SEARCH

SEARCH BY CITATION

Keywords:

  • ORBISAC ;
  • platelet concentrates;
  • RANTES ;
  • sCD62P;
  • sCL40L;
  • TACSI

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Background and Objectives

Buffy-coat (BC)-derived platelet concentrates (PCs) are the predominant product for platelet transfusion in many countries. Two automated systems, OrbiSac and TACSI, have been introduced in blood centres to prepare these PCs, as an alternative to the manual method. We compared the in vitro quality of PCs prepared by both methods during standard storage.

Study Design and Methods

Twenty primary BC pools were split into two parts, which were processed with OrbiSac and TACSI system to obtain OrbiSac PCs (O-PCs) and TACSI PCs (T-PCs), respectively. On days 1, 5 and 7 of standard storage, samples were taken and the following analysed: cell count, metabolic variables, platelet function and content of activation and proinflammatory substances.

Results

Both the OrbiSac and TACSI systems produced PCs that meet the standards for platelet products in terms of platelet and leucocyte content. In vitro evaluation pointed to the similar preservation of platelet metabolism (pH, glucose, bicarbonate and lactate) in O-PCs and T-PCs. Moreover, there were no significant differences between O-PCs and T-PCs as regards the hypotonic shock response or in the platelet aggregation profile. The OrbiSac system caused greater platelet activation, which resulted in higher concentrations of sCD62P, RANTES and sCD40L on the day the PCs were prepared.

Conclusion

The systems OrbiSac and TACSI can be used to produce buffy-coat-derived PCs whose cell content, platelet function and metabolism are similar during standard storage. However, the preparation with the OrbiSac system induces a transient increase in platelet activation and release of proinflammatory substances.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

The transfusion of platelet concentrates (PCs) is an essential tool in modern medicine, allowing surgery, aggressive chemotherapy schemes and the clinical management of patients with particular bleeding disorders [1]. These blood products may be selectively obtained from donors by plateletpheresis or prepared from whole blood donations [2, 3]. The use of each of these two products varies between countries and individual institutions, with an approximately 50:50 ratio in Europe. In most developed countries, except USA, the classical platelet-rich-plasma (PRP) method has been substituted by the buffy-coat (BC) procedure as the routine form to prepare whole blood-derived PCs [2]. While PCs prepared by the BC and the PRP methods are very similar in term of their in vitro properties, survival and efficacy upon transfusions [2-5], the BC method has facilitated automatization and new strategies for PCs preparation, including prestorage pooling to obtain transfusional PCs doses, universal leucodepletion and pathogen inactivation.

The original manual BC method is highly labour-demanding, time-consuming and rends products with a limited and variable platelet content [6]. The development a decade ago of the OrbiSac system by Gambro BCT (which is now part of Terumo BCT, Lakewood, CO, USA) was a significant step forward in the BC process. This automated system replaced the manual BC procedure in many blood centres and helped to standardize and optimize PC preparation [7, 8]. More recently, Terumo has developed the TACSI system that offers the major advantage of automated preparation of six PCs simultaneously. Platelets are known to be affected during preparation and storage under blood banking conditions, in a complex process referred to as platelet storage lesion [9, 10]. Several studies have analysed the quality of BC-derived PCs prepared manually and with OrbiSac [11-16], while there is scarce information on PCs prepared with TACSI [17]. The aim of this study was to evaluate the effect of using OrbiSac or TACSI on the in vitro quality of PCs during storage under standard conditions for up to 5 days.

Material and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Preparation, storage and sampling of platelet concentrates

Standard whole blood units obtained from healthy volunteer donors were processed to separate plasma, red blood cells and BC units, as detailed elsewhere [16]. The day of whole blood donation was considered the day 0, and fractionation of units was performed within 18 h from extraction.

