Vox Sanguinis (2014) 106, 45–54 © 2013 International Society of Blood Transfusion DOI: 10.1111/vox.12074
ORIGINAL PAPER
Impact of constant storage temperatures and multiple warming cycles on the quality of stored red blood cells T. Wagner,1 M.-A. Pabst,2 G. Leitinger,2,3 U. Reiter,4 N. Kozma,1 G. Lanzer1 & B. Huppertz2 1
Department of Blood Group Serology and Transfusion Medicine, Medical University of Graz, Graz, Austria Institute of Cell Biology, Histology and Embryology, Medical University of Graz, Graz, Austria 3 Center for Medical Research (ZMF), Medical University of Graz, Graz, Austria 4 Department of Radiology, Medical University of Graz, Graz, Austria 2
Background Red blood cells (RBCs) are routinely stored in liquid state at temperatures below 6°C, and RBC unit core temperature should not exceed 10°C during transport. Since the critical temperature of 10°C was chosen mostly arbitrarily, this study investigated the effect of both constant temperature settings as well as multiple rewarming cycles on stored RBCs with respect to morphology, biochemical parameters and haemolysis. Materials and Methods Buffy coat-depleted filtered RBCs were used as standard products. RBCs were stored at 1–6°C (reference group, n = 12), 13 and 22°C (test groups, n = 12 each) or stored at 1–6°C and warmed up five times to 10, 13, or 22°C for a period of 24 h each. Various biochemical parameters were measured weekly. RBCs were further investigated using electron microscopy. Results Red blood cells stored constantly at 13 or 22°C showed stable haemolysis rates until day 28 and day 14, respectively. RBCs stored at 1–6°C with five warming-up periods to 10, 13 or 22°C each lasting 24 h (total 120 h) did not exceed the limit of the haemolysis rate at the end of storage. Differently shaped erythrocytes were found in all samples, but more crenate erythrocytes appeared after 42 days of storage independent of temperature profiles.
Received: 20 December 2012, revised 2 July 2013, accepted 4 July 2013, published online 2 August 2013
Conclusion Red cells can be kept at constant temperatures above 6°C without apparent harmful effects at least until day 14, whereas multiple warming cycles for no longer than 24 h at 10, 13 or 22°C with subsequent cooling do not cause quality loss as assessed using the in vitro assays employed in this study. Key words: electron microscopy, quality, RBC, storage, temperature.
Introduction Currently, storage of RBCs is a worldwide routine procedure. Due to ongoing research in the last 100 years, storage periods were extended from hours (whole blood in glass bottles) to 6 weeks (RBC units resuspended in different media) [1]. The regimen to store RBCs is to keep them as cold as possible, but to prevent them from freezing [2–4]. This approach intends to reduce metabolism to a Correspondence: Thomas Wagner, Department of Blood Group Serology and Transfusion Medicine, Medical University of Graz, Auenbruggerplatz 3, A-8036 Graz, Austria E-mail: [email protected]
minimum and keeps cells alive under unphysiological conditions [5–7]. After removal of plasma (needed for plasma fractionation) from whole blood (WB) donations administration of an additive solution, for example salinen, adenine, glucose, mannitol (SAGM) additive solution is required to maintain metabolic activity of red cells, reducing risk of adverse reactions to plasma proteins and infections and permitting a longer shelf life and preserving cell quality during storage. Once the RBC unit is delivered from the blood bank, it should not be reused, mainly because of the lack of quality control. However, providing adequate temperature control, the re-use of these units is acceptable [8, 9]. Radio frequency identification (RFID) technology turned
45
46 T. Wagner et al.
out to be a promising tool to handle this task [10–12]. The technology allows to record temperature and good manufacturing practice (GMP) relevant information with multiple usage and cost reduction [13, 14]. Attempts to reduce discard (which is up to 10%) of blood products due to lack of adequate quality control have been discussed in several studies [15–17]. Additionally, the well-known 30-min rule is a subject of controversies and under various circumstances suggested to be reduced or even extended [18–23]. Various studies have investigated the influence of storage temperatures between -10 and +37°C on quality and survival of transfused red blood cells (RBCs) [5–7, 24], but only little is known about storage temperature of RBC in additive solution [25, 26]. To assess under which conditions RBCs could be reused, after being once released from the blood bank, this study focused to investigate (1) the influence of higher storage temperatures and (2) the effect of multiple warming cycles during storage on morphology, quality and functional parameters of refrigerated stored RBCs during 42 days of storage.
