Standardization of platelet counts--problems and pitfalls.

In a previous work, we have compared two automated techniques for platelet-rich plasma using a thrombocounter or whole blood with the continuous flow instrument Auto-analyzer (AA). More recently, we have tested two instruments using whole blood: the first one, fully automated: Coulter S+; and the other one, semi-automated: Clay-Adams. In the present work, these 4 methods are compared to the phase contrast microscope technique, used as reference. The coefficients of variation ranged between 2 and 13 per cent. The coefficients of correlation between the different methods were between 0.92 and 0.96. Platelet distribution curves for platelet volumes obtained in parallel with platelet-rich plasma and with whole blood in 30 controls, show that platelet population are not significantly different. These instruments count particles and, therefore, necessitate a calibration made by means of platelet standards. Different platelet standards have been studied: these containing latex particles seem to give better results than suspensions of human or animal platelets.