http://france.elsevier.com/direct/TRACLI/

Disponible en ligne sur www.sciencedirect.com

14 (2007) 514–521

Transfusion Clinique et Biologique

Original article

Use of random versus apheresis platelet concentrates

Utilisation des concentres de plaquettes issus

d’apherese ou de sang total

G. Andreu a,*, J. Vasse b, I. Sandid c, R. Tardivel b, G. Semana b

a Institut national de la transfusion sanguine, 6, rue Alexandre-Cabanel, 75739 Paris cedex 15, Franceb ETS de Bretagne, rue Pierre-Jean-Gineste, 35000 Rennes, France

c Agence française de sécurité sanitaire des produits de santé, Paris, France

Available online 15 April 2008

Abstract

The respective use of random (RPC) and apheresis (APC) platelet concentrates is highly heterogeneous among countries, ranging from 10 to 98%RPC in countries supposed to provide a similar transfusion service to patients. Moreover, when considering each country in the past 10 years, one canobserve that some have changed their policy, switching from a majority of APC to RPC or vice versa. This presentation intends to analyse which factorsmay impact such decisions. For many years, the only available platelet component was a RPC obtained from whole blood donation by a twocentrifugation steps process, the ‘‘platelet rich plasma’’ or PRP method. Since the beginning of the 1970s, APCs became available, with in fact manydifferent techniques leading to many APCs that may not be equivalent. Since the end of the 1980s, a new method of RPC preparation was developed,using the buffy-coat (BC–PC), providing a blood component with highly preserved platelet functions as compared to RPCs prepared by the PRPtechnique. Finally, the use of each of these components either native, or leuco-reduced, or suspended in a storage solution, or processed with a pathogeninactivation technique adds new layers of complexity to compare them. Innumerable references can be found in the literature describing in vitrofunctional parameters of platelet concentrates. Although it is clear that BC–RPC retain much more their in vitro functions than PRP–RPC, indicatingthat no one should use the latter any more, it is much more difficult to distinguish differences between other PCs. Conversely, only a very few studieshave been published related to a comparison of clinical efficacy of RPC versus APC, the endpoints being mainly CCI. Similarly to the in vitro studies,although RPC prepared with the PRP method show the lowest CCIs, no clear difference exists between ‘‘modern’’ RPC and APC. Another factor thatmay impact policy decision is the occurrence of adverse reactions in recipients. When considering only comparable data, for example leuco-reducedRPC versus leuco-reduced APC, there is now evidence that the latter is more associated with adverse reactions in recipients: data from hemovigilancein France show that, although no difference is noted for febrile non haemolytic transfusion reactions, nor for bacteria contamination, the incidence ofallergic adverse reactions is about four times higher with APC as compared with RPC. Other aspects may impact the decision: the fact that using APC inplace of RPC reduces the total donor exposure of patients was considered critical in some countries to reduce the risk of transmission of bloodtransmissible disease. Finally, the cost of the components, much higher for APC may be considered.# 2008 Elsevier Masson SAS. All rights reserved.

Résumé

L’utilisation respective des concentrés de plaquettes d’aphérèse (CPA) et des mélanges de concentrés de plaquettes standard (MCPS) est trèshétérogène selon les pays, variant de 10 à 98 % de MCPS dans des pays du Conseil de l’Europe ayant des systèmes de santé équivalents. De surcroît,si l’on considère chaque pays individuellement dans les dix dernières années, certains d’entre eux ont changé de structure de production soit versplus de CPA, soit vers plus de CPS. Cette revue tente d’analyser les facteurs qui peuvent être pris en compte dans le choix des produits plaquettaires.Pendant de nombreuses années, le seul produit plaquettaire disponible était les plaquettes issues de sang total obtenues par la méthode dite duplasma riche en plaquettes (PRP). Depuis le début des années 1970, les techniques de préparation de CPA ont été mises au point. Depuis la fin desannées 1980, une nouvelle technique de préparation des MCPS à partir de la couche leucoplaquettaire a été mise au point, qui permet de mieuxpréserver les fonctions des plaquettes. En dernier lieu, l’utilisation de ces produits sanguins labiles (PSL) soit natifs, soit déleucocytés, soit enfin ensolutions de conservation ou traité par un procédé de réduction des pathogènes ajoute des niveaux de complexité à leur comparaison. De trèsnombreux travaux ont été consacrés aux paramètres fonctionnels in vitro des concentrés de plaquettes. Autant il apparaît clairement une supériorité

* Corresponding author.E-mail address: gandreu@ints.fr (G. Andreu).

