Available online at www.sciencedirect.

com

Transfusion and Apheresis Science 39 (2008) 2935 intl.elsevierhealth.com/journals/tras

Platelet transfusions in clinical practice at a multidisciplinary hospital in North India

Anupam Verma *, Prashant Pandey, Dheeraj Khetan, Rajendra Chaudhary
Department of Transfusion Medicine, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Raebareli Road, Lucknow 226 014, India

Abstract Specialty wise utilization pattern of platelet concentrates (PLT) over a period of 2 months was evaluated prospectively for appropriateness. Overall 4.87 random donor platelets (RDP) (total 1672) units were issued per request. A total of 1101 RDP (66%) were transfused prophylactically against 221 requests (64.4%) while, 571 RDP were transfused for therapeutic (requests = 122, 35.6%) reasons. Twenty-three percent of prophylactic requests and 15% of the therapeutic requisitions were not justied. Most common reason for unjustied prophylactic transfusion was unavailability of pre-transfusion platelet count. Concurrent screening of request forms to ensure optimized PLT usage may further decrease platelet misuse at our center. 2008 Elsevier Ltd. All rights reserved.
Keywords: Platelet transfusion practice; Guidelines; Transfusion trigger; Thrombocytopenia

1. Introduction The increasing demand for platelet (PLT) transfusions in recent years along with increasing awareness about associated risks of blood component transfusions have focused attention on the need for continuously analyzing the actual transfusion practices in hospitals. Platelet transfusions are used for prevention or control of bleeding associated with deciency in platelet number or function [1]. Strategies should be undertaken to minimize the
* Corresponding author. Tel.: +91 522 2668700x2506; fax: +91 522 2668017. E-mail addresses: aver@sgpgi.ac.in (A. Verma), pkpandey2007 @gmail.com (P. Pandey), dheerajkhetan@gmail.com (D. Khetan), rkcchaud@sgpgi.ac.in (R. Chaudhary).

need for platelet transfusions wherever possible and appropriately use this precious human resource as platelets are often in short supply [2]. Inspite of availability of so many practice guidelines and recommendations [35] to support the clinicians in their decisions related to platelet transfusions, the results are less than satisfactory and in fact, platelets are considered to be one of the most misused blood component products [6,7]. At our center, the platelet request forms are not monitored concurrently for indications except for platelet dosage, which is reviewed by the transfusion specialist especially during period of shortages for inventory management and thus there remains a possibility of inappropriate use of this scarce component. In this study we analyzed the utilization pattern of platelet concentrates at our hospital by

1473-0502/$ - see front matter 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.transci.2008.05.013

30

A. Verma et al. / Transfusion and Apheresis Science 39 (2008) 2935

dierent specialties to nd out if there was any misuse so that proper corrective measures can be implemented. 2. Materials and methods The study was carried out at 740-bed tertiary care teaching hospital in the state of Uttar Pradesh, the most populous province of India. This hospital has major medical and surgical superspecialities like hematology including bone marrow transplant (BMT), cardiac surgery, gastrosurgery, critical care medicine (CCM), and nephrology, etc., which are major users of platelets. In the present study all the requests for platelets by various specialties from 15th February to 15th April 2005 (2 month duration) were evaluated prospectively. 2.1. Platelet concentrates (PLT) Platelet concentrates during the study period were prepared either from whole blood donation (random donor platelets, RDP) by buy coat method using the optipress II (Baxter,USA) or from single donor by apheresis (single donor platelets, SDP) technique using cell separator CS-3000 (Baxter healthcare, USA). 2.2. Data collection and analysis Patients details such as age, gender, pre-transfusion platelet counts and other laboratory values
Table 1 Criteria for assessment of platelet transfusion appropriateness Justied usage

