1. Introduction

During blood bank storage, red blood cells (RBCs) undergo some deleterious changes that progressively affect their metabolism, cytoskeleton and membrane [1]. Loss of structural and functional integrity of the RBCs during storage, collectively referred to as red cell storage lesion (RCSL) [2].

Pediatric patients with malignancies or benign hematologic diseases have underlying pathophysiologies leading to increased requirements for transfusion therapy [3]. Packed red blood cell (pRBC) transfusions is common in pediatric intensive care unit (ICU). Studies in adult ICU have shown unfavorable outcomes in patients who received pRBC [4]. Almost all adult and pediatric patients requiring transfusion usually receive blood with oldest compatible unit available given first due to ‘first-in, first-out’ principle [5]. It was suggested that large part of the side effects after RBC transfusion could be associated with prolonged storage of RBCs before transfusion [6].

The oxidative damage to the cytoskeleton and membrane due to reduction in antioxidant enzymes occurring in long storage is involved in increased vesiculation and loss of cation gradients across the membrane [7]. Vesiculation causes irreversible damage to RBCs because once membrane has been lost the RBC cannot regain its original morphology. Vesicles produced by cells can be divided into; (1) Exosomes, 30–100 nanometer (nm). (2) Microparticles (MP), 0.1–1micrometre (μm). (3) Apoptotic bodies, 1–5 μm [8].

Many post-transfusional adverse effects were attributed to microparticles release as, splenic sequestration of old RBCs due to irreversible damage of RBCs membrane via vesiculation [7]. Nitric oxide (NO) scavenging by MP leads to inhibition of NO-dependent vasodilation in the recipients [9]. Also MP initiate thrombin generation in a FXI-dependent manner which initiates pro-Coagulant activity [10]. Expression of phosphatidylserine (PS) on MP, accounts for their proinflammatory activities. Leading to neutrophil priming, which was linked to transfusion-related acute lung injury (TRALI) [11-12].

As erythrocytes become senescent they display molecules on their surface that label them for elimination and one of these molecules is PS. The PS is normally exposed on inner leaflet of cell membrane. When RBCs undergo eryptosis (apoptosis like self destruction as a result of cumulative breakdown of hemoglobin molecule) [13], the PS becomes exposed on outer side of the membrane, becoming detectable by annexin V binding [6].

Efforts have been made to reduce toxic effects of transfusions through leukoreduction [14]. Prestorage leukofiltration avoids alloimmunization, as human leukocyte antigen (HLA) may solubilize from leukocytes during storage, and leukocyte cytokines released in storage, causing febrile non hemolytic transfusion reaction (FNHTR) and can pass through filters. Also leukocytes released enzymes in storage may be detrimental to red cell viability [15]. The study aims to evaluate leukofiltration impact on MP release in storage and PS exposure on old stored RBCs and to re-evaluate storage period length for minimizing adverse outcomes resulting from long storage periods.

2. Materials and Methods

Sample collection and preparation

Twenty volunteers donated twenty blood units in September 2015 in blood transfusion centre of Al-Zahraa hospital after taking written consents for acceptance of using of their bags in research aspects.

Inclusion criteria, Healthy adults 18-65 years with fitting body weight and hemoglobin >13.5g/dl.

Exclusion criteria, Pregnancy, hypertension and hypotension, fever before donation and positive results of standard blood donor infectious disease testing.

Twenty volunteers had withdrawn whole blood (WB) units each formed of (450 mL) in blood bags containing citrate phosphate dextrose adenine anticoagulant (63mL), then 20 pRBCs concentrate were prepared by centrifugation of WB bags in (Thermo Scientific, Waltham, MA, USA) centrifuge by hard spun at room temperature with no brakes within 4 hours of donation then plasma supernatant was separated into satellite bag using a manual separator while keeping HCT level to approximately 70%.

Units were devided to two groups:

Group I (n=10), pRBCs units underwent leukofiltration. By allowing the pRBCs to pass through a closed system leukodepletion filter offered by (Macopharma, Rue Lorthioi, Mouvauux, France). Filter and transfer tubing were sealed off and the bags were separated to obtain the leukofiltered pRBCs. Post-filtration Samples were examined for leukodepletion assurance.

Group II (n=10), left unfiltered pRBCs.

