Authors
Sandra Ramirez-Arcos¹,² | Yuntong Kou¹ | Dilini Kumaran¹ |
Geraldine M. Walsh³| Anita Howell⁴ | Chryslain Sumian⁵ | Stefan Reichenberg⁵ | Quentin Brebant⁶ | Ken McTaggart¹
1.Product & Process Development, Canadian Blood Services, Ottawa, Ontario, Canada
2.Department of Biochemistry, Microbiology and Immunology, University of Ottawa, Ottawa, Ontario, Canada
3.Product & Process Development, Canadian Blood Services, Vancouver, British Columbia, Canada
4.Product & Process Development, Canadian Blood Services, Edmonton, Alberta, Canada
5.Medical Affairs, Macopharma, Mouvaux, France
6.Research and Development, Macopharma, Mouvaux, France
Introduction
Di(2-ethylhexyl) phthalate (DEHP) has been used as the plasticizer of choice in polyvinyl chloride (PVC) storage containers for red blood cell (RBC) concentrates (RBCCs), in part because DEHP was found to decrease the RBC storage lesion by stabilizing RBC membranes [1]. However, phthalates have been identified as potential endocrine disruptors and human carcinogens [2, 3]. The European Union regulators mandate blood storage containers to be DEHP-free from 2030 onwards [4]. To address this new mandate, suppliers of blood bags have been investigating containers plasticized with alternative non-phthalate plasticizers including di(2-ethylhexyl) terephthalate (DEHT) [2]. Furthermore, to mitigate the impact of DEHP removal on RBCC quality, next-generation additive solutions (ASs) such as phosphate-adenine-glucose-guanosine-saline-mannitol (PAGGSM) are being used to maintain RBCC quality [5-8].
Canadian Blood Services’ primarily follows European practices for producing whole blood–derived fresh blood components and typically sources blood collection sets from European manufacturers. Although there is currently no Canadian legislation prohibiting the use of DEHP in medical devices, the adoption of DEHP-free whole blood collection sets in Canada is anticipated. Canadian Blood Services has recently evaluated the in vitro quality of RBCCs collected in non-DEHP whole blood collection sets and stored in PVC-DEHT bags containing PAGGSM. The study showed that RBCC units stored in DEHT/PAGGSM and DEHP/saline-adenine-glucose-mannitol (SAGM) had comparable in vitro quality parameters (e.g., haemolysis), with both storage conditions meeting Canadian Standards Association requirements [5].
In addition to evaluating the impact of alternative plasticizer/AS blood bags on RBCC in vitro quality, it is imperative to assess any impacts to bacterial survival or proliferation in RBCCs. Bacterial contamination of RBCCs remains a rare but serious clinical concern. The US Centres for Disease Control and Prevention (CDC) assessed the incidence of transfusion-transmitted infections associated with contaminated RBCC units between 1998 and 2000 and showed the rate of transfusion-transmitted bacteraemia to be approximately 1 in 500,000 RBCC units with an estimated risk of fatalities in recipients of 1 in 8 million units of RBCC [9].
Currently, there is a void in literature on the impact of DEHP or alternative plasticizers on bacterial growth in blood components. However, it has been reported that DEHP can promote the growth and metabolism of bacteria such as Escherichia coli and Bacillus subtilis in laboratory culture media at low doses (≤150 μg/mL) but displays an inhibitory effect at doses ≥300 μg/mL [10]. Thelliez et al. [4] conducted a study to determine the accumulation of DEHP- and DEHP-free plasticizers, including DEHT, during blood component storage. Results showed a statistically higher DEHP equivalent concentration in RBCCs compared to DEHT, reaching maximum values of 1.85 and 0.86 μg/dm2/mL, respectively, at the end of storage. Therefore, in this study we aimed to investigate whether proliferation of transfusion-relevant bacteria in DEHT-PAGGSM RBCCs was different from bacterial growth in DEHP-SAGM units, impacting product safety.