were obtained from medical records and computer les. Details of platelet transfusion support were obtained from blood bank records. Tentative diagnoses mentioned on the blood requisition forms were used for evaluation. The requisition forms (not the patients) were categorized as therapeutic or prophylactic on the basis of presence or absence of clinical bleed and analyzed specialty wise for total number of units requested per episode, episodes of platelet transfusion per patient, pre-transfusion platelet count and requirement of special procedures like irradiation. The following criteria (Table 1) based on our hospital guidelines for platelet transfusions were used for analysis. 2.3. Statistical analysis All the data was entered in computer (Microsoft Excel, Microsoft Corp., USA) and analyzed using the SPSS (ver. 12.0) software. 3. Results A total of 1746 RDP from 2592 whole blood donations and 48 SDP from same number of apheresis donors were prepared during the study period. A total of 352 requisitions for RDP and 48 for SDP were received from 134 patients (including 28 outside hospital patients), out of which nine requests for RDP were not accepted due to inventory problem. These patients were transfused a total of 1672 RDP and 48 SDP units during the study

Prophylactic indications I. PLT count <10 109/l in stable patients. II. PLT count <20 109/l in patients associated with fever, sepsis, splenomegaly, or undergoing hematopoietic stem cell transplant. III. PLT count <30 109/l in patients receiving anti thymocyte globulin (ATG) therapy or pre splenectomy in adult idiopathic thrombocytopenic purpura (ITP) patients. IV. PLT count <30 109/l in stable premature infants and <50 109/l in a sick infant. V. PLT count <50 109/l in patients undergoing major surgery/invasive procedure/massive transfusion or in critically ill patients on ventilator. VI. Patients without recent, i.e. within 24 h PLT count who had received PLT transfusion(s) within last 2 days at our hospital for therapeutic reason. Therapeutic indications I. Documented platelet function defect in a bleeding patient. II. PLT count <100 109/l in patients of cardiopulmonary bypass (CPB) having non-surgical, unexpected diuse bleed during or in postoperative period or regardless of platelet count if there is diuse signicant bleed within 24 h postoperatively. III. Clinically signicant non-surgical bleeding (suspected platelet bleed). IV. Bleeding patients without recent, i.e. within 24 h PLT count who received PLT transfusion(s) within last 2 days at our hospital for therapeutic reason.

A. Verma et al. / Transfusion and Apheresis Science 39 (2008) 2935

31

period. The median age of patients was 31.3 years (range: 2 days76 years) with male to female ratio of 2.1:1. Specialty wise RDP usage pattern of accepted requisitions is shown in Table 2. Over all, 4.87 RDP units were issued per requisition received. Similarly, specialty wise SDP usage pattern is shown in Fig. 1, an average of 2.4 units were transfused per patient (one per requisition). Hematology department including BMT unit consumed maximum RDP (n = 950) and SDP (n = 32) followed by CCM department, which consumed 310 RDP and 10 SDP units. A total of 1101 (65.8%) RDP units (against 67.4% of the total requests) for prophylactic and 571 (34.2%) RDP units (against 32.6% of the total requests) were used for therapeutic reasons. Of the 106 patients from our hospital, 45.3% (n = 48) received a single episode while 19.8% (n = 21) patients received ve or more episodes of RDP transfusions. Repeat platelet transfusions were primarily seen in patients of hematology and CCM. Whereas majority (15/16) of the cardiac surgery patients were transfused one time only. On comparing pre-transfusion platelet counts (PC) across patients from dierent specialties hematology patients were found to be managed at a lower threshold compared to other patients (Table 3). Over all 65% of prophylactic RDP requisitions had a pre-transfusion PC below 20 109/l and thus were considered as appropriate. Of the rest 77 ($35%) prophylactic requests, pre-transfusion platelet counts were not available for 24 (10.9%). Out of which eight were considered appropriate as these were from patients with documented thrombocytopenic bleeding on previous occasions, the

Number of SDP units transfused

60 50 40 32 30 20 10 0
Hematology (n=15) CCM (n=3) Gastrosurgery Total (n=20) (n=2)

48 Prophylactic Therapeutic Total 31

26 17 6 2 8 10 3 3 6

Number of patients is shown in parentheses

Fig. 1. Platelet (SDP) usage pattern of accepted requisitions.