1246-7820/$ – see front matter # 2008 Elsevier Masson SAS. All rights reserved.doi:10.1016/j.tracli.2008.01.004

G. Andreu et al. / Transfusion Clinique et Biologique 14 (2007) 514–521 515

des MCPS préparés à partir de la couche leucoplaquettaire sur ceux préparés à partir de PRP indiquant que ces derniers ne devraient plus êtreutilisés, autant les différences entre les autres modes de préparation de concentrés de plaquettes ne sont pas aussi nettes. Un petit nombre d’études aété consacré à la comparaison des CPA et des MCPS, le critère principal étant en règle la recirculation plaquettaire corrigée par la surface corporelle(CCI). Là encore, les MCPS préparés selon la méthode PRP apparaissent inférieurs aux CPA, alors que ceux préparés selon la méthode de la coucheleucoplaquettaire leur sont comparables. Un autre facteur à prendre en compte est la survenue chez les receveurs d’effets indésirables. Encomparant les CPA et les MCPS préparés à partir de couches leucoplaquettaires, les premiers sont associés à une fréquence significativement plusélevée de réactions allergiques. D’autres aspects peuvent influencer la décision entre CPA et MCPS : le fait que l’utilisation de CPA permet deréduire le nombre de donneurs pour une même prise en charge thérapeutique et enfin le coût, beaucoup plus élevé pour les CPA.# 2008 Elsevier Masson SAS. All rights reserved.

Keywords: Platelet; Apheresis; Use; Adverse reaction

Mots clés : Plaquette ; Aphérèse ; Utilisation ; Effets indésirables

1. Introduction

The first successful attempt to obtain platelet concentratesfrom whole blood donation using the platelet rich plasma (PRP)technique was realized in the mid 1960s [1], and soon after,some developments lead to the PRP technique almost as it isstill performed in some countries [2]. About 10 years later, theonset of apheresis techniques, initially developed for granu-locytes collection [3], was extended to platelets [4]. Thefollowing main milestone in platelet processing for clinical useoccurred 10 years later again, when platelet concentrates wereobtained from whole blood using the buffy-coat technique [5].

Both PRP and BC platelet concentrates from whole bloodcontain a relatively small amount of platelets, ranging from 0.5to 0.8 1011 platelets. Four to eight units from different donorshave to be used for a single transfusion episode in an adultpatient, most frequently as a pool. Although it is not the purposeof this review to go into detailed blood processing techniques, adiagram published by Murphy [6] provides the maininformation on these two processes. As these PC have notbeen selected on a compatibility basis for the recipient, andhave been prepared by pooling components from donors atrandom (except for ABO group) they have been named‘‘random donor’’ PCs. In the following text, we shall therefore

Fig. 1. Proportion of RPC and APC used in co

consider platelet rich plasma random donor platelet concen-trates (PRP–RPC), and buffy-coat random donor plateletconcentrates (BC–RPC).

Conversely, apheresis platelet collection enables to collectenough platelets from a single donor to obtain a therapeutic unitfor an adult patient, and the donor can be selected on severalcompatibilities criteria, for example HLA or CMV negative,according to the patient’s characteristics. Although somedifferences may be observed according to the cell separatorused, this technique will be referred in this text as apheresisplatelet concentrates (APC).

Today, the respective use of random (RPC) and apheresis(APC) platelet concentrates is highly heterogeneous, rangingfrom 10 to 98% RPC in countries supposed to provide a similartransfusion service to patients, as shown on Fig. 1 [7].Moreover, when considering each country in the past 10 years,one can observe that some have changed their policy, switchingfrom a majority of APC to RPC or vice versa. For instance, inFrance, the use of RPC decreased until 2002, and since then isconstantly increasing (Fig. 2).

This review intends to analyse the factors that may help todecide which type of platelet component can be used as a firstline therapy, taking in account the developments in transfusionmedicine observed in the recent years.

untries of the Council of Europe in 2004.

Fig. 2. Evolution of use of RPC and APC in France from 1994 to 2007 (for 2007, values are a projection from the first six months data).

G. Andreu et al. / Transfusion Clinique et Biologique 14 (2007) 514–521516

2. Historical rationale for a preferential use of APC

When the first apheresis techniques became available, thefollowing arguments have been proposed to favour the use ofAPC.

Quality assessment of in vitro platelet functions were foundto be improved in APC as compared to PRP–PC [8], althoughother found no difference [9].

For an equivalent quantity of platelet transfusions, the use ofAPC requires less donors as compared to RPC (from four to 10times less according to the type of RPC prepared and theamount of platelets collected by apheresis). Therefore, one canexpect a reduction of disease transmission risk through bloodtransfusion. The main concern was initially to reduce the risk ofvirus transmission.

Fig. 3. Comparative study of BC–RPC and PRP–RPC for in vitro quality up to day 7concentration of 10 mg/ml), and the presence of GMP 140 (fluorescence intensity)

It had been also expected that the use of APC could lead tothe reduction of HLA immunization frequency, which was,before the introduction of leukoreduction techniques, extre-mely high in multitransfused patients with severe malignanthaematological diseases, with anti-HLA antibodies in at least30%, and up to 60% of the patients [10,11].

Some authors described that APC were less frequentlyinvolved than PRP–PC in adverse events related to bacteriacontamination [12,13].

