Hold on to your units: Quality of red cell concentrates is only affected after multiple transient warming events Authors: Jayme Kurach, Mackenzie Brandon‐Coatham, Nishaka William, Carly Olafson, Tracey R. Turner, Anika Tahsin Rahman, Behrouz Ehsani‐Moghaddam, Gwen Clarke, Jason P. Acker Published: 2026 ABSTRACT AbstractBackgroundCryopreserved red cell concentrates (RCCs) glycerolized using a high glycerol (40%) method can be stored below −65°C for up to 30 years; however, units may be inadvertently warmed above −65°C due to freezer failures, human errors, or routine inventory management. RCCs known to have experienced transient warming events (TWEs) or storage temperatures ≥−65°C are frequently discarded.Study DesignSix groups of ABO/Rh matched RCCs were pooled‐and‐split (n = 9), glycerolized using the ACP 215 (Haemonetics), and frozen. Three types of TWEs were performed: (1) “Fast” TWEs (warming at room temperature [RT] to −25°C), (2) “Slow” TWEs (exposure to −25°C for 2 h), and (3) “Thaw” TWEs (single warming to RT). Units (n = 6) were exposed to 0 (control), 1, 10, or 30 “Slow” or “Fast” TWEs. RCCs were deglycerolized using the ACP 215, stored at 1–6°C, and tested at 0, 1, 7, and 14 days post‐deglycerolization for red blood cell (RBC) quality.ResultsNo significant differences among RCCs over hypothermic storage were found when exposed to “Fast” or “Slow” single or “Thaw” TWEs, although a significant decrease in the quality of RCCs was observed when units were subjected to 10 or more “Fast” or “Slow” TWEs.DiscussionOur data suggests the single TWEs examined in this study do not significantly impact the quality of RCCs post‐deglycerolization. Based on these observations, TWE exposure considerations should be made on a unit‐by‐unit basis, with additional policy support for unit retention. Abbreviations analysis of variance additive solution‐3 adenine triphosphate Canadian Blood Services citrate‐phosphate‐dextrose citrate‐phosphate‐dextrose‐adenine‐1 Canadian Standards Association hemoglobin high glycerol method hypothermic storage leukoreduced, red cell concentrates mean cell volume mean corpuscular fragility partial pressure of oxygen where hemoglobin is 50% saturated with oxygen polyvinylpyrrolidone red blood cell red cell concentrates saline‐adenine‐glucose‐mannitol temperature monitoring transient warmed transient warming events INTRODUCTION Transient warming events (TWEs) occur when transfusable products are exposed to temperatures higher than their recommended critical storage temperatures for an undefined amount of time. 1 Few studies have looked at the impacts of TWEs on high glycerol method (HGM) cryopreserved RBCs stored at ≤−65°C, but TWEs under these circumstances are likely damaging due to the potential effects of ice recrystallization. However, a recent study by Kurach et al. demonstrated the negligible impact of specific “real‐world” singular TWEs on cryopreserved red cell concentrates (RCCs) quality. 2 Equipment failures led to exposure of frozen RCC units to freezer temperatures ranging from (1) −49°C for 30 min to (2) −30°C for 48 h, with some units experiencing both TWEs. All units met the Canadian Standards Association (CSA) standard criteria for transfusion of leukoreduced, deglycerolized RBCs; however, in vitro quality assessments of RBC deformability, fragility, and oxygen affinity revealed potential differences in the group exposed to multiple TWEs. 2 Due to the great expense associated with the production, storage, tracking, shipping, and processing of cryopreserved RCCs, blood centers must protect inventory from risks to quality. Cryopreserved RCCs are typically collected for patients with rare (occurring in <1 in 1000 individuals) or very rare (occurring <1 in 3000) phenotypes 3 and may be one of the few transfusion options for these patient populations. At Canadian Blood Services (CBS), RBCs are collected, hypothermically stored (HS) for up to 21 days prior to HGM cryopreservation and stored at ≤−65°C for up to 10 years. 4 Common inventory management practices (verification, testing, shipping, and transport of units) may inadvertently lead to TWEs. The compounding impacts and downstream effects of TWE exposures has yet to be ascertained. Two potential drivers of cellular injury during exposure to multiple TWEs are the increase in average ice crystal size due to ice recrystallization and the repeated influx and efflux of water, which promotes osmotic injury. 