LAUSR

Platelet transfusion guidelines for hypoproliferative thrombocytopenia were published in December 2014. In March 2015, our laboratory decreased the platelet count critical value from 20,000 cells/mL to 10,000 cells/mL to coincide with the recommendations for transfusion trigger for bleeding prophylaxis. We sought to determine the impact of these interventions in the platelet utilization for patients with acute leukemia (AL) in our institution. Patients admitted with the three main AL Medicare severity diagnosis-related groups (MS-DRGs) 834 (AL without major operating room procedure with major comorbidity), 837 (chemotherapy with AL as secondary diagnosis or with high-dose chemotherapy agent with major comorbidity), and 838 (chemotherapy with AL as secondary diagnosis with comorbidity or high-dose chemotherapy agent) were included. For each MS-DRG, we collected the number of platelet units transfused per patient and the length of hospital stay (LOS). For the three MS-DRGs combined, we calculated the mean pretransfusion platelet counts. In order to detect trends in utilization, the data were aggregated into 3month periods as follows: July to September 2014 (P1), October to December 2014 (P2), January to March 2015 (P3), April to June 2015 (P4), and July to September 2015 (P5). Analysis of variance and t-tests were used for statistical analysis where appropriate. For MS-DRG 834, the mean number of platelet units per patient progressively decreased from 10.6 in P1 to 4.0 in P5, while the mean LOS decreased from 29.6 days in P1 to 16.2 days in P5 (R1⁄4 0.627); for MS-DRG 837, platelet usage decreased from 8.6 units/patient in P1 to 5.1 units/ patient in P5, and the mean LOS was 24.9 days in P1 and 22.3 days in P5 (R1⁄4 .394); for MS-DRG 838, the mean number of units per patient was 1.0 in P1 and 0.9 in P5, and the LOS went from 10.0 days in P1 to 8.4 in P5 (R 1⁄4 .698). For the combined MS-DRGs, the mean pre-transfusion platelet counts decreased from 15,900 cells/mL in P1 to 12,900 cells/mL in P5 (P < .0001). The mean LOS decreased from 20.2 in P1 to 16.3 days in P5 (P 1⁄4 .46) for all patients with AL, and was positively correlated with the mean number of platelets transfused per patient (R1⁄4 .886), while not correlated with the mean pretransfusion platelet count (R 1⁄4 .001). From this analysis, we conclude that while decreasing LOS likely played a role in the reduced number of platelets transfused per patient admitted AJCP / MEETING ABSTRACTS