3. Reappraisal of the rationale for a preferential use ofAPC

Most of the arguments that had been put forward in favour ofa preferential use of APC 10 to 20 years ago in countries such as

storage: pH, ATP release (in mM/1011 platelets), aggregation with collagen (at a. From reference [15].

G. Andreu et al. / Transfusion Clinique et Biologique 14 (2007) 514–521 517

France or Switzerland, were not supported by robust studiesproviding evidence that APC was actually superior to RPC. It istherefore useful to reconsider them in the light of our presentknowledge and the recent epidemiology situation.

3.1. In vitro and in vivo quality assessment of Plateletconcentrates

3.1.1. Comparison of PRP–RPC and BC–RPCInitially developed in the Netherlands [14], the preparation

of BC–RPC was quickly shown to have many advantages overPRP–RPC, with a quantity of platelets recovered from wholeblood improved by 20 to 30% and a reduced number of residualwhite blood cells after processing. Moreover, platelet in vitrofunctions have been shown to be much better preserved in BC–

RPC: as an example, after seven days of storage, ATP release,aggregation with collagen and GMP140 in BC–RPC werefound to be equivalent to PRP–RPC stored for one day [15].Fig. 3 illustrates these observations, that were confirmed byother studies [16].

In regard to platelet in vivo recovery a single study [17]using labelled platelets showed no difference between BC andPRP–RPC.

It has been also recognized that the use of PRP–RPC wasmuch more associated with adverse reaction in recipient thanBC–RPC [18,19,20], and their occurrence is strongly correlatedto the amount of IL-6 in PC [18]. In addition, no significantincrease in levels of TNF-alpha is observed in BC–PC duringstorage period, while levels increased significantly in PRP–PCon day 1 [21].

Finally, PRP–RPC preparation is fully manual, and has atleast one highly sensible step in term of platelet activation afterthe high speed centrifugation, while BC–RPC can be producedwith an automatic process with a high and reproducible plateletrecovery [22].

Having considered that BC–RPC are of better quality thanPRP, most European countries have adopted this technique [6].

3.1.2. Comparison of RPC and APCAmong the studies comparing BC–RPC and APC in regard

to recovery in patients as measured by corrected countincrement and interval between transfusion, statistically sign-ificant difference was never observed [20,23–26]. Conversely,in the published study comparing PRP–RPC and APC, there is astatistically significant difference in favour of APC [27,28].

In conclusion, there is now a bulk of data showing that ifthere is a superiority of APC versus PRP–RPC, no differencecan be found when APC is compared to BC–RPC.

3.2. HLA immunization

A single but highly informative study compared the extent ofHLA immunization in patients transfused with either PRP–

RPC or APC [29]. No statistical difference has been observedprovided that they are both leucoreduced: 3% of LR–RPCrecipients, 4% of LR-APC recipients, as compared with 13% ofnon leucoreduced RPC recipients.

It is now well established that the main factor leading toHLA immunization is the total amount of leucocytes in bloodcomponents. Therefore, although no study equivalent to theTRAP study exists with BC–RPC, we can speculate that theylead to at least equivalent results, as their leukocyte content isthe lowest among platelet products: according to the 2005results of the French data base for quality control of bloodcomponents, therapeutic units of BC–RPC contain a mean of26 � 103 leucocytes, and APC 40 � 103 leucocytes (Chabanel,personal communication).

We can conclude that there is now good evidence that theprevention of HLA immunization cannot be a criterion tofavour the use of APC as a first line therapy. Of course, the useof HLA compatible APC remains the best solution to provideefficient platelet transfusion in case of HLA immunizedpatients, but this situation concerns nowadays only a smallminority of patients.

3.3. Risk of HIV, HCV and HBV transmission

It is not questionable that the risk of HIV, HCV and HBVtransmission for a given patient is directly related to the numberof donors, and the promotion of APC versus RPC was a cleverinitiative according to the magnitude of transfusion transmittedHIV and HCV infections in the mid 1980s.

Since that time, it is important to note that in most countries,the risk calculated from actual blood donations data decreaseddramatically in the past 20 years, mainly due to the developmentof active policy for donor selection, and to a lesser extent to thedevelopment of highly sensitive detection techniques. In theFrench experience, between 1992 and 2005, the calculated riskhas been reduced by a factor 4.5 for HIV, 15 for HBVand 30 forHCV, to reach a residual risk of 1/2.6 million donations forHIV, 1/6.5 million donations for HCV and 1/1 milliondonations for HBV [30]. Accordingly, in the context of anactive post transfusion serologic surveillance policy for HCVand HIV, in the years 2000 to 2006, five transmissions have beendetected, 1 HCV in 2001, 2 HIV in 2001 and 2002, and 2 HBV in2001 and 2003.

We can conclude that the magnitude of the risk to transmitHIV, HCV and HBV decreased dramatically in the past 20years. However, as the risk has not disappeared, the rationalefavouring APC, although less stringent, may still be consideredand deserves to be analysed as a safety public health measure interm of cost-effectiveness.

