Results of an educational program to promote the use of pooled instead of single-donor platelet transfusions

Abstract

Transfusion of platelets is commonly indicated in the inpatient oncology setting. These platelets are obtained either through apheresis from a single donor or pooled from the whole blood of several donors. The amount of transfused platelets, infection risk, incidence of alloimmunization, and increases in posttransfusion platelet count are similar for these two platelet products. Although single-donor platelets are preferred over pooled platelets in some instances, single-donor platelets are often given regularly, despite a higher cost and more limited donor supply. Oncology fellows at Baylor University Medical Center at Dallas initiated an education campaign regarding the indications for pooled and single-donor platelet transfusions. The quality improvement campaign included seminars led by oncology fellows for nursing personnel and resident housestaff on the two oncology floors, as well as electronic correspondence to attending physicians. The number of pooled and single-donor platelet transfusions on the two floors was recorded for the 3 months after the education campaign (July–September 2011) and compared with the corresponding data from the previous year. Over the 3-month study period after the education campaign, the average percentage of pooled platelets transfused increased to 34.1% from 13.1% for the prior year. Given this increase, the estimated cost benefit over the 3-month study period was $45,000.

On the inpatient oncology floors (6 and 9 Roberts) at Baylor University Medical Center at Dallas (BUMC), prophylactic and therapeutic platelet transfusions are commonly administered. Platelet products are obtained through apheresis from a single donor (SDP) or are pooled from whole blood. To obtain equivalent platelet quantities from whole blood, 5 to 6 units of whole blood–derived platelet concentrates are pooled. SDP obtained through apheresis is of the same quality as pooled random-donor platelets and causes equivalent increases in posttransfusion platelet counts (1). However, there is a significant difference in donor supply and cost, with each unit of SDP and pooled platelets costing BUMC approximately $550 and $275, respectively.

Platelet transfusions are expensive and are associated with a number of side effects, including febrile reactions, transmission of viral and bacterial infections, graft-versus-host disease, hemolysis, and alloimmunization. It has been postulated that the use of SDP instead of pooled platelets would decrease donor exposures and thus reduce the risks of infection and alloimmunization. However, studies have not shown a benefit with using SDP versus pooled random-donor platelets for prevention of alloimmunization, provided platelets are leukocyte reduced (2). SDP that are HLA matched are indicated for patients with a known history of alloimmunization (3). Also, current testing for viral transmission has greatly reduced the incidence of transfusion-related viral infections (4). The risk of bacterial infections with platelet transfusions is increased due to platelet storage at 22°C rather than the 4°C required by red cells. This allows for more rapid growth of bacterial contamination of platelet products, which is usually caused by contamination by skin flora at the time of blood collection. While one study reported a decreased risk of infections with SDP versus pooled platelet transfusion (5), this risk should now be mitigated, as each product routinely undergoes rapid bacterial culture testing prior to transfusion (6).

METHODS

This study was approved by the institutional review board at BUMC. The number of pooled, SDP, and total platelet transfusions given on the inpatient oncology floors (6 and 9 Roberts) was recorded by the BUMC blood bank. The study period included the 3 months following the education intervention (July–September 2011). The period of July to September 2010 was used as a control. The education campaign included presentations given by oncology fellows to nursing personnel on 6 and 9 Roberts as well as internal medicine housestaff. In addition, flyers were posted on 6 and 9 Roberts, and electronic correspondence was sent to attending physicians commonly rotating on those floors. The education campaign initially began in April 2011 but was delayed due to a shortage of pooled platelet products during that time. The primary education campaign took place in a 2-week period prior to the study period. Educational material encouraged the use of pooled platelets rather than SDP, except when histocompatible donors were needed for patients with a known history of alloimmunization.

The percentage of pooled platelet transfusions for July, August, and September 2010 (preinvervention) was calculated, as well as the percentage over the same months in 2011 (postintervention). To compare the 2 years, a two-sample test for proportions was used. A cost-benefit analysis was performed by multiplying the total number of platelets transfused during the study period by the percentage of pooled platelets transfused during the control period from the previous year. This number was then subtracted from the actual number of pooled platelets transfused during the study period to give the estimated number of SDP saved. SDP saved was then multiplied by $275 (the approximate cost difference between SDP and pooled platelets) to determine the cost benefit.

