Novel anticoagulant-preservative solution maintained the hemostatic function of cold stored whole blood for 56 days

Abstract

Background:

Whole blood (WB) is an efficient product for field medical resuscitation because of its unitary composition, tolerance for storage on ice and in field refrigerators, and simplicity of use. We measured quality parameters of a novel 8-week WB storage system.

Study Design and Methods:

Here, 500 mL of WB from healthy donors was collected in 70 mL of CPDA-1, leukoreduced with a platelet-sparing filter, pooled into ABO-compatible two-unit pools, and split into matched pairs of equal volume designated as Test or Control units. Test units received an additional 50 mL of a novel WB preservative solution (APEX units, Hemerus Medical, St Paul, MN). A total of 15 paired WB units were evaluated at Day 0 (D0) and periodically up to Day 56 (D56) of storage at 1–6°C across two centers. Quality testing included cellularity, ATP concentrations, hemolysis, blood gases, metabolites, coagulation factor levels, thromboelastography (TEG), and bacterial culture.

Results:

At D56, APEX units displayed higher RBC ATP concentration (3.14 vs. 2.18 μmol/gHb, p = 0.001), pH (6.53 vs. 6.50, p = 0.01), and higher bicarbonate reserve (8 vs. 5.4, p < 0.0001). D56 APEX units had greater platelet contribution to TEG clot strength (p < 0.01) and better preservation of red cell ATP (p < 0.001). Activities of fibrinogen, factor VIII, factor V, and protein S activity in APEX units remained within the reference levels on D56. No bacterial contamination was detected at the end of storage.

Discussion:

These findings suggest that APEX preserves RBCs effectively and maintains platelet and plasma coagulation functions for up to 56 days.

1 ∣. INTRODUCTION

Hemorrhage remains a leading cause of preventable death in both military and civilian trauma settings, necessitating efficient blood product availability for resuscitation. Whole blood (WB) is the preferred treatment for hemorrhagic resuscitation due to its ability to restore blood volume, oxygen delivery, and coagulation factors in a single product while reducing donor exposure. However, standard WB storage with CPDA-1 is limited to 35 days at 1–6°C, restricting its utility for prolonged operational readiness. Cold-stored WB (CSWB) undergoes storage lesions that impact RBC viability, platelet function, and coagulation factor stability, which can limit hemostatic efficacy over time.^1-7 Mitigating these changes could extend WB shelf life and improve logistics for trauma care, particularly in austere environments.

AS-7 (SOLX, Hemerus Medical LLC, St. Paul, MN) is an FDA-approved RBC additive solution (AS) that was formulated to improve the metabolism of RBCs with the addition of bicarbonate and phosphate at high pH, which increases buffering capacity. AS-7 has been shown to reduce the RBC storage lesions, including reducing hemolysis, decreasing microvesicle formation, improving biochemical parameters, and increasing end of storage in vivo RBC recovery out to 56 days compared with AS-1-stored RBCs. For this study, the expertise gained in the creation of the AS-7 (SOLX) AS was applied to create a new WB anticoagulant-preservative solution (APEX, Hemerus Medical LLC, St. Paul, MN).

In this study, we compared the viability and function of the therapeutically relevant cellular and plasma components of long-term stored, leukoreduced APEX WB prepared with leukoreduced CPDA-1 WB, both utilizing platelet-sparing filters. Our data demonstrate that APEX anticoagulant-preservative solution supports the feasibility of an extended shelf life for WB with acceptable RBC quality, platelet function, and plasma factor stability over conventional storage in CPDA-1 anticoagulant.

2 ∣. STUDY DESIGN AND METHODS

2.1 ∣. Human subject protection

This research was approved by the University of Cincinnati Institutional Review Board, Research Regulatory Compliance Division of the US Army Institute of Surgical Research, the Human Use Protections Committee and Commander of the US Army Research and Development Command, and the US Army Surgeon General.

