Transfusion and Apheresis Science 40 (2009) 99–103

Contents lists available at ScienceDirect

Transfusion and Apheresis Science

journal homepage: www.elsevier .com/ locate/ t ransc i

Comparative study of automated plateletpheresis using five differentapheresis systems in a tertiary care hospital

Rajendra Chaudhary *, Sudipta Sekhar Das, Dheeraj Khetan, Shashank Ojha, Sunil VermaDepartment of Transfusion Medicine, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Raibareli Road, Lucknow 226014, India

a r t i c l e i n f o

Keywords:PlateletpheresisCell separatorDonor safetyPlatelet yield

1473-0502/$ - see front matter � 2009 Elsevier Ltddoi:10.1016/j.transci.2009.01.016

* Corresponding author. Tel.: +91 522 2668700; faE-mail address: rkcchaud@sgpgi.ac.in (R. Chaudh

a b s t r a c t

Different types of cell separators are available nowadays based on either continuous orintermittent flow technology to meet the growing demands for single donor apheresisplatelets. This prospective study compares the five machines used in our center withregard to procedure parameters, product quality and adverse effects on the donor.

A total of 477 plateletpheresis on various machines were performed on eligible donorsover a period of 28 months after taking informed consent. All procedures were performedfollowing the departmental standard operating procedure (SOP) and manufacturer’sinstructions. All donor and procedure related details were obtained from the procedureregister. Statistical analysis was done using the SPSS statistical package (version 12, USA).

The median age, BSA and BMI of our plateletpheresis donors were calculated to be29 years, 1.69 m2 and 23.1 kg/m2, respectively. Analyzing the parameters related to donorcomfort such as donation time (DT), needle time (NT) and processing time (PT), the MCSmachines were not ‘‘donor friendly” compared to Amicus and Fresenius. Platelet yield byAmicus was significantly higher as compared to other cell separators (p < 0.05). Platelet-pheresis associated citrate toxicity was higher with the Amicus and MCS 3p and vasovagalside-effects was observed least with the CS 3000 machine. Though, quality of apheresisproduct in terms of yield is comparable with all the machines, there are differences inthe systems with regard to donor safety, procedure time and donor retention.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

There has been an increasing trend towards the use ofapheresis platelets to support thrombocytopenic patients.Plateletpheresis procedures are thought to be generallysafe without serious complications to the donor and theproducts obtained are preferred over platelets derivedfrom whole blood due to several advantages, such as lim-ited donor exposure and transmission of infections [1–3].

Different types of plateletpheresis machines are avail-able today based on either continuous or intermittent flowtechnology. Although, the principle of separation in thesecell separators is generally same, various technical differ-

. All rights reserved.

x: +91 522 2668017.ary).

ences may affect the quality of the product obtained andtheir adverse effects on the donor [4].

At our center, plateletpheresis is being performed usingfive machines, based on either continuous flow or intermit-tent flow technology. This prospective study was carriedout to compare these machines with regard to procedureparameters, product quality and adverse effects on thedonor.

2. Materials and methods

2.1. Donors and procedures

This prospective study included 477 plateletpheresisprocedures performed on eligible donors aged between19–55 years over a period from May 2005 to August2007 after taking informed consent. All procedures were

100 R. Chaudhary et al. / Transfusion and Apheresis Science 40 (2009) 99–103

performed by the same team of resident doctors underprophylactic calcium supplementation at the dose of250 mg elemental calcium orally before the procedure.The numbers of plateletpheresis procedures performedon each machine are as follows:

1. Fenwal CS3000, version 10.8 (Baxter Healthcare, RoundLake, IL, USA), TNX-6 (chamber), interface offset 6:n = 117.

2. Fenwal Amicus separator, version 2.5, double needle(DN) (Baxter Healthcare Corporation, Deerfield, IL,USA): n = 133.

3. Fresenius separator (COM.TEC), version 4.00.xx, DN(Fresenius HemoCare GmbH, Bad Homburg v.d.H,Germany): n = 63.

4. Haemonetics MCS3p separator (Hemonetics Corpora-tion, Braintree, Massachusetts, USA): n = 97.

5. Haemonetics MCS+ separator (Hemonetics Corporation,Braintree, Massachusetts, USA): n = 67.

All plateletpheresis procedures were performed follow-ing the departmental standard operating procedure (SOP)using closed system apheresis kits and ACD-A anticoagu-lant in the proportion of 1:10–1:12 [5]. The end point ofeach procedure was based on the target yield of 3 � 1011

platelets per unit, maintaining blood flow rate for all col-lections at 50–80 ml/min depending on the machine. Noneof the machines used in present study had on-line leukore-duction filters. Donor’s physiognomic details like age, gen-der, weight and height and plateletpheresis proceduredetails, such as blood volume processed, total anticoagu-lant used, time taken, flow rate and collection efficiencyof machines were recorded for each procedure from theprocedure register. Any adverse effects on donors such ashypocalcaemia and vasovagal reactions were recordedand graded following departmental SOP.