5 , 6 , 7 Glycerol partially mitigates these mechanisms because of its high viscosity and low eutectic point. 4 Recent studies have demonstrated that adding compounds which actively inhibit ice recrystallization to 15% glycerol solutions can improve outcomes following TWE exposures. 8 Due to the concentration‐dependent nature of ice nucleation and growth kinetics in glycerol‐containing solutions, these findings cannot be directly translated. To elucidate the impacts of different types of TWEs which model those encountered during inventory management, the objective of this study was to explore the impacts of the number of TWEs (0, 1, 10, and 30) and the speed of TWEs (“Fast”/“Slow”) on cryopreserved RBC quality. Additionally, to investigate a potential mechanism of damage associated with successive TWEs, ice crystal growth patterns during both “Fast” and “Slow” TWEs following consecutive TWE exposures were determined. We hypothesized that singular TWEs would have minimal effect on RBC quality, whereas multiple TWEs would impart significant damage on RBCs that would be dependent on the rate of thawing and refreezing. MATERIALS AND METHODS RCC unit preparation and glycerolization Six groups of nine ABO/Rh matched citrate‐phosphate‐dextrose (CPD)‐SAGM (Macopharma, France) leukoreduced RCC units (n = 54) were obtained from the CBS Blood4Research program (Vancouver, British Columbia). RCC units were processed using the buffy coat method (RBC filtration or top/bottom method). Each group of ABO/Rh matched RCC units (n = 9) were pooled and split into 9 RCC units. Donor demographics within each pool can be found in Table S1. The RCC units were glycerolized 7 days post‐collection on the ACP 215 closed‐system processor (Haemonetics, Massachusetts, USA) placed in metal cassettes, frozen in a −80°C freezer, and stored at ≤−65°C. 4 Controlled transient warming event exposures From each pool, one of the 9 units was assigned to each experimental TWE group: temperature monitoring (TM), non‐transient warmed (control, No TWE), single “Fast” TWE, 10 “Fast” TWEs, 30 “Fast” TWEs, single “Slow” TWE, 10 “Slow” TWEs, 30 “Slow” TWEs, and “Thaw” TWE (Figure 1). The TM unit was used alongside the experimental units to monitor the internal temperature of the unit during TWEs (Table S2). The “Fast” TWE RCC units were exposed to room temperature 1, 10, or 30 times, until the TM unit reached −25°C, and were then returned to ≤−65°C. The “Slow” TWE RCC units were exposed to −20°C in a −20°C freezer 1, 10, or 30 times, for 2 h, and then returned to ≤−65°C. The “Thaw” TWE RCC unit was completely thawed at room temperature and then returned to ≤−65°C. Schematic representation of the pool and split approach utilized to assess the effects of transient warming events (TWEs) on the quality of glycerolized LR‐RCCs. Red cell concentrates (RCCs) were exposed to either no TWE, single or multiple TWE(s) at specific warming rates, deglycerolized and stored for in vitro testing for 2 weeks. MCV, mean cell volume; RBC, red blood cells; TM, temperature monitoring. Unit deglycerolization RCC units were thawed in a 37°C water bath and deglycerolized on the ACP 215 (Haemonetics, Massachusetts, USA). 4 The deglycerolized RBCs were resuspended in AS‐3 (Haemonetics, Massachusetts, USA) and HS for 14 days. RBC quality assessment RCC quality was assessed post‐deglycerolization at 0, 1, 7, and 14 days. A hematology analyzer (DxH 520, Beckman Coulter, Co. Clare, IE) was used to obtain hemoglobin (Hb) concentration and mean cell volume (MCV). Adenine triphosphate (ATP) concentration was measured using a previously described commercially available, enzyme‐based, spectrophotometric assay (DiaSysDiagnostic Systems GmbH, Holzheim, Germany). 9 p50 was assessed using an automated blood‐oxygen analyzer (Hemox‐Analyzer Model B, TCS Scientific, New Hope, USA). 10 , 11 , 12 Extracellular potassium measurements were performed by an accredited external laboratory (Alberta Precision Laboratories, Edmonton, Alberta, Canada) using an automated chemistry analyzer (Roche Cobas c503, Roche Diagnostics). RBC hemolysis, spun hematocrit, total and supernatant Hb were completed as previously described. 