Abstract

Thromboelastography methods have been used to predict or monitor treatment of haemophilia patients with recombinant activated factor VII (rFVIIa). However, neither of the two thromboelastographic methods (ROTEM and TEG) has as yet been validated. This multi-centre, randomised trial compared both methods in terms of intra- and inter- patient variability following in vivo and ex vivo rFVIIa administration to haemophilia A and B patients with and without inhibitors. Patients ((3)16 years old) received the same intravenous rFVIIa dose (45, 90 or 180 microg/kg) twice, 1-12 weeks apart. Blood samples were collected pre-dose and 15, 60, 120 and 240 minutes post-dose for ROTEM and TEG analysis. Pre-dose samples were also spiked ex vivo with rFVIIa (0. 6, 1. 2 or 2. 4 microg/ml), to correspond to the three in vivo doses. Twenty-six haemophilia A and four haemophilia B patients were enrolled. A significant treatment effect was observed with in vivo rFVIIa (p<0. 05) with more pronounced effects in inhibitor (n=14) versus non-inhibitor (n=16) patients. There was a strong positive correlation between ROTEM and TEG parameters. Intra- and inter-patient variation was large for all thromboelastography parameters at all time points and rFVIIa doses. Intra-patient variation was generally lower for non-inhibitor than inhibitor patients, and lower following ex vivo spiking versus in vivo rFVIIa administration. In conclusion, there was a clear effect of rFVIIa on all thromboelastography parameters, but the large intra- and inter-patient variability following in vivo rFVIIa administration renders the use of our method unsuitable for dose-response prediction for haemophilia patients in the clinical setting.

Abstract

Fibronectin, an opsonic glycoprotein, is known to bind fibrinogen and fibrin. Microaggregate debris contained in stored bank blood is composed of degenerating platelets, leukocytes, and fibrin strands. The debris ranges in size from 10 to 160 micrometers. This study examined the effect of transfusion of 2 units of stored red blood cells, containing varying amounts of macroaggregate debris, on in vivo levels of fibronectin. Anemic outpatients were selected at random to receive blood transfusion through either a 170 micrometer standard blood filter or one of four microaggregate filters. A sixth group received saline-washed red blood cells. Results showed that the greatest drop in posttransfusion levels of fibronectin (39 microgram/ml) were found in the group that received blood through a 170 micrometer standard filter and thus received the greatest amount of microaggregate debris (p less than 0.05; N = 50). The patients who received the least microaggregate debris, those receiving washed red cells, showed the smallest decrease in posttransfusion levels of fibronectin (9 microgram/ml); this decrease was not significant (p greater than 0.05; N = 50). Patients receiving blood through 20 to 25 micrometer microaggregate blood filters showed a fall in posttransfusion levels of fibronectin (10 to 15 microgram/ml) intermediate between that found for the two groups described above; this decrease was not significant (p greater than 0.05; N = 35). The decrease in posttransfusion levels of fibronectin found after administration of blood through a 40 micrometer microaggregate screen filter (31 microgram/ml) was significantly different from pretransfusion levels (p less than 0.05; N = 20). Data obtained from this study suggest that transfusion of the microaggregate debris contained in 2 units of stored bank blood can lower in vivo levels of fibronectin. We conclude that if maintaining high levels of fibronectin is shown to be of value in the treatment of critically ill patients, removal of microaggregate debris from any blood transfusions required by these patients would be warranted.