In each of twenty different series of experiments, 10 ABO compatible fresh BC units, having approximately 18–20 h from whole blood donation, and 600 ml of SSP+ platelet additive solution (MacoPharma, Mouvaux, France) were pooled in a large-volume bag (ref.BB*T200BM, Terumo Europe, Leuven Belgium) using a sterile connecting device (TSCD®-II, Terumo Europe). Each of these primary BC pools (1157·1 ± 10·4 ml; Haematocrit 22·5 ± 0·7%; 874·5 ± 82·2 platelets/μl; n = 20) were divided into two transfer bags (ref.720433, Laboratorios Grifols S.A, Barcelona, Spain) of similar volume (578·5 ± 5·2 ml) containing cells in approximately 30% of plasma and 70% of SSP+. One of the two bags, which was chosen at random, was sterile connected to an OrbiSac BC system platelet processing set (Terumo BCT ref.50000) and processed with OrbiSac to obtain a leucodepleted OrbiSac PCs (O-PCs). The second bag of each primary BC pool was sterile connected to a TACSI Pl kit (Terumo BCT ref. TF-CSPB3SY02) and processed with the TACSI system for the preparation of leucodepleted TACSI PCs (T-PCs). In both cases, processing was carried out following the manufacturer's instructions, except for the addition of SSP+ to the BC pool which was performed manually as indicated above. The platelet storage containers of both the OrbiSac BC and TACSI system are made of n-butyryl, tri n-hexyl citrate (BTHC)-plasticized PVC.

After removing the excess of air and foam, all PCs (O-PCs and T-PCs) were left undisturbed for about 1 h, and then stored for up to 7 days in a standard blood bank PCs incubator set at 22°C and with moderate flat-bed shaking (60 cycles/min.). On days 1, that is the day of preparation of PCs, 5 and 7 of storage, samples (10 ml) were aseptically drawn from the PCs through a sampling injection spike (OriGen Biomedical, Austin, TX, USA) and assayed as described below. All the PCs were tested for sterility by a microbiological culture performed at day 7 which was negative in all cases.

Test and assays

Cell counts and mean platelet volume (MPV) in the PCs samples were determined by means of an electronic particle counter (STKS, Coulter Electronics, Hialeah, FL, USA).

Residual leucocytes in these leucodepleted PCs were counted by flow cytometry using Leukocount™ (Becton Dickinson, San Jose, CA, USA) according to the manufacturer's instructions.

For the platelet functional studies, samples of PCs were centrifuged (1000 g, 10 min), the platelet poor supernatant removed for later use, and the platelet pellet was gently resuspended in the appropriate volume of identical aliquots of AB fresh-frozen plasma, stored at −80°C and defrosted at 37°C just before use, to obtain a platelet suspension containing 300 × 109 platelet/L. Hypotonic shock response (HSR) and platelet aggregation assays in the platelet suspensions obtained from the PCs were performed as described previously [16]. The agonists and concentrations used in aggregations were as follows: 1·6 mm arachidonic acid (AA) (DiaMed, Cressier, Switzerland), 25 μm thrombin receptor activating peptide (TRAP) (Sigma-Chemical, Madrid, Spain), 1·25 mg/ml ristocetin (Sigma-Chemical) and 10 μg/ml collagen (Menarini Diagnostics).

For the rest of experiments, the platelet poor supernatants obtained were centrifuged again (1500 g, 10 min) to precipitate residual cell fragments. Any impurities were rejected, and the recovered supernatant (PC supernatants) was stored frozen at −80°C until use to measure metabolic parameters and quantify proinflammatory markers.

The pH in the PCs supernatants was measured at 22°C using a pH meter (M220, Corning Incorporated, Corning, NY, USA). Bicarbonate and lactate levels were determined in an ABL 825 Blood Gas Analyser (Radiometer, Copenhagen, Denmark), and glucose was measured by Advia 1800 Chemistry Analyser (Siemens Healthcare Diagnostics, Tarrytown, NY, USA. Finally, the levels of soluble P-Selectin (sCD62P), regulated upon activation, normal T-cell expressed and secreted chemokine (RANTES, also known as CCL5] and soluble CD40 ligand (sCD40L) were measured in the PCs supernatants with the BD™ cytometric bead array (CBA) (Becton Dickinson) and using a BD FACSCalibur™ (Becton Dickinson) flow cytometer. The specific CBA kits, Human soluble P-Selectin (sCD62P) Flex Set, Human sCD154 (sCD40 ligand) Flex Set and Human RANTES Flex Set, were used following the manufacturer's instructions.