9 , 13 , 14 RBC morphology index scores were determined using fixed and stained blood smears. 9 , 15 Osmotic hemolysis was induced by incubating phosphate buffered saline washed RBCs in a hypotonic buffer. 16 , 17 Osmotic fragility was completed by exposing RBCs to a series of saline solutions and determining the mean corpuscular fragility (MCF, the concentration which produces 50% hemolysis). 9 Deformability and osmotic gradient ektacytometry RBC deformability and osmotic gradient ektacytometry (Osmoscan) was completed using a Laser Assisted Optical Rotational Red Cell Analyzer (LORRCA, RR Mechatronics Manufacturing B.V., The Netherlands). For deformability, RBCs were diluted 9:1000 in polyvinylpyrrolidone (PVP) isotonic solution and exposed to shear stresses ranging from 0.95 to 30.00 Pa at 37 ± 2°C. Deformability curves were transformed using an Eadie‐Hofstee linearization technique to obtain the EImax (maximum RBC elongation) and K EI (shear stress required to reach ½ of the EImax). 18 For Osmoscan, RBCs were diluted 1:25 in PVP isotonic solution and exposed to an osmotic gradient (50–500 mOsm/kg) at a constant shear stress (30 Pa, 37 ± 1°C). Osmoscan elongation index values (EImin, EImax, and EIhyper) were determined from the resulting curve. Osmoscan EImax and EImin were used in the calculation of delta (Δ) Elongation = (EImax − EImin). 19 Cryomicroscopy One microliter of RBCs in 40% glycerol (Glycerolyte 57, Fresenius Kabi AG, Germany) and SAGM (Macopharma, France) was distributed homogeneously between two 12 mm glass coverslips. Coverslips were transferred onto a Linkam FDCS196 convection cryomicroscope stage (Linkam Scientific Instruments, Tadworth, England) on a Nikon Eclipse 80i inverted microscope (Nikon, Tokyo, Japan). The temperature profiles (Table S3) were input into Linksys 32 Software (Version 2.4.3, Linkam Scientific, Tadworth, England). Samples were nucleated at −20°C using a copper probe pre‐cooled in liquid nitrogen and frozen at 0.7°C/min (to emulate the freezing rate of a full‐scale unit) to −80°C prior to being subject to TWE. Images were captured with a Nikon LU Plan 20× objective lens (Nikon, Tokyo, Japan) using a Hamamatsu charged‐coupled device camera (C4742‐80, Hamamatsu, Japan) interfaced with NIS Elements AR Software (Version 3.22.4, Nikon, Tokyo, Japan). Ice crystal area was calculated using the NIS Software manual area quantification tool. A Nikon LU Plan 50× objective was used for representative images (as in Figure 2) (Nikon, Tokyo, Japan). The effects of controlled transient warming events on ice recrystallization. Quantified crystal area and representative images after zero, one, two, and three (A) “Fast” or (B) “Slow” transient warming events (TWEs) (n = 3). Scale bar on representative images is 50 μm. Analysis of variance (ANOVA) tests were used to show statistical significance (****p < .0001, **p < .01, *p < .05). Statistical analysis Statistical analysis was performed using GraphPad Prism v.10.2.3 (GraphPad Software Inc., La Jolla, CA, USA) and SAS 9.4 Maintenance 8 (SAS Institute Inc., Cary, NC, USA). Mixed model analysis for repeated measurements with multiple comparisons was performed to estimate the effects of the number of TWEs (0, 1, 10, or 30) and post‐deglycerolization storage time (0, 1, 7, and 14 days) and their impact on RBC quality parameters. Comparisons between the No TWE and TWE groups were completed. A p‐value of p < .05 was considered statistically significant. RESULTS Multiple “Slow” TWEs had a more significant impact on the hematocrit, hemolysis, and MCV than multiple “Fast” TWEs The Canadian Standards Association (CSA) standards for Hb (≥35 g/unit in 90% of the units tested), hematocrit (≤80% in 90% of the units tested) and hemolysis (≤0.8% in 90% of the units tested) for deglycerolized RBCs were assessed across all experimental groups. 