Materials and methods Preparation of blood cells
temperature is close to the widely accepted temperature of 10°C for transportation purposes and from daily experience we know, that there is a peak near 13°C when using temperature loggers for transportation, which suggests routine noncompliance with the regulations. Choosing 22°C reflects standard handling temperature when, for example, cross-matching is performed. To examine the impact of multiple warming and cooling during storage on the quality of RBCs, units were continually stored at 1–6°C (n = 12) for 42 days as control group, while further RBCs were continually stored at 1–6°C, but weekly exposed to 10°C (n = 10), 13°C (n = 10) or 22°C (n = 10), for a period of 24 h, summing up to five periods of exposure to raised temperatures with a total of 120 h for each unit. In previous studies, using different temperature measurement devices to estimate both core and surface temperature, it has been shown that RBCs take many hours to completely reach the environmental temperature [16, 18, 19, 23, 29]. Therefore and to possibly provoke a measurable effect, a 24-h time slot was used. For the refrigerated storage (1–6, 10, 13°C) specially adapted refrigerators and for room temperature storage the routinely used chamber for overnight storage of whole blood (WB) donations were used. Except on day 3, samples were taken weekly, on day 7, 21, 28, 35 and 42 RBCs subsequently exposed to 10, 13 and 22°C for 24-h.
Whole blood was obtained from healthy volunteer blood donors according to the Austrian regulations for blood donation and after informed consent. All donors agree that their donation may be used for research or quality procedures as this is requested by Austrian law [27, 28]. Collection was performed in triple bags (LCR5 filtration set) containing citrate–phosphate–dextrose (CPD) in the primary bag, SAG-M in the satellite bag (MacoPharma; LAB. Pharmaceutiques, Tourcoing, France) and centrifuged at 4000 9 g for 10 min at 20°C. RBCs and plasma were separated from buffy coat fraction and transferred into the satellite containers using an automated separator (Compomat G4; NPBI, Amsterdam, The Netherlands). RBCs were leukoreduced using LCR5 leukoreduction filter (MacoPharma; LAB. Pharmaceutiques) within 30 min after separation. The final volume of stored RBCs was 300–320 ml and the residual white blood cells (WBC) count below 1 9 10E6 per unit.
where fHb is the free haemoglobin, HctRBC the haematocrit and HbRBC the total haemoglobin of the RBC unit.
RBC storage
Osmotic resistance
To examine the impact of storage temperature on the quality of RBCs, units were continually stored at 1–6°C (n = 12) as control, while further RBCs were stored at 13°C (n = 12) or 22°C (n = 12) for 42 days. The rationale to choose 13°C as a temperature of interest was that this
Osmotic resistance was measured to determine the resistance of erythrocytes against hypotonic solutions. Dilution series using reagents with decreasing hypotonic concentrations were performed. Twelve tubes were prepared containing NaCl solutions ranging from 09 to
In vitro measurements of metabolic parameters Samples were taken under aseptic conditions, with sample site couplers from different donors and analysed on day 3, 7, 14, 21, 28, 35 and 42. Potassium, pH, lactate, glucose, haematocrit, haemoglobin and free haemoglobin were determined at room temperature using a blood gas analyzer (Cobas b 221; Roche, Graz, Austria) Total blood count was performed at room temperature using an automatic counter (Sysmex SE-9500; M€ uller, Vienna, Austria). Haemolysis rate (HR) was calculated according to the following equation [30]: HR½% ¼ ð100 HctRBC Þ ½% fHb ½g=dl = HbRBC ½g=dl
2013 International Society of Blood Transfusion Vox Sanguinis (2014) 106, 45–54
Temperature dependent quality of RBCs 47
01%, with 09 as the physiological condition. In each tube, 50 ll of RBCs was inoculated and incubated for 30 min. The tubes were rotated and subsequently centrifuged for 5 min at 1500 g. The supernatant was measured by photometry at 540 nm.
