The potential use of autologous thrombocytic growth factors to accelerate bone regeneration requires improved methods of isolating platelet-rich plasma (PRP). In addition to discontinuous cell separation, a second method by which PRP is produced at the point-of-care has now become available. In this study, growth factor levels in PRP from these two sources were compared. Whole blood was drawn from 115 healthy donors (73 males, 42 females) aged 21 - 62 years (mean 36, SD 10). The PRP was separated by the blood bank (BB) using the discontinuous cell separation method or at the 'point-of-care' by the so-called 'buffy coat' method (analogous to the Curasan PRP Kit). Growth factor content differed significantly for TGF-beta1 (BB 268.65+/-70.77 ng/ml, Curasan 95.02+/-60.67 ng/ml (sign test P<0.001)) and PDGF-AB (BB 133.59+/-46.26 ng/ml, Curasan 233.70+/-111.86 ng/ml (P<0.001)), while the content of IGF-I (BB 85.37+/-25.58 ng/ml, Curasan 101.72+/-47.7 ng/ml (P<0.160)) showed no significant difference. The higher thrombocyte count in the BB PRP (BB 1434300+/-351960/ microl, Curasan 908.500+/-492.30/microl) seems to result in higher TGF-beta1 levels, while the higher leukocyte count in the Curasan PRP (BB 160+/-320/ microl, Curasan 30130+/-12500/microl) seems to result in higher PDGF-AB levels. The similar IGF-I levels in the two preparations might merely reflect similar amounts of plasma in the PRP produced by each approach.