Abstract 3698


Thromboelastography (TEG) is a method of analyzing coagulation commonly used to assess pre- and perioperative bleeding risk. We have been interested in the role of cell-derived microparticles (MP) in coagulation and in this study, examined correlations between TEG parameters and MP levels as described in Methods.


Patients: Whole citrated blood from a total of 43 patients was examined by TEG (Haemoscope TEG 5000), within 5 hours of drawing. The study subjects consisted of 12 ITP, 5 TTP, 26 thrombocytopenia (TP) of various causes and 21 healthy controls. TEG: Parameters analyzed were maximum amplitude (MA), clot strength (G), lag time (R), time to formation of clot of amplitude 20 mm (k), rate of growth (a), coagulation index (CI). MP Assay: MP from platelets (PMP) were defined flow cytometrically by CD31+/CD41+; from red cells (RMP) by glycophorin A (CD235); from endothelia (EMP) by CD62E; and from leukocytes (LMP) by CD45. In addition, we employed generic markers, lectin FITC-Ulex europaeus (Ulex) as a measure of “total MP”; and annexin V+ (AnV+) MP, a marker of procoagulant phospholipids. Platelet counts were obtained from CBC at the clinical laboratory.


We assessed correlations between each TEG parameter, and each MP parameter, as well as platelet counts. Only those which were significant (p<0.05) are reported below. (1) RMP levels correlated positively and strongly with lag time, R (r=0.737, p<0.0001). LMP levels correlated positively with k (r=0.553, p=0.0001) moderately. Interestingly, we found moderate but consistent positive correlation between PMP levels and MA, G, and CI. These values were r=0.314, r=0.317, and r=0.346, respectively, and p<0.05 in each case. (2) Normal control values had an average G of 7.86 ±1.70 Kd/cm2 (mean ±SD), in contrast to ITP, TTP, and TP patients which had average G's of 8.62 ±3.96, 10.34 ±1.97, and 5.96 ±3.98 Kd/cm2, respectively. However, no statistically significant differences of means between patient groups and controls were found for any parameter in these small populations. (3) Platelet count was found to have a moderate correlation with G in TP patients (r=0.427, p<0.05) and with k (r= -0.459, p=0.018), but neither variable (G, k) correlated with platelet counts in ITP or TTP, possibly owing to the smaller populations in those groups. (4) Examining all patients combined for correlation between platelet count and G, we found r=0.633 (p<0.0001), confirming previous studies suggesting a relation between platelet count and G in both healthy and diseased patients. High platelet count was reasonably well correlated to high PMP (r=0.506, p<0.001). This may indicate that platelet activation as well as platelet count influences TEG, as it has been shown that PMP levels are a proxy of platelet activation.


We conclude that MP contributes significantly to multiple variables in TEG profiles of patients with ITP, TTP, and TP, some positively, some negatively; and that there is wide variability in MP levels among these diseases. Platelets clearly also modulate TEG parameters, although the interplay between platelets and MP in forming a stable clot bears further research. PMP have the largest array of effects, positively correlating with MA, G, and CI, and negatively correlating with R+k, while having no significant effect on the other parameters. The data suggest that nearly all the MP affect the TEG parameters significantly. The most commonly affected parameter was MA, which is correlated to EMP, LMP, PMP, and Ulex. G and CI were also frequently affected by various MPs. This study is limited by the small patient population, various medications, and varying degree of severity of diseases.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.