20 The CSA standard for Hb was met for the singular “Fast”/“Slow” TWE groups (Table 1). The CSA standard for hematocrit was met for all experimental groups (Table 1). Significant decreases in hematocrit were observed in 30 “Fast” TWEs (Days 0/1/7/14: p < .0001) and in all “Slow” multiple TWE groups (10 TWE: Day 0: p < .05; Day 1: p < .001; Day 7: p < .0001; Day 14: p < .01; 30 TWE: Days 0/1/7/14: p < .0001) (Table 2). The CSA standard for hemolysis was met in the No TWE, “Thaw,” and singular “Fast” TWE groups, while 83% of units in the singular “Slow” TWE group and 0% of units in the 10/30, “Fast”/“Slow” groups met the criteria (Table 1). In the “Fast” TWE groups, significant increases in hemolysis were observed following 10 TWEs (Day 1: p < .01) and 30 TWEs (Days 1/7/14: p < .0001) (Table 2). In the “Slow” TWE groups, significant increases in hemolysis were identified after 10 TWEs (Day 1: p < .001; Day 7: p < .0001; Day 14: p < .01) and 30 TWEs (Day 0: p < .05; Days 1/7/14: p < .0001) (Table 2). TABLE 1Percentage (%) of units in each experimental group (n = 6/group) that meet the CSA standards for hemoglobin, hematocrit, and hemolysis for deglycerolized red blood cells (RBCs).Experimental groupHemoglobin a HematocritRBC hemolysis b No TWE83100100Thaw and refreeze831001001 “Fast” TWE10010010010 “Fast” TWE33100030 “Fast” TWE010001 “Slow” TWE1001008310 “Slow” TWE0100030 “Slow” TWE01000Abbreviations: TWE, transient warming events. a Hemoglobin values are from 24 h post‐deglyerolization. b RBC hemolysis values are from expiry at 14 days post‐deglycerolization. Percentage (%) of units in each experimental group (n = 6/group) that meet the CSA standards for hemoglobin, hematocrit, and hemolysis for deglycerolized red blood cells (RBCs). Abbreviations: TWE, transient warming events. Hemoglobin values are from 24 h post‐deglyerolization. RBC hemolysis values are from expiry at 14 days post‐deglycerolization. TABLE 2Median and interquartile ranges for hematocrit, hemolysis, and mean cell volume (MCV) for red cell concentrates exposed to “Fast” or “Slow” transient warming events (TWEs). Multiple comparisons tests were used to show statistical significance relative to the No TWE group (****p < .0001, ***p < .001, **p < .01, and *p < .05).AssayDayNo TWEThaw and refreezeSingle “Fast” TWETen “Fast” TWEsThirty “Fast” TWEsSingle “Slow” TWETen “Slow” TWEsThirty “Slow” TWEsHematocrit (%)049 (48–52)50 (44–53)49 (48–52)47 (46–50)32**** (29‐34)51 (48–53)46* (44–47)27**** (26‐29)148 (47–52)48 (44–56)49 (46–50)44** (43–47)29**** (26‐30)49 (47–52)41*** (40–45)22**** (19‐27)746 (45–49)48 (41–51)44 (42–46)44 (42–46)28**** (24‐30)48 (45–51)41**** (40–42)19**** (16‐22)1447 (45–50)48 (40–49)48 (46–49)45 (43–47)28**** (23‐33)51 (46–56)41*** (40–44)20**** (16‐22)Hemolysis (%)00.19 (0.16–0.24)0.21 (0.17–0.31)0.23 (0.16–0.28)0.26 (0.22–0.28)1.6 (1.5–3.7)0.24 (0.20–0.27)0.41 (0.27–0.44)5.2* (3.6–8.9)10.17 (0.15–0.21)0.23 (0.21–0.40)0.21 (0.17–0.31)0.32** (0.28–0.41)2.2**** (2.0–5.7)0.18 (0.12–0.26)0.47*** (0.35–0.65)7.8**** (4.4–13)70.31 (0.28–0.36)0.33 (0.27–0.46)0.39 (0.32–0.42)0.81 (0.74–0.87)4.2**** (3.6–9.3)0.41 (0.37–0.50)1.1**** (0.95–1.4)12**** (8.4–20)140.39 (0.35–0.42)0.47 (0.43–0.55)0.63 (0.52–0.68)1.8 (1.5–1.9)8.5**** (5.2–17)0.60 (0.49–0.76)2.4** (1.9–2.8)17**** (12‐31)MCV (fL)096 (95–99)96 (94–98)95**** (94–97)99**** (97–100)103**** (98–104)96 (94–98)102 (100–105)104**** (100–106)195 (94–98)95 (94–96)94**** (93–96)94* (93–95)96 (94–97)94**** (93–96)95**** (93–98)100* (97–104)794 (93–95)93 (92–95)92**** (91–94)95 (92–95)99**** (97–102)94 (92–95)97**** (94–99)102**** (100–108)1495 (94–97)94 (92–96)93**** (92–95)98 (96–99)105**** (102–108)94 (93–97)102**** (98–104)109**** (107–114) Median and interquartile ranges for hematocrit, hemolysis, and mean cell volume (MCV) for red cell concentrates exposed to “Fast” or “Slow” transient warming events (TWEs). Multiple comparisons tests were used to show statistical significance relative to the No TWE group (****p < .0001, ***p < .001, **p < .01, and *p < .05). The number and TWE rate had a significant impact on the MCV (p < .0001) (Table 2). One TWE produced significant decreases in the MCV (“Fast”: Days 0/1/7/14: p < .0001; “Slow”: Day 1: p < .0001) (Table 1). Thirty “Fast” and multiple “Slow” TWEs produced significant increases in the MCV (30 “Fast”: Days 0/7/14: p < .0001; 10 “Slow” TWEs: Days 1/7/14: p < .0001; 30 “Slow” TWEs: Days 0/7/14: p < .0001; Day 1: p < .05) (Table 2). Irrespective of TWE rate, changes in ATP, p50, extracellular potassium, RBC morphology, and osmotic hemolysis were more pronounced following