In the previous study [12], we found that PRP and A-PRF preparations exert distinguishable actions on periosteal cell proliferation. Because both IL-1β and IL-6 are known to be produced by WBCs [23], and because WBCs are not included in PRGF preparations, we thought that the bi-phasic effects of PRP preparations may be attributed to WBCs. Furthermore, if WBCs are highly concentrated in A-PRF, it is expected that IL-1β and IL-6 are concentrated at higher levels to exert negative effects at higher doses of A-PRF extracts in this study. As Choukroun intended [3], we observed that WBCs, as well as platelets, were highly concentrated in A-PRF preparations. Similarly, WBCs were found to be concentrated in CGF preparations. In addition, the inflammatory cytokines were not exceptionally concentrated at higher levels in PRP preparations, and no strong positive correlation between WBC counts and pro-inflammatory cytokine levels was observed in PRP preparations (data not shown). Therefore, we speculate at present that the negative effects of PRP preparations at higher doses may not be due to these pro-inflammatory cytokines or WBCs. However, at the same time, we are concerned how accurately the indirect subtraction method determines WBC and platelet (PLT) counts, because this method does not count any possible adhesion-dependent loss of blood cells or nonuniformity of cell distributions especially in red thrombus. Thus, further studies are needed to perform to obtain convincing evidence to explain this discrepancy.