The pRBCs units were stored at 2-4°C. Samples were withdrawn from pRBCs units in aseptic manner at day-0, -14, -21, and at day -28 (around one week before final expiration date) and underwent standard quality control measures along storage.

Methods

1- Flow-cytometry

Flow-cytometry was conducted in Allergy and immunology centre-AL-Azhar university on multi color FACSCalibur (BD, Biosciences, San jose, USA). CellQuest Pro software (BD Biosciences, San jose, USA) was used for data analysis. Compensation setting was established before acquiring the samples using color calibrite beads (BD, Biosciences, San jose, USA, LOT 5093879). The optimal concentration for each dye used in flow was determined by titration experiments. Unstained samples were acquired to detect the sample auto-florescence. Mouse IgG2a FITC and IgG1 PE Controls (BD, Biosciences, San jose) were obtained for non specific binding detection. All dyes were applied gently to the Vortex before using to avoid any clumping. Filtered solutions were used to avoid any noise signals.

a- Microparticles separation and detection, After centrifugation of units samples at 1,500xg for 20 minutes twice to remove cells and apoptotic debris ≥ 1μm. The supernatant plasma became MP-enriched plasma. The microparticles were pelleted from 1 ml of supernatant MP-enriched plasma by another centrifugation at 10,000xg for 60 minutes at 4°C this method of separation of MPs was simulating Dinkla et al. 2014 method [16]. The majority of exosomes < 100 nm should remain in the supernatant while the microparticles will form a pellet. Then microparticles pellets of 20 μL were stored at -80°C till processing.

For processing; frozen pellets were thawed, and the 20 μL of microparticles suspension of each sample was pipetted in TrueCount tube (BD Biosciences, San jose, CA, USA, LOT 57221) containing standardized number of fluorescent beads = (50433) and incubated with 5μL of RBCs specific marker, the glycophorin A of a cluster differentiation 235a- pythoerythrine conjugated (CD235a-PE) antibody (Immunotech, Beckman Coulter, Marsellia, France, LOT 22) for 10 minutes at room temperature in dark then add 200 μL of filtered buffered saline for acquiring the sample the acquisition. Acquisition was discontinued after 5,000 events in relative region [6].

Microparticles were identified by morphological gating according to their light scattering profiling. After setting forward scatter/ side scatter on logarithmic amplification gating was set to one logarithm lower than that shown with intact RBCs, as MP size gate ranging from (0.1-1 μm). Only microparticles showing positivity for CD235a were quantified. We did not use annexin to recognize MP as only about one third of MP expose PS [17] (Figure 1).

Quantification of microparticles per microlitre (/uL) was calculated as follow [6],

Figure 1. [A] The FS/SS graph illustrating gating strategy for microparticles detection by morphological gatting. The microparticle gate was defined as region 1 [R1] lying one logarithm lower than that known for RBCs and only events included in R1 were considered as microparticles events and tested for antyCD235a, while R2 represented absolute count beads gate. [B] The plots represented FL1/CD235a considering population in upper left quadrant positive for CD235a at day-0, -14, -21, -28 for a [LR] unit while TrueCount beads appearing in small regions lying in upper right quadrant. [C] Overlay histograms representing successive samples at day-0, -14, -21, -28 for a non-leukoreduced (NLR) unit [C-1] and a LR unit [C-2] respectively, representing anty CD235a/count and considering the population after 2nd log channel, as positive

b- Phosphatedylserine exposure on intact RBCs membrane at day 28 evident by binding to annexin V; immediately after sample withdrawal, RBCs were washed with ice cold phosphate buffer saline (PBS) and centrifuged for 3 minutes at 1500xg at 4°C. after discarding the supernatant RBCs pellet was resuspended in ice cold 1X binding buffer containing calcium (Ca²⁺) for final concentration of cells =5 X 10⁶ ml then taking 50 μL of diluted sample and incubating it with 2.5 μL of annexin V-fluorescein isothiocyanate conjugated (annexin-FITC) (Immunotech, Beckman Coulter, Marsellia, France, LOT FI14006, PN IM2375) for 30 min on ice in the dark. Then adding 10 μL of anti-CD235a-PE (Immunotech, Beckman Coulter, Marsellia, France, LOT 22) the RBCs specific marker and incubate for another 10 minutes in room temperature in the dark before washing the sample once, then resuspend in 400 μl ice cold buffer. Analyze by flow after acquiring the sample within 30 minutes of final preparation. A total of 100,000 cells were counted. Setting forward scatter/ side scatter on logarithmic amplification. Only events in the size range of RBCs were gated for examining co-expression of anti-CD235a -PE and annexin V FITC.

