The effect on canine erythrocyte osmotic fragility under different temperature variations and storage times

Abstract

Background

The erythrocyte osmotic fragility (EOF) test has been used to determine the integrity of human and animal red blood cell (RBC) membranes, and to observe the mechanical quality of blood cells meant for transfusion. Existing studies proposed 1–6˚C as an optimal range of blood bag storage temperature to minimize risk of hemolysis. However, temperature fluctuations during blood bag delivery may induce denaturation of RBC membrane, resulting to in vitro hemolysis, and subsequent compromise of RBC function and viability. The purpose of this study was to determine the effect of time and temperature on the quality of canine whole blood (WB) used for transfusion. The specific aim is to utilize EOF analysis to determine alterations of osmotic fragility under different temperatures at selected storage time points. Fifteen healthy dogs were recruited to donate WB (250 mL) collected in CPDA-1 (citrate, phosphate, dextrose, and adenine). Fifteen segments of blood samples from each blood bag were stored at 4˚C. Over a period of 35 days, samples were analyzed at 3 different time points: days 0, 28 and 35. On each testing day, 5 blood segments of each blood bags were separately stored at 1, 4, 10, 25 and 35˚C for 24 h before EOF analysis.

Results

Findings of our study showed slightly increased osmotic resistance amongst storage temperatures, 1, 4 and 10˚C, but increased osmotic fragility were observed at 25 and 35˚C through all testing days. When storage time points were compared, the mean corpuscular fragility (MCF) of samples did not significantly differ regardless of temperature; however, when considering both temperature and time, samples stored over 28 days and incubated at 35˚C showed the most significant differences.

Conclusion

Canine whole blood with CPDA-1 stored over 28 days and exposed to temperature over 25˚C may have resulted in the assumed loss of membrane integrity, causing hemolysis. The observed outcome of the study implies that EOF underwent statistically significant influence by increasing temperatures and was only affected in lesser degrees by storage time.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12917-025-05162-4.

Background

Demand of blood products continually rises to accommodate patients of various medical conditions, especially those involved with severe trauma and blood loss. The critical nature of such patients creates the pressure for blood products to be readily provided, which initialized the concept of blood preservatives, and the scrutiny of appropriate environment to keep blood products at its best condition throughout storage and transport, or prior to transfusion. However, viability of blood cells decreases inside blood bags, leading to hemolysis and accumulation of metabolites; a phenomenon referred as ‘storage lesion’ which is considered a major issue in the field of transfusion medicine [1–3].

Many factors are considered when addressing the poor survival of red blood cell (RBC). Temperature and storage duration are major contributors to reduced RBC survival, highlighting the importance of strict standards for proper blood storage and transportation [4–7]. The use of additive solutions is also an appropriate option to preserve RBC quality for longer period [8]. Diseases of hematology such as sickle-cell anemia, beta-thalassemia, and parasites of blood cells also produce abnormal RBC fragility [9, 10], prompting specific regulations to only accept donor candidates that are clinically healthy. While age, breed, species, and genetic factors may be ruled in as intrinsic determinants of RBC membrane integrity [11–13], their significance lacks description in veterinary transfusion medicine, and therefore aren’t typically addressed when evaluating the reason of poor storage survival of RBC.

The EOF test is considered as a diagnosis method of hereditary spherocytosis [14, 15] and screening of beta-thalassemia in humans [16, 17] by utilizing different concentrations of saline solution to detect abnormally fragile RBC. In veterinary medicine, EOF was used to observe RBC membrane fragility across species [12, 18–21]. A study proposed normal range of fragility in healthy dogs as 0.38–0.55% NaCl [22], which provided reference for more recent studies to evaluate loss of RBC viability in canine blood bags [23].

The purpose of this study is to determine the effect of time and temperature on the quality of canine blood used for transfusion. The specific aim is to use EOF analysis to determine alterations of osmotic fragility under varying temperatures at selected storage time points.

Materials & methods

Study population

The blood samples used in this study were convenient samples collected from 15 canine donors who met the requirements of the Veterinary Transfusion Medicine Center (VTMC) in National Pingtung University of Science and Technology (NPUST), Taiwan; between the ages of 1 and 8 years old, weighed over 20 Kg, clinically healthy, and not medicated for any conditions [24]. Vaccination status and deworming routines were confirmed, followed with measurement of temperature, pulse, and respiration (TPR) values. Auscultation, palpation, and evaluation of hematology and biochemistry profiles were performed to confirm healthy status. Prior to blood collection, donors were assessed to be negative for Ehrlichia canis, Ehrlichia ewingii, Ehrlichia chaffeensis, Anaplasma phagocytophilum, Anaplasma platys, Borrelia burgdorferi, and Dirofilaria immitis (IDEXX, SNAP 4Dx Plus). Post-donation nested PCR were performed to further ensure the donors were free from vector-borne diseases (Rickettsia spp., Babesia gibsoni, Babesia vogeli, Anaplasma platys, Ehrlichia canis), and bacterial culture with goat blood agar was used to validate the aseptic collections.

