Aspects of platelet storage

Pergamon

09553s&?q94poo37-9

Transfus. Sci. Vo1.15, No.4, pp. 351355, 1994 Copyri t @ 1994 Elsevier Science Ltd Printed in $ reat Bntam. .’ All rights reserved 0955-3886/94 $7.00 + 0.00

Aspects of Platelet Storage Claes F. HGgman

n The proper storage of platelets requires sufficient knowledge of the behaviour of these cells, made by nature to react instantly. It is important not to activate them during preparation and storage and to maintain oxidative metabolism at the highest possible degree. Optimally, the citrate concentration in the storage medium should be 810 mmol/L. The addition of acetate either to the anticoagulant or to a platelet additive solution gives the potential for improved platelet storage. In order to evaluate the clinical efficiency of platelet concentrates (PCs), corrected count increment is to be recommended for frequent use. The in vitro bleeding time seems to be a valuable supplement, both for research and clinical purposes. Bacterial contamination is a threat which can be diminished by using an appropriate technique for preparation and storage. Rapid automated bacterial culture makes it possible to detect contamination which may be particularly important if the shelf life of PCs is being extended beyond the present 5 days. n

INTRODUCTION This paper is a brief review discussing some new data in the light of more

established ones with respect to the storage of platelets. Reference will also be made to different methods of preparation of the platelet concentrates (PCs) and to the possibility of improving safety. ANTICOAGULATION AND IMMEDIATE POST-COLLECTION TREATMENT Good mixing during blood collection is necessary in order to get a minimum of thrombin formation which would cause platelet activation. Fibrinopeptide A can be measured as a test of quality, i.e. that the staff does its job well and that, in the case of using mechanical mixers, these are suitably adjusted for their purpose.’ As has been emphasized by Bode2 platelet activation is one of the major events which may lead to poor viability and function. The citrate concentration in CPD plasma, ~20 mmol/L, is unnecessarily high for anticoagulation. An optimal anticoagulant should contain only half resulting in of this concentration, improved FVIII stability and better platelets.3 The initial pH in ACD or CPD whole blood is suitably lowered to avoid platelet activation. Measures which may increase pH should be avoided if the preparation steps that follow have activating power, such as centrifugation. PCs prepared from buffy coat (BC) which is stored at room temperature in a poorly

352

Transfus. Sci.

Vol. 15, No. 4

gas permeable container, such as PL 146, will have higher pH when prepared soon after collection than after 24 h-7.1 vs 6.8-due to accumulation of lactic acid and COZ. In a container with higher gas permeability such as PL 2209 the difference will be less than 0.1 pH unit, when mixed during storage (own observations, unpublished), but the consumption of the plasma bicarbonate will be equally high, about 40% of initial. PREPARATION OF PC The traditional method of preparing PC is to allow the whole blood unit to rest for 2-6 h, centrifuge it to obtain plateletrich plasma (PRP), concentrate the platelets into a pellet in the plastic bag by centrifugation and resuspend after a resting period of l-2 h. When this procedure is performed soon after blood collection, platelet activation frequently occurs and the platelets behave badly during storage.4 Also, when BC is used as the source of platelets, holding for a couple of hours results in a better yield than when centrifugation is done very soon after blood collection.’ Some apheresis procedures concentrate the platelets without, others with, pelleting of the platelets, the former procedure being more gentle on the platelets and with a smaller risk of activation than the second. STORAGE CONDITIONS The platelets derive 85% of their adenosine triphosphate (ATP) regeneration by oxidative metabolism through the tricarboxylic acid (TCA) cycle and only 15% from glycolysis-the former being much more effective than the latter.6 Gas permeable containers which allow oxygen transfer, and mixing of the PC during storage are therefore necessary to maintain, as far as possible, oxidative metabolism during storage. The size and gas permeability of the container is critical for platelet storage. Each type of container has certain limits concerning how many and how few