3.4. Bacteria contamination

A great amount of work has been done in the past 10 yearsthat resulted in a drastic reduction of this risk.

In countries where bacteria detection has been introduced,one can expect that the relative risk of APC and RPC shouldconverge to similar levels.

In other countries where bacteria detection has not beenimplemented, we can now look to the results of hemovigilance.In France, for the years 2003–2004, the incidence of adverse

Table 1Volume of plasma in platelet concentrates

Platelet concentrate (adult therapeutic unit) Volume of plasma/PC Number of donors Volume of plasma/donor

Extremes (ml) Relative value Extremes (ml) Relative value

BC–RPC in additive solution 100–125 1 5 20–25 1BC–RPC in plasma 200–300 2.2 5 40–60 2PRP–RPCa 300–350 2.9 5 60–70 2.5APC in additive solution 150–200 2.2 1 150–200 7APC in plasma 300–450 3.3 1 300–450 15

a Values extrapolated from the data of Arnold et al. [27].

G. Andreu et al. / Transfusion Clinique et Biologique 14 (2007) 514–521518

reaction related to bacteria contamination is 1/50,368 RPC and1/ 60,591 APC, the difference being not significant [31].

Therefore, we can conclude that using modern whole bloodprocessing techniques which prevent bacteria contamination[32] and rigorous skin cleansing procedure before bloodcollection, there is no more matter to favour APC to preventadverse reaction related to bacteria contamination.

4. Additional information to consider in comparingRPC and APC

The introduction of leukoreduction in the 1990s and itsuniversal adoption in many countries, of platelet additivesolutions in the past 10 years, and eventually of pathogenreduction, all techniques applying to both APC and RPC, arenew inputs that must be considered in comparing the respectiveadvantages of RPC and APC. In the field of diseasetransmission, the vCJD case deserves to be considered. Finally,three other information should be considered and shall beshortly discussed: (i) the impact of preparing RPC on the meanhemoglobin content of red blood cell concentrates; (ii) thecomparison of adverse reactions in blood donors giving wholeblood or giving apheresis platelets, and the cost of these bloodcomponents.

4.1. Leukoreduction

The use of universal leukoreduction, implemented in agrowing number of countries, is a major decision to improvetransfusion safety, by at least reducing drastically the incidenceof HLA immunization [33], and adverse reactions in patients[20,34].

In the specific case of comparing different plateletconcentrates, we should avoid all studies involving nonleukoreduced components, due to the fact that not only thenon leukoreduced PRP–RPC have much more leukocytes thanBC–RPC, but also apheresis procedures are extremelyheterogeneous in term of initial leukocyte content.

4.2. Platelet additive solutions

Since the first attempts to use platelet additive solutions,many improvements have been done, and although in vivorecovery of platelets is slightly decreased, many advantages arerelated to their use [35]. Among them, a significant reduction

of allergic transfusion adverse event is clearly demonstrated[36–40].

Platelet additive solution is easy to implement for BC–RPCand cell separators have been adapted to use them as well.Conversely, they are not well adapted to PRP–PC, an additionalfact pleading for the suppression of this historical component.

Table 1 illustrates the plasma content of platelet concen-trates with and without platelet additive solution. Values arederived from the national blood components qualitycontrol database of EFS for the year 2005 (A Chabanel,personal communication), except for PRP–RPC, which are notprepared in France. PRP data have been calculated from theinformation provided by Arnold [27]. In clear, not only BC–

RPC total volume of plasma is the lowest by a factor 2 to 3, butthe volume of plasma from individual donors is reduced by afactor 7 to 15.

We can conclude that the inability to use additive solutionsfor PRP–RPC is an additional argument to ban them, and thatthe expected beneficial effect for patients of reducing plasmavolume from individual donors is much more marked in thecase of BC–RPC.

4.3. Adverse reactions in recipients

For many years, we had only a few information on theadverse reactions comparing RPC and APC. Moreover, manystudies compared leukoreduced with non leukoreduced plateletconcentrates and cannot be taken in account nowadays. Atpresent however, in countries like France, we can use data fromhemovigilance, assuming that the notification of adverse eventsis done in similar conditions for recipients of BC–RPC or APC,and that there is no specific clinical criteria for using BC–RPCor APC. This is the case in Brittany, and we presented theseresults recently [39].

Between 2003 and 2006, the blood transfusion centre ofBritanny had delivered 1,275 BC–RCP without additivesolution (referred as RPC AS�), 8,206 BC–RCP with additivesolution (referred as RPC AS+), 25,698 APC without additivesolution (referred as APC AS�), and 3,525 APC with additivesolution (referred as APC AS+). Apart from the use ofcompatible APC in HLA immunized patients, the criterion toselect APC or RPC was only the availability of an ABOcompatible transfusion. In the same way, components inadditive solution were used progressively as the productionshifted from plasma to additive solution.