2- KX21 Hematology analyzer (Sysmex, Kobe, Japan) for measuring mean corpuscular volume, (MCV) and mean corpuscular hemoglobin concentration (MCHC) of pRBCs at day-0, -14, -21, -28.

3- AVL-9180 Electrolyte analyzer (Roche diagnostics, Switzerland) for assessing level of potassium (K+), in sample supernatant after centrifugation at 1,500xg for 20 min at day -0, -14, -21,- 28. Pre-dilution of samples with K+ > 10 mmol/l was done.

Statistics

Data was statistically described in terms of mean ± standard deviation (±SD). Student (t) test and one way analysis of variance (ANOVA) were used for comparison of quantitative data. Correlation between various variables was done using Pearson correlation. Statistical significance was set at P ≤ 0.05. Analysis of data was carried out using Statistical Package for the Social Science; (SPSS) (Inc., Chicago, IL, USA) release 16 for Microsoft Windows (2006).

3. Results

Correlation studies

Our study revealed no significant correlation in Group-I between MP formation and K level at; day-14 (r=0.5, P=0.2), day -21 (r=0.3, P= 0.3), day-28 (r=-0.06, P=0.9). Nor Group II; day-14 (r=0.4, P=0.2), day-21 (r=0.2, P= 0.6), day-28 (r=0.5, P=0.1).

The MP number in Group I showed no significant correlation with MCV at day 14 (r=-0.4, P=0.2), day-21 (r=-0.3, P=0.4), day-28(r=-0.2, P=0.5). Also in Group I there was no significant correlation seen between MP and MCHC at day -14 (r=0.3, P=0.3), day-21 (r=0.1, P=0.7), day-28 (r=0.3; P=0.3).

The MP number in Group II showed no significant correlation with MCV at day-14 (r=-0.3, P=0.4), day-21 (r=-0.2, P=0.6), day-28 (r= 0.004; P=1). Also in Group I there was no significant correlation seen between MP and MCHC at day-14 (r=0.3, P=0.4), day-21 (r=-0.4, P=0.2), day-28 (r = -0.4; P= 0.2).

Correlation between K level with MCV at day-14, day-21, day-28 in Group I showed no correlation of significant importance (r = 0.3; P= 0.4), (r=-0.3; P= 0.4) and (r = -0.2; P= 0.6) respectively.

No significant correlation was seen between K level with MCV in Group II at day-14, day-28 (r=-0.5; P=0.1), (r=–0.4; P=0.2) while in day-21 showed negative significant correlation as (r= -0.6; P=0.05).

The K+ level in Group I showed no correlation of significant importance with MCHC at day- 14, day-21, day-28 (r = 0.3; P= 0.4), (r = 0.1; P= 0.7) and (r = -0.5; P= 0.2) respectively.

In Group II significant correlation was seen between K+ level and MCHC only at day-14 (r=- 0.6; P=0.5), while no significant correlations were shown at day-21 and day-28 (r=0.4; P=0.3), (r= -0.03; P= 0.9) respectively.

Annexin bound cells percentage in Group I significantly correlated with K+ at day -28 (r= 0.7; P= 0.03) (Figure, 5), while annexin bound cells percentage in Group I did not show any significant correlation with MP, MCV and MCHC at day -28 (r = 0.3; P= 0.5) (r = -0.5; P= 0.1) and (r = -0.5; P= 0.1) respectively.

Significant correlation was seen between annexin bound cells percentage and MP in Group II at day-28 (r=0.6; P= 0.05) (Figure, 6). while no significant correlation was seen between annexin bound cells percentage and MCV, MCHC and K+ at day-28 in Group II (r =- 0.4; P= 0.2) (r = -0.2; P= 0.3) and (r = -0.6; P= 0.8).

