ORIGINAL RESEARCH
Evaluating safety and cost-effectiveness of platelets stored
in additive solution (PAS-F) as a hemolysis risk mitigation strategy
Monica B. Pagano,1 Brennan L. Katchatag,2 Shiva Khoobyari,1 Mark Van Gerwen,1 Nina Sen,2
N. Rebecca Haley,3 Terry B. Gernsheimer,4 John R. Hess ,2 and Ryan A. Metcalf 5
BACKGROUND: Platelet inventory constraints can
result in minor ABO incompatibility and possible
hemolysis. The aims of this study were to determine the
reduction of isoagglutinin in titers of platelets stored in
additive solution (PAS) and compare its safety,
efficiency, and cost-effectiveness with full-volume and
plasma-reduced platelets.
STUDY DESIGN AND METHODS: Isoagglutinin titers
were performed in paired whole blood donor samples
and apheresis platelets collected in PAS (PAS-PLT)
aliquot samples by the tube method.
RESULTS: A total of 149 pairs of donor/platelet
samples were tested: 75 group O, 59 group A, and
15 group B. For group O donor samples, the median
anti-A IgG and IgM were 64 and 16, respectively, and
the median anti-B IgG and IgM were 64 and
16, respectively. For group O PAS-PLT samples the
mean anti-A IgG and IgM, and anti-B IgG and IgM were
32 and 8, and 16 and 8, respectively. For group A donor
samples, the mean anti-B IgG and IgM was 8 in both
cases; and both titers decreased to 2 in PAS-PLT. For
group B donor samples, mean anti-A IgG and IgM was
16 in both cases; and both titers decreased to 4 in PAS-
PLT. PAS-PLT demonstrated a net reduction in cost and
improved efficiency when compared to plasma reduction.
The use of PAS-PLT resulted in a 40% reduction of
allergic transfusion reactions.
CONCLUSION: The use of PAS decreases plasma
isoagglutinin titers, transfusion reactions, and is cost-
effective when compared to routine plasma reduction as
a strategy to mitigate hemolysis risk from minor
incompatible platelet transfusion.
P
latelet inventory management is a challenge that
all transfusion services face on a daily basis, prin-
cipally due to their relatively short shelf life of
5 or sometimes 7 days. Several strategies have
been developed to efficiently manage platelet inventory to
provide safe platelet components in a timely manner and
prevent shortages. One common practice is to transfuse
platelets based on expiration date (first-in, first-out) cross-
ing the ABO group barrier, and allowing the transfusion of
ABO-incompatible plasma present in the platelet compo-
nents up to a certain volume.
Although rare, acute intravascular hemolytic transfusion
reactions have been caused by ABO isoagglutinins after the
transfusion of minor ABO-incompatible apheresis platelets.1
Hemolytic transfusion reactions may occur when one or more
of the following conditions are present: the plasma contains a
high-titer antibody, large volumes of plasma are transfusd, or
small children and infants are the recipients of the transfu-
sions.2 Some mitigation strategies to reduce the risk of hemo-
lysis include 1) limiting the allowed amount of incompatible
plasma (by limiting the number of incompatible platelet
ABBREVIATIONS: PAS = platelet additive solution; PAS-PLT =
apheresis platelets collected in platelet additive solution.
From the 1Department of Laboratory Medicine, Division of
Transfusion Medicine, University of Washington, Seattle,
Washington; 2Department of Laboratory Medicine, Division of
Transfusion Medicine, Harborview Medical Center, Seattle,
Washington; 3Bloodworks Northwest, Seattle, Washington;
4
Department of Medicine, Division of Hematology, University of
Washington, Seattle, Washington; and the 5ARUP, University of
Utah, Salt Lake City, Utah.
Address reprint requests to: Monica B. Pagano, MD, Blood
Bank, University of Washington Medical Center, 1959 NE Pacific,
Seattle, WA 98195; e-mail: monibea@uw.edu
Received for publication September 3, 2018; revision received
November 11, 2018, and accepted November 29, 2018.
doi:10.1111/trf.15138
© 2018 AABB
TRANSFUSION 2018;00;1–6
TRANSFUSION 1
PAGANO ET AL.
units/volume given to a patient during a specified period, or
reducing the plasma content in the unit); 2) establishing a
level of isoagglutinin titer perceived to be safe; and 3)
avoiding certain high-risk hemolysis combinations such as
the use of type O platelets for type A recipients.3 Limiting the
amount of incompatible plasma, although reasonable, is not