Adenosine triphosphate (ATP) To determine ATP concentration (lM/dl) in red blood cells, an enzymatic test (ATP Hexokinase FS; DiaSys, Vienna, Austria) was performed according to the manufacturer’s instructions.
2,3-Diphosphoglycerate (2,3 DPG) To determine 2,3 DPG concentration (mM/l) in red blood cells, an enzymatic test (UV-testkit; Roche, Graz, Austria) was used according to the manufacturer’s instructions.
assay (ELISA) reader. Reading was adjusted to the background absorbance of reaction mix and to haemolysis, by using the supernatant without reaction mix.
Scanning electron microscopy Red blood cells were diluted in an equal amount of a mixture of 25% glutaraldehyde and 2% paraformaldehyde in 01M cacodylate buffer, pH 74, and a drop of this mixture was placed on alcian-blue coated cover slips and fixed for 30 min at room temperature in the same fixative. After washing in 01M cacodylate buffer, RBCs were postfixed for 30 min at room temperature in 2% OsO4. After dehydration, RBCs were dried using a critical point dryer, sputtered and examined in a scanning electron microscope (DSM 950; Carl Zeiss AG, Oberkochen, Germany).
Transmission electron microscopy Assessment of free total protein Total free protein concentrations were determined according to Lowry et al. Briefly, 30 ll of 1:10 prediluted plasma samples or BSA protein standards were incubated with 750 ll of Lowry reagent and incubated for 10 min. After the addition of 1:2 prediluted Folin reagent and a further incubation step, the absorbance was measured in a BioPhotometer (Eppendorf, Vienna, Austria).
Assessment of free and degraded spectrin Thirty microgram of total free protein was separated using Nu polyacrylamide gel electrophoresis (NuPAGE), 4–12% Bis-Tris gels (Invitrogen, Carlsbad, CA, USA) and transferred to 045 lm nitrocellulose membranes (Hybond ECL; Amersham BioSciences UK Ltd. Buckinghamshire, UK) using a semidry blotting system. Membranes were probed for spectrin with mouse monoclonal antispectrin antibody (clone RG/26, 1 lg/ml; Santa Cruz Biotechnology, Santa Cruz, CA, USA), and the detection was performed by WesternBreeze chemiluminescent immunodetection system (Invitrogen, Lofer, Austria) according to the manufacturer’s instructions.
Red blood cells stored at constant temperatures were either encapsulated in 7% gelatine at 37°C or they were prefixed with 025% glutaraldehyde for 20 min and then encapsulated in alginate beds. Both types of preparations were then fixed at room temperature in a mixture of 25% glutaraldehyde and 2% paraformaldehyde in 01M cacodylate buffer, pH 74 for 2 h. After washing with 01M cacodylate buffer, RBCs were postfixed for 120– 150 min at room temperature in 2% OsO4, dehydrated in a series of ascending ethanol concentrations and propylene oxide and embedded in TAAB epoxy resin (TAAB Laboratories Equipment Ltd., Berkshire, UK). Blocks were cut with a Leica Ultracut UCT microtome (Leica Microsystems, Vienna, Austria). Toluidine-blue-stained semithin sections were scanned with a tissue faxs AxioImager Z.1 (Carl Zeiss MicroImaging GmbH, G€ ottingen, Germany) using a software, Tissue Faxs (TissueGnostics GmbH, Vienna, Austria) and the number of differently formed erythrocytes were counted in standardized fields (n = 6) with a 609 objective. Ultrathin sections were stained with uranyl acetate and lead citrate and examined at 50 kV in an electron microscope (EM 902; Carl Zeiss AG, Oberkochen, Germany). No differences between preparations encapsulated in gelatine or alginate were observed.