2.2 ∣. Collection and processing

Thirty units of CPDA-1 WB were successfully collected, pooled in ABO identical groups of 2, the pools split into 15 sets of paired units, and the APEX preservative solution added to one member of each pair. Seven pairs were retained at the Hoxworth Blood Center (HBC) for subsequent testing, and 8 pairs were sent to the US Army Institute of Surgical Research in San Antonio (ISR), taking advantage of slightly different testing methods available at each site. Units were sampled on days 0, 7, 14, 21, 35, 42, and 56, with the removal of about 50 mL at each sampling point. The resulting volume decreased from about 590 to 240 mL for the APEX bags and 540 mL to 190 mL for the CPDA-1 bags over the duration of storage. All collected units were successfully tested and reported. In some cases, WB was centrifuged at 200× g for 10 min to isolate platelet-rich plasma (PRP) for testing. Additional centrifugation at 2000× g was performed for plasma and red cell sample isolation, respectively.

2.3 ∣. Hematology analysis

Complete blood counts were obtained using an ADVIA 2120i Hematology System (Siemens, Saint Paul, MN) or a Sysmex Hematology Analyzer XE-2100D with a blood center module (Sysmex, Lincolnshire, IL). Red blood cell count, total WB hemoglobin, platelet count, mean corpuscular volume (MCV), and mean platelet volume (MPV) were reported. Free plasma hemoglobin was determined using a HemoCue AB Plasma/Low Hb analyzer (Daraher, Washington, District of Columbia). Ten μL of PPP was added to a HemoCue 201 microcuvette and analyzed for free hemoglobin level (in mg/dL). Spun hematocrit levels were calculated via the Packed Cell Volume method.^8,9 A small volume of WB was drawn into a microcapillary tube and centrifuged (Haematokrit 200; Hettich, Beverly, MA) for 5 min according to local site procedures. A digital (Iris CritSpin; Eindhoven, Netherlands) or manual hematocrit reader was then utilized to assess the packed cell volume and produce a microhematocrit percentage.

2.4 ∣. Metabolic indices and biochemistries

WB and PRP biochemistry values were measured with an i-STAT point-of-care blood analyzer (Abbott Laboratories, Chicago, IL) using CG4+ and CG8+ cartridges or an ABL835 FLEX chemical analyzer (Rockville, MD) as per the manufacturer's instructions. Glucose, Na⁺, K⁺, Lactate, pH, pCO₂, pO₂, HCO₃ values on WB samples and pH values on PRP samples were reported at each testing time point. An aliquot (1 mL) of PRP was lysed using 9% Triton-X100 lysis solution (Sigma Aldrich; St. Louis, MO) for Total Platelet lactate dehydrogenase (LDH) analysis. LDH values were obtained using a Dimension EXL 200 Integrated Chemistry System (Siemens Medical Solutions USA, Malvern, PA) or similar. The LDH ratio was calculated by obtaining LDH values of PRP samples and PPP samples.

2.5 ∣. Whole blood ATP

RBC ATPs were determined spectrophotometrically, as previously described.¹⁰

2.6 ∣. Coagulation function

Thromboelastography (TEG) can assess clotting capabilities in WB, platelet-rich plasma (PRP), and plateletpoor plasma (PPP), allowing for the evaluation of coagulation factor activity separately from platelet function. We analyzed the coagulation function of WB and PPP samples monitored at both sites using thromboelastography (TEG 5000; Haemonetics, Boston, MA) with kaolin to initiate intrinsic pathway coagulation. Briefly, 1 mL of blood sample (WB or PPP) was mixed with 40 μL of kaolin solution and added to a TEG cup containing 20 μL of 0.2 M CaCl₂. R-time (in min), maximum amplitude (MA; in mm), α-angle (in°), K-time (in min), and lysis at 30 minutes (LI-30; in %) were all recorded.

2.7 ∣. Coagulation factor analysis

At the ISR site, previously frozen plasma samples were assessed for coagulation viability using a STA-R Max (Diagnostica Stago, Parsippany, NJ) coagulation analyzer. Fibrinogen, Factor V, and Factor VIII levels were recorded, along with Protein S activity.