Study design

Every blood bag from each canine donor provided 15 blood segments, which were then separated into 3 groups and stored at 4˚C. Each group represents a test day: day 0, 28 and 35. The EOF test was conducted after samples were incubated for 24 h (Fig. 1).

Fig. 1.

Experimental design

Blood sample collection

The WB samples were collected at NPUST-VTMC. Canine donors were stabilized through physical restraint, and no sedatives or anesthetics were used. Prior to blood collection, venipuncture area was trimmed and disinfected. From each donor, 250 mL of whole blood from the external jugular vein was transferred into a blood bag containing 35 mL of CPDA-1 (JMS, Hiroshima, Japan) in a low-pressure suction chamber (Animal Blood Resources International, California, USA).

After collection, blood was run into its tube and separated to 15 segments, each 5 cm in length which is equivalent to about 1 mL of blood (Fig. 2). All blood samples were stored at 4˚C for the duration of the study until analysis.

Fig. 2.

From each blood bag, whole blood was run into its tube and separated to 15 segments

NaCl solution Preparation

Different concentrations of NaCl solutions were prepared by mixing sodium chloride 0.9% with distilled water to create 50 mL solution of each concentration: 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.55, 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9% NaCl. A tube of 50 mL distilled water was used to represent 0% NaCl.

EOF test

On each testing days (day 0, 28 and 35), 5 segments of WB samples were separately incubated at 1, 4, 10, 25 and 35˚C for 24 h. WB segments were cut at the end and the contents were gently transferred into a microtube, then separated from plasma to yield RBCs. The RBCs were washed by mixing with 4.8 ml of 0.9% NaCl and centrifugation at 2516 x g for 3 min before removing the supernatant. The wash process was repeated 3 times.

The 4µL of washed RBC was added to separate microtubes containing 1 mL of the varying NaCl solutions. The mixture was gently vortexed and incubated at room temperature (25˚C) for 30 min. Subsequent to incubation, the samples were centrifuged at 1300 x g for 10 min. The obtained supernatant was examined spectrophotometrically (Bio-Tek µQuant Universal Microplate Spectrophotometer), set at optical density 540 nm [25]. The absorbance values were calculated by referencing the absorbance when 100% of RBCs are hemolyzed in distilled water, and the absorbance when no hemolysis occurred in 0.9% NaCl. The mean corpuscular fragility (MCF) or H₅₀ graph was made to visualize the concentration of NaCl that produces 50% hemolysis [10].

Statistical analysis

All collected data was analyzed using Statistical Package for Social Sciences (SPSS) version 26. One-way repeated measures ANOVA was performed to test for significant differences among means of the individual groups (temperature or time). Two-way repeated measures ANOVA was used to determine for any interactions between the two parameters (temperature and time). Difference of means showing p < 0.05 was considered significantly different.

Graphs were plotted with NaCl concentration (%) on the X-axis and hemolysis (%) on the Y-axis.

Results

Osmotic fragility

The EOF curve displays the NaCl content ranging from 0.1 to 0.9%, with a total of 17 different concentrations. Each concentration shows specific percentage of hemolysis. Fifteen conditions; three time points and five varying temperatures were used to create hemographs showing the differences between day (D) 0, 28 and 35, and temperature (T) 1, 4, 10, 25 and 35˚C. The average MCF/H₅₀ of all 15 conditions were determined by calculating the average MCF from the OD values of each blood samples of 15 dogs. (Table 1).

Table 1.

Average mean corpuscular fragility (MCF) for each storage conditions

Day/Temperature	T1	T4	T10	T25	T35
D0	0.47 ± 0.046	0.46 ± 0.047	0.47 ± 0.046	0.47 ± 0.041	0.49 ± 0.038
D28	0.44 ± 0.032	0.44 ± 0.033	0.44 ± 0.032	0.47 ± 0.046	0.49 ± 0.031
D35	0.44 ± 0.052	0.43 ± 0.046	0.45 ± 0.050	0.48 ± 0.058	0.52 ± 0.051

D: Day.

T: Temperature (˚C)

Temperature related differences of MCF

The averages of OD values from all 15 samples in each storage conditions were converted to hemolysis percentage, and expressed in a form of curve graphs. Figure 3 visualizes the trend of EOF after incubation in 5 varying temperatures across 3 storage time points.On the initial testing day (D-0), EOF of samples incubated at 35˚C shifted slightly to the right of 0.5% NaCl, verifying that H₅₀ occurred earlier and at higher concentration of NaCl. This indicates an increased EOF on day 0 at 35˚C (Fig. 3a).

Fig. 3.