2.2. Sampling

2.2.1. Donor’s pre-donation hematological valuesTo measure the pre-donation hematological values of

all donors as a part of the screening process, whole blood(WB) samples in EDTA vials were collected just beforethe procedure and parameters such as platelets (PLT),hemoglobin (Hb), hematocrit (Hct), WBC count weremeasured.

2.2.2. Sampling from unitApproximately 1 ml samples from each bag were col-

lected in EDTA (K2 EDTA) after thoroughly stripping thesegment to ensure a representative product of the bag.The samples were then mixed thoroughly by means ofmechanized blood mixer (Techno FAB, Baroda, India) for15 min and then subjected to measurements of parameterssuch as pH, Hct, PLT count, RBC and WBC counts.

2.2.3. MeasurementsHematological values of both donor and bag samples

were obtained using a routinely calibrated automated cellcounter (Micros 60, ABX diagnostics, France). The pH of allplatelet units were measured by a calibrated portable pH

meter (Toshniwal inst. mfg. Pvt. Ltd, Ajmer, India) follow-ing the departmental SOP. Residual WBC content of thebag was measured by Nageotte chamber hemocytometercounts [6]. Swirling in platelet units was assessed visuallyand documented as ‘present’ or ‘absent’.

2.3. Statistical analysis

Statistical analysis was done using the SPSS statisticalpackage (version 12, USA). All results were calculated asmean ± SD and a ‘p’ value of <0.05 was considered statisti-cally significant. Mean values were compared using the un-paired or paired Student’s t test as appropriate. Definitions,abbreviations and formulae used in the study to obtain do-nor and procedure details are described in Table 1.

3. Results

The median age of our plateletpheresis donors was29 years with a male to female ratio of 19:1. The medianBSA and BMI were calculated to be 1.69 m2 and 23.1 kg/m2 respectively. Mean pre-apheresis platelet count of thedonor population was 207.8 � 106/ml with a mean Hb va-lue of 13.9 gm/dl (Table 2).

Table 3 describes the details of plateletpheresis proce-dures which also reflects the safety and comfort of the do-nors. Analyzing DT, NT and PT the MCS machines were theslowest compared to Amicus and Fresenius (p < 0.01). TheAmicus was even faster than the Fresenius in terms ofMDR and MRR, but this was not statistically significant(p > 0.05). The volume of whole blood processed byCS3000 was significantly higher when compared to theother four machines (p < 0.05).

The mean volume of products obtained by CS 3000(210.6 ml) was significantly lower than the volume ob-tained by the other machines (p < 0.05) with a net plateletyield of 2.91 � 1011 which is comparable to the MCS andFresenius. Platelet yield by Amicus was significantly higheras compared to other cell separators (p < 0.05). WBC con-tamination in the unit was lowest with the Fresenius com-pared with others (Table 4). Plateletpheresis associatedcitrate toxicity was higher with the Amicus and MCS 3pand vasovagal side-effects was observed least with the CS3000 machine. But none of these comparative calculationswere statistically significant (p P 0.193) (Table 5).

4. Discussion

Plateletpheresis is now becoming a routine procedurein most of the blood centers in developing countries. Bloodcenters use different cell separators, each having its ownmechanism of operation, needing frequent validation to as-sure production of acceptable products. We at our centerperform plateletpheresis procedures using five cell separa-tors. The majority of plateletpheresis donors in the presentstudy were males (Table 1) as females were mostly de-ferred due to either low hemoglobin or poor venous access.However the representation of female donors can be in-creased if the Hb cutoff is lowered from the existing crite-ria of 12.5 gm/dl without compromising donor safety as

Table 1Definitions, abbreviations and formulae used in the present study.

