Upon activation, many cells shed components of their plasma membranes as microparticles. Depending on the methods of preparation and analyses, microparticle counts may vary significantly between laboratories, making data analyses and clinical correlations challenging. To assess how variations in sample preparation affect microparticle measurements, blood samples from 13 healthy, adult volunteers were labeled with Annexin V, cell-specific antibodies, and antibodies against tissue factor (TF). Data were acquired and analysed using an EPICS XL-MCL flow cytometer. Annexin V(+) monocyte-, platelet-, endothelial-, or erythrocyte-derived microparticles accounted for 10.4%, 38.5%, 43.8%, and 7.3% of the total number of microparticles (13.7 +/- 3.0 x 10(3)/ml of whole blood), respectively. A similar distribution of cell types was seen for TF(+) microparticles (6.3 +/- 2.6 x 10(3)/ml of whole blood). No statistical difference was noted in microparticle distribution using either 19- or 21-gauge needles. Elevated levels of platelet- and erythrocyte-derived microparticles were detected in heparin and PPACK-anticoagulated samples as compared to samples anticoagulated with ACD or sodium citrate (P < 0.05, student's t-test). Additional centrifugation was critical for removing platelet contamination, which significantly affected microparticle counts. Finally, Annexin V(+) and TF(+) microparticles were significantly reduced upon sample storage at low temperatures. Microparticle levels are significantly affected by variations in sample preparation and storage. These results illustrate the need to standardize assay protocols in order to obtain consistent measurements. Our studies further optimize sample preparation for microparticle detection.