Erythrocyte osmotic fragility (EOF) curve of temperature related difference on (a) day 0, (b) day 28, and (c) day 35

On D-28, samples incubated at T-1, T-4, and T-10 did not only continue to maintain a H₅₀ within normal range (0.38–0.55% NaCl), but they also exhibited a slight shift to the left, suggesting enhanced osmotic resistance at these particular conditions. Samples incubated at T-25 on D-28 showed no statistically significant difference compared with those incubated at T-4 (p = 0.079), nor with the samples incubated at T-25 on D-0 (H₅₀ ≤ 0.5% NaCl, p = 0.791). This indicates that osmotic stability was preserved up to D-28, even after 24 h of incubation at 25 °C. T-35 on D-28 showed a statistically significant difference in comparison to T-4; H₅₀ was apparent even at higher saline concentrations (p = 0.004) (Fig. 3b; Table 2).

Table 2.

Comparison of mean corpuscular fragility (MCF) at different temperatures on each testing time

Day 0		Day 28		Day 35
Temperature	p-value	Temperature	p-value	Temperature	p-value
T1 vs. T4	0.854	T1 vs. T4	0.554	T1 vs. T4	0.582
T1 vs. T10	0.919	T1 vs. T10	0.903	T1 vs. T10	0.854
T1 vs. T25	0.775	T1 vs. T25	0.019*	T1 vs. T25	0.098
T1 vs. T35	0.136	T1 vs. T35	0.001*	T1 vs. T35	0.000*
T4 vs. T10	0.775	T4 vs. T10	0.639	T4 vs. T10	0.462
T4 vs. T25	0.639	T4 vs. T25	0.079	T4 vs. T25	0.027*
T4 vs. T35	0.094	T4 vs. T35	0.004*	T4 vs. T35	0.000*
T10 vs. T25	0.854	T10 vs. T25	0.026*	T10 vs. T25	0.142
T10 vs. T35	0.165	T10 vs. T35	0.001*	T10 vs. T35	0.000*
T25 vs. T35	0.228	T25 vs. T35	0.262	T25 vs. T35	0.012*

D: Day

T: Temperature (℃)

*: p-value < 0.05

On D-35, the shift of EOF curve seemed mild at T-25 and T-35 (Fig. 3c). However, the MCF of T-25 and T-35 was statistically significant compared to other temperatures (Table 2). Samples of T-25 and T-35 of D-35 required higher saline concentrations to reach H₅₀ (Table 1), supporting the implication of earlier hemolysis.

Time related differences of MCF

Figure 4 visualizes the trend of EOF after incubation in 3 different storage time across 5 temperature variations. The MCF of control group 4˚C had no statistically significant differences among test days D-0, 28 and 35 (D-0 vs. D-28, p = 0.307; D-28 vs. D-35, p = 0.729; D-0 vs. D-35, p = 0.171), although osmotic resistance can be seen throughout the 35 day period, as shown by the shift of EOF curve towards the left (Fig. 4b). Samples incubated at T-1 and T-10 also showed a left shift on D-28 and D-35 (Fig. 4a, c). Despite that, the MCF on D-28 and D-35 did not show statistical difference compared to D-0 at optimal storage temperature (T-1, T-4, and T-10) (Supplementary Table 1).

Fig. 4.

Erythrocyte osmotic fragility (EOF) curve of time related difference on (a) 1˚C, (b) 4˚C, (c) 10˚C, (d) 25˚C, and (e) 35˚C. Observation of time related difference was made by using 4˚C as the control reference

Shown in Fig. 4d and e, the EOF curves did not show obvious shift. Differences of MCF at T-28 and T-35 were non-significant, regardless of test days (Supplementary Table 1). Overall, no significant differences were seen in the MCF of every temperature in all test days.

MCF comparison between different storage time and temperature conditions

Two-way repeated measures ANOVA was used to determine the differences between test conditions and each group’s p-value is used to determine significance (Supplementary Table 2). Blood samples stored over 28 days at T-35 appeared to yield the most significantly higher MCF compared to all other test days and storage temperatures (Supplementary Table 2).

Discussion

The data obtained from this study demonstrated that temperature is a crucial factor which must be strictly controlled to maintain good quality of RBCs. The influence of temperature was observed throughout the study in the form of increasing fragility along with the rise of storage temperature. While samples of our study exhibited stable fragility at D-0 on all incubation temperatures, samples stored until D-28 and D-35 showed higher fragility at T-25 and T-35. Thirty-five days is the recommended shelf life of canine WB stored in CPDA-1 [26, 27]. Our result supports 35 days as the maximum shelf life of canine whole blood stored in CPDA-1, and exposure of 35-day-old WB to temperature over 25 ˚C will increase RBC fragility.