multiple TWEs Significant decreases in ATP concentrations were identified following multiple “Fast” TWEs (10 TWEs: Day 7/14: p < .0001; 30 TWEs: Day 1/7/14: p < .0001) and “Slow” TWEs (10 TWEs: Day 14: p < .0001; 30 TWEs: Day 7/14: p < .0001) (Table 3). p50 was the highest following 10 “Slow” TWEs (Day 7: p < .01) (Table 3). Thirty “Slow” TWEs resulted in lower p50 values (Days 1/14: p < .0001) (Table 3). Potassium was elevated for a single “Fast” TWE at Day 14 (p < .05) and for a single “Slow” TWE beyond Day 1 (Day 1/7/14: p < .0001) (Table 3). Following multiple “Fast” TWEs, significant increases in potassium were found on Day 1 (10 TWEs: p < .0001; 30 TWEs: p < .01), and Day 14 (30 “Fast” TWEs: p < .05) (Table 3). Following multiple “Slow” TWEs, significant increases in potassium were found at all testing time points (Day 0: 10 TWEs: p < .0001; 30 TWEs: p < .001; Day 1: 10 TWEs: p < .0001; 30 TWEs p < .001; Day 7: 10 TWEs/30 TWEs: p < .001: Day 14: 30 TWEs: p < .0001) (Table 3). TABLE 3Median and interquartile ranges for ATP, p50, potassium, morphology and osmotic hemolysis measurements for red cell concentrates exposed to “Fast” or “Slow” transient warming events (TWEs). Multiple comparisons tests were used to show statistical significance relative to the No TWE group (****p < .0001, ***p < .001, **p < .01, *p < .05).AssayDayNo TWEThaw and refreezeSingle “Fast” TWETen “Fast” TWEsThirty “Fast” TWEsSingle “Slow” TWETen “Slow” TWEsThirty “Slow” TWEsATP (μmol/g Hb)14.6 (4.2–5.4)5.1 (4.8–5.4)4.3 (4.1–4.5)4.1 (3.8–4.4)3.5**** (2.8–3.6)4.3 (4.0–5.2)3.8* (3.4–4.2)3.8 (3.1–4.1)72.6 (2.3–3.1)2.6 (2.4–3.0)2.3 (2.2–2.5)1.5**** (1.4–2.1)0.8**** (0.6–1.4)2.6 (2.1–2.9)1.3 (1.1–1.6)0.1**** (−0.4–0.7)141.8 (1.7–2.0)1.3 (1.1–1.9)1.5 (1.3–1.9)0.9**** (0.6–1.1)0.3**** (−0.3–0.7)1.3 (1.1–1.9)0.4**** (0.0–0.9)0.4**** (0.2–0.8)p50 (mmHg)121.1 (20.8–21.4)19.0 (18.7–19.8)19.3**** (19.2–19.9)19.1**** (18.6–19.9)19.6*** (18.2–20.2)20.9 (19.7–21.2)20.9 (20.5–22.3)19.2**** (18.7–19.8)718.5 (18.1–19.5)20.5 (19.3–20.9)19.4 (18.9–19.7)19.2 (18.9–19.5)18.3 (17.6–19.8)19.8 (18.7–20.0)21.9** (20.7–23.5)18.9 (18.3–19.1)1419.1 (18.9–19.4)20.3 (18.6–20.6)18.7 (17.2–18.8)17.8**** (17.3–18.2)18.2**** (17.7–18.5)17.7 (17.4–18.8)19.8 (18.9–20.2)17.2**** (16.6–17.6)Potassium (mmol/g Hb)00.013 (0.012–0.013)0.015 (0.014–0.018)0.023 (0.016–0.028)0.046 (0.040–0.055)0.052 (0.036–0.061)0.022 (0.014–0.033)0.059**** (0.052–0.069)0.050*** (0.044–0.062)10.054 (0.044–0.060)0.073 (0.059–0.082)0.089 (0.072–0.106)0.155**** (0.092–0.175)0.129*** (0.111–0.134)0.106**** (0.102–0.136)0.158**** (0.142–0.160)0.09*** (0.083–0.104)70.148 (0.134–0.160)0.177* (0.167–0.186)0.178 (0.083–0.211)0.211 (0.089–0.230)0.152 (0.122–0.166)0.230**** (0.207–0.247)0.192**** (0.180–0.210)0.111**** (0.084–0.120)140.193* (0.177–0.205)0.221 (0.212–0.230)0.245* (0.233–0.259)0.239 (0.221–0.248)0.147* (0.109–0.174)0.246**** (0.223–0.262)0.197 (0.186–0.209)0.109**** (0.089–0.120)Morphology index173 (69–81)69 (63–74)77 (75–84)72 (69–76)82 (71–88)67 (62–75)68 (65–77)66 (56–71)754 (51–63)63 (58–65)63**** (59–69)57** (50–63)35**** (32‐41)59 (58–60)48 (41–59)24**** (20‐32)1450 (43–56)48 (44–52)47 (44–55)45 (43–47)15**** (15‐22)46 (42–48)30**** (29‐33)13**** (11‐19)Osmotic hemolysis (%)133 (31–36)45 (41–47)38**** (37‐40)41**** (37–43)53**** (47–57)37**** (37–43)51**** (48–57)53**** (49–58)733 (32–37)35 (33–39)35 (31–39)45**** (41–49)59**** (53–69)33 (28–39)51**** (46–58)69**** (65–74)1434 (33–40)38 (37–42)42* (38–43)55**** (53–64)69**** (61–78)43**** (40–53)60**** (57–67)80**** (78–83)Abbreviation: Hb, hemoglobin. Median and interquartile ranges for ATP, p50, potassium, morphology and osmotic hemolysis measurements for red cell concentrates exposed to “Fast” or “Slow” transient warming events (TWEs). Multiple comparisons tests were used to show statistical significance relative to the No TWE group (****p < .0001, ***p < .001, **p < .01, *p < .05). Abbreviation: Hb, hemoglobin. Thirty TWEs (“Fast”/“Slow”) led to significantly lower morphology index scores beyond Day 7 (p < .0001) (Table 3). Ten “Slow” TWEs had a significant impact on RBC morphology at Day 14 (p < .0001) (Table 3). Significant differences in osmotic hemolysis were found on every testing day in units exposed to multiple TWEs (p < .0001) (Table 3). Osmotic fragility and RBC rheological parameters were significantly impaired following 10 “Slow” TWEs and 30 “Fast” TWEs The MCF was significantly increased following multiple “Slow” TWEs (10 TWEs: Days 7/14: p < .0001; 30 TWEs: Days 1/7/14: p < .0001) (Table S4). Significant decreases in the EImax were seen following Day 14 after 10 “Slow” (p < .05) and 30 “Slow”/“Fast” TWEs (p < .0001) (Table S4). One “Slow” TWE led to an increase in the EImax at Days 1 (p < .01) and 7 (p < .001) (Table S4). Significant increases in RBC rigidity were identified at Day 1 following a single “Slow” TWE (p < .05) and 30 “Fast” TWEs (p < .01) (Table S4). A transient decrease in K EI was identified at Day 7 following 10 “Slow” TWEs (p < .001) (Table S4). Significant decreases in the ΔElongation were identified at Day 7 in 10 “Fast”/“Slow” TWE groups (p < .0001) (Table 4). Thirty TWE exposures lead to a significant decrease in ΔElongation at all points (“Fast” TWE: Days 1/7/14: p < .0001; “Slow” TWE: Days 1/7: p < .0001; Day 14: p < .01) (Table 4). The osmotic gradient EImax was significantly lower following multiple TWEs (10 “Slow”: Day 14: p < .0001; 30 “Slow”: Days 1/7/14: p < .0001; 30 “Fast”: Days 7/14: p < .0001) (Table 4). Significant increases in EImin were identified following multiple “Fast”/“Slow” TWEs (10 “Fast”/“Slow”: Day 7: p < .0001; 30 “Fast”: Day 1/7/14: p < .0001; 30 “Slow”: Day 1/7: p < .0001) (Table 4). Significant decreases in EIhyper were identified following 30 “Fast” TWEs after Day 7 (Days 7/14: p < .0001) (Table 4). Significant decreases in EIhyper were seen following multiple “Slow” TWEs (10 TWEs: Days 7/14: p < .0001; 30 TWEs: Day 1: p < .01; Days 7/14: p < .0001) (Table 4). TABLE 4Median and interquartile ranges for rheological measurements for red cell concentrates exposed to “Fast” or “Slow” transient warming events (TWEs). Multiple comparisons tests were used to show statistical significance relative to the No TWE group (****p < .0001, **p < .01, *p < .05).AssayDayNo TWEThaw and refreezeSingle “Fast” TWETen “Fast” TWEsThirty “Fast” TWEsSingle “Slow” TWETen “Slow” TWEsThirty “Slow” TWEsΔ Elongation10.41 (0.39–0.42)0.38 (0.36–0.40)0.41 (0.39–0.42)0.38 (0.36–0.40)0.37**** (0.36–0.38)0.39 (0.36–0.40)0.38 (0.35–0.38)0.37**** (0.36–0.38)70.41 (0.40–0.42)0.39 (0.38–0.40)0.39 (0.38–0.41)0.37**** (0.36–0.39)0.34**** (0.34–0.35)0.39 (0.38–0.40)0.35**** (0.34–0.36)0.34**** (0.32–0.36)140.39 (0.37–0.41)0.39 (0.38–0.39)0.38 (0.34–0.39)0.35 (0.34–0.36)0.29**** (0.28–0.31)0.36 (0.34–0.37)0.34 (0.32–0.35)0.30** (0.30–0.33)EImax 10.601 (0.600–0.601)0.602 (0.602–0.605)0.603 (0.590–0.605)0.604 (0.598–0.605)0.598 (0.581–0.603)0.601 (0.597–0.603)0.598 (0.597–0.600)0.599**** (0.598–0.598)70.600 (0.598–0.600)0.598 (0.597–0.599)0.598 (0.579–0.601)0.597 (0.581–0.600)0.592**** (0.590–0.594)0.599 (0.598–0.599)0.593 (0.592–0.594)0.587**** (0.585–0.59)140.592 (0.590–0.593)0.594 (0.592–0.594)0.588 (0.566–0.593)0.588 (0.581–0.590)0.557**** (0.541–0.567)0.592 (0.586–0.594)0.582**** (0.581–0.586)0.566**** (0.563–0.574)EImin 10.190 (0.283–0.206)0.22 (0.202–0.244)0.189 (0.186–0.202)0.218 (0.205–0.235)0.228**** (0.219–0.230)0.211 (0.203–0.247)0.223 (0.215–0.251)0.227**** (0.221–0.236)70.188 (0.183–0.198)0.203 (0.196–0.223)0.20* (0.193–0.212)0.221**** (0.21–0.227)0.251**** (0.243–0.256)0.208 (0.20–0.217)0.241**** (0.238–0.258)0.248**** (0.233–0.267)140.197 (0.187–0.225)0.206 (0.203–0.217)0.21 (0.201–0.222)0.236 (0.227–0.241)0.258**** (0.252–0.270)0.227 (0.219–0.245)0.249 (0.237–0.264)0.262 (0.247–0.269)EIhyper 10.301 (0.300–0.301)0.301 (0.301–0.303)0.302 (0.295–0.303)0.302 (0.299–0.302)0.299 (0.291–0.302)0.301 (0.299–0.302)0.299 (0.299–0.301)0.300** (0.299–0.300)70.300 (0.299–0.300)0.299 (0.299–0.300)0.299 (0.290–0.300)0.299 (0.291–0.300)0.296**** (0.296–0.297)0.300 (0.299–0.300)0.297**** (0.297–0.298)0.294**** (0.293–0.295)140.296 (0.295–0.297)0.297 (0.296–0.297)0.294 (0.283–0.296)0.294 (0.291–0.295)0.279**** (0.271–0.284)0.296 (0.293–0.297)0.291**** (0.291–0.293)0.283**** (0.282–0.287) Median and interquartile ranges for rheological measurements for red cell concentrates exposed to “Fast” or “Slow” transient warming events (TWEs). Multiple comparisons tests were used to show statistical significance relative to the No TWE