Fig. 4. Frequency per 1000 platelet concentrates of allergic and febrile nonhemolytic adverse reactions observed with BC–RPC and APC without (PAS�)or with (PAS+) platelet additive solution.

G. Andreu et al. / Transfusion Clinique et Biologique 14 (2007) 514–521 519

Adverse reactions have been notified according to theorganization of the national hemovigilance network [41].Allergic adverse reactions were observed with two RCP AS�,two RCP AS+, 153 APC AS� and 11 APC AS+. Febrile nonhemolytic reactions were observed with one RPC AS�, 12 RPCAS+, 40 APC AS� and five APC AS+ (Fig. 4). These data shownot only the significant decrease of allergic reaction linked withadditive solution as already mentioned, but also a statisticallysignificant difference of allergic reactions between APC andRPC with or without additive solution, in favour of RPC:P = 0.04 without and P = 6.7 � 10�7 with additive solution.Conversely, similar incidence of febrile non haemolyticreactions is not surprising due to the leukoreduction.

Similar results have been observed at the French nationallevel in 2005 [40].

We can conclude that APC are more prone to elicit allergicadverse reaction in recipients than BC–RPC, and that the use ofthe later may reduce their incidence by a factor over three in thecase of components prepared in plasma, and by a factor over 10in the case of components prepared in additive solution.

4.4. Pathogen reduction

In the last 10 years, significant progress has been done, andat present one pathogen reduction process developed by Cerus(Intercept) is authorized in most European countries, a secondone developed by Navigant Biotechnology (Mirasol) is welladvanced in the process of marketing authorization and a thirdone developed by MacoPharma is on the way to reach theclinical evaluation stage.

One of the main interest of such technique is to be proactivein preventing the deleterious impact of the introduction of a newvirus or parasite, as it happened in the past few years with WestNile virus in North America and Chikungunya virus in theIndian Ocean. Moreover, as all the existing techniques areefficient on most bacteria species, they add a new level of safetyfor bacteria contamination.

As most of theses techniques involve a specific physico-chemical treatment, a long-term follow-up of recipients is

needed to ensure their full safety. Up to now, the growingexperience in large scale use of pathogen reduction did notrevealed any specific adverse reaction or event, and its use inmany countries is expected to increase in the next years.

The implementation of such techniques would also abolishthe advantage of APC over RPC in term of transfusion safetyand virus transmission.

4.5. Risk of vCJD

vCJD became a major public health concern worldwide inthe recent years, and the countries expected to have the mostimportant impact are UK and France. It is now clearlyestablished that the blood transfusion can be a way of human tohuman transmission of this disease [42].

Accordingly, the reduction of the number of donors for asimilar therapeutic unit was considered at least in France and inthe UK as a preventive measure, leading to recommend the useof APC in preference to RPC [43].

Pathogen reduction technologies adapted to the vMCJ agentare not enough developed today to expect an available processin the next few years. However, due to the better knowledge ofthe specific infectivity of each blood fraction on one side and onthe evolution of platelet production on the other side, the simplerecommendation to favour APC made in 2000 deserves to beevaluated again. Firstly, it is established that not all the fractionsof blood may contain the pathologic prion: it is not found inplatelet per se, but is found mostly in leukocytes and in plasma[44]. Secondly, these recommendations had been done beforethe implementation of platelet additive solutions.

Therefore, nowadays, the question to which experts shouldanswer in comparing RPC and APC for their respective risk totransmit vMCJ is the following: is a product prepared from asingle donor containing 150 to 200 ml of plasma and 40,000leukocytes safer than a product from five donors containingfrom each one 20 to 25 ml of plasma and 5000 leukocytes.

4.6. Impact of BC–RPC on hemoglobin content of redblood cell concentrates

Preparing BC–RPC is not neutral in regard to thehemoglobin content of the corresponding red blood cellconcentrate. Again, the National database of blood componentsquality control of EFS may provide a fair evaluation. In 2005,red blood cell concentrates prepared from whole bloodfiltration, that is, with no BC processing contained a meanof 57 g of hemoglobin. Conversely, red blood cell concentratesprepared along with BC processing contained a mean of 51 g ofhemoglobin [45]. Although the two types of productionsprovide components complying with the generally acceptednorm of a minimum of 40 g of hemoglobin, there is no doubtthat overall, a red cell therapeutic unit prepared along with aBC–RPC contain 10% less hemoglobin than the equivalentproduct with no BC–RPC processing.

The overall consequences of such a difference is difficult toassess in practice, not only due to the difficulty in organizing aspecifically designed trial, but also related to the fact that, as a

G. Andreu et al. / Transfusion Clinique et Biologique 14 (2007) 514–521520

counterpart, red blood cell concentrates prepared according tothe buffy-coat technology contain significantly less residualleukocytes after leukoreduction, with a median of 24,000leukocytes per unit as compared to 77,000 [45].

4.7. Adverse reactions in blood donors

Although apheresis donation shares many adverse reactionswith whole blood donation, some reactions such as thoserelated to citrate toxicity are specific to apheresis.