Previous studies also demonstrated higher level of hemolysis over increased temperature [3, 4]. Higher temperature results in accelerated metabolic process of RBC (e.g. glycolysis), which may rapidly deplete nutrients, accumulate metabolic waste, and promote acidification within the blood bag; as the consequence, they compromise the shelf life of whole blood or packed red blood cells (pRBCs) [3, 28]. Hess [28] also reported that glycolysis in RBC occurs 20 times slower at 3˚C than 37˚C, which may explain the significant shift of hemolysis curve to the right at 25˚C and 35˚C in our study. Samples stored at 1–10˚C in our study showed no significant hemolysis.

RBC viability is well-maintained until the 35th day as long as temperature is maintained between 1˚C to 10˚C. As the significant shift of EOF was only observed after the fourth week of storage (D-28), we may point out that storage time also affects EOF. However, significant increase of EOF only began when samples stored over four weeks were exposed at 25˚C or 35˚C for 24 h. Therefore, it is reasonable for various blood bank regulations to set a mandatory storage temperature of 1˚C to 6˚C, and transport temperature of 1˚C to 10˚C to delay RBC degradation [3, 29].

At optimal cold storage, increased EOF may not be due to membrane loss, but rather, the accumulation of osmotically active substances within RBC renders it hyperosmolar and sensitive to extreme hypoosmotic challenge [30]. However, a recent prospective study showed that over 51% of 35 to 42 day-old pRBC units treated with saline, adenine, dextrose, and mannitol (SAGM) experienced hemolysis over 0.8%, which is the limit of in-bag hemolysis set by European Directorate for the Quality of Medicines and Healthcare (EDQM) [31]. Hence, it is still recommended to test the degree of hemolysis in pRBCs beyond shelf-life, prior to transfusion [29, 31].

Better resistance to osmotic stress was observed at T-1, T-4 and T-10. At D-28, T-1 showed improved resistance to hemolysis, and at D-35, it maintained an MCF lower than the initial testing day (D-0). Similar observations were seen at samples of T-4 and T-10. In short, temperature range 1˚C to 10˚C is optimal to maintain RBC membrane integrity, and fluctuations within the proposed range will not greatly affect quality of blood components. The findings of this study supported the standard range of transport temperature, 1˚C to 10˚C as proposed by AABB [29]. Storage and transport of blood bags should be maintained within this recommended range as to not compromise the RBC quality.

Our results indicate that RBC fragility is increased after 28 days of storage, and caution must be taken to prevent canine whole blood over 28-days old from being exposed to temperature of 25˚C or higher for over 24 h.

While our study may provide reference of optimal storage temperature and period for whole blood, this study does not address other variables such as the type of blood product, the type of additive solutions used, and errors within blood processing; all of which may fairly influence RBC osmotic fragility. Furthermore, blood samples used in this study were stored in segments and not in the primary bag. The difference of in vitro environment may affect erythrocyte viability and the resulting MCF. These variables must be considered upon performing EOF and result interpretation.

Additionally, this study specifically examines EOF as a parameter of RBC viability and function, and therefore unable to explain for any possible influence of blood cell metabolites in our results. During storage, blood cells continue to perform glycolysis to generate ATP as source of energy to maintain its function, resulting in ATP depletion, and loss of normal membrane transport [6, 32]. The complexity and variations of factors influencing EOF prompts for further study of biochemical alterations in blood products under varying storage conditions.

Conclusion

When the right storage conditions are met, whether during transport or storage at the blood bank, slight fluctuations in temperature should not significantly influence RBC viability. Storage at 4 °C and transport at 10–25 °C does not negatively affect MCF. However, prolonged storage causes poor RBC resistance towards higher temperature, which results to higher susceptibility of hemolysis. Blood banks should always be aware and strive for optimal storage conditions to be met.

Abbreviations

AABB

American association of blood banks

CPDA-1

Citrate phosphate dextrose adenine-1

EDQM

European Directorate for the Quality of Medicines and Healthcare

EOF

Erythrocyte osmotic fragility

H₅₀

50% hemolysis

PCR

Polymerase chain reaction

pRBC

Packed red blood cell

RBC

Red blood cell

SAGM

Saline adenine glucose mannitol

TPR

Temperature, pulse, respiration

WB

Whole blood

MCF

Mean corpuscular fragility

Authors’ contributions

YT: Supervision, study design, statistical analysis, review and editing of manuscript; DWA: Experimentation, statistical analysis, writing of original manuscript; WL: Review and editing of manuscript.

Funding

This work was supported by grants NSTC 112-2627-M-020001 and NSTC 113-2627-M-020-001 from the National Science and Technology Council, Taiwan.

Data availability

No datasets were generated or analysed during the current study.

Ethics declarations

Ethics approval and consent to participate

The use of animals in this study were approved by the local Institutional Animal Care and Use Committee (Approval No.: NPUST-109-065). Informed consent was obtained from owners of the animals that participated in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.