Table 1. Comparison between Group I and Group II units in mean values of parameters in corresponding day-0, -14, - 21, -28 of storage

Table 2. Comparison between rates of change in parameters means among each group of units at different periodic storage points at day-0, -14, -21, -28

Figure 2. Comparison between Group I and Group II in mean values of microparticles x105/uL released at corresponding day- 0, -14, -21, and -28 revealed significant increase in Group II, (P=0.05), (P<0.0001), (P<0.0001), and (P<0.0001) respectively. Also comparison between rate of change in mean values of microparticles released among each group at different periodic storage points at day-0, -14, -21, -28 revealed significant difference among Group I (P<0.0001) and among Group II ( P<0.0001)

Figure 3. Comparison between Group I and Group II in mean values of potassium level mmol/l at corresponding day -0,-14, -21, -28 revealed significant increase in Group II only at day-14, -21,and -28, (P<0.0001), (P<0.0001), and (P<0.0001) respectively. Also comparison between rate of change in mean values of potassium level among each group at the different periodic storage points at day-0, -14, -21, -28 revealed significant difference among Group I ( P<0.0001) and among Group II ( P<0.0001)

Figure 4. Comparison between Group I and Group II in mean values of percentage of annexin bound cells at day 28 revealed significant differences (P<0.0001)

Figure 5. Correlation between annexin bound cells percentage and K+ level in Group I at day-28 (r= 0.7; P= 0.03)

Figure 6. Correlation between annexin bound cells percent and MP release in Group II at day-28 (r=0.6; P= 0.05)

4. Discussion

In storage, the RBC membrane becomes unstable, echinocytes form and membrane is lost by vesiculation [7]. Microparticles are small vesicles of less than 1 μm in size, containing a subset of proteins derived from their parent cells [6].

In our study mean number of MPs detected in Group I after leukofiltration at day-0 was 1.5x10⁵±0.3×10⁵/uL and by day-28 reached 4x10⁵±0.4×10⁵/uL. Our values were higher than those reported by Grisendi et al, 2015 who examined saline, adenine, glucose, mannitol (SAG-M) suspended RBCs leukodepleted by filtration and the mean number of MPs detected in their study depending on morphological gate was 1.17×10⁵±0.067×10⁵/uL at day-0 and by day-42 reached 2.58×10⁵±0.34×10⁵/uL. This increase in MPs numbers in our study could be explained by differences in storage media in both studies as we added no SAG-M. Grisendi et al also compared the numbers of MPs evaluated depending on the morphological gate only with numbers obtained after using carboxyfluorescein diacetate succinimidyl ester (CFSE) staining which can stain only closed vesicles and exclude cell fragments from gating and the latter numbers were less than numbers given depending on morphological gate alone. Grisendi and his college explained that due to ability of CFSE to exclude non specific events like cellular fragments which were negative for CFSE while the morphological gate alone could not exclude those non specific events [6].

Our study revealed continuous elevation in the mean levels of MP/uL as the storage period increase reaching >2 folds increase in Group I from day-0 to day-28 and around 4 fold increase in Group II. Which points that the influence of storage period elongation on MP generation was less noticed in Group I that underwent leukofiltration.

The comparison between Group I and Group II in mean values of MP/ul revealed significant increase in Group II at day-0 (P=0.05), day-14 (P<0.0001), day-21 (P<0.0001), day-28 (P<0.0001). This data was in agreement with Sugawara et al, 2010 who examined platelet derived microparticles (PDMP) in WB and concluded that leukofiltration of WB lowers prestorage PDMP which remained low throughout 35 days of storage while in contrast, PDMP increase significantly in unfiltered WB [18].

These data in the current study, was not in agreement with Nollet et al, 2013 who reported that significant differences did not emerge after comparing values of red cell derived microparticles (RDMP) between plasma aphaeresis stored without leukofiltration, with leukofiltered plasma [19]. But this could be due to the difference in type of units examined as we examined pRBCs units not plasma.

The current study revealed successive increase in the mean levels of K+ reaching around 9 and 10 folds increase from day-0 to day-28 in Group I and Group II respectively (Figure. 3). Which could be explained by that RBC membrane provides permeability barrier, enabling cells to maintain different concentrations of ions internally, but this barrier is not perfect and ions may leak. In circulation this minor leak is constantly corrected by adenosine triphosphatase (ATPase) which pumps potassium into cell in exchange for sodium and maintains potassium gradient. Donated RBCs are stored at 4°C, and at 4°C the ATPase has limited functionality, even before ATP becomes limited [7]. Also it was reported that oxidation of the cytoskeleton due to decreased synthesis of glutathione may exacerbate cation leak by weakening the membrane. Lipids are also oxidized and lipid hydroperoxides may permit a deformation- dependent leak of cations [20].