Data from the literature suggests that the risk of reactionsrequiring hospitalization is greater in case of apheresis [46].Similar data have been described in the French nationalexperience of donor hemovigilance: among whole blooddonations 70/million required a medical consultation or a stayin hospital, as compared to 107/million apheresis donations(P = 0.02) [47].

We can conclude that in term of donor safety, there is noargument to promote apheresis donation if an equivalentcomponent is already available by whole blood processing.

4.8. Cost of APC and RPC

4.8.1. Cost versus pricing policyAn interesting fact is that there is no parallelism between the

actual cost and the pricing policy. UK and France may beregarded as two opposite examples of pricing policy for plateletconcentrates. In France a complex calculation involving amongother the actual platelet content and the number of units pooledlead to different prices for a same quantity of platelets in RPC orAPC – for a content of 3 � 1011, a RPC costs 217 s and an APCcosts 314 s – while in the UK, the same therapeutic unit costsca 300 s, whatever the type (APC or RPC) of preparation is.

It is clear that the actual cost of preparing plateletconcentrates is more important for APC. However, it may bedifficult to evaluate accurately what is the magnitude of thedifference, as the cost accounting methodologies can differaccording to the institutions and/or the countries.

4.8.2. Financial impact of APC/RPC policy in FranceIn France for the year 2006, EFS delivered 189,243 APC

containing a mean of 4.5 � 1011 platelets and 42,590 BC–RPCcontaining a mean of 3.9 � 1011 platelets, at the approximatecost of 100.5 million s, (89 for APC and 11.5 for BC–RPC).Overall, there was 1,030 � 1014 platelets used at the mean priceof 97 s per 1011 platelets.

We know that even if we estimate that there is no reason toadopt APC as a first line component, there is a remaining needfor APC to maintain a pool of HLA typed donors that we canestimate at 20% of the total platelet need. In such a hypothesis,the overall cost of platelet transfusion would be 75.5 million s,that is, a cost of 73 s per 1011 platelets. Therefore, in a costeffectiveness analysis, the overall difference of 25 millions srepresents the cost to be balanced by the expected advantages ofthe present policy of APC upon RPC. Moreover, in thehypothesis of implementing at the country level a pathogenreduction technique at an individual cost of 60 s would cost in

the latter hypothesis 14.5 million s, leading to a decrease of10.5 million s per year in the expenses for platelet transfusionas compared to the present situation.

5. Conclusion

In the past 20 years, many changes occurred in bloodtransfusion, leading to reassess the respective advantages ofAPC and RPC.

The development of BC–RPC technology is a major step inplatelet processing from whole blood that provides so manyimprovements over the PRP–RPC methods introduced 40 yearsago, that we can reasonably conclude the latter should bebanned from modern blood transfusion services.

Among all the advantages initially attributed to APC, theonly one that remains today is the reduction of donor exposure,even though its strength is weaker due to the dramaticreduction of known viruses transmission risk. It is noteworthythat in the perspective of the generalization of pathogenreduction technology, its impact would be restricted to thevCJD risk.

Regarding recipients adverse reactions, all recent dataprovided by specifically addressed studies and hemovigilancenetwork as well, show an advantage for BC–RPC. Moreover, onthe donor side, there is at least no advantage for APC.

It is clear that the use of APC has the consequence to providered blood cell concentrates containing ca 10% less hemoglobin,with the potential disadvantage that they contain significantlymore residual leukocytes after the leukoreduction process [45].

We did not consider in this review the immunomodulatoryeffect of platelet transfusion, due to the lack of studyspecifically designed to compare RPC and APC. In this matteragain, the balance is probably between the donor exposure onone side, and the quantity of the blood fractions that may play arole, mainly leukocytes and plasma from individual donors.

Finally, looking at the cost of production not as a simple costaccounting exercise, but as the neat invoice to the public healthorganization for platelet transfusions, provides the major part ofthe numerator of a classical cost-benefit study, which remainsand deserves to be done due to the changing environmentdescribed in this review.

References

[1] Freireich EJ, Kliman A, Gaydos LA, et al. Response to repeated platelettransfusion from the same donor. Ann Intern Med 1963;59:277–87.

[2] Mourad N. A simple method for obtaining platelet concentrates free ofaggregates. Transfusion 1968;8:48.

[3] McCredie KB, Freireich EJ, Hester JP, Vallejos C. Leukocyte transfusiontherapy for patients with host-defense failure. Transplant Proc Sep1973;5(3):1285–9.

[4] Ts’ao CH, Wirman JA, Ruder EA. Altered in vitro functions of plateletsprepared by the haemonetics blood processor. J Lab Clin Med Aug1975;86(2):315–25. Links.

[5] Pietersz RN, Reesink HW, Dekker WJ, Fijen FJ. 1: preparation ofleukocyte-poor platelet concentrates from buffy coats. II. Lack of effecton storage of different plastics. Vox Sang 1987;53(4):208–13.