remaining 16 requisitions (including ve from outside our hospital) were considered as inappropriate. Other reasons for inappropriate prophylactic requisitions are summarized in Table 4. Thus transfusion of 245 (14.65%) RDP against 51 (23%) prophylactic requisitions was not justied (Fig. 2). Appropriateness of therapeutic RDP transfusion was determined on the basis whether PC was deranged or not in the presence of clinical bleed. Transfusion of 93 (16.3%) RDP units against 18 (14.7%) therapeutic requisitions were considered inappropriate as pre-transfusion PC in these cases was not deranged (Fig. 2). Majority of inappropriate therapeutic platelet transfusions were to patients with deranged coagulation prole and therefore should have been transfused fresh frozen plasma and not the platelets. During the study period, a total of 74 RDP, but no SDP units, were discarded 47 RDP units due to

Table 2 Specialty wise platelet usage pattern (RDP) of accepted requisitions (n = 343) Specialty (no. of patients) Requisitions (per patient) RDP issued Prophylactic Total Hematology (46) CCMa (15) CVTSb (16) Gastrosurgery (08) Nephrology (13) Gastromedicine (06) Radiotherapy (02) Outside (28) Total (134)
a b

Therapeutic Total 213 125 84 48 44 34 0 23 571 Average (/request) 5.6 5.92 6 3 5 4.9 0 1.4 5.05

Total (avg./req)

Average (/request) 5.2 5.97 5.7 3.8 3.8 4.6 2 2.9 4.8

183 (4) 52 (3.5) 17 (1) 28 (3.5) 19 (1.5) 10 (1.7) 3 (1.5) 31 (1.1) 343 (2.6)

754 185 17 46 38 14 06 41 1101

967 (5.28) 310 (5.96) 101 (5.94) 94 (3.4) 82 (4.3) 48 (4.8) 06 (02) 64 (2.06) 1672 (4.87)

CCM = critical care medicine. CVTS = cardiovascular and thoracic surgery.

32

A. Verma et al. / Transfusion and Apheresis Science 39 (2008) 2935

Table 3 Prophylactic RDP requisitions and pre-transfusion platelet count across patients of dierent specialties PLT count (109/l) <10 1020 2150 51100 >100 NA Total HM, n (%) 49 77 07 12 (33.8) (53.1) (4.9) CCM, n (%) 06 04 11 07 01 02 31 (19.3) (12.9) (35.5) (22.5) (3.2) (6.5) CVTS, n (%) 01 (33.3) 01 (33.3) 01 (33.3) 03 gastroenterology; GS, n (%) 06 (50) 05 (41.7) 01 (8.3) 12 GM = medical NP, n (%) 01 05 02 02 10 (10) (50) (20) (20) GM, n (%) 02 (66.4) 01 (33.3) 03 RT, n (%) 03 (100) 03 OS, n (%) 03 04 02 05 14 (21.4) (28.5) (14.3) Total, n (%) 58 86 37 15 01 24 (26.2) (38.9) (16.7) (6.8) (0.4) (10.9)

(8.2)

(35.7)

145

221 RT = radiotherapy;

HM = hematology; OS = outside.

GS = surgical

gastroenterology;

NP = nephrology;

Table 4 Reasons of inappropriate prophylactic RDP transfusions Reasons No. of requests (n = 51) 16 10 05 06 04 04 02 04 RDP Tx (n = 245) 74 43 28 30 22 19 11 18