Materials and Methods
Figure 1 describes the experimental approach followed in this study
Experimental design for comparative bacterial growth in di(2-ethylhexyl) phthalate (DEHP)-saline-adenine-glucose-mannitol (SAGM) red blood cell concentrates (RBCCs) and di(2-ethylhexyl) terephthalate (DEHT)-phosphate-adenine-glucose-guanosine-saline-mannitol (PAGGSM) RBCCs. The diagram describes the process of whole blood (WB) collection and pooling into DEHP-free containers followed by splitting into either DEHT or DEHP bags before production and storage of RBCCs as either DEHT/PAGGSM or DEHT-SAGM units. RBCCs were stored overnight at room temperature (RT), then subjected to baseline in vitro quality and sterility testing prior to spiking with transfusion-relevant bacteria and storage for 43 days under refrigeration with weekly sampling to determine bacterial counts. rWBC, residual white blood cells.
Whole blood collection and RBCC production
Whole blood was collected from volunteer donors at Canadian Blood Services’ netCAD Blood4Research facility (Vancouver, BC, Canada). All donors gave general consent to donate blood for research and specific consent to having their donation drawn into an unlicensed DEHT/PAGGSM blood bag set. Paired ABO-matched whole blood units were collected into DEHT/PAGGSM whole blood bag sets (Macopharma, Mouvaux, France; Ref# PRORQT4-A 500 mL) anticoagulated with citrate-phosphate-dextrose (CPD). On the same day as collection, paired whole blood units were pooled into a DEHP-free PVC/citrate bag (CompoStop flex, Ref# PD51600, Fresenius Kabi, Homburg, Germany) and then split evenly into one DEHT/PAGGSM bag and one DEHP/SAGM whole blood set (Macopharma; Ref# LQT710X 500 mL) from which the anticoagulant had been removed. Whole blood was stored overnight at room temperature, and leukoreduced RBCCs were produced on day 1 post collection from each of the two whole blood units using a top/bottom RBC filtration process with the same centrifugation and extraction programmes, as described previously [5]. For split units in DEHT sets, RBCs were extracted into PAGGSM AS; for split units in DEHP sets, RBCs were extracted into SAGM AS.
RBCC baseline testing and bacterial spiking
Following filtration and prior to refrigerated storage (day 1 post collection), RBCCs were tested for baseline in vitro quality parameters including haemoglobin per RBCC (g/unit), haematocrit (L/L), mean cell volume (fL), % haemolysis and residual white blood cells (rWBC), as described previously [5]. RBCCs were stored at 1–6°C within 24 h of collection. On day 1 post collection, RBCCs were shipped refrigerated overnight to the Microbiology Lab (Ottawa, Ontario). Upon receipt, on day 2 post collection, baseline sterility testing was performed with the BACT/ALERT 3D system (bioMérieux, Marcy-l’Étoile, France) in the Microbiology Lab following established protocols at Canadian Blood Services [11]. Paired DEHT/PAGGSM and DEHP/SAGM RBCC units were inoculated with one of four transfusion-relevant bacteria: the fast growing facultative Gram-negative Yersinia enterocolitica PEI-A-105 and Serratia liquefaciens PEI-A-184, slow growing facultative Gram-positive Listeria monocytogenes PEI-A-199, and the aerotolerant anaerobe Gram-positive Cutibacterium acnes BPN-BT-19195. These bacterial strains were chosen based on previous studies showing that they are good models to evaluate bacterial proliferation and survival during RBCC storage [12, 13]. Units were spiked to a final bacterial load of ~100 CFU/mL for the facultative species and to a target load of ~1000 CFU/mL of C. acnes, as this bacterium survives but does not proliferate in RBCCs. These bacterial inocula were chosen to allow tracking of bacterial survival/proliferation during RBCC storage. On Day 2 post collection, spiked units were stored at 1–6°C for 43 days and sampled on days 0, 7, 14, 21, 28, 35 and 43 for bacterial enumeration. Bacterial identification of selected samples was performed at the end of storage to confirm the identity of the inoculated organisms. The study was performed in triplicate for each bacterial species.