based on clear evidence and the maximum amount of plasma
that can be considered safe is unknown. Current studies have
not successfully defined a safe isoagglutinin titer, and it has
been questioned whether antibody titers can predict risk of
hemolysis.4
Transfusion support for patients undergoing hemato-
poietic stem cell transplantation represents a unique chal-
lenge. ABO incompatible hematopoietic stem cell
transplants are associated with worse outcomes when com-
pared to ABO-compatible transplants, but the implications
associated with transfusion of plasma that is incompatible
with the recipient endothelium are not known.5’6 At our
institution, the ABO type of the graft and the ABO type of
the recipient (expressed on the endothelium) are consid-
ered when selecting a unit of platelets for transfusion. Plate-
let units that are not compatible with both donor and
recipient are plasma reduced to 100 mL to minimize the
risk for the graft and potential risks for the endothelium
even after engraftment. Similar volume reduction strategies
have been shown to be expensive and resource intensive,
which was part of the impetus for this study.
Platelet additive solution (PAS) is a balanced crystalloid
solution with nutrients and a favorable pH for platelet stor-
age.7 Apheresis platelets collected in PAS (PAS-PLT) contain
approximately 33% of the plasma content of regular aphere-
sis platelets, resulting in a lower incidence of adverse
events.8,9 This may be related to reduced plasma protein
content and lower human leukocyte antigen antibody con-
tent in addition to limiting ABO incompatibility, which has
been demonstrated in PAS-C.10,11
We hypothesized that units collected in PAS-F would
contain lower isoagglutinin titer levels by a dilutional effect,
and that implementation of PAS-F in lieu of the current vol-
ume reduction strategy would be a more efficient use of
TABLE 1. Donor and PAS platelets isoagglutinin titers for groups O, A and B*
Anti-A
Anti A IgG
Donor sample 64 (16–562)
PAS platelets 32 (4–128)
Group A
Anti B IgG
Donor Sample 8 (1–64)
PAS Platelets 2 (1–16)
* Results are presented as median and range.
2 TRANSFUSION
resources, cost-effective, and safe. The objectives of this
study were to characterize the ABO isoagglutinin titers in
apheresis units collected in PAS, to compare isoagglutinin
titers from the donor with the respective PAS collected
platelet, to describe the incidence of adverse events, and to
evaluate whether the use of PAS-PLT would be more effi-
cient and cost-effective.
MATERIALS AND METHODS
This study was considered nonhuman subject research,
and institutional review board approval was not required.
The local blood center provided blood samples from indi-
viduals donating apheresis platelets collected in PAS.
These samples were routinely collected for infectious dis-
ease testing, and were used to test for isoagglutinin titers
after all infectious disease testing was completed. Samples
obtained for quality control from the PAS apheresis units
were also provided by the local blood center and tested for
isoagglutinin titers (PAS-PLT). Donor and PAS-PLT sam-
ples were stored at 4 C and tested within 20 days of
collection.
Serologic tube testing was performed, and to avoid
interexaminer variation during testing, all testing was per-
formed by the same individual throughout the study. Serial,
twofold dilutions of plasma (1, 2, 4, 8, 16, 32, 64, 128, ≥256)
were prepared for both IgM and IgG antibodies. Samples
were tested for anti-A and anti-B IgM isoagglutinins by tube
method following 15 minutes’ incubation at room tempera-
ture. Anti-A and anti-B IgG isoagglutinins were tested by
tube method after 15 minutes’ incubation at 37 C and add-
ing anti-human globulin. The titer was interpreted as the
reciprocal of the highest dilution that yields a 1+ macro-
scopic reaction.
Information collected in the database included sample
collection date, unit type, unit number, ABO type, testing
date, reagent lot number, anti-A IgM titer, anti-B IgM titer,
anti-A IgG titer, anti-B IgG titer, and the technologist per-
forming the test.
Group O
Anti-B
Anti A IgM Anti B IgG Anti B IgM
16 (2–256) 64 (4–256) 16 (4–64)
8 (1–64) 16 (1–64) 8 (1–16)
Group B
Anti B IgM Anti A IgG Anti A IgM
8 (2–64) 16 (4–64) 16 (4–64)
2 (1–16) 4 (1–16) 4 (1–16)