platelets can be stored without deteriWhen the upper limit is oration.’ exceeded there is a risk of switching to anaerobic metabolism and rapid acidification of the PC because of lack of oxygen. When the lower limit is exceeded less protons and CO, are formed; too much CO, may diffuse out of the bag resulting in high pH, microvesiculation and platelet destruction.4 Pyruvate, obtained during glycolysis, does not appear to be used for the formation of acetylcoenzyme A in the TCA cycle.“-” The natural fuel in platelet oxidative metabolism is not known, fatty acids have been proposedi1o of practical interest and significance is that acetate can be used as a component of the storage medium.gJ1 An artificial medium for platelet storage (platelet additive solution, PAS) containing acetate was first described by Adams et al.,” using Plasma Lyte A, although at that time the function of acetate in the metabolism of stored platelets was not established. Metabolic acidification of the storage medium (usually plasma) impairs the platelets and limits PC shelf life. Morphological and other structural changes occur, first reversible, later irreversible, at a decreased pH. Attempts to store platelets without glucose, in order to decrease glycolysis and lactate formation, have generally not been successfu1.13The presence of acetate in the storage medium stabilizes pH and can be used to maintain bicarbonate concentration at the initial level. This is probably achieved through a combination of consumption of bicarbonate for buffering with release of COZ, and generation of the oxidative bicarbonate in metabolism.‘“~L’ Citrate, binding Ca2+ at a sufficient degree to avoid platelet activation and thrombin formation, has to be present, when the medium contains plasma coagulation factors. When the citrate concentration is lowered to 4 mmol/L by dilution with saline, clotting will occur; at 8 mmol/L there is no clotting unless provoked by other factors.

Aspects of Platelet Storage 353

Gulliksson et ~1.‘~ found that by diluting BCs with saline before preparation of PC, the pH was much more stable at subsequent storage of the PCs than if these had been prepared using an additive solution containing 30 mmol/L citrate. A systematic follow-up revealed that lactate formation was only 0.060.08 mmol/day/lO” platelets, when the citrate concentration was 8 mmol/L in the storage medium, but twice that at the normal citrate concentration of 20 mmol/L.‘5 Thus, 40% plasma in saline seems to be an improved storage medium as compared to plasma and many of the PASS in current use. However, the use of saline as a diluent for plasma necessitates thorough standardization, since clotting is a risk if the citrate concentration falls below 8 mmol/ L. Some poor results with Plasma Lyte A obtained by our group and others may be explained by an unsuitable mixture of this solution and plasma, leading to clotting problems. If 0.5 CPD is used for blood collection the PAS must contain citrate. Gulliksson et al. also showed that this glycolysis stimulating effect of citrate could be abolished if acetate was included in the medium at an optimal concentration of 30 mmol/L-which is close to that present in Plasma Lyte A. However, combination of a low citrate and an optimal acetate concentration did not result in lower lactate production than when either of them were present in optimal concentrations alone. Favourable results have been obtained with a saline-acetate-citrate medium as a diluent for BCs, prepared from CPD whole blood as well as 0.5 CPD whole blood. This additive solution is now under clinical evaluation. Jonsson, t6 using thrombocytapheresis with Haemonetics PCS Plus, has used ACD with addition of acetate as an anticoagulant. The resulting plasma contains 1530 mmol/L acetate. Excellent storage conditions have been obtained. The pH has been kept stable during PC storage for 12 days and the platelets showed very small deterioraiS

15:4-D

tion of aggregation response to agonists such as ADP. The PC shelf life of 10 days is now tested in clincial trials. The results were similar to those described by Bertolini et al.” concerning PC from BC. CLINICAL EFFICACY Corrected count increment (increment x body surface :platelet dose) (CCI) is often used in a regular evaluation of the efficacy of platelet transfusions. The major aim is to detect platelet refractoriness in multiple transfused patients but the technique can also be used to evaluate the efficacy of PCs. Both apheresisPCs and BC-PCs showed higher CCIs when stored up to 2 days after preparation (= fresh), than when stored for 3-S days (stored). In these studies plasma and a mixture of plasma (30% J and a PAS (70%) with NaCl(70 mmol/L), KC1 (10 mmol/L), Na phosphate (5 mmol/L), Na citrate (30 mmol/L) and mannitol (30 mmol/L) were used as storage media, respectively. I7 Hopefully, the new storage media will improve the quality of stored platelets. The function in vivo of transfused platelets can be tested with the in vitro bleeding time (IVBT).‘* We have used this technique for 4 years and have obtained correspondence between CC1 and IVBT in about 70% of the tests. In the remaining 30% either a sufficient CC1 but without lowered IVBT, or no CC1 but a lowered IVBT, have been encountered. The lack of response of IVBT after platelet transfusion seems to be due to patient-related factors in some instances and to poor quality PC in others.“,” BACTERIAL

CONTAMINATION

Transfusion transmitted bacterial infection (TTBI) is a complication which is probably more common than transfer of viral infection, although more available for treatment, if diagnosed early.2’ The microbes may enter into the blood component from ( 11the donor’s blood, (2) the

354

Transfus.

