 Claes F, Eriksson L, Hedlund K, et al.: The bottom and top system: a new technique for blood component preparation and storage. Vox Sang 1988; 55:211–217 5 Pietersz RNI, Reesink HW, de Korte D, et al.: Storage of leukocyte-poor red cell concentrates: filtration in a closed system using connection device. Vox Sang 1989; 57:29–36 6 Sheffield WP, Bhakta V, Jenkins C, et al.: Conversion to the buffy coat
7
8
9
10
11
method and quality of frozen plasma derived from whole blood donations in Canada. Transfusion 2010; 50:1043–1049 Van der Meer PF, Pietersz RNI, Hinloopen B, et al.: Automated separation of whole blood in Top and Bottom bags into components using the compomat G4. Vox Sang 1999; 76:90–99 Thomas S, Beard M, Garwood M, et al.: Blood components produced from whole blood using the Atreus processing system. Transfusion 2008; 48:2515–2524 Lagerberg JW, Salado-Jimena JA, L€ of H, et al.: Evaluation of the quality of blood components obtained after automated separation of whole blood by a new multiunit processor. Transfusion 2013; 53:1798–1807 Johnson L, Winter K, Kwok M, et al.: Evaluation of the quality of blood components prepared using the Reveos automated blood processing system. Vox Sang 2013; 105:225–235 Sandgren P, Hild M, Sj€ ofin A, et al.: Storage of Buffy-Coat-derived platelets
© 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
12
13
14
15
16
in additive solutions; in vitro effects on platelets prepared by the novel TACSI system and stored in plastic containers with different gas permeability. Vox Sang 2010; 99:341–347 Plaza EM, Cespedes P, Fernandez H, et al.: Quality assessment of buffycoat-derived leucopleted platelet concentrates in PAS-plasma, prepared by the OrbiSac or TACSI automated system. Vox Sang 2013; 106:38–44 Harboe M: A method for determination of hemoglobin in plasma by nearultraviolet spectrophotometry. Scand J Clin Lab Invest 1959; 11:66–70 de Korte D, Kleine M, Korsten HG, et al.: Prolonged maintenance of 2,3 diphosphoglycerate acid and adenosine triphosphate in red blood cells during storage. Transfusion 2008; 48:1081– 1089 Guide to the Preparation, Use and Quality Assurance of Blood Components, 16th edn. Strasbourg, France, EDQM, 2011. Gulliksson H, van der Peer PF: Storage of whole blood overnight in different
8 A. Lotens et al.
blood bags preceding preparation of blood components: in vitro effects on red blood cells. Blood Transfus 2009; 7:210–215 17 Thomas S, Hancock V, Cardigan R: Repeated short-term warming of red blood cell concentrates has minimal effect on their quality. Vox Sang 2012; 103:113–121 18 Van der Meer PF, Cancelas JA, Cardigan R, et al.: Evaluation of overnight hold of whole blood at room temperature before component processing:
effect of red blood cell (RBC) additive solutions on in vitro RBC measures. Transfusion 2011; 51(Suppl):15s–24s 19 Kurup PA, Arun P, Gayathri NS, et al.: Modified formulation of CPDA for storage of whole blood, and of SAGM for storage of red blood cells, to maintain the concentration of 2,3-diphosphoglycerate. Vox Sang 2003; 85: 253–261 20 Van Der Meer PF, Pietersz RN: Overnight storage of whole blood: a comparison of two designs of butane-
1,4-Diol cooling plates. Transfusion 2007; 47:2038–2043 21 Moog R, Reichenberg S, Hoburg A, et al.: Quality of methylene blue-treated fresh-frozen plasma stored up to 27 months. Transfusion 2010; 50:516– 518 22 Tardivel R, Bois S, Vignoli C, et al.: Automatisation de la preparation des produits sanguins labiles. Transfus Clin Biol 2009; 16:175–178
© 2014 International Society of Blood Transfusion Vox Sanguinis (2014)