Descriptive statistics were performed using computer
software (Stata, StataCorp).
RESULTS
PAS platelet titers
A total of 149 pairs of donor and PAS platelet samples were
tested, consisting of 75 (50.3%) group O, 59 (39.6%)
group A, and 15 (10.1%) group B test sets. The median and
Fig. 1. (A) Group O IgG, IgM isoagglutinin titers distribution. (B) Group A IgG, IgM isoagglutinin titers distribution. (C) Group B IgG, IgM
isoagglutinin titers distribution.
PAS ISOAGGLUTININ TITERS
range of isoagglutinin titers for groups O, A, and B are
shown in Table 1. Isoagglutinin titer distribution is pre-
sented in Fig. 1. The anti-A and anti-B IgM mean isoaggluti-
nin titers for group O PAS were 65% lower than the
corresponding mean for the donor group. The IgG anti-A
and IgG anti-B isoagglutinin titers for group O PAS were
64% and 69% lower than the donor mean, respectively. For
non–group O PAS, the IgM (anti-A and anti-B) and IgG
(anti-A and anti-B), were 62% and 66% lower than the
donor mean.
TRANSFUSION 3
PAGANO ET AL.

Fig. 1. Continued.

Cost and efficiency analysis comparing PAS versus Transfusion reactions

plasma reduction The incidence of transfusion reactions was statistically sig-
Each volume reduction procedure takes 92 minutes of labo- nificantly lower when full volume-PAS platelets were trans-
ratory staff time to complete, which was incorporated in the fused compared to full-volume platelets stored in 100%
analysis of cost differences (Table 2). Extrapolating from the plasma (p = 0.003; Table 3). Volume-reduced platelets did
6-month evaluation period, the expected annual number of not show a significant difference in transfusion reactions
volume reduction procedures was 1018. The annual when compared with full-volume platelets stored in 100%
expected hours spent performing volume reductions was plasma (p = 0.297). When analyzing specific transfusion
1561, amounting to 0.75 of a full-time equivalent. reaction types, PAS platelets led to a reduction in only aller-
At our institution, PAS platelets were $19.58 more gic transfusion reactions when compared with full-volume
expensive than the average cost of non-PAS platelets platelets in 100% plasma (Table 4).
(including whole blood platelet concentrate and apheresis
platelets). We calculated the cost associated with the vol-
ume reduction procedure to be $65.32. This calculation
does not include potential productive activities that the per-
DISCUSSION
sonnel would dedicate as a consequence of not performing To our knowledge, this is the first study demonstrating the
these procedures. The cost savings per unit (replacing reduction of ABO isoagglutinin titers of groups O, A, and B
plasma reduction with PAS) is $45.74. Annual cost savings apheresis platelets collected in PAS-F compared to donor iso-
translated to $47,000. agglutinin titers. As predicted, there was an approximately
two-thirds reduction in IgM and IgG titers of donor sample
to PAS platelets, consistent with the 65% replacement of
TABLE 2. Estimated time and cost for plasma plasma with PAS. In addition, the use of PAS decreases the
reduction procedure and cost saving per unit incidence of allergic transfusion reactions, and it is more
Time (min) Costs (US$) cost-effective and operationally efficient when compared with
Add transfer pack 5 3.6 reducing the amount of plasma by centrifugation.
Sterile welder blade use 1
Drain product to transfer pack 2
A variety of studies have been conducted to evaluate and
Balance product and QC 10 describe the risks of hemolysis due to transfusion of ABO-
Centrifuge 20 incompatible plasma and/or platelet products.2,12,13 Interest-
Remove plasma and rest product 25
Modify in LIS, relabel product, and verify 5
ingly, there is not a standard definition of what is considered
Resuspend product 5 a “low-titer” plasma product established in the United States;
Agitate product 20 however, the AABB standards do require “a policy concerning
Total time 92
Total supply costs 4.6
transfusion of components containing significant amounts of
Staff time with benefits 60.7 incompatible ABO antibodies or unexpected red cell anti-
Total costs per volume reduction 65.3 bodies.” A survey sent out by the College of American
Additional cost for PAS 19.6
Cost saving per unit 45.7
Pathologists revealed that 83% of North American laboratories
LIS = laboratory informatics system; PAS = platelet additive solu-
have a policy addressing the prevention of hemolytic transfu-
tion; QC = quality control. sion reactions from ABO-minor-mismatched platelet transfu-
sions.3Mitigation strategies used by different facilities include