Unavailability of pre-transfusion PC and/or clinical details Pre-transfusion PC > 50 109/l Leukemia with PC > 20 109/l Stable aplastic anemia with PC > 20 109/l Fever and/or sepsis with PC > 20 109/l Minor invasive procedure with PC > 50 109/l Pre splenectomy in ITP with PC > 30 109/l Chronic DIC without bleed

a reactive screen for transfusion transmitted infections, 12 RDP for quality control and 15 RDP units for outdating. The irradiated platelets (71 RDP and 14 SDP) were demanded for 3 BMT patients and 5 pre-BMT patients for prevention of graft versus host disease. 4. Discussion Clinical transfusion audit helps to identify current pattern of usage and areas of improvement. The transfusion committee should monitor and audit the blood utilization practices in a hospital. This should be regularly evaluated to improve the blood component utilization and facilitate policy making in a referral center. Concerns regarding risks of blood transfusion and non-availability of

Total requests for RDP

Prophylactic 64.4% (n= 221)

Therapeutic 35.6% (n= 122)

Appropriate 77% (n=170)

Inappropriate 23% (n=51)

Platelet count, coagulation profile

coag 70.5% (n=86)

N

PC

PC

Abn. coag 14.7% (n=18)

PC coag 1.6% (n=2)

N N

N PC Abn. coag 13% (n=16)

Appropriate 85.3% (n=104)

Inappropriate 14.7% (n=18)

Fig. 2. Appropriateness of RDP transfusion: prophylactic versus therapeutic.

A. Verma et al. / Transfusion and Apheresis Science 39 (2008) 2935

33

blood components make it crucial to optimize platelet transfusions and reduce wastage. We carried out a prospective study of platelet components usage at our center having major super specialized medical and surgical branches during 2 month duration. In this study, unlike other studies for assessment of platelet usage [711], requisition forms and not the patients were analyzed under two categories therapeutic or prophylactic on the basis of platelet count and presence or absence of clinical bleed. This was due to the fact that transfusion in a given patient may have been categorized as therapeutic at one time and as prophylactic at some other instance, thus making it dicult to allocate patients between these two categories. Over all, 4.87 RDP units were transfused per requisition (per episode) received (Table 2). This is in accordance with the published guidelines of transfusing 1 unit of RDP per 10 kg body weight as majority of our patient population were adults [12]. In a study from Australia, a pool of 4 units of RDP was transfused per episode in 73% of their patients [7]. Another analysis of platelet utilization reported a median of 4.5 units of whole blood derived platelets being transfused prophylactically per episode [11]. Patients with hematological diseases accounted for 34% of the patients but used 58% of the platelets (Table 2) that is hematology patients were the major users of platelet concentrates in our study. Similar ndings have been reported by McCullough et al. [8], in their study of 243 patients and by Schoeld et al., in their study covering 14 Australian hospitals [7]. In our study we have observed that 65.8% of RDP units were transfused for prophylactic while 34.2% were transfused for therapeutic purpose (Table 2). This is also in agreement with published data [8,11]. Of the prophylactic requests for RDP, pre-transfusion counts were not available for 8.2% of our inpatients; higher as compared to Eikenboom et al. [13] who reported unavailability of pre-transfusion counts in 4.3% of their in-patient platelet transfusions. Sixty-three percent (of the episodes for which pre-transfusion counts were available) of the platelet transfusions were with pre-transfusion counts of more than 10 109/l, the most accepted transfusion trigger for prophylactic platelet transfusion. Amongst the dierent specialties, hematology patients were found to be managed at a lower threshold compared to other patients (87% with