[6] Murphy S. Platelets from pooled buffy coats: an update. Transfusion2005;45:634–9.

G. Andreu et al. / Transfusion Clinique et Biologique 14 (2007) 514–521 521

[7] Van der Poel C. The use of labile blood components in relation to optimaluse and risk. EuroSAT meeting, Paris, October 4, 2007.

[8] Sloand EM, Yu M, Klein HG. Comparison of random-donor plateletconcentrates prepared from whole blood units and platelets prepared fromsingle-donor apheresis collections. Transfusion 1996;36:955–9.

[9] Turner VS, Hawker RJ, Mitchell SG, Symour Mead AM. Paired in vivoand in vitro comparison of apheresis and ‘‘recovered’’ platelet concen-trates stored for 5 days. J Clin Apheresis 1994;9:189–94.

[10] Eernisse JG, Brand A. Prevention of platelet refractoriness due to HLAantibodies by administration of leukocyte-poor blood components. ExpHematol Jan 1981;9(1):77–83.

[11] Andreu G, Dewailly J, Leberre C, Quarre MC, Binet ML, Tardivel R, et al.Prevention of HLA immunization with leukocyte-poor packed red cellsand platelet concentrates obtained by filtration. Blood 1988;72(3):964–9.

[12] Ness PM, Campbell-Lee SA. Single donor versus pooled random donorplatelet concentrates. Curr Opin Hematol 2001;8:392–6.

[13] Ness P, Braine H, King K, et al. Single-donor platelets reduce the risk ofseptic platelet transfusion reactions. Transfusion 2001;41:857–61.

[14] Piertersz RN, de Korte D, Reesink HW, et al. Preparation of leukocyte-poor platelet concentrates from buffy-coats. III. Effect of leukocytecontamination on storage conditions. Vox Sang 1988;55:14–20.

[15] Bertolini F, Rebulla P, Porretti L, Murphy S. Platelet quality after 15 daystorage of platelet concentrates prepared from buffy coats and stored in aglucose-free crystalloid medium. Transfusion 1992;32:9–16.

[16] Fijnheer, Pietersz RNI, de Korte D, et al. Platelet activation duringpreparation of platelet concentrates: a comparison of the platelet-richplasma and the buffy coat methods. Transfusion 1990;30:634–8.

[17] Keegan T, Heaton A, Holme S, et al. Paired comparison of plateletconcentrates prepared from platelet-rich plasma and buffy coats usinga new technique with 111 and 51 Cr. Transfusion 1992;32:113–20.

[18] Muylle L, Wouters E, Peetermans ME. Febrile reactions to platelettransfusion: the effect of increased interleukin 6 levels in concentratesprepared by the platelet-rich plasma method. Transfusion 1996;36(10):886–90.

[19] Oksanen K, Ebeling F, Kekomaki R, Elonen E, Sahlstedt L, Volin L, et al.Adverse reactions to platelet transfusions are reduced by use of plateletconcentrates derived from buffy coat. Vox Sang 1994;67:356–61.

[20] Anderson NA, Gray S, Copplestone JA, et al. A prospective randomizedstudy of three types of platelet concentrates in patients with haematolo-gical malignancy: corrected platelet increments and frequency of nonhaemolytic febrile transfusion reactions. Transfus Med 1997;7:33–9.

[21] Chaudhary R, Aggarwal A, Khetan D, Dayal R. Cytokine generation instored platelet concentrate: comparison of two methods of preparation.Indian J Med Res 2006;124:427–30.

[22] Larsson S, Sandgren P, Sjodin A, et al. Automated preparation of plateletconcentrates from pooled buffy coats: in vitro studies and experiences withthe OrbiSac system. Transfusion 2005;45:743–51.

[23] Strindberg J, Berlin G. Transfusion of platelet concentrates – clinicalevaluation of two preparations. Eur J Haematol 1996;57:307–11.

[24] Ericksson L, Kristensen J, Olsson K, et al. Evaluation of platelet functionusing the in vitro bleeding time and corrected count increment of trans-fused platelets. Comparison between platelet concentrates derived frompooled buffy coats and apheresis. Vox Sang 1996;70:69–75.

[25] Akkök CA, Brinch L, Lauritzsen GF, Solheim BG, Kjeldsen-Kragh J.Clinical effect of buffy-coat vs. apheresis platelet concentrates in patientswith severe thrombocytopenia after intensive chemotherapy. Vox Sang Jul2007;93(1):42–8.

[26] Vasse J, Tardivel R, Gaucheron S, Andreu G, Semana G. Surveillancestudy of the transfusion efficiency after transfusions of apheresis plateletconcentrates and pooled platelet concentrates. ISBT regional meeting,Madrid, July 7, 2007.

[27] Arnold DA, Heddle NM, Kulczycky M, et al. In vivo recovery and survivalof apheresis and whole blood-derived platelets: a paired comparison inhealthy adults. Transfusion 2006;46:257–64.