Statistical analyses
Since this was a pool-and-split study comparing paired units under both conditions (DEHP/SAGM and DEHT/PAGGSM), statistical analyses were based on the within-pair differences. The in vitro quality parameters were first assessed for normality with the Anderson–Darling test. Then, either a paired t-test was used to compare day 1 in vitro quality parameters that were normally distributed in DEHP/SAGM and DEHT/PAGGSM RBCC, or a Wilcoxon signed-rank test was used for the comparison of day 1 in vitro quality parameters that were not normally distributed in the two groups of units. For each bacterial species, bacterial counts and growth curve slopes were compared between DEHP/SAGM and DEHT/PAGGSM RBCC over 43 days of storage. Additionally, growth curve slopes of each repetition of L. monocytogenes grown in RBCCs stored in DEHP/SAGM and DEHT/PAGGSM were compared between days 0 and 7 of storage using the two-tailed paired t-test. A p-value <0.05 was considered significant.
Results
In vitro quality of RBCC on day 1 of storage for DEHP/SAGM and DEHT/PAGGSM RBCCs
In vitro quality parameters including haemoglobin and haematocrit were not statistically significantly different between DEHP/SAGM and DEHT/PAGGSM units (p = 0.889 and p = 0.224, respectively, Table 1). The mean cell volume was significantly higher in DEHP/SAGM RBCCs compared to the DEHT/PAGGSM units (p = 0.024), in line with our previous findings [5]. This difference is driven by the AS, which have different tonicities [14] as has been seen in previous studies of DEHT/PAGGSM RBCCs [7]. Haemolysis and rWBC were also significantly higher in DEHT/PAGGSM RBCCs in comparison with DEHP/SAGM units (p = 0.000 and p = 0.011, respectively) (Table 1). Higher rWBC levels align with our previous report [5]; however, this level of haemolysis early in storage in DEHT/PAGGSM is unexpected and is higher than early storage haemolysis in DEHT/PAGGSM RBCC reported by other studies [6, 7]. Despite these differences, all DEHP/SAGM and DEHT/PAGGSM RBCCs produced for the study had acceptable in vitro quality results on day 1 as per Canadian Standards Association requirements, and were deemed suitable for comparing bacterial viability and growth between the two types of units.
| Quality parameter, mean (SD)a | DEHP/SAGM | DEHT/PAGGSM | Anderson–Darling normality testb | Paired t-test data (p-value) | Wilcoxon signed-rank test (p-value) |
|---|---|---|---|---|---|
| Haemoglobin per unit (g/unit) | 54 (4) | 54 (4) | <0.005 | NA | 0.889 |
| Haematocrit (L/L) | 0.618 (0.017) | 0.613 (0.018) | 0.087 | 0.224 | NA |
| Mean cell volume (fL) | 97.3 (4.3) | 96.8 (4.2) | <0.005 | NA | 0.024 |
| Haemolysis (%) | 0.09 (0.03) | 0.17 (0.12) | 0.277 | 0.000 | NA |
| rWBC (×106 cells/unit) | 0.031 (0.138) | 0.140 (0.877) | NAc | NA | 0.011 |
- Note: Typical DEHP/SAGM RBCCs at Canadian Blood Services have 55 ± 6 g/unit of haemoglobin, 0.67 ± 0.03 L/L haematocrit and 0.0597 ± 0.1165 × 106 rWBC [20]. RBC haemolysis should be ≤0.8% [17].
- Abbreviations: DEHP, di(2-ethylhexyl) phthalate; DEHT, di(2-ethylhexyl) terephthalate; NA, not applicable; PAGGSM, phosphate-adenine-glucose-guanosine-saline-mannitol; rWBC, residual white blood cell; SAGM, saline-adenine-glucose-mannitol; SD, standard deviation.