Lactate dehydrogenase (LDH) assay Lactate dehydrogenase activity was measured using an LDH cytotoxicity detection kit (Takara Bio Inc., Otsu Shiga, Japan). The assay was performed according to the manufacturer’s protocol. Hundred microlitre of supernatant sample was mixed with 100 ll reaction mixture in a 96-well plate. The absorbance of the samples was read at 492 nm/620 nm using an enzyme-linked immunosorbent 2013 International Society of Blood Transfusion Vox Sanguinis (2014) 106, 45–54
Statistical analysis Statistical evaluation was performed using NCSS 2007 software (NCSS, LLC, Kaysville, UT, USA). Metabolic parameters of all groups were expressed as medians together with their 95% confidence limits. Medians of pH, potassium, glucose, lactate, haemoglobin, haematocrit, free haemoglobin and haemolysis rate were compared
48 T. Wagner et al.
with the control group medians using the Wilcoxon ranksum test for difference in medians. Group medians of 2,3DPG, ATP and osmotic resistance parameters were compared by Kruskal–Wallis one-way analysis of variance and Dunn’s multiple-comparison test. Significance level was generally set to P < 005.
Results Metabolic parameters at different, constant temperatures during storage During storage, the units underwent different biochemical and structural alterations. Single measurements of the biochemical parameters at day 3 showed minimal metabolic changes between the reference and the test groups. Levels of pH showed a slight decrease, while free haemoglobin levels increased during storage (Table 1). Potassium level increased the most in units stored at 22°C from the 7th day and had leaked out by the day 21 reaching tenfold higher levels at day 42 than on day 3 in the warm stored group. Free haemoglobin and the haemolysis rate of RBCs stored constantly at 22°C remained low and stable until day 14, increased slightly on day 21 and faster thereafter. In contrast, RBCs stored constantly at 13°C had haemolysis rates within normal range until day 35, but increased at the end of storage. The highest level of free haemoglobin in either filtered or nonfiltered RBCs at the end of 42-day storage should be
ORIGINAL PAPER
Impact of constant storage temperatures and multiple warming cycles on the quality of stored red blood cells T. Wagner,1 M.-A. Pabst,2 G. Leitinger,2,3 U. Reiter,4 N. Kozma,1 G. Lanzer1 & B. Huppertz2 1
Department of Blood Group Serology and Transfusion Medicine, Medical University of Graz, Graz, Austria Institute of Cell Biology, Histology and Embryology, Medical University of Graz, Graz, Austria 3 Center for Medical Research (ZMF), Medical University of Graz, Graz, Austria 4 Department of Radiology, Medical University of Graz, Graz, Austria 2
Background Red blood cells (RBCs) are routinely stored in liquid state at temperatures below 6°C, and RBC unit core temperature should not exceed 10°C during transport. Since the critical temperature of 10°C was chosen mostly arbitrarily, this study investigated the effect of both constant temperature settings as well as multiple rewarming cycles on stored RBCs with respect to morphology, biochemical parameters and haemolysis. Materials and Methods Buffy coat-depleted filtered RBCs were used as standard products. RBCs were stored at 1–6°C (reference group, n = 12), 13 and 22°C (test groups, n = 12 each) or stored at 1–6°C and warmed up five times to 10, 13, or 22°C for a period of 24 h each. Various biochemical parameters were measured weekly. RBCs were further investigated using electron microscopy. Results Red blood cells stored constantly at 13 or 22°C showed stable haemolysis rates until day 28 and day 14, respectively. RBCs stored at 1–6°C with five warming-up periods to 10, 13 or 22°C each lasting 24 h (total 120 h) did not exceed the limit of the haemolysis rate at the end of storage. Differently shaped erythrocytes were found in all samples, but more crenate erythrocytes appeared after 42 days of storage independent of temperature profiles.