4 TRANSFUSION
 PAS ISOAGGLUTININ TITERS

TABLE 3. Incidence of transfusion reactions comparing PAS platelets versus platelets stored in plasma
Platelet component Components transfused FY17 Q1 and Q2 Transfusion reactions, n Transfusion reaction rate (%) p
Full volume, plasma 5204 162 3.11
Reduced volume, plasma 486 11 2.26 0.297
Full volume, PAS 748 9 1.2 0.003
Reduced volume, PAS 23 1 NA
Washed 35 0 NA
Totals and cumulative rate 6496 183 2.82
FY = fiscal year; NA = not applicable; PAS = platelet additive solution; Q1 = first quarter; Q2 = second quarter.

TABLE 4. Incidence of transfusion reaction by type of platelet (non-PAS vs. PAS)

Non-PAS, n Incidence (%) PAS, n Incidence (%) z score p value
Allergic 89 1.71 4 0.53 2.42 0.008
FNHTR 28 0.54 4 0.53 0.01 0.49
Other 6 0.12 0 0 0.93 0.17
Unrelated/underlying 39 0.74 2 0.27 1.49 0.07
Total 162 3.11 10 1.34 2.92 0.002
FNHTR = febrile nonhemolytic transfusion reaction; PAS = platelet additive solution.

provision of ABO-compatible plasma products, physician noti-
fication, volume reduction, washed platelets, antibody screen-
ing for high-titer ABO antibodies, pooled platelets of assorted
ABO donor types, and limited transfusions of out-of-group
plasma products. The risk of a clinically significant acute
hemolytic transfusion reaction associated with group O donor
platelets transfused to non–group O recipients is relatively
low, estimated to be about 1 in 9000 (0.01%) in one study and
0.05% in another study.1,14 These reactions are likely rare due
to the neutralization of the anti-A or anti-B antibodies by the
recipient’s soluble or endothelial-based A and B antigens
and/or the dilution of the incompatible ABO antibodies in the
recipient’s plasma volume. However, because our service pro-
vides blood components for a large hematopoietic stem cell
transplant program, situations in which the endothelium and
plasma lack the implicated ABO antigen that is expressed on
the RBC surface (i.e., major incompatible hematopoietic stem
cell transplant) may be more common. Given that isoaggluti-
nin titers are not reliable in predicting hemolysis, it is our
practice to reduce the plasma volume of platelets not to
exceed 100 mL of incompatible plasma. The use of PAS has a
double benefit: the total plasma content is similar to a platelet
unit that has been concentrated to 100 m, resulting in less
ABO antibody and plasma protein exposure, reducing risks of
hemolytic and allergic reactions, respectively.
PAS platelets are best known for decreasing the inci-
dence of allergic transfusion reactions. Other applications of
PAS platelets have been suggested to reduce adverse
transfusion reactions associated with circulatory overload,
dyspnea, and transfusion-related acute lung injury, but defin-
itive evidence remains insufficient.10 PAS platelets may have
clinical and manufacturing advantages: The formulation of
PAS can be adjusted to optimize energy metabolism and/or
minimize activation of stored platelets; improvements to
platelet viability at Day 5 or greater are possible; plasma vol-
ume reduction may minimize plasma component–associated
transfusion reactions; and pathogen reduction technology
may be applied to inactivate bacterial growth or facilitate its
growth for detection.
However, there may be some limitations to the use of
PAS platelets, such as decreased count increment and for
patients who are massively bleeding and require plasma com-
ponents to control plasma factor deficiencies in this setting.15
Overall, the implementation of PAS-F platelets in place
of volume reduction for out-of-group platelet transfusions is
an effective strategy on multiple fronts. First, patients
requiring out-of-group platelets due to inventory constraints
would receive a lower dose of anti-A and anti-B per transfu-
sion, which is supported by the titer reductions demon-
strated here. Second, this study confirms prior findings that
PAS platelets reduce the incidence of allergic transfusion
reactions. Third, the implementation of PAS-F platelets
would lead to net annual cost savings and would substan-
tially improve operational efficiency. The findings from this
study are likely to be translatable to other institutions using
volume reduction strategies, particularly those performing a
large number of platelet transfusions annually.
There are some limitations to this study. The age of the
donor population is unknown and can affect the titer-level
evaluation due to possible effects of lowered ABO isoagglu-
tinin levels in an increasingly older population base. How-
ever, we have no reason to suspect this would not reflect
our current donor population. Ideally, there should be no
delay in testing from the time of collection but as much as
16 days of delay occurred due to staff paid time off or other
obligations of the study group. Provision of fresh RBC
reagents and minimal change in lots to reduce titer value
variation is ideal but could not be avoided; smaller batches

