Quality Validation of Platelets Obtained From the Haemonetics and Trima Accelautomated Blood-collection Systems

Background: Platelet transfusion is required to treat haemo-oncology or trauma patients. Platelet apheresis (PPH) performed with apheresis equipment has increased rapidly in recent years. Leucocyte-reduced platelet apheresis (LRPH) can reduce the risk of platelet refractoriness and febrile nonhemolytic transfusion reactions (FNHTRs) for transfusion. Accordingly, this study aimed to investigate and compare the platelet metabolic and functional responses between PPH performed with Haemonetics and LRPH performed with Trima Accel cell separator. Methods: The qualities of platelets collected through PPH and LRPH were evaluated in terms of visual appearance, morphology, platelet-aggregation changes, metabolic activities, and bacterium-screening test during 5-day storage. Statistical analyses included two-sample t-test and generalised estimating equation(GEE) method. Results: During 5-day storage in LRPH, residual leucocytes were all <1.0×10 6 , and the parameters of platelet function were as follows: platelet aggregated to agonists such as adenosine 5′-diphosphate (ADP) and collagen, and the extent of shape change and pO 2 showed no statistically signicant difference between PPH and LRPH. The hypotonic shock reaction (HSR) on days 0, 1, and 3 were signicantly higher in LRPH than in PPH (71.78±6.92 vs. 64.10±7.42; p=0.002; 71.53±8.98 vs. 62.96±9.84; p=0.007; 68.05±7.28 vs. 57.76±6.80; p<0.0001, respectively). Values of mean platelet volume (MPV) were statistically larger in PPH than in LRPH on days 0, 1, and 3. On day 5, the swirling score was higher in LRPH than in PPH. The mean lactate levels had no statistically signicant difference between PPH and LRPH. Moreover, no growth was observed through bacterium-screening test conducted on 40 samples. Conclusion: Comparison of LRPH and PPH products 11 the 20). Quality control tests for the residual leucocytes of LRPH collected with Trima Accel version 5.0 were performed by manually counting the cells A study has shown that skin are the major in One strategy is to use a for The volume this bag is then used for routine laboratory-screening and testing. The collected sets from two instruments in this work included pouch. The quality-control were determined through least ≥ and had ≥ during storage. Moreover, the residual leucocytes of LRPH were < 5 × 10 6 per nal product for 95% of units sampled.

If donors have been fasting, we asked them to eat a meal or offered them snacks, water, drink, or milk before donation. Almost all PPH and LRPH donors in Taiwan were male. From June to September 2013, 40 eligible male donors were recruited to participate in this study.

Data collection
Donors were individually subjected to the blood collection with Haemonetics or Trima Accel device. Vital signs included systolic and diastolic blood pressure, and pulse was monitored and recorded at the beginning and end of the procedure. The donors were also monitored for adverse events during the procedures. The following data were entered into the cell-separator program for both devices: donor height, weight, gender, hematocrit, and pre-count of PLT values. Welltrained apheresis staff performed all the procedures. The body mass index (BMI), de ned as the body mass divided by the square of body height, was also recorded.

Instruments and quality control
A Haemonetics MCS + ED cell separator (Braintree, MA, USA) was used with a targeted yield of 6 × 10 11 PLTs for double units. The whole-blood ow was set to 85 mL/min, which can be modi ed by the collection staff depending on the donors' venous condition. The acid citrate dextrose-A (ACD-A) anticoagulant (AC)/whole-blood ratio was 1:10 (n = 20). The second device used for PLT collection was Trima Accel software version 5.0 (Caridian BCT, Lakewood, CO, USA) with a targeted yield of 6 × 10 11 PLTs for double units. The whole-blood ow was 75-100 mL/min the, and the ACD-A AC/whole-blood ratio was 1:11 (n = 20). Quality control tests for the residual leucocytes of LRPH collected with Trima Accel version 5.0 were performed by manually counting the cells in the Nageotte chamber as described elsewhere [9]. A previous study has shown that resident skin ora are the major bacterial contamination in PCs [10]. One strategy to reduce such contamination is to use a diversion bag for the rst aliquot of the donation. The volume of this diversion bag is then used for routine laboratoryscreening tests, such as blood-type and infection-disease testing. The collected apheresis sets from the two instruments in this work included the diversion pouch. The quality-control criteria for the product were determined through standard operation procedure (SOP). At least 90% of units sampled contained ≥ 3.0 × 10 11 PLTs, and at least 90% of units had a pH ≥ 6.2 during 5-day storage. Moreover, the residual leucocytes of LRPH were < 5 × 10 6 per nal product for 95% of units sampled.