Parameters Abbreviation SI unit Definitions/formulae

Body surface area [20] BSA m2 pheight (cm) �weight (kg)/3600

Body mass index [12] BMI Kg/m2 weight/height2

Total blood volume [21] TBV ml Male(M):2740 ml/m2, Female(F):2370 ml/m2

Platelet mass(�1011) [22] PLT-M – (PLT count � 106/ml) � (TBV in ml)Pre-donation time [12] p min Time for kit installation, priming & programmingDonation time [12] DT min Time from start until reinfusionNeedle time [12] NT min DT + reinfusion time + post-procedure sample collection timeProcessing time [12] PT min Time from set up to final product (PT) = NT + pWhole blood processed WBP ml Volume processed from start until reinfusion – citrate volumeMaximum draw rate MDR (ml/min) Highest rate of WB withdrawal from donor in a procedureMaximum return rate MRR (ml/min) Highest rate of blood infused to donor during a procedureTotal ACD used ACD ml Total ACD consumed during the procedure as displayed by machineCollection efficiency of machine [5] CE % PLT yield/total PLT processed � 100

Table 2Physiognomic and pre-procedure haematimetric values of donors(N = 477).

Donor characteristic Values

Age (range) 29 (19–55)Gender (M:F) 19:1Weight (kg) 67.4 ± 10.45Height (m) 1.63 ± 2.9BSA (m2) 1.69 ± 1.6BMI (kg/m2) 23.1 ± 1.1TBV(ml) M:4658 ± 437, F:3807 ± 419Platelet count (�106/ml) 207.8 ± 40.7PLT-M (�1011) 9.1 ± 1.59WBC count (�106/ml) 6.8 ± 0.9Hb (g/dl) 13.9 ± 1.3Hct (%) 40.2 ± 3.6

All physiognomic values of donor expressed as median ± SD.All haematimetric values expressed as mean ± SD.

R. Chaudhary et al. / Transfusion and Apheresis Science 40 (2009) 99–103 101

suggested by others [7,8]. Low BSA and BMI of our donorpopulation might be attributed to the natural and regionalvariations of the physiognomic and demographic charac-teristics of the donors. Also, pre-apheresis haematemetricvalues of donors such as PLT count and hemoglobin werein the lower normal side which has already been reportedfrom our center [9].

Table 3Procedure characterization of automated plateletpheresis (N = 477).

Parameters CS 3000 MCS 3p(n = 117) (n = 97)

Vein access double singleACD:WB 1:10 1:12ACD 304.3 ± 57.7a,b 274.6 ± 38.1MDR 60 50MRR 60 50DT 80.8 ± 17.3a,b 96.8 ± 15.6NT 94.4 ± 16.9a,b 105.7 ± 15.1PT 119.7 ± 15.7 122.4 ± 14.3WBP 3667.1 ± 698.8a,b,d,e 2965.1 ± 357.9

All values expressed as mean ± SD.a Significant compared with MCS 3P.b Significant compared with MCS+.c Significant compared with CS 3000.d Significant compared with Amicus.e Significant compared with Fresenius.

All machines used in the present study consistently pro-vided optimized platelet products. The CS 3000 processed asignificantly higher volume of whole blood to obtain thetarget yield of P3 � 1011 PLT per unit with a mean collec-tion efficiency of 56.3% which is lower compared to theother devices. This can be attributed to the older versionof the machine, because newer machines were more effi-cient in collecting platelets with the Amicus having a max-imum mean CE 66.9%. This observation also compliedfavorably with previous data [10–12]. However CE valuediffers from center to center and might not be directlycomparable because the donor population varies and PLTyield and post-procedure PLT count are other factors thatinfluence the CE making this variable not really suitablefor comparison of apheresis systems [12]. However wedon’t expect such variations in the current unicentricstudy.

In context to the DT, NT and PT of various cell separa-tors used, the MCS systems were the slowest of all devices.These differences might be because of their intermittentflow technique and slow flow rate of 50–60 ml/min. AsDT and NT directly concern the apheresis donors thereforethese are also important parameters to evaluate the safetyand comfort of healthy donors.

Though, plateletpheresis is a safe procedure withoutsignificant complications, at times symptoms due to citrate

MCS+ Amicus Fresenius(n = 67) (n = 133) (n = 63)

single double double1:12 1:10 1:10281.3 ± 47.3 319.4 ± 65.8a,b,c 308 ± 69.9a,b

50 80a,b 70a,b

50 80a,b 70a,b

100.3 ± 24.3 61.6 ± 14.8a,b,c 67.3 ± 13.5a,b,c

109.9 ± 22.7 68.2 ± 12.8a,b,c 78.1 ± 13.3a,b,c

126.5 ± 22.1 83.1 ± 11.7a,b,c 92.6 ± 11.9a,b,c

2917.5 ± 521.1 2956.7 ± 653.3 3169.6 ± 837.1

Table 4Productivity of various aphaeresis systems for plateletpheresis (N = 477).

