Improving technology for collecting platelets by apheresis: five-year experience in one blood center

Over the past decade, newly introduced methods for apheresis platelet collection have led to increased collection yields. This has resulted in "splitting," which allows transfusion of 2 patients from 1 high-yield collection. Although many small studies exist, no large studies have described the impact of methodological changes on routine blood center collections. We constructed a database containing selected parameters from 45,224 apheresis collections spanning July 1997 to April 2002, using Gambro BCT Spectra (Lakewood, CO), Fenwal CS-3000+ (Baxter Healthcare Corp, Fenwal Division, Deerfield, IL), and Baxter Amicus instruments. A Baker 9110+ hematology analyzer (Bio Chem Immunosystems, Inc., Allentown, PA) was used for platelet counting. Monthly average collection yields, distribution yields (product platelet contents after splitting), and split rates (the fraction of donations which may be split) were determined. The monthly mean collection yield and split rate correlated very closely. Both rose throughout the study period. The split rate climbed from 25% to 70% by study end. However, mean monthly distribution yields decreased by 7% because split and unsplit platelet yields both rose as split rates rose. Overcollections with the Amicus correlated with underestimation of donors' true preprocedure platelet counts during machine programming. Undercollections occurred in donors with low counts and, with-single needle Amicus, microcytic platelet collection. These results may assist in the optimization of an apheresis program. Increased collection yields correlated with cell separator type, dual-needle access, donor platelet count >250 x 10(9)/L, programming with true preprocedure platelet counts and capacity for triple product preparation from collection yields exceeding 2-bag storage capacity.