- a Data are reported as mean (SD), except for rWBC which is reported as median (max), as the data are not normally distributed due to limitations of limits of detection.
- b The Anderson–Darling test was used to assess normality, with a p-value >0.05 indicating normal distribution.
- c Default non-normal distribution due to the skewed nature of the data.
Proliferation of psychrotrophic fast-growing bacteria in DEHP/SAGM and DEHT/PAGGSM RBCCs
For Y. enterocolitica and S. liquefaciens, there was no significant difference in the bacterial growth rate between DEHP/SAGM and DEHT/PAGGSM RBCCs (p = 0.95 and p = 0.87, respectively) (Figure 2). Both species grew to a bacterial load of approximately 107 CFU/mL by day 7 of RBCC storage, a load that is considered clinically significant [15]. The growth of S. liquefaciens plateaued at ~108 CFU/mL, while that of Y. enterocolitica plateaued to 109 CFU/mL by day 14 of storage.
Figure 2
Bacterial growth in red blood cell concentrates (RBCCs) stored in di(2-ethylhexyl) phthalate (DEHP)/saline-adenine-glucose-mannitol (SAGM) and di(2-ethylhexyl) terephthalate (DEHT)/phosphate-adenine-glucose-guanosine-saline-mannitol (PAGGSM) containers. The graph describes bacterial growth over 43 days of RBCC storage under refrigeration. Psychrotrophic species Yersinia enterocolitica, Serratia liquefaciens and Listeria monocytogenes proliferated in RBCCs, while the aerotolerant anaerobe C. acnes remained viable with no increase in bacterial counts until day 43 of storage. L. monocytogenes grew more slowly than the other two psychrotrophic species, with a significant decline in counts at days 0–7 of storage in DEHT/PAGGSM RBCC compared to DEHP/SAGM units. No differences in bacterial counts were observed for any of the species at the end of RBCC storage. *p = 0.03.
Proliferation of psychrotrophic slow-growing L. monocytogenes in DEHT/PAGGSM RBCCs and DEHP/SAGM RBCCs
As shown in Figure 2, counts of L. monocytogenes decreased more in DEHT/PAGGSM RBCCs than in DEHP/SAGM RBCCs from day 0 to day 7 of storage, resulting in significantly different growth curve slopes (p = 0.03). However, after day 7, the bacterium proliferated in both types of RBCCs, reaching similar counts (~107 CFU/mL) on day 43 of storage.
Survival of the aerotolerant anaerobic C. acnes in both DEHT/PAGGSM and DEHP/SAGM RBCCs
C. acnes counts remained at approximately 103 CFU/mL until the end of RBCC storage, with no significant differences observed between DEHT/PAGGSM and DEHP/SAGM RBCC units (p = 0.08) (Figure 2).
Discussion
Data obtained in this study with selected transfusion relevant bacteria suggest that there is no increase in the safety risk of RBCC storage in DEHT containers with PAGGSM compared to DEHP/SAGM stored-RBCCs.
We showed that the fast-growing psychrotrophic species Y. enterocolitica and S. liquefaciens displayed the same growth dynamics in DEHP/SAGM and DEHT/PAGGSM RBCC, reaching clinically relevant levels by day 7 of storage with no significant differences in bacterial counts. These species have been implicated in septic transfusion reactions involving contaminated RBCCs [16] and our results showed that the risk posed by DEHT/PAGGSM RBCC units to transfusion patients is not different from that posed by current units stored in DEHP/SAGM.