Statistical analysis

The results of the different variables measured in the PCs are presented as mean ± standard deviation. The Kolmogorov–Smirnov test was used to check for the normal distribution of data. Differences among variables between O-PCs and T-PCs at each storage day were assessed using a paired t-Student's test or the Wilcoxon test, as appropriate. P-values of <0·05 were accepted as statistically significant. For each group of PCs (O-PCs or T-PCs), any changes in variables occurring during storage were analysed by means of a one-way anova for repeated measures and Bonferroni correction test. These statistical analyses were performed using IBM spss Statistics version 15.0 for Windows (IBM Corporation, New York, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Volume and cell content of PCs

The general properties of the PCs are summarized in Table 1. As shown, the volume of the O-PCs was about 30 ml greater than that of the T-PCs. Both processing systems, that is OrbiSac and TACSI, provided approximately 80% recovery in the PCs of the platelets contained in the primary pools of BC. Consequently, the total platelet content in O-PCs and T-PCs was very similar, although platelet density was slightly but significantly higher in T-PCs (Table 1). Of interest is the fact that the platelets in T-PCs showed a higher MPV than the platelets in O-PCs, a difference that remained significant throughout storage. Finally, there was no significant difference between O-PCs and T-PCs as regards the total amount of contaminating leucocytes (Table 1). The highest contamination was 0·192 × 106 and 0·153 × 106 leucocytes in O-PCs and T-PCs, respectively.

Table 1. Characteristics of the platelet concentrates (PCs) obtained with the OrbiSac (O-PCs) and TACSI (T-PCs) systems, from homologous pools of identical ABO buffy-coat units suspended in SSP+ (70%) and plasma (30%)
 O-PCsT-PCs
  1. MPV, mean platelet volume.

  2. All values are mean ± SD from 20 different units.

  3. a

    Denotes P < 0·01 with respect to O-PCs according to paired t-test.

Volume (ml)380·7 ± 6·9349·9 ± 8·7a
Platelets
(×109/unit)351·9 ± 39·9360·6 ± 33·5
(×109/l)925·1 ± 109·61031·7 ± 104·2a
MPV (fl)6·65 ± 0·436·89 ± 0·40a
Recovery (%)80·5 ± 9·882·4 ± 6·7
Leucocytes
(×106/unit)0·04 ± 0·050·03 ± 0·04

Metabolic parameters and functional assays

The metabolic behaviour of the homologous PCs obtained from buffy-coat pools in plasma SSP+ by means of the OrbiSac (O-PCs) or with TACSI (T-PCs) system was evaluated during storage by measuring the pH and the concentrations of glucose, bicarbonate and lactate in the respective PC supernatants. As shown in Table 2, the pH remained around 7·2 in all the PCs throughout the storage period. As expected for these platelet products, both glucose and bicarbonate gradually decreased during storage, while the lactate level increased steadily to about 12 mm on day 7. All these metabolic parameters changed to a similar extent in O-PCs and T-PCs during storage. However, slight differences (statistically significant but probably meaningless from a functional point of view) were determined between T-PCs and O-PCs regarding these parameters, especially on the day of preparation (day 1) (Table 2).

Table 2. Evolution during 7 days of storage of platelet metabolism parameters and hypotonic shock response (HSR) in homologous platelet concentrates (PCs) obtained with OrbiSac (O-PCs) or with TACSI (T-PCs) systems, from pools of identical ABO buffy-coat units suspended in SSP+ (70%) and plasma (30%)
ParameterDay 1Day 5Day 7
O-PCsT-PCsO-PCsT-PCsO-PCsT-PCs
  1. Values are mean ± standard deviation from 20 different PCs. A paired t-test was used to compare values.

  2. a

    Denotes P < 0·05 with respect to O-PC on each day of storage.

  3. b

    Denotes P < 0·05 with respect to Day 1 in each type of PCs (O-PC or T-PCs).

pH7·20 ± 0·047·23 ± 0·05a7·30 ± 0·07b7·33 ± 0·07b7·24 ± 0·107·27 ± 0·10a
Glucose mg/dl149·6 ± 9·0154·2 ± 6·5a102·9 ± 11·5b101·1 ± 10·0b76·2 ± 21·3b70·8 ± 19·5a,b
Bicarbonate (mm)7·28 ± 0·537·76 ± 0·40a6·56 ± 0·85b6·84 ± 1·42b5·65 ± 1·15b5·78 ± 0·83b
Lactate (mm)5·02 ± 0·954·62 ± 0·51a9·71 ± 1·57b9·34 ± 2·12b11·35 ± 2·67b12·06 ± 2·19b
HSR (%)95·6 ± 11·099·7 ± 15·786·3 ± 8·7b88·6 ± 9·6b87·7 ± 14·283·8 ± 10·5b

