Cryopreserved platelets washed with a dialysis machine for dimethyl sulphoxide removal

Cryopreserved platelets (cPLTs) can be stored for years and are mainly used in military settings. However, the commonly used cryoprotectant dimethyl sulphoxide (DMSO) has toxic side effects when utilized in high quantities. We developed a novel method to aseptically remove DMSO from thawed cPLTs by dialysis.

efficacy of our method remains to be determined. However, the function of the platelets declined 24 h after washing, making them unsuitable for transfusion.
Keywords cryopreserved platelets, dialysis method, quality evaluation INTRODUCTION Platelets are critical in blood homeostasis and clotting [1,2], and transfusion of platelets and their products can alleviate clinical and trauma-related injuries [3,4]. However, the Food and Drug Administration (FDA) stipulates that the shelf life of platelets stored by conventional shaking methods at 22 ± 2 C is only 5 days [5,6].
Cryopreservation at À80 C can extend the storage period of platelets by at least 2 years [7][8][9]. A literature review evaluated the efficacy and safety of platelets cryopreserved with dimethyl sulphoxide (DMSO) from 1972 to 2013. Valeri's classic preparation method was cited, where platelets were frozen in 6% DMSO and centrifuged to remove DMSO plasma before freezing at À65 C without a post-thaw wash (post-TW). No significant adverse events were reported in any of the studies. The recovery of cryopreserved platelets (cPLTs) was approximately half that of fresh platelets, but the survival rate decreased only slightly. Platelets are essential in haemostasis, and no significant adverse events caused by DMSO have been reported in numerous studies [10].
We developed a dialysis machine that can remove DMSO from cPLTs once they are thawed through a combination of molecular convection and diffusion [11,12]. The cryoprotectant permeates through the fibre membrane of the dialyser, and the platelets flow back into the blood bag through the hollow fibre column. This cycle is repeated to effectively wash the platelets. The advantage of this method is that it quickly removes DMSO, minimizes the risk of contamination and avoids platelet damage caused by centrifugation.

Ethics, species and intervention
The study was conducted in accordance with the Declaration of Freezing, reconstitution and washing of the platelets A 27% DMSO solution was prepared with 54.75 mL of normal saline (Chengdu Qingshan Likang Pharmaceutical, Sichuan) and 20.25 mL of DMSO (Sigma, USA) and stored at 4 C. Platelets (262 ± 7 mL) were cryopreserved by adding 75 mL of 27% DMSO into the platelet bag at a constant speed. The final volume in the bag was 336 ± 7 mL, and the concentration of DMSO was 5%-6%. After standing for 10 min at room temperature, the bag was stored at À80 C for 1 week. The cPLTs were thawed in a water bath at 40 C for 10 min. The platelets were washed for 5, 10, 15, 20 and 25 min in the dialysis machine, the platelet supernatant was collected and the osmotic pressure was measured. The volume of the post-TW platelets was 244 ± 63 mL.

Dialysis procedure
Two blood pumps were used as the dialysis machine's thrust, and DMSO, lactic acid and other harmful substances were separated from the platelets with a dialysis column (Fresenius, FX8) in the sealed pipeline ( Figure 1). DMSO was separated from the platelets through molecular convection and diffusion and displaced from the dialyser due to the osmotic pressure. The separated DMSO was collected in a sealed pipeline and expelled quickly. Platelets could not penetrate through the dialysis column, so they returned to the platelet bag with the aid of a blood pump. The washing liquid was isotonic normal saline

Sample evaluation
Osmotic pressure: The supernatant from 1 mL of platelets was centrifuged at 1500 Â g for 5 min. Aliquots (20 μL) of platelet supernatants were taken to measure the osmotic pressure using a crystal osmometer (Advanced Company, USA) according to the recommended protocol. The following formula is used to calculate the DMSO residue: O l , last osmotic pressure of the post-TW platelet samples; O i , initial osmotic pressure of the pre-freeze platelet samples; O t , osmotic pressure of the thawed frozen platelets before washing.
Platelet parameters: The platelet counts, platelet distribution width (PDW) and mean platelet volume (MPV) were measured using a blood cell counter (Beckman, USA).
Recovery rates: Platelet recovery after dialysis was evaluated in terms of counts and volume using the following formula.
C l , last platelet count after dialysis; C i , initial count in the pre-freeze platelet sample; V l , final volume of the platelet sample after dialysis; V i , initial volume of the pre-freeze platelet sample.
Aggregation: Platelets were resuspended in platelet-poor plasma (PPP) prepared by centrifugation at 1500 Â g for 5 min, and the cell density was adjusted to 3 Â 10 8 /mL. Then, the platelets were allowed to rest for more than 60 min at room temperature before conducting the experiments. Platelet aggregation was determined by the turbidi-

Statistical analysis
All experimental data were processed by SPSS 21.0 statistical software. One-way ANOVA was used to compare groups, and a value of p < 0.05 was considered to indicate statistical significance. All data are presented as the mean ± standard deviation (SD).