2.8 ∣. RBC morphology

Isolated RBCs (20 μL) were fixed using 1 mL of 0.5% Glutaraldehyde solution (pH 6.0). Fixed RBC samples were stored refrigerated until RBC morphology analysis. RBC morphology was assessed by light microscopy using glutaraldehyde according to the method of Usry and colleagues.¹¹

2.9 ∣. Bacterial testing

At the end of storage (day 56), WB samples were evaluated for the presence of bacteria and fungi. WB samples (10-mL) were aseptically sampled into anaerobic and aerobic blood culture bottles (BacT/ALERT, Becton Dickinson) and sent to a third-party testing lab (LabCorp or Quest Diagnostics).

2.10 ∣. Statistics

Data were managed in customized spreadsheets (Excel, Microsoft, Redmond, WA) and analyzed using statistical software (GraphPad Prism, La Jolla, CA). Paired Student's t-tests were used to determine significant differences between CPDA-1 and APEX units at each storage time point (p < 0.05).

3 ∣. RESULTS

3.1 ∣. Effects of volume differences between cells stored in CPDA-1 and APEX

The concentration measures between APEX and CPDA-1 units are affected by dilution inherent to the anticoagulant–preservative systems. Control units contain 70 mL CPDA-1, whereas APEX units receive an additional 50 mL of the novel WB preservative solution. Based on a WB collection volume of 500 mL and an average hematocrit of 44%, the expected reduction in concentration is 8.1% for WB and 12.5% for the supernatant measures. This dilution factor should be considered when comparing concentration-dependent parameters between the two groups.

Values of measurements such as spun hematocrit, supernatant K⁺, and free hemoglobin were consistently lower in APEX units throughout storage due to the additional 50 mL of preservative solution (Figure 1). Specifically, spun hematocrit was reduced by about 8.5%, reflecting dilution across the entire unit, while free hemoglobin concentration and supernatant potassium were reduced by about 13.5%, as the extra fluid represents a larger fraction of the supernatant volume.

FIGURE 1.

Effects of volume differences between cells stored in CPDA-1 and APEX. (A) Spun hematocrit (HCT). (B) Supernatant potassium (K⁺). (C) Plasma hemoglobin levels. Data visualized as mean ± SD for n = 15 units. ^†p < 0.05, ^‡p < 0.01, ^§p < 0.001.

3.2 ∣. General measures of metabolism

Similar to the trend observed with spun hematocrit, supernatant potassium, and free hemoglobin, glucose concentration was significantly lower in APEX units compared with CPDA-1 on day 0 (p < 0.0001)—a direct result of the increased dilution of plasma with the preservative solution. Glucose concentrations in the supernatant fell during storage, with a concurrent rise in lactate (Figure 2A,B). Lactate levels at days 42 through 56 were significantly higher in APEX-stored WB when compared with their contemporaneous controls (Figure 2B). In the context of relative alkalinization (Figure 2C), these results support the existence of increased anaerobic glycolytic flux in APEX units.

FIGURE 2.

Glucose levels, lactate production, and pH decline during storage in CPDA-1 and APEX. (A) Glucose, (B) lactate, and (C) pH were measured in stored CPDA-1 and APEX units for 56 days. Data visualized as mean ± SD for n = 15 units. ^‡p < 0.05, ^‡p < 0.01, ^§p < 0.001.

3.3 ∣. Red blood cell analytes

We chose to evaluate RBC ATP concentrations in stored WB units, as improved RBC ATP values have been shown to correlate with improved in vivo recovery.¹² Extended storage in the APEX system led to comparatively higher RBC ATP concentrations after week three of storage, ending with concentrations of 3.14 ± 1.04 μM/g Hb (~60% of day 0 values; Figure 3A). In CPDA-1, the RBC ATP concentrations fell to 2.18 ± 0.80 μM/g Hb, approximately 40% of starting values. At day 56, RBC ATP levels were 64.2 ± 20.2% of day 0 values (Figure 3A; p < 0.001 vs. CPDA-1 at day 56). Morphology scores were comparable between CPDA-1 and APEX storage systems at individual sites (Figure 3B); however, as expected with an observer-dependent test, the values obtained at the two testing sites were disparate, particularly at later time points. Hemolysis was progressive under both conditions of storage, at both sites, and well within the allowed 1% under all conditions tested (Figure 3C).