[28] Gurkan E, Patah PA, Saliba RM, Ramos CA, Anderson BS, Champlin R,et al. Efficacy of prophylactic transfusions using single donor apheresis

platelets versus pooled platelet concentrates in AML/MDS patientsreceiving allogeneic hematopoietic stem cell transplantation. Bone Mar-row Transplant 2007;1–4.

[29] The Trial to Reduce Alloimmunization to Platelets Study Group. Leuko-cyte reduction and ultraviolet B irradiation of platelets to prevent alloim-munization and refractoriness to platelet transfusions. N Engl J Med1997;337:1861–9.

[30] Pillonel J, Le Marrec N, Girault A, David D, Laperche S. Epidemiologicalsurveillance of blood donors and residual risk of blood-borne infections inFrance, 2001 to 2003. Transfus Clin Biol Jul 2005;12(3):239–46. Links.

[31] Rebibo D, Hauser L, Zorzi P, Legras JF, Afssaps, Slimani A, et al.International forum: hemovigilance. Vox sang 2006;90(3):207–41.

[32] Högman CF, Gong J, Eriksson L, Hambraeus A, Johansson CS. Whitecells protect donor blood against bacterial contamination. Transfusion Sep1991;31(7):620–6.

[33] Andreu G, Dewailly J. Prevention of HLA alloimmunization by usingleukocyte-depleted components. Curr Stud Hematol Blood Transfus1994;29–40.

[34] Enright H, Davis K, Gernsheimer T, McCullough JJ, Woodson R, SlichterSJ. Factors influencing moderate to severe reactions to PLT transfusions:experience of the TRAP multicenter clinical trial. Transfusion2003;43:1545–52.

[35] Gulliksson H, AuBuchon JP, Cardigan R, et al. Storage of platelets inadditive solutions: a multicentre study of the in vitro effects of potassiumand magnesium. Vox Sang 2003;85:199–205.

[36] Kerkhoffs JL, Eikenboom JC, Schipperus MS, van Wordragen-VlaswinkelRJ, Brand R, Harvey MS, et al. A multicenter randomized study of theefficacy of transfusions with platelets stored in platelet additive solution IIversus plasma. Blood 2006;108:3210–5.

[37] de Wildt-Eggen J, Nauta S, Schrijver JG, van Marwijk Kooy M, Bins M,van Prooijen HC. Reactions and platelets increments after transfusion ofplatelet concentrates in plasma or an additive solution: a prospective,randomised study. Transfusion 2000;40:398–403.

[38] Snyder E, McCullough J, Slichter SJ, Strauss RG, Lopez-Plaza I, Lin JS,et al. Clinical safety of platelets photochemically treated with amotosalenHCl and ultraviolet A light for pathogen inactivation: the SPRINT trial.Transfusion 2005;45:1864–75.

[39] Andreu G, Vasse J, Herve F, Tardivel R, Semana G. Introduction of plateletadditive solutions in transfusion practice. Advantages, disadvantages andbenefit for patients. Transfus Clin Biol 2007;14:100–6.

[40] Rebibo D, Simonet M, Slimani A. Étude nationale sur les effets indésira-bles receveurs liés aux concentrés de plaquettes avec et sans solution deconservation. Transfus Clin Biol 2007;14:213–4.

[41] Andreu G, Morel P, Forestier F, Debeir J, Rebibo D, Janvier G, et al.Haemovigilance network in France: organization and analysis of imme-diate transfusion incident reports from 1994 to 1998. Transfusion2002;42:1356–64.

[42] Llewelyn CA, Hewitt PA, Knight RSG, Amar K, Cousens S, Mackenzie J,et al. Possible transmission of variant Creutzfeldt-Jakob disease by bloodtransfusion. Lancet 2004;363:417–21.

[43] Alpérovitch A, Barin F, Bégaud B, Béloucif S, Billette de Villemeur T,Brown P, et al. Analyse du risque de transmission de la nouvelle variantede la maladie de Creutzfeldt-Jakob par le sang et ses dérivés. Recom-mandations, Agence française de sécurité sanitaire des produits de santé,11 décembre 2000.

[44] Brown P, Rohwer RG, Dunstan BC, MacAuley C, Gajdusek DC, DrohanWN. The distribution of infectivity in blood components and plasmaderivatives in experimental models of transmissible spongiform encepha-lopathy. Transfusion 1998;38:810–6.

[45] Chabanel A, Carrat F, Begue S, Masse M, Perrault MP, Andreu G. Qualityof leucoreduced red blood cell concentrates: 5 years of follow-up inFrance. Vox Sang 2008;97:41–7.

[46] Winters JL. Complications of donor apheresis. J Clin Apheresis Jul2006;21(2):132–41.

[47] Rebibo D, Hauser L, Simonet M, Slimani A. Hémovigilance donneur :aspects nationaux et internationaux. Transfus Clin Biol 2007;14:173.