Received: 20 December 2012, revised 2 July 2013, accepted 4 July 2013, published online 2 August 2013
Conclusion Red cells can be kept at constant temperatures above 6°C without apparent harmful effects at least until day 14, whereas multiple warming cycles for no longer than 24 h at 10, 13 or 22°C with subsequent cooling do not cause quality loss as assessed using the in vitro assays employed in this study. Key words: electron microscopy, quality, RBC, storage, temperature.
Introduction Currently, storage of RBCs is a worldwide routine procedure. Due to ongoing research in the last 100 years, storage periods were extended from hours (whole blood in glass bottles) to 6 weeks (RBC units resuspended in different media) [1]. The regimen to store RBCs is to keep them as cold as possible, but to prevent them from freezing [2–4]. This approach intends to reduce metabolism to a Correspondence: Thomas Wagner, Department of Blood Group Serology and Transfusion Medicine, Medical University of Graz, Auenbruggerplatz 3, A-8036 Graz, Austria E-mail: [email protected]
minimum and keeps cells alive under unphysiological conditions [5–7]. After removal of plasma (needed for plasma fractionation) from whole blood (WB) donations administration of an additive solution, for example salinen, adenine, glucose, mannitol (SAGM) additive solution is required to maintain metabolic activity of red cells, reducing risk of adverse reactions to plasma proteins and infections and permitting a longer shelf life and preserving cell quality during storage. Once the RBC unit is delivered from the blood bank, it should not be reused, mainly because of the lack of quality control. However, providing adequate temperature control, the re-use of these units is acceptable [8, 9]. Radio frequency identification (RFID) technology turned
45
46 T. Wagner et al.
out to be a promising tool to handle this task [10–12]. The technology allows to record temperature and good manufacturing practice (GMP) relevant information with multiple usage and cost reduction [13, 14]. Attempts to reduce discard (which is up to 10%) of blood products due to lack of adequate quality control have been discussed in several studies [15–17]. Additionally, the well-known 30-min rule is a subject of controversies and under various circumstances suggested to be reduced or even extended [18–23]. Various studies have investigated the influence of storage temperatures between -10 and +37°C on quality and survival of transfused red blood cells (RBCs) [5–7, 24], but only little is known about storage temperature of RBC in additive solution [25, 26]. To assess under which conditions RBCs could be reused, after being once released from the blood bank, this study focused to investigate (1) the influence of higher storage temperatures and (2) the effect of multiple warming cycles during storage on morphology, quality and functional parameters of refrigerated stored RBCs during 42 days of storage.
Materials and methods Preparation of blood cells
temperature is close to the widely accepted temperature of 10°C for transportation purposes and from daily experience we know, that there is a peak near 13°C when using temperature loggers for transportation, which suggests routine noncompliance with the regulations. Choosing 22°C reflects standard handling temperature when, for example, cross-matching is performed. To examine the impact of multiple warming and cooling during storage on the quality of RBCs, units were continually stored at 1–6°C (n = 12) for 42 days as control group, while further RBCs were continually stored at 1–6°C, but weekly exposed to 10°C (n = 10), 13°C (n = 10) or 22°C (n = 10), for a period of 24 h, summing up to five periods of exposure to raised temperatures with a total of 120 h for each unit. In previous studies, using different temperature measurement devices to estimate both core and surface temperature, it has been shown that RBCs take many hours to completely reach the environmental temperature [16, 18, 19, 23, 29]. Therefore and to possibly provoke a measurable effect, a 24-h time slot was used. For the refrigerated storage (1–6, 10, 13°C) specially adapted refrigerators and for room temperature storage the routinely used chamber for overnight storage of whole blood (WB) donations were used. Except on day 3, samples were taken weekly, on day 7, 21, 28, 35 and 42 RBCs subsequently exposed to 10, 13 and 22°C for 24-h.