Platelet functional status

To compare the platelet functional status in O-PCs and T-PC during standard storage, HSR and aggregometry experiments were carried out. The HSR assays demonstrated that the capacity of platelets to recover shape after hypotonic stress during storage remained good (above 80%) and similar in O-PCs and T-PCs (Table 2). Figure 1 displays the changes in the platelet aggregation profile of these PCs. On day 1, the maximum platelet aggregation values with each of the four agonists tested were above 70% in both O-PCs and T-PCs, and close to the normal aggregation response of healthy subjects with these doses of agonists (not shown). This aggregation response of PCs gradually decreased as storage progressed, more abruptly in the case of the less potent agonists collagen and arachidonic acid. Importantly, no significant differences were observed between O-PCs and T-PCs in the aggregation response to any of the agonists at any stage during storage (Fig. 1).

image

Figure 1. Profile of platelet aggregation response in homologous platelet concentrates (PCs) obtained with OrbiSac (O-PCs) or TACSI (T-PCs) during 7 days of storage. Homologous PCs were obtained with OrbiSac (O-PCs) (image_n/vox12072-gra-0002.png) or with TACSI (T-PCs) (image_n/vox12072-gra-0001.png) systems from the same pools of identical ABO buffy-coat units suspended in SSP+ (70%) and plasma (30%). At days 1, 5 and 7 of storage under standard blood bank conditions, samples of the PCs were collected, the PCs supernatants were removed, and the platelets were resuspended in AB plasma at 300 × 109 cells/l. As described in Material and Methods, platelet aggregation was induced with 1·6 mm arachidonic acid, 25 μm TRAP, 10 μg/ml collagen and 1·25 mg/ml ristocetin, and measured as maximal change in light transmission. All data in the plots are the mean ± standard deviation from 20 experiments. # Denotes P < 0·05 with respect to day 1 in each type of PCs (O-PC or T-PCs).

Download figure to PowerPoint

Mediators of inflammation in supernatants

Table 3 summarizes the evolution of activation/proinflammatory markers in the PCs during storage. Of note was the fact that on the day of preparation (day 1), the concentrations of sCD62P, RANTES and sCD40L were slightly but significantly higher in O-PCs than in their homologous T-PCs. As shown in Table 3, this difference decreased as storage proceeded; the concentration of these bioactive compound reaching similar level in both types of PC on day 7.

Table 3. Changes in the pro-inflammatory profile, during 7 days of storage, in homologous platelet concentrates (PCs) obtained with OrbiSac (O-PCs) or with TACSI (T-PCs) systems, from the same pools of identical ABO buffy-coat units suspended in SSP+ (70%) and plasma (30%)
ParameterDay 1Day 5Day 7
O-PCsT-PCsO-PCsT-PCsO-PCsT-PCs
  1. Values are mean ± standard deviation from 20 different PCs. Paired t-test or Wilcoxon test, as appropriate, was used to compare values.

  2. a

    Denotes P < 0·05 with respect to O-PC on each day of storage.

  3. b

    Denotes P < 0·05 with respect to Day 1 in each type of PCs (O-PC or T-PCs).

sCD62 (mg/ml)28·77 ± 7·8821·47 ± 6·23a46·77 ± 7·41b45·35 ± 4·44b46·29 ± 8·71b45·00 ± 4·09b
RANTES (mg/ml)22·70 ± 12·1416·63 ± 6·14a58·87 ± 3·95b61·80 ± 4·49a,b62·49 ± 3·29b64·04 ± 1·18b
sCD40L (mg/ml)1·24 ± 0·760·63 ± 0·33a3·70 ± 1·05b4·24 ± 1·01a,b4·18 ± 1·50b4·61 ± 0·79b

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Because the preparation of PCs from BCs is a laborious task that requires multiple steps, semi-automatic systems such as OrbiSac and TACSI have increasingly been introduced in blood centres during the last decade, providing benefits in term of organization and the quality of PCs [18, 19].