RESULTS

The absolute numbers of pooled and SDP transfused on 6 and 9 Roberts are shown in the Table. A total of 1197 platelet units were transfused during the 6 months studied. Of these, 312 were pooled and 885 were SDP; 788 were transfused on 6 Roberts and 409 were transfused on 9 Roberts. During the control period, 443 units of platelets were transfused on oncology floors, and 55 (12.4%) were pooled platelets (confidence interval [CI] 9.7%–15.8%). During the study period, 754 units of platelets were transfused, and 257 (34.1%) were pooled platelets (CI 30.7%–37.5%). The absolute difference in percentage of pooled platelets transfused between 2010 and 2011 was 21.7% (CI 17.1%–26.2%), which was statistically significant (P < 0.001). The increase in pooled platelets transfused was seen on both 6 and 9 Roberts. There was no increase in platelet transfusion reactions or adverse events related to transfusion during the study period months.

Table.

Pooled and single-donor platelets transfused on 6 and 9 Roberts

		6 Roberts			9 Roberts			Total
Group	Month	SDP	Pooled	Percent Pooled	SDP	Pooled	Percent > Pooled	SDP	Pooled	Percent Pooled
Control	July 2010	94	10	9.6%	50	3	5.7%	144	13	8.3%
	Aug 2010	115	15	11.5%	37	4	9.8%	152	19	11.1%
	Sep 2010	63	15	19.2%	29	8	21.6%	92	23	20.0%
Study	July 2011	108	53	32.9%	54	22	27.9%	162	75	31.6%
	Aug 2011	77	66	46.2%	74	27	26.7%	151	93	38.1%
	Sep 2011	105	67	39.0%	79	22	21.8%	184	89	32.6%

The cost-benefit analysis revealed that approximately 94 units of pooled platelets would have been transfused during the study period had transfusions continued at the control rate of 12.4%. When this number was subtracted from the actual number of pooled platelets transfused during the study period, 164 units of SDP were saved. When multiplied by the average cost difference between DP and pooled platelets, the cost benefit was approximately $45,000 during the study period.

DISCUSSION

A large proportion of platelets transfused at BUMC are given to oncology patients. Our experience revealed that transfusion of SDP rather than pooled platelets was occurring at a high rate on oncology floors at BUMC. This was likely occurring due to a number of factors, including a previous system in which there was a significant cost benefit for apheresis platelets due to expenses related to bacterial testing of each platelet concentrate in a unit of pooled platelets. This cost benefit, however, has been abrogated by the ability to test the whole pooled unit of platelets only once. Currently, at BUMC, each unit of SDP costs the hospital $550, compared with $275 for pooled platelets. A cost-effectiveness study showed that the estimated cost per quality-adjusted life year of using SDP as opposed to pooled platelet therapy in stem cell transplant patients ranged from $168,700 to $519,822 (7).

In addition to cost, another factor that may have played a role in the culture of SDP use over pooled platelet use at BUMC was the belief among medical personnel that pooled platelets would result in a higher incidence of alloimmunization due to more donor exposures. However, no studies have shown a benefit to using leukoreduced SDP compared with leukoreduced pooled random-donor platelets to reduce alloimmunization (2). All platelet products at BUMC are universally leukoreduced.

Another factor influencing the increased use of SDP over pooled platelets was likely fear of increased infection with pooled platelets. Certainly, there is an increased risk of bacterial contamination of platelets due to storage at 22°C rather than 4°C. Ness et al found an increased risk of septic platelet transfusion reactions in pooled platelet products versus SDP (5). However, this retrospective observational study was performed prior to the mandate by the American Association of Blood Banks and College of American Pathologists for universal bacterial testing of all blood products (8, 9). Other more recent data show an equivalent low risk of bacterial contamination with pooled platelets and SDP when screening bacterial cultures are used (10). Concern over an increased risk of viral transmission with pooled platelets is not as strong as it once was, given the advent of routine nucleic acid testing of blood for HIV, human T-cell lymphotrophic virus, hepatitis C, and hepatitis B (4).

Some variability in our data set warrants further analysis. The increased overall number of platelets transfused during our study period compared to the control period was likely due to the transition of bone marrow transplant patients onto 6 and 9 Roberts in spring 2011. In addition, the higher percentage of pooled platelets given during the study period may be attributable to a previous educational campaign launched in April 2011. This campaign was delayed when there was a shortage of pooled platelet products in that month.

Despite an effective campaign to decrease the amount of SDP transfused on oncology floors at BUMC, most platelet products are still SDP. This is likely due to the ingrained culture and habits of using SDP at BUMC. Further increases in pooled platelet use are still warranted, however, as an appropriate goal for pooled platelet use should be closer to 90%. This could be attained through more extensive and personalized education for attending physicians, as our education campaign only included electronic correspondence for this group. Individualized feedback related to platelet-prescribing history for attending physicians and comparison to their peers may be a way to improve these practices. In summary, pooled platelets are as safe and effective as single-donor platelets and also have cost-benefit advantages. Therefore, continued efforts to promote the use of pooled platelets are worthwhile.