Visual appearance
When discoid PLTs are exposed to a light source, swirling is caused by the re ection of light. This swirling can be used to evaluate PLT function. Herein, swirling was assessed by visual inspection, and we de ned the scores as follows: 0, no swirling; 1, limited swirling; and 2, maximum swirling.

Laboratory testing
Pre-and post-apheresis samples were subjected to complete blood counts (CBCs) by using an automated blood-cell counter (Sysmex KX-21N, Sysmex Corp., Kobe, Japan). PLT quality was examined through morphological, biochemical, and functional changes on days 0, 1, 3, and 5. The PLT count and mean PLT volume (MPV) were tested. The potential marker of PLT reactivity was used to measure MPV. PLT shapes ranged from discoid to spheroid, and the MPV increased (swelling) in correlation with reduced PLT survival. Loss of swirling was associated with loss of PLT function.
PLT-aggregation changes were examined by light transmission aggregometry (Chrono-log 490-4D) in response to adenosine 5′-diphosphate (ADP) and collagen (CHRONO-LOG, Leiden, the Netherlands)-induced aggregation. For ADP-induced aggregation, the nal concentration of PLT rich plasma (PRP) was 300 × 10 3 µL, and the corresponding PLT poor plasma served as the control. About 450 µL of PRP was placed in a reaction cup under continuous stirring at 37 °C, and then 50 µL of ADP ( nal concentration = 30 µM) was added to the PRP sample. PLT-aggregation activity was then tested. About 490 µL of PRP was placed in a reaction cup under continuous stirring at 37 °C, and then 10 µL of collagen ( nal concentration, 20 µg/µL) was added to the PRP sample.
Collagen-induced aggregation activity was then tested.
Hypotonic shock response (HSR) and extent of shape change (ESC) were explored to evaluate PLT survival. pO 2 , pCO 2 , glucose, and lactate tests were conducted to examine PLT storage. HSR was measured according to dynamic light transmittance through light transmission aggregometry and calculated as follows: HSR=[(Y-a)/(Y-X)] × 100%, where X is the transmittance of normal-saline-diluted PRP, Y is the maximum transmittance of deionized-water-diluted PRP, and "a" is the minimum transmittance of deionized-water-diluted PRP.
PLT shape change, as one of the earliest morphological indicators of PLT activation, was subsequently examined. ESC was tested by adding 10 µL of 0.1 M ethylenediaminetetraacetate to PRP. Subsequent aggregation was blocked by adding 10 µL of 0.1 mM ADP. The change in shape was measured by light transmission aggregometry.
The bacterium-screening tests for all PPH and LRPH were conducted in 2007 in Taichung Blood Center. About 8 mL of sample was collected from each PLT unit ≥ 24 h after blood donation and inoculated in an automated blood-culture system (BacT/ALERT®, bioMérieux, Marcy l'Etoile, France). CO 2 production was monitored to determine bacterial growth, which changed the color from grey to yellow at the bottom of the culture bottle as detected with a gas-permeable sensor [11]. Each sample was placed in an aerobic culture bottle inoculated in the automated culture system for up to 24 h at 35-37 °C. APs were released to the hospital if no growth was observed.

Statistical analysis
Continuous variables were reported as the mean ± standard deviation. The two-sample t-test was used for continuous variables in bivariate analysis.
To further assess PLT function on different days, i.e., to compare the results of days 0 with those of day 1,3 and 5, the method of generalised estimating equations was used by considering the dependence. A p value less than 0.05 was considered as statistically signi cant. All analyses were performed with SAS version 9.4 (SAS Institute Inc, Cary, NC, USA).