Interestingly, slow-growing L. monocytogenes had a sharp decrease in bacterial counts in DEHT/PAGGSM RBCC units from days 0 to 7 of storage; however, the bacterial load was comparable in both DEHT/PAGGSM and DEHP/SAGM RBCCs at the end of storage, indicating a similar safety risk for transfusion patients. The reduction in bacterial counts during early storage in DEHT/PAGGSM RBCC is intriguing and could be strain-specific or could be due to an effect of the plasticizer in combination with the change in the AS. Although there were statistically significant differences in mean cell volume, haemolysis and rWBC between DEHT/PAGGSM and DEHP/SAGM RBCCs prior to bacterial spiking (Table 1), it is unknown if any of these factors contributed to the unexpected observation in the decline of L. monocytogenes counts, which merits further investigation. Importantly, despite the observed differences, all in vitro quality parameters met the current Canadian Standard CAN/CSA:Z902-25 [17].
Sandy et al. [10] have shown that DEHP concentrations ≤150 μg/mL promote growth of E. coli and B. subtilis, an effect that was not observed in our study. However, these species do not typically survive/proliferate in RBCCs. To our knowledge, no similar studies have been conducted with DEHT. According to Thelliez et al. [4], the concentration of DEHP leached into RBCCs by day 49 of storage is approximately 1.85 μg/dm2/mL of RBCCs, which equates to 30.2 μg/mL in the PVC-DEHP bags. This appears to be insufficient to influence survival or proliferation of any of the bacteria tested in the assays described herein.
Our studies could be complemented by testing other bacterial species that can degrade phthalates, such as Pseudomonas fluorescens [18]. Furthermore, it would be interesting to run a similar comparative study in RBCCs stored in containers that have other phthalate-free plasticizers such as di-isononyl-cyclohexane-1,2-dicarboxylate (DINCH) and/or n-butyryl-tri-n-hexyl citrate (BTHC), which have also been shown to be promising candidates to replace DEHP in blood collection and storage containers [19].
Overall, our study, the first to provide bacterial safety data in DEHP-free blood bags, supports the substitution of DEHP bags with alternative plasticizers for RBCC storage without compromising the safety of transfusion patients in terms of bacterial contamination risk.
Acknoledgements
The authors thank volunteer blood donors and the Canadian Blood Services NetCAD Blood4Research Facility (Vancouver, BC, Canada) for collecting and processing red blood cell concentrate units used in this study. Blood bags were supplied by Macopharma.
S.R.-A., G.M.W., A.H. and K.M. conceived the study with input from C.S., S.R. and Q.B.; Y.K. and D.K. executed the study and assisted with data analysis; and S.R.-A. wrote the manuscript, which was reviewed by all co-authors.
Conflict of interest statement
This study was conducted in collaboration with Macopharma. C.S., S.R. and Q.B. are Macopharma employees. All other authors declare no conflicts of interest.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
1. Bider R-C, Lluka T, Himbert S, Khonder A, Qadri SM, Sheffield WP, et al. Stabilization of lipid membranes through partitioning of the
blood bag plasticizer di-2-ethylhexyl phthalate (DEHP). Langmuir. 2020;36:11899–907.
2. Hirosawa N, Yano K, Suzuki Y, Sakamoto Y. Endocrine disrupting effect of di-(2-ethylhexyl)phthalate on female rats and proteome analyses of their pituitaries. Proteomics. 2006;6:958–71.
3. David RM, Moore MR, Finney DC, Guest D. Chronic toxicity of di(2-ethylhexyl)phthalate in mice. Toxicol Sci. 2000;58:377–85.
4. Thelliez A, Sumian C, Chazard E, Reicheberg S, Lecoeur M, Decaudin B. Migration of di(2-ethylhexyl) phthalate, diisononylcyclohexane-1,- 2-dicarboxylate and di(2-ethylhexyl) terephthalate from transfusion medical devices in labile blood products: a comparative study. Vox Sang. 2023;118:533–42.
5. Stephenson T, Howell A, Olafson C, Sumian C, Reicheberg S, Brebant Q, et al. In vitro quality of whole blood-derived red cell concentrates collected, processed and stored in a blood bag set plasticized with di (2-ethylhexyl) terephthalate. Vox Sang. 2025;120:1103–12.