Sci.

Vol. 15, No. 4

donor’s skin, (3) the donor’s leucocytes or (4) the environment via leaks in the container system. Protection against bacterial growth may be obtained through (1) unsuitable growth conditions (e.g. aerobic environment for anaerobic bacteria), (2) bacterial killing by antibodies and/or complement, or (3) phagocytosis of bacteria by leucocytes in the collected blood. In spiking experiments we have demonstrated that leucocytes in the collected whole blood act by removing bacteria which have been inoculated into the blood.” This is probably a very potent protective mechanism, active in certain bacterial species but not in others. During the last 3 years we have found in our university hospital three cases of TTBI transmitted by PC. In one of them pneumococci were the causing agent. Contamination from the donor’s blood seemed most likely. Bacteraemia had probably been present in the donor and the bacteria had not been killed either in vivo or in vitro by the antibacterial mechanisms in the blood. It seems very unlikely that pneumococci had been introduced from the environment at preparation. Bacillus cereus and Enterobacter aerogenes caused the other two complications. The likely causes were a leaky seal made by the Haemonetics 312 device in one case and a faulty plastic clamp, damaging the tubings, in the other. Yersinia enterocolitica has been the focus of interest in the US during recent years after a series of deaths caused by stored red cells. This species is very sensitive to complement and is rapidly phagocytosed when added to freshly collected whole blood.” However, intracellular killing in the phagocytes is sometimes insufficient and the likely mechanism of transfer from donor to recipient is via donor leucocytes disintegrating in the red cell unit after a few weeks of storage with bacterial release and growth. 24 Transmission of Yersinia via PC is highly unlikely and has not been reported to our knowledge.

Gram-negative rods like E. coli and Enterobacter species grow rapidly in PC, in spiking experiments to clinically dangerous concentrations within l-2 days. Staphylococcus epidermidis is regularly present on the human skin and is one of the most commonly encountered contaminants in PCs. It grows slowly and, in most of our spiking experiments, did not multiply above the initial bacterial concentration until after 4 days of storage. They are readily phagocytosed, but the procedure may take several species are hours.22,25 Propionibacterium present in normal skin and can be found in PCs at an incidence of l-4% (own observations, unpublished). They do not grow, however, because of the aerobic environment, since they are anaerobic. Thus, they lack clinical significance in context with platelet transfusion. As shown above, leaks in the system of plastic bags/tubings is a clearcut risk factor. Emphasis should be made on proper quality control, in this case inspection of each seal made with the sterile connector device and checking of all instruments after purchase. In a recent study we have demonstrated that a novel technique for blood culture, BacT/Alert (Organon Teknika, Brussels, Belgium), detected the presence of living bacteria within 6-16 h.25 Spiking experiments were performed with 6 strains of bacteria, S. epidermidis, S. aureus, Ps. aeruginosa, B. cereus, E. aerogenes and Str. sanguis, all of them candidates for PC contamination. The study indicates that this may be a way of improving the safety of platelet transfusions with respect to TTBI. Acknowledgements of the data referred to in this pa er have been obtained in collaboration with my co-war Rers Lars Eriksson and fian Gon to whom I want to ex ress my warm thanks. I am than a fuJ also to Hans Gulh I! sson for kindly sharing with me some of his as yet unpublished data.

Many

REFERENCES 1. Pflugshaupt R, Kurt G: FPA content-a criterion of quality for plasma as factor VIII source. VOX Sung 1983; 45:224-232.