Whole blood was obtained from healthy volunteer blood donors according to the Austrian regulations for blood donation and after informed consent. All donors agree that their donation may be used for research or quality procedures as this is requested by Austrian law [27, 28]. Collection was performed in triple bags (LCR5 filtration set) containing citrate–phosphate–dextrose (CPD) in the primary bag, SAG-M in the satellite bag (MacoPharma; LAB. Pharmaceutiques, Tourcoing, France) and centrifuged at 4000 9 g for 10 min at 20°C. RBCs and plasma were separated from buffy coat fraction and transferred into the satellite containers using an automated separator (Compomat G4; NPBI, Amsterdam, The Netherlands). RBCs were leukoreduced using LCR5 leukoreduction filter (MacoPharma; LAB. Pharmaceutiques) within 30 min after separation. The final volume of stored RBCs was 300–320 ml and the residual white blood cells (WBC) count below 1 9 10E6 per unit.
where fHb is the free haemoglobin, HctRBC the haematocrit and HbRBC the total haemoglobin of the RBC unit.
RBC storage
Osmotic resistance
To examine the impact of storage temperature on the quality of RBCs, units were continually stored at 1–6°C (n = 12) as control, while further RBCs were stored at 13°C (n = 12) or 22°C (n = 12) for 42 days. The rationale to choose 13°C as a temperature of interest was that this
Osmotic resistance was measured to determine the resistance of erythrocytes against hypotonic solutions. Dilution series using reagents with decreasing hypotonic concentrations were performed. Twelve tubes were prepared containing NaCl solutions ranging from 09 to
In vitro measurements of metabolic parameters Samples were taken under aseptic conditions, with sample site couplers from different donors and analysed on day 3, 7, 14, 21, 28, 35 and 42. Potassium, pH, lactate, glucose, haematocrit, haemoglobin and free haemoglobin were determined at room temperature using a blood gas analyzer (Cobas b 221; Roche, Graz, Austria) Total blood count was performed at room temperature using an automatic counter (Sysmex SE-9500; M€ uller, Vienna, Austria). Haemolysis rate (HR) was calculated according to the following equation [30]: HR½% ¼ ð100 HctRBC Þ ½% fHb ½g=dl = HbRBC ½g=dl
2013 International Society of Blood Transfusion Vox Sanguinis (2014) 106, 45–54
Temperature dependent quality of RBCs 47
01%, with 09 as the physiological condition. In each tube, 50 ll of RBCs was inoculated and incubated for 30 min. The tubes were rotated and subsequently centrifuged for 5 min at 1500 g. The supernatant was measured by photometry at 540 nm.
Adenosine triphosphate (ATP) To determine ATP concentration (lM/dl) in red blood cells, an enzymatic test (ATP Hexokinase FS; DiaSys, Vienna, Austria) was performed according to the manufacturer’s instructions.
2,3-Diphosphoglycerate (2,3 DPG) To determine 2,3 DPG concentration (mM/l) in red blood cells, an enzymatic test (UV-testkit; Roche, Graz, Austria) was used according to the manufacturer’s instructions.
assay (ELISA) reader. Reading was adjusted to the background absorbance of reaction mix and to haemolysis, by using the supernatant without reaction mix.
Scanning electron microscopy Red blood cells were diluted in an equal amount of a mixture of 25% glutaraldehyde and 2% paraformaldehyde in 01M cacodylate buffer, pH 74, and a drop of this mixture was placed on alcian-blue coated cover slips and fixed for 30 min at room temperature in the same fixative. After washing in 01M cacodylate buffer, RBCs were postfixed for 30 min at room temperature in 2% OsO4. After dehydration, RBCs were dried using a critical point dryer, sputtered and examined in a scanning electron microscope (DSM 950; Carl Zeiss AG, Oberkochen, Germany).
Transmission electron microscopy Assessment of free total protein Total free protein concentrations were determined according to Lowry et al. Briefly, 30 ll of 1:10 prediluted plasma samples or BSA protein standards were incubated with 750 ll of Lowry reagent and incubated for 10 min. After the addition of 1:2 prediluted Folin reagent and a further incubation step, the absorbance was measured in a BioPhotometer (Eppendorf, Vienna, Austria).