pre-transfusion counts below 20 109/l, Table 3). A similar proportion of prophylactic transfusions were reported in a study by Cameron et al. [14] where only 5.3% of patients from hematology oncology and BMT units were transfused with pre-transfusion counts of 50 109/l, while 85.7% had pre-transfusion counts of 20 109/l. However, unlike our ndings Eikenboom et al. [13] found only 68% of their hematology patients being transfused with pre-transfusion counts below 20 109/l. In India, blood component therapy is still in its infancy, not many are aware of the various guidelines available for platelet usage and the associated risks of platelet transfusion, therefore there are more chances of inappropriate platelet usage which leads to waste of the precious human resource that is already scarce in developing countries. In our study we found that 69/343 (20%) requests were inappropriate (Fig. 2). This is in accordance with Hui et al. [15] who reported that 88% PLT were prescribed in an appropriate manner. However, Schoeld et al. [7] from Australia has observed 33% inappropriate platelet transfusions. This dierence in appropriate transfusion may be due to dierence in patient population or use of dierent guidelines for assessment of appropriateness. We have observed that inappropriate transfusions were more when platelets were transfused for prophylaxis as compared to when transfused to control bleeding (Fig. 2). Higher misuse for prophylactic transfusion may be due to the overestimation of the immediate risk and the tendency of the clinicians to transfuse at the safer side of transfusion trigger adjudged for a particular patient. Another reason, that may not be applicable to developed nations where there is no shortage of platelet concentrates, is the concern of clinicians to preoccupy the scarce platelets for their patients, in case the transfusion service fails to meet the demand when a further decrease in the platelet count of the patient is anticipated. This was reected in increased requisitions received during weekends both for RDP and SDP and during night hours (data not shown). Not many studies have discussed the misuse of platelet transfusion under therapeutic as well as prophylactic categories, majority of the workers have elaborated either on the therapeutic or on the prophylactic misuse only. Moreover, the reasons for inappropriate prophylactic platelet transfusion, as shown in our study (Table 4) are also not available to compare our ndings. Majority of these inappropriate requests belonged to hematology which is

34

A. Verma et al. / Transfusion and Apheresis Science 39 (2008) 2935

also the main user of platelets at our institute. Another dierence in platelet transfusion practice at our center is that the RDP were given without consideration to ABO and Rh compatibility in almost all adolescent and adult cases except in two leukemia patients admitted in hematology where ve each ABO matched RDP were transfused when the platelet counts did not increase with ABO mismatch platelets besides these patients could not aord to have apheresis platelets. SDP were transfused only after ABO compatibility in 100% cases. The PLT expiry rate was relatively low which is reected by the fact that only 15 of the total 1746 RDP prepared during the study period were discarded which is much lower compared to expiry rates of 5.86.4% quoted by Q-Probes study from USA [16]. PLT outdating has been found out to be the main reason for discarded PLT in various studies [1619]. One interesting nding in this study was the observation that SDP were transfused within 1 day and RDP till 5 days after collection. However, during periods of shortages like during weekends, holidays the outside demands were given less priority compared to our hospital demands for platelets and RDP with 12 days storage were issued. As per our department policy in view of anticipated platelets requirement, RDP were stocked during periods of shortages by preparing more number of platelets from whole blood donations. A limitation of our study is the absence of data to allow the inference about the ecacy of PLT transfusions, as post-transfusion platelet counts are generally not available for majority of our patients receiving RDP. A post-transfusion platelet count should be obtained within 24 h in all cases to assess the response to platelet products and to document any refractoriness to platelet transfusion that actually can bring any change in the treatment policy. In view of our lack of data for platelet transfusion response assessment, we are now monitoring our hematology patients for post-transfusion platelet counts for SDP where the product yield is readily available as a part of quality control. In conclusion, information concerning the platelet preparation, utilization and its wastage was collected with a view to improve the existing blood transfusion practices and our study shows that platelet transfusion therapy at our institute is satisfactory, in the sense, guidelines are being followed in majority of prophylactic as well as therapeutic platelet transfusions (Fig. 2). This is in contrast to