Sample characteristics
The demographic characteristics and CBC data of a total 40 male donors (N = 20 per group) are shown in Table 1
The bowl was centrifuged well, and the head of the bowl was rmly pressed down on to fully seat it in the centrifuge chuck. The dual pump tubing around the "Blood" and "Transfer "pumps was looped in preparation for the pump autoload sequence. The single pump manifold was snapped into place, and the AC tubing was looped around the AC pump.
The system collected the PLT layer in a small volume of plasma to obtain concentrated PLT products. If the PLT yield was 6 × 10 11 or above, Haemonetics recommends opening the clamps on both PLT storage bags. The MCS + UPP disposable set is shown in Fig. 1.
To separate whole blood into PCs by using Trima Accel system, a continuous-ow centrifuge was used. Whole blood was collected from the blood donor and mixed with an AC. The mixture was pumped into the channel that was a plastic pathway set in the centrifuge lter with a well-designed groove. Five pumps drew and returned the blood from the donor by using the Trima Accel system. A cassette that was part of the tubing set guided the ow of blood and products.
A leucoreduction system (LRS) was used in Trima Accel to minimize the leucocyte content of LRPH products. Following the principle of elutriation, the chamber enabled the separation of PLTs from leucocytes through the technology of saturated, uidized particle-bed ltration. The LRS chamber trapped the leucocytes, which can solve the problem of low-yield PLT collection for LRPH. The tubing set, cassette overview, and LRS are shown in Fig. 2. The system can consistently collect LRPH products with residual leucocytes < 1.0 × 10 6 .

PLT qualities
The results of comparing PLT qualities between PPH (MCS+) and LRPH (Trima Accel) products during storage days 0, 1, 3, and 5 are presented in

Strengths And Limitations
This study compared the metabolic and functional responses (e.g., visual appearance and bacterium-screening test responses) between the PPH and LRPH products of the Haemonetics and Trima Accel automated blood-collection systems and found that additional processe are needed to reduce leucocytes and obtain high-quality PLTs.
The transfusion of PLTs signi cantly increased in 2013 [20]. Since then, signi cant improvements have been made in terms of the e ciency and quality of PLT pheresis to offer high donor comfort. The use of apheresis PLT concentrates was intended to reduce the risk of antibody-positive TRALI [21] .
A previous study has shown that IL-1beta, IL-6, and IL-8 levels were signi cantly higher in post-leucocyte-reduced PPHs than in pre-leucocyte-reduced ones during storage, indicating that prestorage leukoreduction blood components reduced the transfusion-associated adverse reactions [22]. Febrile nonhaemolytic transfusion reactions (FNHTRs) are the most frequent adverse reactions for blood transfusion [23]. The risk rates of FNHTRs are higher for transfusion PLT and non-leucocyte-reduced blood products than for leucocyte-reduced ones [24,25]. Many studies have also reported that using universal leucocyte-reduced blood components can reduce the rates of [26], which is a considerable problem for patients and hospitals [27]. Different PPH technologies can affect bacterial contamination. A higher rate of positive bacterial culture results is associated with a higher rate of apheresis PLT septic transfusion reactions [28].
The results of bacterium-screening test showed no bacterial growth using the two different devices in this work.
We also compared the data of day 0 with those of days 1,3,and 5 in the PPH and LRPH groups. For PLT aggregation activity and PLT survival test, we observed that the values of ADP, collagen and ESC statistically decreased on day 1, 3, and 5 compared with those on day 0 in both groups ( Table 2). The performance of PLT storage indicated that the glucose level statistically decreased but the lactate level statistically increased compared with those on day 0 in both groups (Table 2). pH was signi cantly higher on days 1, 3, and 5 than on day 0 in both groups (in Table 2). The residual leucocytes of LRPH signi cantly decreased on days 1, 3, and 5 compared with those on day 0 ( Table 2).
No growth was observed through bacterium-screening tests of the 40 samples. The PPH and LRPH products met the criteria of quality control according to the SOP (data not shown). All 40 participants had no adverse events during the collection processes.

Main ndings
This study aimed to compare the metabolic and functional responses between the PPH and LRPH products of the Haemonetics and Trima Accel automated blood-collection systems, respectively. We assessed the quality of the products and subsequently stored condition in terms of PLT aggregation activity and PLT survival and performance during storage in PH and LRPH from different instrument in both groups during the 5-day storage. Data showed the ability of PLT to aggregate to ADP and collagen, with no difference between PPH and LRPH from different instruments during days 0, 1, 3, and 5. Moreover, the leucocyte-depletion process for LRPH collection (Trima Accel) did not affect the functional capacity and metabolic responses of the PCs. Furthermore, LRPH have better PLT survival, pH value as PLT components quality indicator and smaller MPV than PPH during days 0, 1 and 3 .