Assessment of free and degraded spectrin Thirty microgram of total free protein was separated using Nu polyacrylamide gel electrophoresis (NuPAGE), 4–12% Bis-Tris gels (Invitrogen, Carlsbad, CA, USA) and transferred to 045 lm nitrocellulose membranes (Hybond ECL; Amersham BioSciences UK Ltd. Buckinghamshire, UK) using a semidry blotting system. Membranes were probed for spectrin with mouse monoclonal antispectrin antibody (clone RG/26, 1 lg/ml; Santa Cruz Biotechnology, Santa Cruz, CA, USA), and the detection was performed by WesternBreeze chemiluminescent immunodetection system (Invitrogen, Lofer, Austria) according to the manufacturer’s instructions.
Red blood cells stored at constant temperatures were either encapsulated in 7% gelatine at 37°C or they were prefixed with 025% glutaraldehyde for 20 min and then encapsulated in alginate beds. Both types of preparations were then fixed at room temperature in a mixture of 25% glutaraldehyde and 2% paraformaldehyde in 01M cacodylate buffer, pH 74 for 2 h. After washing with 01M cacodylate buffer, RBCs were postfixed for 120– 150 min at room temperature in 2% OsO4, dehydrated in a series of ascending ethanol concentrations and propylene oxide and embedded in TAAB epoxy resin (TAAB Laboratories Equipment Ltd., Berkshire, UK). Blocks were cut with a Leica Ultracut UCT microtome (Leica Microsystems, Vienna, Austria). Toluidine-blue-stained semithin sections were scanned with a tissue faxs AxioImager Z.1 (Carl Zeiss MicroImaging GmbH, G€ ottingen, Germany) using a software, Tissue Faxs (TissueGnostics GmbH, Vienna, Austria) and the number of differently formed erythrocytes were counted in standardized fields (n = 6) with a 609 objective. Ultrathin sections were stained with uranyl acetate and lead citrate and examined at 50 kV in an electron microscope (EM 902; Carl Zeiss AG, Oberkochen, Germany). No differences between preparations encapsulated in gelatine or alginate were observed.
Lactate dehydrogenase (LDH) assay Lactate dehydrogenase activity was measured using an LDH cytotoxicity detection kit (Takara Bio Inc., Otsu Shiga, Japan). The assay was performed according to the manufacturer’s protocol. Hundred microlitre of supernatant sample was mixed with 100 ll reaction mixture in a 96-well plate. The absorbance of the samples was read at 492 nm/620 nm using an enzyme-linked immunosorbent 2013 International Society of Blood Transfusion Vox Sanguinis (2014) 106, 45–54
Statistical analysis Statistical evaluation was performed using NCSS 2007 software (NCSS, LLC, Kaysville, UT, USA). Metabolic parameters of all groups were expressed as medians together with their 95% confidence limits. Medians of pH, potassium, glucose, lactate, haemoglobin, haematocrit, free haemoglobin and haemolysis rate were compared
48 T. Wagner et al.
with the control group medians using the Wilcoxon ranksum test for difference in medians. Group medians of 2,3DPG, ATP and osmotic resistance parameters were compared by Kruskal–Wallis one-way analysis of variance and Dunn’s multiple-comparison test. Significance level was generally set to P < 005.
Results Metabolic parameters at different, constant temperatures during storage During storage, the units underwent different biochemical and structural alterations. Single measurements of the biochemical parameters at day 3 showed minimal metabolic changes between the reference and the test groups. Levels of pH showed a slight decrease, while free haemoglobin levels increased during storage (Table 1). Potassium level increased the most in units stored at 22°C from the 7th day and had leaked out by the day 21 reaching tenfold higher levels at day 42 than on day 3 in the warm stored group. Free haemoglobin and the haemolysis rate of RBCs stored constantly at 22°C remained low and stable until day 14, increased slightly on day 21 and faster thereafter. In contrast, RBCs stored constantly at 13°C had haemolysis rates within normal range until day 35, but increased at the end of storage. The highest level of free haemoglobin in either filtered or nonfiltered RBCs at the end of 42-day storage should be
