University of Washington/Seattle Children's/BloodworksNW, Washington, United States
Background/Case Studies: Some studies have found a decreased corrected count increment (CCI) in patients receiving psoralen-treated platelets in 65% platelet additive solution/35% plasma (PR-PAS) compared to conventional platelets in 100% plasma (PLT-P). No studies have reported increments with psoralen-treated platelets in plasma (PR-P). This study compares CCIs for PR-PAS, PR-P, and PLT-P transfusions at a tertiary care pediatric hospital.
Study
Design/Methods: Platelet components issued from September 1, 2021 to January 24, 2023 with the exception of washed platelets were retrospectively evaluated. PR-PAS were collected by Amicus separator (Fresenius Kabi) whereas PLT-P and PR-P were collected by Trima Accel (Terumo BCT). CCI calculations used the DuBois method for body surface area and assumed 3x10^11 platelets per apheresis unit. For aliquots, the percentage of platelet volume aliquoted from the original unit was multiplied by 3x10^11 to calculate the number of platelets transfused. The 4-hour and 24-hour CCIs used post-transfusion platelet count within 5 hours and 12-25 hours after issue, respectively, and pre-transfusion platelet count within 6 hours before issue. Generalized linear mixed models were used to assess differences between platelet product types.
Results/Findings: A total of 7309 platelet transfusions in 573 patients were evaluated: 4266 PLT-P, 2129 PR-PAS, and 914 PR-P. The median 4-hour CCI with PR-PAS was 2747 (interquartile range (IQR) 0, 6477) compared to 11686 (IQR 5364, 19391) with PR-P and 9905 (IQR 4581, 16980) with PLT-P (Figure A.1). For 24-hour CCI, the values were PR-PAS 1329 (IQR 1400, 4703), PR-P 5374 (IQR 1097, 11670), PLT-P 6292 (IQR 1964, 12780) (Figure A.2). Modeling CCIs, accounting for patient clustering effects, found 4-hour CCI for PR-PAS at 3960 (95% confidence interval (CI) 2935, 4985), significantly lower than PR-P (11999; 95% CI 10153, 13911) and PLT-P (11874; 95% CI 10645, 12924) with p < 0.001 for both (Figure A.3). The pattern was the same for 24-hour CCIs (PR-PAS 2681 [95% CI 2030, 3199], PR-P 7065 [95% CI 5996, 8232], PLT-P 7444 [95% CI 6757, 8349]) with CCI of PR-PAS significantly lower than PR-P and PLT-P (p < 0.001; Figure A.4). When limiting to the 317 patients with hematology/oncology diagnoses or hematopoietic progenitor cell transplant, a similar pattern was observed. Median age in days of PR-PAS was 4 (IQR 4, 5) compared to 5 (IQR 4, 5) for PR-P and 7 (IQR 6, 7) for PLT-P. There was no statistical difference in percentage of patients receiving major incompatible transfusions. Platelet transfusion interval was not significantly different between groups (p=0.67). Conclusions: PR-PAS showed significantly reduced CCI compared to PLT-P or PR-P. PR-P have a similar CCI compared to PLT-P. Further study is needed to understand whether lower CCI is associated with apheresis collection system (e.g., Amicus versus Trima), platelet additive solution, and/or psoralen-treatment.
Importance of research: This is the first study to report corrected count increment (CCI) for psoralen-treated platelets in plasma (PR-P) and provides foundation for future work on impact of apheresis collection instrument, platelet additive solution (PAS), and psoralen-treatment on platelet increment and utilization. Our study finds that psoralen-treated platelets in PAS-C showed significantly reduced CCIs compared to conventional platelets in plasma (PLT-P) or PR-P. PR-P demonstrated comparable CCI to PLT-P.