Comparison with literature data and possible explanations
Thrombin receptor activator for peptide 6 (TRAP-6) induces PLT aggregation. The present study showed that PCs collected from Trima Accel device responded signi cantly less to TRAP-6 than did MCS + PLTs [12]. This nding showed that a higher-quality PLT concentrates with least activated PLTs can be obtained by using Cobe Spectra and Trima Accel than by using MCS + and Amicus machines [13].
Our study also showed the ability of PLTs to aggregate to ADP and collagen, with no difference between PPH and LRPH form different instruments.
Various studies have evaluated PLT quality during processing and storage of different PLT pheresis instruments. The biological and morphologic changes were associated with PLT procoagulant function [14].
The present work indicated that HSR values on days 0, 1, and 3 were signi cantly higher in LRPH than in PPH but not on day 5. ESC also had no statistically signi cant difference between the two groups during on day 5.
The metabolic activities of PPH and LRPH were measured by the pH, pCO 2 , pO 2 , glucose, and lactate levels. Evaluation of the pCO 2 and pO 2 changes in PPH and LRPH showed that pCO 2 increased and pO 2 decreased or increased. The differences in pO 2 probably depend on the sample [15].
The mean pH on days 0, 1, 3 and 5 were higher than 7.0 in both PPH and LRPH from the different devices. At pH 6.0, the PLTs changed from discoid to spherical. Thus, pH was an indicator of stored-PLT functionality. The pH of PLT products must remain ≥ 6.2 during storage under agitation at 20-24 °C according to the American Associated Blood Bank and our SOP. pH decreases because of increased lactic acid production through the glycolysis pathway.
Our study showed that glucose levels decreased and lactate increased during the 5-day storage in PPH and LRPH products. Previous studies also support our data [16,17].
The present research further showed that the process for LRPH collection (Trima Accel) did not affect the functional capacity of PLTs compared with PPH collection (Haemonetics MCS + ED). PLT activation decreased during storage because of WBC removal from PCs [18] During PC storage, the biochemical, morphological, and functional changes, known as PLT storage lesion, should be detected properly and early because these are the most important aspects of PC transfusion [19] .
Our study has several limitations. First, we did not monitor the bacterium-screening tests for PPH and LRPH to expired days (5-days). The availability of PCs was delayed for up to 48 h after blood collection for bacteria testing. We followed the SOP to incubate 24 h (only to day 2) by using routine test conditions for PPH and LRPH distribution to the hospital. Second limitation was the relatively small number of analyzed PPH and LRPH samples. Third limitation was that based on the criteria for donating PPH and LRPH products, the mean BMI of our participants was > 24 kg/m 2 . BMI > 24 kg/m 2 is de ned as overweight according to the Health Promotion Administration of the Ministry of Health and Welfare of Taiwan. The mean age of our participants was ≥ 40 years old (middle-age was de ned as 40-65 years). We also cannot analyze the association of PLT activity with different subgroups of BMI and age, respectively.
The hameovigilance system is very important for public health practitioners to understand the adverse reactions of transfusion and prevent bacterial contamination in PLTs [29]. Bacterial transmission remains a major challenge in PLT transfusion. A previous study has shown that additive solution substitutes for plasma in PLTs at 4 °C storage are necessary [30].

Conclusions
Our ndings showed that LRPH collected form the Trima Accel automated blood-collection system required additional processes to reduce leucocyte and had no disadvantageous for PLT qualities compared with PPH products collected form Haemonetics device. The MPV and swirling score were better in LRPH than in PPH collected by Haemonetics machine. All products met the criteria of quality control, and no bacterial growth was observed through bacterium-screening tests.
Patients with clinical indications such as dyserythropoiesis, chemotherapy, thalassemia, bone-marrow transplantation, organ transplantation, and immunode ciency, as well as those who had experienced two times of fever or chills and adverse reactions caused by leucocytes, should be transfused with LRPH products according to the National Health Insurance regulations in Taiwan. Further study should be conducted to investigate additional methods for reducing the adverse events caused by leucocytes and to compare the outcomes of more apheresis instruments with consideration of the safety of donors, products and recipients. Scheme of an MCS+ UPP disposable set in the Haemonetics system Scheme of tubing set, cassette overview, and LRS in the Trima Accel system