Parameters CS 3000 MCS3p MCS+ Amicus Fresenius(n = 117) (n = 97) (n = 67) (n = 133) (n = 63)

Volume (ml) 210.6 ± 20.6a,b,d,e 274.5 ± 29.3 269.8 ± 23.7 256.5 ± 43.7a,c 259.4 ± 19.9pH 7.1 ± 0.43 7.03 ± 0.46 7.07 ± 0.47 7.05 ± 0.51 7.03 ± 0.39Yield (�1011) 2.91 ± 0.75 2.88 ± 0.81 2.81 ± 0.72 3.43 ± 1.1a,b,c,e 3.08 ± 1.4b

WBC (�108) 5.13 ± 0.53 6.17 ± 0.71 4.7 ± 0.66 2.7 ± 0.79 a,b,c 1.9 ± 0.64a,b,c

Hct (%) 0.4 ± 0.34 0.4 ± 0.41 0.5 ± 0.39 0.3 ± 0.29 0.3 ± 0.37CE 56.3 ± 13.6 59.7 ± 8.6 59.3 ± 8.9 66.9 ± 5.2c 63.6 ± 4.9

All values expressed as mean ± SD.a Significant compared with MCS 3P.b Significant compared with MCS+.c Significant compared with CS 3000.d Significant compared with Amicus.e Significant compared with Fresenius.

Table 5Adverse reactions encountered in donors during plateletpheresis (N = 477).

Reactions n (%) CS3000 MCS3p MCS+ Amicus Fresenius Total(n = 117) (n = 97) (n = 67) (n = 133) (n = 63)

Citrate toxicity 11 (9.4) 11 (11.3) 6 (8.9) 15 (11.3) 6 (9.5) 49 (10.2)Hematoma 5 (4.3) 4 (4.2) 3 (2.6) 6 (4.5) 3 (4.7) 21 (4.4)Vasovagal 3 (2.6) 4 (4.2) 3 (2.6) 6 (4.5) 3 (4.7) 19 (4.0)

Grade I 3 2 3 5 3 16 (3.4)Grade II 0 2 0 1 0 3 (0.6)Grade III 0 0 0 0 0 0

Percent (%) 16.2 19.6 17.9 20.3 19.0

102 R. Chaudhary et al. / Transfusion and Apheresis Science 40 (2009) 99–103

toxicity may be discomforting to the donor and thereforeearly recognition and its prevention is important toencourage repeat donations. We observed such symptomsin 10.2% of our donors, which is in accordance with previ-ous studies [13–15]. Prophylactic calcium supplementa-tion has been recommended to prevent this complicationby many workers and that has also been implemented atour center since May 2006 [14,15]. ACD-A load in donorsundergoing Amicus plateletpheresis was significantlyhigher compared to other machines. Despite prophylacticcalcium supplementation, the Amicus caused the highestnumbers of citrate-related reactions. Though this observa-tion was not statistically significant however rapid drawrate as well as return rate of citrated blood to donors mighthave contributed to these side-effects.

The mean volume of final product was higher in theMCS systems and this might be due to increased numberof cycles required to achieve desired yield of 3 � 1011/unitor the poor concentration capacity of the MCS systems.However this increased plasma volume is of little signifi-cance except in neonates or across blood group transfu-sions. Mean yield of platelets in the final product wasfound to be comparable in all the machines (Table 3) withred cell contamination much below the allowable limits[8,16]. According to the American Association of BloodBanks (AABB) 2005, 90% of SDP must contain P3 � 1011

platelets per unit while the Council of Europe, 2007, rec-ommends platelet count P2 � 1011 per unit [17,18]. Theselevels have been determined to provide desired haemo-static platelet doses to the recipient. Adequate dose ofplatelets reduces the transfusion frequency, cost and less

number of donor exposures. The efficacy of different cellseparators with regard to leukodepletion varies consider-ably. Newer machines with leukoreduction program likeGambro Trima Accel give a consistently leukodepletedproduct while machines of old version like CS 3000 andMCS have a 95% failure rate [19]. Though the apheresis de-vices used in our study, such as the Amicus and FreseniusCOM.TEC are suitable for optimized platelet collection,these machines are not designed to provide leukoreducedproduct.