Concerning Group I units those underwent leukofiltration, K+ levels were, at day-0 = (3±0.5) mmol/l, at day-14 = (8.2±1.7) mmol/l at day-21 = (20.2±3.3) mmol/l and at day-28 = (27.6±2.1) mmol/l. Those values were less than those given by Chaudhary and, Katharia, 2012 when they studied buffy coat removed RBCs and the mean K+ levels obtained by them were; on day-0 (5.16±1.2) mmol/l; on day-14 (15.7±3.0) mmol/l; and on day-28 (35.1±4.6) (mmol/l), which may reflect efficacy of leukofiltration over buffy coat removal [21].

Comparison between Group I and II in mean values of K+ levels revealed significant increase in Group II at day-14 (P<0.0001), day-21 (P<0.0001), day-28 (P<0.0001) while day-0 showed no significant increase (P=0.9). This data was in agreement with Sonker et al, 2014 who stated that leukofiltered pRBC showed lesser elevation of K+, at end of storage period than their unfiltered counterpart and that leukofiltration appeared to produce RBC units with less cellular contamination and less release of intracellular enzymes [15].

Kamel et al, 2010 stated that leukoreduction by buffy coat removal had no impact on potassium nor PS exposure [22]. These conflicting results could be explained by that leukoreduction by buffy coat removal removes only 70–80% of leukocytes. And any given leukocyte content <5 X10⁶/unit is considered as leukoreduced [15]. So leukocytes toxic effect can still exist while leukofiltration performs far better efficacy of leukocyte depletion.

The current study revealed continuous increase of MCV during storage in both groups as the storage period increased. Previous studies explained that continuous increase in MCV in storage by the presence of cation leak, as redistribution of cations leads to increased RBCs uptake of water causing swelling. In circulating RBCs loss of cation gradients and cell swelling may activate the Gardos channel, resulting in potassium loss and cell shrinkage, but this mechanism requires calcium which is reduced in stored RBCs due to the presence of citrate [23]. Citrate in storage solutions chelates calcium needed for Gardos channel activity [24].

Flatt et al, 2014 clarified that during storage some cells lose their membrane by vesiculation, becoming smaller and spherocytic while others are swollen by water intake due to cation leak, so the rate of swelling must be higher to show overall increase in MCV by end of storage [7].

Blasi et al, 2012 stated that by day 21 of storage the osmotic fragility of RBCs is increased and more than 50% of the cells display non-discocyte morphology. By day 35 of storage the morphology of about 25% of RBCs is irreversibly altered; the cells have lost membrane through vesiculation and become spherocytic [25].

Comparison between both groups in mean values of MCV revealed significant increase at day-28 (P=0.02) in Group II with mean MCV value = (96.2±6.8) fl while in Group I which underwent leukofiltration MCV was (90±1.4) fl. Which agrees with Phelan et al, 2010 study that pointed that leukocyte activity on red cell membranes during storage caused distortion of RBCs morphology [26].

Comparison between both groups in means of MCHC revealed significant decrease in Group II at day-28 (P<0.0001), as MCHC mean was (31.1±1.2) mg/dL in Group I but in Group II was (28.6±1.6) mg/dL While no significant differences were seen in two groups at day-14 (P=0.2), day-21 (P=0.09).

A previous study by Dinkla et al, 2014 stated that PS exposure on stored RBCs remains quite low, only beginning to increase around day-28 of storage [16]. So day-28 was the time chosen in our study to examine PS exposure and results revealed significant increase in mean percentage of annexin V bound cells in Group II which was (10.7±1.7)% while in Group I was only (3.9±1.5)%.

Lu et al, 2011 used lactadheren which showed better efficacy to detect PS exposure in comparison with annnexin. Lactadherin detected 1.5% PS-positive stored RBCs vs 0·5% for annexin V after 14 days of storage, which significantly reached 18.4 vs 4.5% after 42 days of storage. The study stated also that incubation at 37°C in fresh heparinised plasma partially reversed PS exposure of RBCs stored for 14 days but had no effect on cells stored for 42 days [27].