6. Graminske S, Puca K, Schmidt A, Brooks S, Boerner A, Heldke S, et al. In vitro evaluation of di(2-ethylhexyl)terephthalate-plasticized polyvinyl chloride blood bags for red blood cell storage in AS-1 and PAGGSM additive solutions. Transfusion. 2018;58:1100–7.
7. Larson L, Sandgren P, Ohlsson S, Derving J, Friis-Christensen T, Daggert F, et al. Non-phthalate plasticizer DEHT preserves adequate blood component quality during storage in PVC blood bags. Vox Sang. 2021;116:60–70.
8. Lotens A, Najdovski T, de Valensart N, Cellier N, Sumian C, Brebant Q, et al. In vitro quality assessment of red blood cells and plasma units in DEHT/SAGM bags and DEHT/PAGGSM bags during storage. Vox Sang. 2025. https://doi.org/10.1111/vox.70151. Online ahead of print.
9. Buerger CS, Jain H. Infectious complications of blood transfusion. StatPearls. Treasure Island (FL): StatPearls Publishing; 2025. Available from: https://www.ncbi.nlm.nih.gov/sites/books/NBK585035/. Last accessed 4 Nov 2025.
10. Sandy EH, Yao J, Zheng S, Gogra AB, Chen H, Zheng H, et al. A comparative cytotoxicity study of isomeric alkylphthalates to metabolically variant bacteria.
J Hazard Mater. 2010;182:631–9.
11. Mastronardi C, Perkins H, Derksen P, denAdmirant M, RamirezArcos S. Evaluation of the BacT/ALERT 3D system for the implementation of in-house quality control sterility testing at Canadian Blood Services. Clin Chem Lab Med. 2010;48:1179–87.
12. Ramirez-Arcos S, Kou Y, Cayer M-P, De Grandmont M-J, Girard M, Cloutier M. The impact of red blood cell manufacturing variables on bacterial growth dynamics: a pilot study. Vox Sang. 2019;114: 478–86.
13. Prax M, Spindler-Raffel E, McDonald CP, Bearne J, Satake M, Kozakai M, et al. Establishment of transfusion-relevant bacteria reference strains for red blood cells. Vox Sang. 2021;116:692–701.
14. Zehnder L, Schulzki T, Goede JS, Hayes J, Reinhart WH. Erythrocyte storage in hypertonic (SAGM) or isotonic (PAGGSM) conservation medium: influence on cell properties. Vox Sang. 2008;95:280–7.
15. Jacobs MR, Good CE, Lazarus HM, Yomtovian RA. Relationship between bacterial load, species virulence, and transfusion reaction with transfusion of bacterially contaminated platelets. Clin Infect Dis. 2008;46:1214–20.
16. Ramirez-Arcos S, Goldman M. Bacterial contamination in “practical transfusion medicine”. 6th ed. Bognor Regis: John Wiley & Sons, Ltd; 2022.
17. Canadian Standards Association. CAN/CSA-Z902–25. Blood and Blood Components. Toronto: CSA Group; 2025.
18. Zeng F, Cui K, Li X, Fu J, Sheng G. Biodegradation kinetics of phthalate esters by Pseudomonas fluoresences FS1. Process Biochem. 2004; 39:1125–9.
19. Vermeulen C, den Besten G, van den Bos AG, Go M, Gouwerok E, Vlaar R, et al. Clinical and in vitro evaluation of red blood cells collected and stored in a non-DEHP plasticized bag system. Vox Sang. 2022;117:1163–70.
20. Circular of Information for the use of human blood components. Red blood cells, leukocytes reduced. Available from: https://www.blood.ca/ sites/default/files/IM-00005_Rev_1.pdf. Last accessed 18 Dec 2025.
This study is published on Vox Sanguinis: Bacterial proliferation is comparable in red blood cell concentrates stored in DEHT/PAGGSM and DEHP/SAGM containers – Ramirez‐Arcos – Vox Sanguinis – Wiley Online Library