TRANSFUSION 5
PAGANO ET AL.
of sample testing may shorten delay of testing and optimize
the use of fresh RBC reagents.
In conclusion, replacing routine volume reduction of
ABO-incompatible platelets for transfusion with PAS-F
platelets can yield multiple benefits, including improved
safety, efficiency, and cost-effectiveness. We conclude that
full-volume PAS platelets are an effective strategy to replace
plasma reduction of minor incompatible platelets, including
for hematopoietic stem cell transplant recipients.
CONFLICT OF INTEREST
The authors have disclosed no conflicts of interest.
REFERENCES
1. Berseus O, Boman K, Nessen SC, et al. Risks of hemolysis due
to anti-A and anti-B caused by the transfusion of blood or
blood components containing ABO-incompatible plasma.
Transfusion 2013;53(Suppl 1):114S-23S.
2. Josephson CD, Mullis NC, Van Demark C, et al. Significant
numbers of apheresis-derived group O platelet units have
“high-titer” anti-A/A,B: implications for transfusion policy.
Transfusion 2004;44:805-8.
3. Fung MK, Downes KA, Shulman IA. Transfusion of platelets
containing ABO-incompatible plasma: a survey of 3156 North
American laboratories. Arch Pathol Lab Med 2007;131:909-16.
4. Karafin MS, Blagg L, Tobian AA, et al. ABO antibody titers are
not predictive of hemolytic reactions due to plasma-
incompatible platelet transfusions. Transfusion 2012;52:2087-93.
5. Vaezi M, Oulad Dameshghi D, Souri M, et al. ABO incompati-
bility and hematopoietic stem cell transplantation outcomes.
Int J Hematol Oncol Stem Cell Res 2017;11:139-47.
6 TRANSFUSION
6. O’Donghaile D, Kelley W, Klein HG, et al. Recommendations
for transfusion in ABO-incompatible hematopoietic stem cell
transplantation. Transfusion 2012;52:456-8.
7. Ringwald J, Zimmermann R, Eckstein R. The new generation of
platelet additive solution for storage at 22 degrees C: develop-
ment and current experience. Transfus Med Rev 2006;20:
158-64.
8. Tobian AA, Fuller AK, Uglik K, et al. The impact of platelet
additive solution apheresis platelets on allergic transfusion
reactions and corrected count increment (CME). Transfusion
2014;54:1523-9 quiz 2.
9. Cohn CS, Stubbs J, Schwartz J, et al. A comparison of adverse
reaction rates for PAS C versus plasma platelet units. Transfu-
sion 2014;54:1927-34.
10. Weisberg SP, Shaz BH, Tumer G, et al. PAS-C platelets contain
less plasma protein, lower anti-A and anti-B titers, and
decreased HLA antibody specificities compared to plasma
platelets. Transfusion 2018;58:891-5.
11. Surowiecka M, Zantek N, Morgan S, et al. Anti-A and anti-B
titers in group O platelet units are reduced in PAS C versus
conventional plasma units. Transfusion 2014;54:255-6.
12. Quillen K, Sheldon SL, Daniel-Johnson JA, et al. A practical
strategy to reduce the risk of passive hemolysis by screening
plateletpheresis donors for high-titer ABO antibodies. Transfu-
sion 2011;51:92-6.
13. Harris SB, Josephson CD, Kost CB, et al. Nonfatal intravascular
hemolysis in a pediatric patient after transfusion of a platelet
unit with high-titer anti-A. Transfusion 2007;47:1412-7.
14. Cid J, Harm SK, Yazer MH. Platelet transfusion - the art and
science of compromise. Transfus Med Hemother 2013;40:
160-71.
15. Dudaryk R, Hess AS, Varon AJ, et al. What is new in the blood
bank for trauma resuscitation. Curr Opin Anaesthesiol 2015;28:
206-9.