Aspects of Platelet Storage

2. Bode AP: Platelet activation may explain the storage lesion in platelet concentrates. Blood Cells 1990; 16:109-126. 3. Griffin B, Bell K, Prowse C: Studies on the procurement of blood coagulation factor VIII: in viva studies on blood components prepared in half-strength citrate anticoagulant. VOX Sung 1988; 54:193198. 4. Solberg C, Holme S, Little C: Morphological changes associated with pH changes during storage of platelet concentrates in first generation 3-day container. VOXSang 1986; .50:71-77. 5. Kretschmer V, Biermann E, Loh H: Separation of platelet concentrates (PCs) from buffy coat (BC) using the bottom and top (BAT) drainage system. Trunsfus Sci 1990; 11:363-366. 6. Kilkson H, Holme S, Murphy S: Platelet metabolism during storage of platelet concentrates at 22°C. Blood 1984; 64:406-414. 0: The platelet 7. Wallvik J, Akerblom storage capability of different plastic containers. VOXSung 1990; 58:40-44. 8. Guppy M, Whisson ME, Sabaratnam R, Withers P, Brand K: Alternative fuels for platelet storage: a metabolic study. VOX Sung 1990; 59: 146-152. 9. Bertolini F, Murphy S, Rebulla P, Sirchia G: Role of acetate during platelet storage in a synthetic medium. Transfusion 1992; 32:152-156. 10. Murphy S: Preparation and storage of platelet concentrates, in Rossi EC, Simon TL, Moss GS (eds): Principles of Transfusion Medicine. Baltimore, Williams & Williams, 1991, pp.205-213. 11 Bertolini F, Rebulla P, Porretti L, Murphy S: Platelet quality after 15-day storage of platelet concentrates prepared from buff$ coats and stored in a glucosefree crystalloid medium. Transfusion 1992; 32:9-16. 12. Adams GA, Swenson SD, Rock G: Survival and recovery of human platelets stored for five days in a non-plasma medium. Blood 1986;67:672-675. 13. Holme S: Effect of additive solutions on platelet biochemistry. Blood Cells 1992; 18:421-430. 14. Gulliksson H, Sallander S, Pedajas I, Christenson M, Wiechel B: Storage of platelets in additive solutions: a new method for storage using sodium chloride solution. Transfusion 1992; 32:435440.

355

H: Storage of platelets in 15. Gulliksson additive solutions: the effect of citrate and acetate in in vitro studies. Trunsfusion 1993; 33:301303. 16. Jonsson S: Adding acetate to cell separator ACD allows for good hemostatic capacity and shelf-life of platelets to 12 days, convincingly supported by bacteriological culture. 9th Sci Congr Eur Sot Hemuphersis. Aberdeen, Scotland, 12-15 Sept, 1993, abstr.47. H, 17. Eriksson L, Shanwell A, Gulliksson Hijgman CF, Svensson LA, Kristensen J, Berg B: Platelet concentrates in an additive solution prepared from pooled buffy coat. In vivo studies. VOX Sung 1993; 64:133-138. 18. Kretschmer V, Schikor B, Sohngen D, Dietrich G: In vitro bleeding test-a simple method for the sensitive detection of the aspirin effect on platelet function. Thromb Res 1989; 56:593-602. J: 19. Hogman CF, Eriksson L, Kristensen Leukocyte-depleted platelets prepared from pooled buffy coat. Post-transfusion increment and ‘in vitro bleeding time’ using the Thrombostat 4000/2. Trunsfus Sci 1993; 14:3539. 20. Kristensen J, Eriksson L, Olsson K, Killander A, Hogman C: Functional capacity of transfused platelets estimated by the Thrombostat 400012. Eur \ Huemut. 1993; 51:152-155. 21. Goldman M, Blajchman MA: Blood product-associated bacterial sepsis. Trunsfus Med Rev 1991; 5:73-83. 22. Hogman CF, Gong J, Eriksson L, Hambraeus A, Johansson CS: White cells protect donor blood against bacterial contamination. Transfusion 1991; 31:620626. 23. Gong J, Hogman CF, Hambraeus A, Johansson CS, Eriksson L: Transfusiontransmitted Yersiniu enterocoliticu infection. Protection through buffy coat removal and failure of the bacteria to gorw in platelet-rich or platelet-poor plasma. VOXSung 1993; 65:42-46. 24. Hogman CF, Gong J, Hambraeus A, Johansson CS, Eriksson L: The role of white cells in the transmission of Yersiniu enterocoliticu in blood components. Transfusion 1992; 32:654-657. 25. Gong J, H@rnan CF, Lundholm M, Gustafsson I: Novel automated microbial screening of platelet concentrates. APMIS 1994; 102:72-78.