FIGURE 3.

Measures of RBC ATP concentration, morphology index, and hemolysis fraction during 56 days of storage in CPDA-1 and APEX. (A) ATP concentrations of stored RBCs. (B) Morphology index. (C) Hemolysis (%). Data for (A) and (B) are visualized as mean ± SD for n = 15 units. (C) is a Box and Whisker Plot; one paired outlier with high, consistent hemolysis in both Test and Control arms throughout storage was excluded as non-evaluable. ^†p < 0.05, ^‡p < 0.01, ^§p < 0.001.

3.4 ∣. Platelet analytes

Platelet number (Figure 4A) was reduced from the normal 250 × 10³/μL after “platelet sparing” leukoreduction. At the HBC site, platelet count remained around 130 × 10³/μL throughout the entire 8-week storage duration, whereas at the ISR site, platelet count plummeted to 21.9 ± 7.1 × 10³/μL in APEX units by day 56 of storage. The variability observed in platelet counts across time points resulted from instrument-specific limitations in measuring platelets over extended storage periods. Automated hematology analyzers have limited validation for accurate platelet enumeration beyond routine storage durations due to changes in platelet morphology and activation states. Despite differences in platelet concentration, the platelets' ability to pull the fibrin mesh tight, as measured by TEG Maximum Amplitude (TEG MA), was consistent across sites (Figure 4B). Furthermore, better preservation of TEG MA function in APEX samples was observed compared with CPDA-1 controls.

FIGURE 4.

Measures of platelet concentration and function during 8 weeks of WB storage in CPDA-1 or APEX. (A) Platelet count. (B) Maximum Amplitude. (C) Platelet clotting activity. Data visualized as mean ± SD for n = 15 units. ^†p < 0.05, ^‡p < 0.01, ^§p < 0.001.

Mean differences in TEG MA values between WB and PPP at Day 0 and Day 56 were used to estimate the effect of cellular components in clot generation. The difference of these values (i.e. Mean WB TEG MA—Mean PPP TEG MA) was estimated as the contribution of “platelet clotting activity” to the WB TEG MA. The remaining platelet clotting activity was then expressed as a percentage of the Day 0 PPP MA for both Test and Control at Day 0 and Day 56 timepoints (Figure 4C). It was observed that the APEX group preserved a greater percentage of platelet clotting activity when compared with the CPDA-1 control. APEX WB maintained about 39% of platelet clotting activity when tested on Day 56 as compared with only 12% for the paired Control units.

3.5 ∣. Plasma coagulation factor levels

Plasma samples were obtained from stored blood units and assessed for coagulation factor activities. Consistent with the increased AS, fibrinogen, factor V, factor VIII, and Protein S levels were lower in the APEX units compared with their contemporaneous CPDA-1 controls (Figure 5A-D). Fibrinogen concentration (Figure 5A) remained relatively stable throughout the 8 weeks of storage, with values of both CPDA-1 and APEX units remaining within normal range. Decreasing activity of labile Factor V (Figure 5B) was observed with cold storage over the 8 weeks and in both groups, but all values remained above 40% activity. Factor VIII activity (Figure 5C) was dramatically reduced after the first week of storage but remained stable throughout the remainder of storage. Equivalently, decreasing activity of Protein S (Figure 5D) was observed among CPDA-1 and APEX units.

FIGURE 5.

Plasma coagulation factors and anticoagulant protein S measures of stored WB in CPDA-1 and APEX. (A) Fibrinogen, (B) factor VIII, (C) factor V, and (D) Protein S were determined using STA-R Max. Data visualized as mean ± SD for n = 15 units. Significant differences between groups (p < 0.05) at each time point are represented by: ^†p < 0.05, ^‡p < 0.01, ^§p < 0.001.

3.6 ∣. Bacterial cultures at the end of storage

High volume aerobic and anaerobic cultures performed at the end of storage revealed no growth despite the repeated sampling.