Some studies have shown that PCs prepared with OrbiSac have a higher platelet content and better conserve the aerobic metabolism of platelets, compared with those prepared manually. In contrast, OrbiSac seems to induce slightly higher platelet activation than manual processing [8, 11-13, 20]. This finding was attributed to the higher shear stress affecting platelets during automatic processing and may be of some concern, even if it has no clinical relevance [12, 13].

To the best of our knowledge, only one comparison of BC-derived PCs prepared with OrbiSac or TACSI has been reported [17]. In this work, Sandgren et al. compared homologous PCs prepared with TACSI but stored in two different types of PVC container (differing in the plasticizer) and used nonhomologous PCs prepared with OrbiSac as a reference group. They found no significant variations between the groups regarding platelet content and MPV, and noted slight differences during storage in metabolic, function and activation parameters between PCs prepared with TACSI and those obtained with OrbiSac [17].

In contrast to the previous study [17], our current work was designed to confirm that any potential differences between O-PCs and T-PCs during 7 days of storage are independent of the platelet source. Thus, high-volume BC pools were divided into half, and each part was processed with TACSI or OrbiSac to produce homologous PCs, which were then stored together under standard blood bank conditions. Platelet recovery (>80%) was almost identical with the two systems, and both devices rendered PCs (>3 × 1011 platelets, <106 leucocytes, pH≈7·2) that easily meet European standards for these blood products [21]. Moreover, there were negligible differences in term of metabolic (pH, glucose consumption, bicarbonate and lactate levels) and functional (HSR, aggregation response) parameters between O-PCs and T-PCs. All the parameters for both types of PCs displayed a similar degree of impairment during storage to that we previously found in manually prepared BC-derived PCs [16].

The most noticeable finding in this study refers to the activation and proinflammatory markers released by these O-PCs and T-PCs, which have not been previously compared.

The platelet surface expression of P-selectin (CD62P) is an established marker of platelet activation during the storage of PCs [9]. Soluble CD62P is released by proteolytic cleavage of CD62P, and it may be a risk factor for venous thromboembolism [22, 23]. The proinflammatory C-C chemokine RANTES is also released as a result of platelet activation during processing and storage of PCs, and the passive transfusion of RANTES in PCs might be associated with inflammatory reactions as well as allergic reactions [24, 25]. As regards sCD40L, which is thought to be mainly derived from platelets [26], its accumulation in PCs may favour febrile reactions due to the induction of cyclooxygenase-2 expression and TRALI by promoting the activation of polymorphonuclear leucocytes and endothelial damage [27, 28].

Interestingly, we found that T-PCs contain significantly lower levels of sCD62, RANTES and sCD40L on the day of preparation, although this difference disappeared as storage progressed. Thus, it seems that processing BC pools with the OrbiSac system induces slightly higher platelet activation than processing with TACSI, although this is a transient phenomenon which evens out during storage. Because most PCs are transfused before day 3 of storage, at least in or centre, reduced amounts of these platelet-derived biological response modifiers may represent an advantage of T-PC over O-PCs.

Worth mentioning is the finding that the levels of sCD62P, RANTES and sCD40L in the PCs following semi-automatic preparation with either OrbiSac or TACSI were slightly but significantly higher than those recently determined in platelet PCs obtained manually [16]. This finding reflects that of Vetlesen et al. [13], who also observed higher platelet activation and RANTES accumulation in PCs prepared with OrbiSac in compassion with units prepared manually.

In conclusion, our evaluation of the processing of homologous platelets with TACSI and OrbiSac suggests that, under controlled conditions, the two methods of production of BC-derived PCs are equivalent in terms of cell recovery and the preservation of in vitro platelet function and metabolism during storage. Compared with TACSI, the OrbiSac system seems to expose platelets to mechanical stress, resulting in transiently higher concentrations of platelet-derived proinflammatory substances in the PCs. Given the vast experience in the transfusion of PCs prepared with OrbiSac, this difference is most likely to be of no clinical relevance.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

Research of the authors' group is funded by the Instituto de Salud Carlos III (ISCIII, PI10/02594), RECAVA RD12/0042/0050 (ISCIII and FEDER) and Fundación Séneca (07703/GERM/07). ISG has a fellowship from ISCIII (FI10/00535).

Conflict of interest statement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References

This study was sponsored by Terumo® Europe España SL.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Material and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest statement
  9. References