group (****p < .0001, **p < .01, *p < .05). Ice crystal size increased following each TWE regardless of TWE rate “Slow” TWEs had a significant effect on ice crystal area following TWEs resulting in a larger crystal size relative to the “Fast” TWE (p < .0001) (Figure 2A,B). Irrespective of TWE type, a significant increase in the ice crystal area was identified after each TWE (p < .0001 in all cases, except 1–2 “Fast” TWE: p < .0032 and 2–3 “Slow” TWE: p = .0491) (Figure 2A,B). Looking at the change in crystal size between each TWE, the difference was greatest following the first TWE, irrespective of TWE type (Δ0‐1 “Slow” TWE: 462 μm2; Δ0‐1 “Fast” TWE: 255.7 μm2; Δ1‐2 “Slow” TWE: 338 μm2; Δ1‐2 “Fast” TWE: 120.3 μm2; Δ2‐3 “Slow” TWE: 144.4 μm2; Δ2‐3 “Fast” TWE: 313 μm2). DISCUSSION This study investigated the impact of controlled TWEs on RBCs frozen in 40% glycerol and stored at ≤−65°C. “Fast” TWEs (warming rates = 2.5–2.8°C/min) model situations whereby cryopreserved RCCs may be exposed to room temperature while being checked, shipped, or packed during inventory management. “Slow” TWEs (warming rates = 0.65–0.68°C/min) model situations like freezer failures / incorrect freezer placements. One, 10, or 30 TWEs were chosen based on a worst‐case exposure to a TWE once per year over the 10‐year or potentially 30‐year storage lifetime. Overall, the outcomes reported align with previous studies which support that not all TWEs are detrimental to the in vitro quality of cryopreserved RCCs. 2 , 21 Previously, we reported on the negligible impact of singular TWEs caused by well‐described freezer failures. 2 RCCs continued to meet CSA criteria and had similar in vitro quality to control RCCs, whereas in the current study this did not hold. 8 Compared to the present study, where core unit temperatures were recorded, only freezer temperatures were previously reported. A much larger temperature differential was experienced in the present study, leading to a higher potential for RBC injury. The CSA standards serve as a clinical benchmark that products must meet to be deemed suitable for transfusion. 20 These criteria were not consistently met for all experimental groups, including the control group. Similar inconsistencies have been found throughout the literature; methodological, experimental variables, donor factors, and small sample size could be implicated. 2 , 22 , 23 , 24 , 25 , 26 Hemolysis and potassium increase due to injury incurred during HS. 9 , 27 Valeri et al., who simulated a single freezer failure condition, found mean hemolysis was higher (5.3 ± 1.5%) and potassium results were lower than anticipated, with no impact on ATP, p50, or 24‐h post‐transfusion survival. 21 Our hemolysis results after 10 “Slow” TWEs were similar to those found in Valeri's study; however, potassium concentrations remained high. Differences between the two studies which may have contributed to the faster accumulation of damage due to a single TWE include: whole blood centrifugation versus buffy coat processing, storage pre‐glycerolization in citrate‐phosphate‐dextrose‐adenine‐1 (CPDA‐1) versus CPD/SAGM, open‐system versus closed‐system glycerolization, and post‐deglycerolization storage in sodium chloride glucose wash solution versus AS‐3. 9 , 21 , 27 Significant declines in ATP concentration and p50 following multiple TWEs were similar to quality degradation in expired HS RBC units. 9 , 13 The tight interplay between these two variables within RBCs is evident in the literature and was altered similarly by TWEs. 28 RBC morphological declines were more severe than those recorded in HS studies. 29 These parameters changed significantly with the increasing number of TWEs; however, they were less sensitive to TWE rate. Mean corpuscular volume, hemolysis, osmotic fragility and rheological parameters appear more sensitive at identifying changes caused by TWE rate differences. Rheological parameters are influenced by the membrane and cytoskeleton composition, surface area: volume ratio, and cytoplasmic viscosity. 