higher misuse of fresh frozen plasma as shown by a previous study from our center [20]. This dierence in misuse may be due to easy availability of FFP compared to PLT which have a shorter shelf life and is not thought to be suitable for volume replacement unlike FFP. The distinctive feature of our study is the analysis of requisition forms and not of the patients which eliminates the indecisiveness of repeatedly allocating same patient into these two dierent categories, viz. therapeutic and prophylactic, during his/her hospital stay. However, there are still some lacunae in transfusion practice at our center such as lesser SDP usage and inappropriate transfusions mainly of prophylactic RDP. There is need for increasing the pool of plateletpheresis donors and considering use of appropriate pharmacotherapy in specic situations besides platelet transfusion. It is anticipated that periodic lectures regarding appropriate component usage and concurrent monitoring of requisition forms can help in improving rational PLT usage at our center. References
[1] Schier CA, Andweson KC, Bennett CL, Bernstein S, Elting LS, Goldsmith M, et al. Platelet transfusion for patients with cancer: clinical practice guidelines of the American Society of Clinical Oncology. J Clin Oncol 2001;19:151938. [2] Schlossberg HR, Herman JH. Platelet dosing. Transfus Apher Sci 2003;28:2216. [3] Murphy MF, Brozovic B, Murphy W, Ouwehand W, Waters AH. British committee for standards in haematology, working party of the blood transfusion task force: guidelines for platelet transfusions. Transfus Med 1992;2:3118. [4] American society of anesthesiologists task force on blood component therapy. Practice guidelines for blood component therapy: a report by the American society of anesthesiologists task force on blood component therapy. Anesthesiology 1996;84:73247. [5] British committee for standards in haematology. Blood transfusion task force. Guidelines for the use of platelet transfusions. Br J Haematol 2003; 122:1023. [6] Siela BH, Scott EP, Connelly DP. Design and preliminary evaluation of an expert system for platelet request evaluation. Transfusion 1991;31:6006. [7] Schoeld WN, Rubin GL, Dean MG. Appropriateness of platelet, fresh frozen plasma and cryoprecipitate transfusion in New South Wales public hospitals. Med J Aust 2003;178:11721. [8] McCullough J, Steeper TA, Connelly DP, Jackson B, Huntington S, Scott EP. Platelet utilization in a university hospital. JAMA 1998;259:24148. [9] Zimmermann R, Buscher M, Linhardt C, Handtrack D, Zingsem J, Weisbach V, et al. A survey of blood component use in a German university hospital. Transfusion 1997;37: 107583.

A. Verma et al. / Transfusion and Apheresis Science 39 (2008) 2935 [10] Zimmermann R, Handtrack D, Zingsem J, Weisbach V, Neidhardt B, Glaser A, et al. A survey of blood utilization in children and adolescents in a German university hospital. Transfus Med 1998;8:18594. [11] Greeno E, McCullough J, Weisdorf D. Platelet utilization and the transfusion trigger: a prospective analysis. Transfusion 2007;47:2015. [12] Consensus development panel: platelet transfusion therapy. JAMA 1987;257:1777-80. [13] Eikenboom JCJ, van Wordragen R, Brand A. Compliance with prophylactic platelet transfusion trigger in haematological patients. Transfus Med 2005;15:458. [14] Cameron B, Rock G, Olberg B, Neurath D. Evaluation of platelet transfusion triggers in a tertiary-care hospital. Transfusion 2007;47:20611. [15] Hui CH, Williams I, Davis K. Clinical audit of the use of fresh-frozen plasma and platelets in a tertiary teaching hospital and the impact of a new transfusion request form. Intern Med J 2005;35:2838.

35

[16] Novis DA, Renner S, Friedberg RC, Walsh MK, Saladino AJ. Quality indicators of fresh frozen plasma and platelet utilization: three college of American pathologists Q-probes studies of 89,81,796 units of fresh frozen plasma and platelets in 1639 hospitals. Arch Pathol Lab Med 2002;126:52732. [17] Munksgaard L, Albjerg L, Lillevang ST, Gahrn- Hansen B, Georgsen J. Detection of bacterial contamination of platelet components: six years experience with BacT/ALERT system. Transfusion 2004;44:116673. [18] Sullivan MT, Wallace EL. Blood collection and transfusion in the United States in 1999. Transfusion 2005;45: 1418. [19] Ledman RE, Groh N. Platelet production planning to ensure availability while minimizing outdating. Transfusion 1984;24:5323. [20] Chaudhary R, Singh H, Verma A, Ray V. Evaluation of fresh frozen plasma usage at a tertiary care hospital in North India. ANZ J Surg 2005;75:5736.