4 ∣. DISCUSSION

Here we demonstrate that leukoreduced CPDA-1 WB stored with 50 mL APEX preservative solution maintained improved WB in vitro quality parameters up to 56 days (8 weeks). We have previously demonstrated that AS-7 RBC additive solution can store RBCs (substantially free of platelets and minimal amount of plasma) for up to 8 weeks and obtained CE mark for 8 weeks storage.¹² Storing RBCs in a greater volume of plasma with increased buffering capacity is expected to allow the APEX RBCs to store even longer. Not previously demonstrated is that platelets stored in the presence of RBCs, with their ability to deliver oxygen and buffer pH changes, allowed the storage of platelets with preservation of substantial clotting function for the full 8 weeks. Substantial plasma coagulation protein function was also preserved for the full 8 weeks of conventional 1–6°C storage.

These findings build on established knowledge and highlight the novel capacity of APEX to extend WB functionality beyond conventional storage durations. Meryman has demonstrated that RBCs can be stored at 1–6°C for 6–9 months if the storage solution is changed regularly to prevent the pH from falling to levels that inhibit glycolysis.¹³ Bicarbonate buffering has a similar, though more limited, effect and depends critically on the CO₂ permeability of the storage container,¹⁴ resulting in the activation of anaerobic glycolytic flux, as the activity of phosphofructokinase is exquisitely sensitive to higher levels of hydrogen ions.¹⁵

Platelet storage is also critically dependent on oxygen availability and pH.¹⁵ However, the recognition of a significant contribution of platelets to the TEG MA after 8 weeks of storage is new, while the persistence of plasma coagulation protein function in CPDA-1 WB for 5 and 6 weeks is well described.^16,17

Limitations of this study include repeated sampling from the bags and subsequent temperature cycling. It is possible that the repeated sampling may have led to increases in biochemical analytes including potassium as well as increased hemolysis, as observed by others.¹⁸ An additional limitation of this study is that this work was all in vitro and performed on a limited number of WB units. We chose a pool-and-split study design for this study to minimize inter-donor variability, providing a more controlled comparison between CPDA-1 and APEX^™ while maintaining statistical power with a limited sample size. Although WB is typically stored as individual units, this methodology is widely used in feasibility studies before moving to licensure trials, which will assess storage effects on individual donors, allowing for an efficient preliminary evaluation of storage performance while reducing donor-dependent variability. Pooling the units, even in groups of two, reduces the effects and detection of donors whose metabolism leads to individual poor storage. Despite these limitations, a mean RBC ATP concentration of 3.1 μM/g Hb for APEX WB is generally considered sufficient, suggesting an average in vivo recovery rate exceeding 80%. In prior clinical studies of SOLX, AS-7 RBCs stored for 8 weeks demonstrated a mean ATP level of 3.1 μmol/g Hb, correlating with an average in vivo RBC recovery of 82%.¹²

Notably, this study highlights the preservation of RBCs and clotting capacity of platelets up to 8 weeks of storage with APEX WB, representing a potentially critical advancement for trauma care and resuscitation in extended storage scenarios. Further confirmatory experiments utilizing both leukoreduced and non-leukoreduced stored APEX WB units are warranted. Nevertheless, the goal of 8-week WB storage appears feasible and worth pursuing in clinical trials.

Abbreviations:

AS

additive solution

ATP

adenosine triphosphate

C

celsius

CO₂

carbon dioxide

CPDA-1

citrate phosphate dextrose adenine blood additive solution

CSWB

cold-stored whole blood

D0

day 0

D56

day 56

DoD

Department of Defense

Hb

hemoglobin

HBC

Hoxworth Blood Center

HCO₃

bicarbonate

HCT

hematocrit

ISR

US Army Institute of Surgical Research

K⁺

potassium

LDH

actate dehydrogenase

LI-30

lysis index at 30 minutes

MA

maximum amplitude

MCV

mean corpuscular volume

pCO₂

partial pressure of carbon dioxide

PLT

platelet

pO₂

partial pressure of oxygen

PPP

platelet poor plasma

PRP

platelet rich plasma

RBC

red blood cell

SD

standard deviation

TEG

thromboelastography

WB

whole blood