Sparrow et al, 2006 study proved that PS was not detected at the external membrane of young or old RBCs during storage but increased levels of annexin V were detected in supernatant of RBCs stored in presence of leukocytes, with significantly greater supernatant levels found for old RBCs compared to young RBCs [28].

Our study revealed that as storage period increase in both groups, the levels of both MP and K were building up along lengthening of the storage periods (Figure, 2) and (Figure, 3).

Our data agrees with Liu et al, 2014 study which stated that morphological changes of RBCs during storage include, decreased MCHC, varied MCV, and reduced integrity of the erythrocyte membrane with formation of MP [29].

This progressive increase in, morphological disturbance of intact RBCs, MP generation and K leak with lengthening of storage period may be the answer of previous observational studies evaluated critically ill individuals, including cardiac surgery [30], pediatric intensive care [31], and trauma patients [32], and found associations between transfusion of “older” RBCs (defined as >14 days of storage) and increased mortality, multiple organ failure (MOF), and sepsis [33]. This may throw some light on why safer transfusion accompanies fresher pRBCs units as older units carry higher levels of MPs which was already tightened to many post-transfusion side effects [7-12]. While Lacroix et al, 2015 stated that Transfusion of fresh red cells, as compared with standard-issue red cells, did not decrease the 90-day mortality among critically ill adults [34].

Putting in consideration that the rate of increase in MP and K+ levels at day-14 in the current study although still showing significant increase comparing to day-0, but still was more stable and steady increase than that sharp rise shown at day-21 and day-28. This was in agreement with other researchers who considered that the cut-off point for the definition of old RBCs was that older than 14 days of storage [35].

The mean values of MP and K were always significantly less in leukofiltered group compared to its unfiltered counterpart. Denoting the efficacy of leukofilteration in limiting the MP release and K leak (Figure, 2) and (Figure, 3).

Also the significant decrease in values of annnexin bound cells high lighten the effect of leukofilteration in limiting apoptotic changes in leukofiltered units (Figure, 4).

The current study revealed no significant correlation in Group I nor in Group II between MPs formation and K+ level, which could be explained by that cation leak does not have direct impact on release of vesicles during storage and this data was in agreement with Flatt et al, 2014 study who stated that vesiculation seems to occur mainly due to oxidation of cytoskeleton/spectrin which weakens spectrin–actin-protein 4.1 interactions. Also storage media which efficiently combat oxidative stress produce fewer vesicles and MPs generation occur partly as result of cation leak [7]. Also Burger et al, 2013 stated that K+ leakage primes erythrocytes for PS exposure. PS exposure will lead to vesiculation in prolonged storage [36].

A significant correlation was seen in Group I at day 28 between annexin and K+ (r= 0.7; P=0.03), which could be explained by, as when RBC leaks cations, the intracellular potassium levels decrease, scrambling may increase and more PS may be exposed. Together these effects initiate creating protrusions on cell membrane [36]. Also Wolf et al, 2009 previously proved that high intracellular potassium concentration in fresh RBCs inhibits lipid scrambling activity [37] (Figure, 5).

A significant correlation was seen in Group II only at day 28 between annexin and MPs formation (r= 0.6; P= 0.05). Mandal et al, 2005 statated that Fas-related signaling molecules such as Fas-associated death domain and caspase 8 are present in vesicles and the amount of vesicle-associated Fas and caspase 3 increases with storage time. Fas is thought to activate caspases which inhibit flippase activity and cause PS exposure [38]. Also Wolf et al, 2009 stated that PS exposure may be a prerequisite for the formation of echinocytes and RBC vesiculation [37]. Still no significant correlation was seen in Group I (r =0.3; P= 0.5) which could be related to leukofiltration effect (Figure, 6).

5. Conclusions

In conclusion, we reported that leukofiltration reduced microparticles generation, levels of potassium leak during storage and percentage of cells exposing phosphatedylserine on their membrane. Storage period length markedly influenced the generation microparticles of and potassium leak especially after 14 days which may provide insights on why safer transfusion may go along with fresher units transfusion, and that may provide clinical relevance in how to minimize transfusion related side effects.

Further studies should be carried out to aid in the development of new strategies to prevent red cell storage lesions and their side effects on the transfused patients.

ACKNOWLEDGMENTS

We thank Dr Mona H. Alrayes, MD. Head master of Blood Bank centre of Al-Zahraa hospital for her support during the development of the study.