Platelet osmotic pressure and recovery rate
The physiological osmotic pressure of human plasma is 280-320 mOsm/kg. The osmotic pressure of the pre-freeze platelets was similar at 297 ± 9 mOsm/kg. In contrast, the osmotic pressure of the platelet supernatant increased to 1423 ± 33 mOsm/kg due to the presence of DMSO, and that of thawed platelets after 30 min of washing was slightly higher than the physiological value at

Functional indicators
The amount of lactic acid produced by the pre-freeze platelets  Table 2.
The coagulation reaction time of the pre-freeze platelets was 9.55 ± 2.47 min, which decreased to 4.80 ± 0.89 min in the post-TW group and increased again to 8.53 ± 2.32 min in the 24-PTW group (Table 3).
T A B L E 2 Blood gas parameters.
Platelet transfusion and preservation are of considerable clinical significance [17], and platelet-free antishock transfusion increases the risk of diffuse intravascular coagulation [3,18]. In 2001, the Dutch Military Blood Bank confirmed that the combination of À80 C frozen platelets, plasma and red blood cells (RBC) was safe when treating (military) trauma injuries, and at least as effective as standard blood products. The higher the use of platelets and plasma, the lower the mortality rate in military theatre [10]. The limited shelf life and other technical difficulties of platelets stored at 22 ± 2 C have propelled the use of cPLTs. Platelets are cryopreserved using 6% DMSO, and cPLTs can be stored at À80 C for up to 2 years [7]. In vitro studies have shown that cPLTs express hypercoagulability markers and have thrombolytic activity [7,19], and in vivo experiments indicate that cPLTs can stop bleeding [20]. production test and thromboelastography were improved in the cPLT recipients after transfusion. In Reade's study [22,23], a scheme was established to evaluate the therapeutic effects of cPLTs and traditional liquid-stored platelets on surgical bleeding involving high-risk cardiothoracic surgical patients. There were no differences between the two groups in terms of potential harm, including deep venous thrombosis, myocardial infarction, respiratory function, infection and renal function. Therefore, cPLT products can be a suitable substitute for platelets in many applications [24,25].
Removal of DMSO from cPLTs is a critical step to minimize toxic-  The release of β-TG and PF4 from post-TW platelets increased in a time-dependent manner in this study. β-TG and PF4 are specific proteins contained in the α granules that are released when platelets are activated, and their release could reflect the specific release reaction of platelets. It was found that platelets released approximately 50% β-TG when stored at room temperature for more than 5 days [30].
Cold storage of whole blood for 6 h will cause platelet aggregation and the release of more β-TG and PF4 [31]. The reason for the reduction in β-TG release in the 24-PTW group was thought to be because the β-TG release from frozen platelets decreases along with the reduction in the activation function of platelets. The coagulation reaction time (R-time) was used as a measure to evaluate the comprehensive effect of coagulation factors during coagulation activation. The R-time did not extend in the TEG experiment of PPP, indicating that there was no significant loss of PPP during the washing process.
Compared with the post-TW group, the R-time of the platelets in the 24-PTW group was prolonged due to the lack of coagulation factors.
MA reflects the maximum strength and hardness of the fibrin/platelet clot [32], and the main influencing factors on this parameter are platelets (80%) and fibrin (20%). The MA of PPP was significantly decreased compared with that of platelets. The main reason for the MA reduction was thought to be due to the loss of fibrinogen (function) and the reduced number of platelets. The morphological alterations caused by cryopreservation include significant shape changes from discoid to spherical, as well as an increase in the number of platelet prostheses [33], which is consistent with our observations ( Figure 5). In summary, cryopreservation, thawing, washing and aggregation of the platelets affected their quality, and 24-PTW platelets stored at room temperature had decreased product contents due to cell death [34].
We produced this machine and have applied for the relevant patents and software copyrights. This machine is in the preclinical research stage and can be used by institutions that still have an inventory of or freeze unconcentrated platelet products in 5%-6% DMSO plasma and require post-thaw centrifugation to the wash products before transfusion and for liquid nitrogen freezing of unconcentrated products. This machine can also be used in military environments as a reserve for frozen platelets. Therefore, we developed a draft of the special consumables needed and the operating procedures. By establishing standardized procedures for freezing platelets based on this device, we aim to standardize the clinical preparation of platelets to promote the clinical application of this instrument. Overall, this novel dialysis method can circumvent the problem of excessive DMSO residue, and the recovery is very similar to previously published reports of cPLTs [10]. This method is a novel solution for the post-thaw removal of excess DMSO from frozen platelets and is probably more practical than post-thaw removal by dilution and centrifugation. Furthermore, the short washing time, enclosed design of the machine and automated operation are conducive to aseptic infusion. The dialysis machine can also be used to remove glycerol and DMSO from cryopreserved RBCs and stem cells [35,36]. Next, we will focus on evaluating the quality of post-TW platelets in animal models to determine the clinical potential of this method.
In conclusion, we developed a novel dialysis method to effectively remove DMSO from cPLTs under aseptic conditions while maintaining platelet quality. The clinical efficacy of our method remains to be ascertained.