30 , 31 , 32 Ten “Slow” TWEs lead to more dehydrated, osmotically fragile RBCs, with a decreased ability to elongate, whereas the same outcomes were not seen until 30 “Fast” TWEs. Similar MCF values are seen in patients with hereditary spherocytosis and expired HS RCCs. 9 , 33 A potential mechanism of damage may be due to the repeated osmotic exchange of water with consecutive TWEs. As temperatures rise above the eutectic point during a TWE, the unfrozen fraction expands, leading to water influx and solute dilution. During refreezing, the opposite occurs, causing osmotic efflux. These fluctuations force the cell to adjust its volume at a time when low temperatures and high solute concentrations in the unfrozen fraction increase rigidity and reduce elasticity of the plasma membrane needed for effective volume regulation, potentially leading to membrane collapse and the formation of sub‐lethal membrane pores. 34 , 35 , 36 , 37 In this study, the relevance of this injury mechanism was not assessed, nor can it be inferred from pre‐deglycerolization hemolysis or supernatant potassium levels, as these parameters were not measured. However, viewing the effects of a TWE as causing sub‐lethal membrane injury helps contextualize the results observed with multiple TWEs of different rates. An increase in average crystal size due to ice recrystallization is believed to be the primary source of injury during TWEs. 8 , 38 Although the exact mechanism—whether mechanical or osmotic stress—remains unclear, numerous cell systems have shown close correlations between the extent of recrystallization and post‐thaw survival. 8 , 39 , 40 Cryomicroscopy images show recrystallization is most extensive after one TWE in 40% glycerol; however, recrystallization is more pronounced following “Slow” TWEs, keeping with the Gibbs–Thomson thermodynamic principle. 41 Exposure to multiple TWEs proved significantly more damaging, and given the likely plateau in ice crystal growth early in the sequence of TWEs, it is unlikely that this damage can be attributed solely to recrystallization. It is important to note that the solutions used for crystal growth evaluations did not contain cells, meaning their ionic composition and glycerol concentration likely differed from those found in the supernatant and cytoplasm of cryopreserved RBCs. As a result, while these findings provide valuable insight into recrystallization as it relates to the speed and frequency of TWEs, they are unlikely to fully replicate real‐world conditions. In agreement with previous studies, this study shows the negligible impact of single TWEs. 2 , 21 We also provided additional evidence that multiple TWEs beyond 10 “Slow” and 30 “Fast” TWEs have significant impacts on RBC in vitro quality. Multiple “Slow” TWEs have a higher propensity to cause damage in a fewer number of TWEs than “Fast” TWEs. Limitations of this study include the inability to pinpoint the exact number of TWEs which cause critical damage accumulation. Previous studies have suggested up to five TWEs of different warming rates and length of exposures will have minimal impacts on RBC, 8 but these conclusions cannot be fully translated to a full‐size cryopreserved RCC. These results provide evidence that all TWEs are unique and highlight the need to accurately track and record exposure events to fully understand their potential impacts on frozen inventory. This study provides prospective data that can inform evidence‐based decision‐making policies on large‐scale inventory management and aid those who are assessing whether a cryopreserved RCC should be saved following TWE exposure. AUTHOR CONTRIBUTIONS JK was responsible for data collection, data analysis, and data interpretation, in addition to writing and reviewing the manuscript. MB‐C was involved with data analysis and data interpretation in addition to manuscript writing and reviewing. CO assisted with study design, data collection, and manuscript preparation. TRT was integral to the conception and design of the study and reviewed the manuscript. ATR collected and analyzed data, while NW played a role in the data collection and manuscript writing process. BE‐M was involved with the analysis and interpretation of the data collected in this study. GC and JPA were integral to study design, data interpretation, securing project funding, and manuscript reviewing. CONFLICT OF INTEREST STATEMENT The authors have disclosed no conflicts of interest. Supporting information Data S1. Supporting Information.