FINGERSTICK BLOOD-SAMPLES IN PLATELET DONOR SCREENING - RELIABILITY AND IMPACT ON PREDICT YIELD PROGRAMS

Although widely used, the reliability of fingerstick platelet counts f or determining donor eligibility and for use with plateletpheresis pre dict yield programs has not been established. We compared platelet cou nts obtained from fingerstick vs. venous samples in several aspects of apheresis platelet collection. Analysis of 25 paired fingerstick and venous predonation samples demonstrated a poor correlation between pla telet counts (r(2) = .43), with fingerstick counts having a 20% lower mean value (P < .05). The effect of using fingerstick vs. venous predo nation platelet counts with apheresis instrument predict yield calcula tions to obtain target yields was determined. Mean yields collected us ing fingerstick/predict yield were 12% (Fenwal CS3000 PLUS) and 15% (H aemonetics MCS+) higher than venous/predict yield units (P < .05). The coefficients of variation (CV) of fingerstick/predict yield and venou s/predict yield collections were comparable (15% vs. 14% [CS3000] and 23% vs. 21% [MCS+], respectively), indicating that possible difference s in accuracy between fingerstick and venous platelet counts had littl e effect on the variability of predict yield collections. A retrospect ive analysis of the CV of 100 fingerstick/predict yield units vs. 100 units collected by processing standard volumes showed no difference: 2 2% vs. 20% (F = 0.99, CS3000), and 22% vs. 24% (F = 0.89, MCS+), respe ctively. We conclude that fingerstick platelet counts are systematical ly lower and correlate poorly with venous counts, though their use sel dom results in false disqualification of donors. We also conclude that fingerstick count/predict yield collections do not produce more consi stent yields of platelets than standard volume collections. (C) 1997 W iley-Liss, Inc.