Though, quality of the product in terms of yield is com-parable with all the machines used in the study, there aredifferences with regard to donor convenience issues suchas DT, NT and time required for the whole procedure. Theseparameters have a direct bearing on the donor recruitmentand retention. Due to stringent donor selection criteria andinconvenience caused to the donors because of lengthyapheresis procedures, the donor pool is declining in manyof the centers. The emphasis is therefore on encouragingrepeat plateletpheresis donations with maximum donorsafety. For instance, newer machines such as Amicus andFresenius are fast with minimum DT and NT, therefore,more number of procedures can be performed in less timewith minimal donor discomfort.

References

[1] Rock G, Sutton DMC. Apheresis: man versus machine. Transfusion1998;38:625–36.

[2] Robinson EA. Donor and therapeutic apheresis. In: Taues RL, editor.Autotransfusion: therapeutic principles and trends. Detroit: GeorgyAppleton; 1997. p. 12–23.

R. Chaudhary et al. / Transfusion and Apheresis Science 40 (2009) 99–103 103

[3] Pomper GJ, Chai LI, Synder EL. Platelet transfusion and alternatives.In: Simon TL, Dzik WH, Synder EL, Stowell CP, Strauss RG, editors.Rossi’s principles of transfusion medicine. 3rd ed. Philadelphia, USA:Lippincott Williams and Wilkins; 2002. p. 232–47.

[4] Gutensohn K, Bartsch M, Kuehnl P. Flowcytometric analysis ofplatelet membrane antigens during and after continuous-flowplateletpheresis. Transfusion 1997;37:809–15.

[5] Chaudhary R, Sekhar Das S, Agarwal P, Shanker Shukla J. Qualitysystems in automated plateletpheresis in hospital-based bloodtransfusion service in north India. J Clin Apher 2005;20(2):81–5.

[6] Lutz P, Dzik WH. Large-volume hemocytometer chamber foraccurate counting of white cells (WBCs) in WBC reduced platelets:validation and application for quality control of WBC-reducedplatelets prepared by apheresis and filtration. Transfusion1993;33:409–12.

[7] Fraser JL, Whatmough A, Uhl L, Kruskall MS. Lowering thehemoglobin cutoff for female plateletpheresis donors. Transfusion1998;38:855–9.

[8] Patel AP, Kaur A, Patel V, Shah D, Kanvinde S, Prajapati S, et al.Comparative study of plateletpheresis using Baxte CS 3000 plus andHaemonetics MCS 3p. J Clin Apher 2004;19:137–41.

[9] Chaudhary R, Das SS, Khetan D, Sinha P. Effect of donor variables onyield in single donor plateletpheresis by continuous flow cellseparator. Transfus Apher Sci 2006;34:157–61.

[10] Moog R, Muller N. Results with new software for plateletpheresissystem. Transfus Sci 1994;9:236–9.

[11] Burgstaler EA, Pineda AA, Bryant SC. Prospective comparison ofplateletpheresis using four apheresis systems on the same donors. JClin Apher 1999;14:163–70.

[12] Picker SM, Radojska SM, Gathof BS. Evaluation of concurrentcollection of in-line filtered platelets and packed red blood cells bymulticomponent apheresis with three last-generation apparatuses.Vox Sang 2006;91:47–55.

[13] Boogaerts Ma. Side effects of haemaphersis. Transfus Med Rev1987;1:186–94.

[14] Strauss RG. Mechanism of adverse effects during hemapheresis. JClin Apher 1996;11:160–4.

[15] Szymanki IO. Ionized calcium during plateletpheresis. Transfusion1978;18:701–8.

[16] Berthholf MF, Mintz PD. Comparison of plateletpheresis using twocell separators and identical donors. Transfusion 1989;29:521–3.

[17] Brecher ME. Technical manual. 15th ed. USA: American Associationof Blood Banks; 2005.

[18] Guide to the preparation, use and quality assurance of bloodcomponents. 13th ed. Strasbourg, France: Council of EuropePublishing; 2007.

[19] Murphy S. Platelet transfusion therapy. In: Loscalzo J, Schafer AI,editors. Thrombosis and hemorrhage. Philadelphia: BlackwellScientific Publication; 1998. p. 1119–34.

[20] Mosteller RD. Simplified calculation of body-surface area. N Engl JMed 1987;317:1098.

[21] Shoemaker WC. Fluids and electrolytes in the acutely ill adult. In:Shoemaker WC, Ayres S, Grenvik Am, et al., editors. Textbook ofcritical care. Philadelphia: WB Saunders Co.; 1989. p. 1130.

[22] Goodnough LT, Ali S, Despotis G, Dynis M, DiPersio JF. Economicimpact of donor platelet count and platelet yield in apheresisproducts: relevance for emerging issues in platelet transfusiontherapy. Vox Sang 1999;76:43–9.