Abstract

Thromboelastography (TEG) is a leading candidate for monitoring bypassing agent therapy in hemophilia. It is not clear, however, which pre-analytic methods provide the most sensitivity in assessing the response to clotting factor concentrates. Previous studies have utilized kaolin (K) or recombinant human tissue factor (TF) as activators, and native blood (NB) or citrated blood (CB) as the collection method. Utilizing blood samples from 16 hemophiliacs (8 with and 8 without inhibitors), we compared NB to CB and K to 2 dilutions of TF [1:17,000(TF1) and 1:42,000(TF2)] with and without ex vivo addition of recombinant FVIIa (rFVIIa) and recombinant factor VIII (rFVIII). For patients with inhibitors, 3 experiments were performed for each of the above comparisons: no added factor and addition of rFVIIa 1.5mcg/ml and 4.5 mcg/ml. For patients without inhibitors (all FVIII deficient), 5 experiments were performed: no added factor, rFVIII 0.5 units/ml, rFVIII 1 units/ml, and rFVIIa 1.5 and 4.5mcg/ml. This initial analysis focused on the R parameter (reaction time) which is the time to initial fibrin formation and is the parameter that is most sensitive to the addition of clotting factor concentrates. The R values were analyzed via log transformed random effects ANOVA. All R values are expressed as geometric means. See table for results. There was no significant difference in the R time of NB and CB for any of the experiments. For inhibitor patients following the addition of rFVIIa, the R time was significantly longer for K than for TF1 and TF2 allowing a clearer distinction in response to various concentrations of rFVIIa. This is demonstrated as a significant difference between the 2 concentrations of rFVIIa with the higher concentration demonstrating a shorter R. For patients without inhibitors, activation with K led to a significantly longer R relative to TF1 and TF2 as well with both placebo and rFVIIa. Importantly, when adding a non-activated factor such as rFVIII, R was equivalent for K and TF2 both of which were significantly longer than TF1. In conclusion, 1) we recommend CB as the collection method as it is more practical in clinical settings, and 2) activation with K demonstrates a longer R than TF1 or TF2, resulting in increased sensitivity to addition of 2 clinically relevant concentrations of rFVIIa in both the inhibitor and non-inhibitor patients. Thus we suggest that K is the best activator when analyzing response to rFVIIa in hemophilia patients.

  Observed (Predicted) Geometric Mean R (mins) for CB 
Patient Type Factor Kaolin TF1 TF2 
*p<0.001 for K versus TF1, K versus TF2, and TF1 versus TF2. 
*p<0.001 for rFVIIa 1.5 mcg/ml versus 4.5 mcg/ml for all activation methods. 
#p<0.03 for K versus TF1 for all factors. 
#p<0.001 for K versus TF2 for placebo and both concentrations of rFVIIa 
Inhibitor* rFVIIa 1.5 mcg/ml 50.6 (50.1) 14.7 (14.7) 25.1 (25.3) 
 rFVIIa 4.5 mcg/ml 37.3 (32.1) 11.1 (9.4) 19.4 (16.2) 
No Inhibitor# Placebo 18.2 (17.6) 9.1 (9.1) 13.6 (13.9) 
 rFVIII 0.5 U/ml 7.7 (7.7) 6.0 (6.0) 8.1 (8.1) 
 rFVIII 1.0 U/ml 7.1 (7.1) 5.8 (5.8) 7.6 (7.6) 
 rFVIIa 1.5 mcg/ml 12.7 (11.3) 5.8 (5.9) 10.3 (8.9) 
 rFVIIa 4.5 mcg/ml 9.8 (11.0) 5.8 (5.7) 8.3 (8.7) 
  Observed (Predicted) Geometric Mean R (mins) for CB 
Patient Type Factor Kaolin TF1 TF2 
*p<0.001 for K versus TF1, K versus TF2, and TF1 versus TF2. 
*p<0.001 for rFVIIa 1.5 mcg/ml versus 4.5 mcg/ml for all activation methods. 
#p<0.03 for K versus TF1 for all factors. 
#p<0.001 for K versus TF2 for placebo and both concentrations of rFVIIa 
Inhibitor* rFVIIa 1.5 mcg/ml 50.6 (50.1) 14.7 (14.7) 25.1 (25.3) 
 rFVIIa 4.5 mcg/ml 37.3 (32.1) 11.1 (9.4) 19.4 (16.2) 
No Inhibitor# Placebo 18.2 (17.6) 9.1 (9.1) 13.6 (13.9) 
 rFVIII 0.5 U/ml 7.7 (7.7) 6.0 (6.0) 8.1 (8.1) 
 rFVIII 1.0 U/ml 7.1 (7.1) 5.8 (5.8) 7.6 (7.6) 
 rFVIIa 1.5 mcg/ml 12.7 (11.3) 5.8 (5.9) 10.3 (8.9) 
 rFVIIa 4.5 mcg/ml 9.8 (11.0) 5.8 (5.7) 8.3 (8.7) 

Disclosures: Young:Novo Nordisk: Consultancy, Honoraria, Research Funding.

Author notes

Corresponding author