OPTIMIZATION OF A NEW CONTINUOUS UV ASSAY FOR THE DETERMINATION OF BLOOD-COAGULATION FACTOR-XIII ACTIVITY IN HUMAN PLASMA

The new photometric assay described by Fickenscher et al. (Thromb. Hae mostas. 65 (1991) 535-540) for the determination of factor XIII facili tates the diagnosis of factor XIII deficiency. In spite of easy handli ng, this test should be used critically. Patients with hyperfibrinogen aemia showed factor XIII activities of less than 20%, whereas with an optimized assay we found normal factor XIII values. Also, the use of a fixed period of incubation for the analysis is questionable, because the period of constant reaction rate occurs earlier and is shorter wit h high factor XIII activities and later and longer with low factor XII I activities. A linear relation between factor XIII activity and signa l only exists up to 80% of activity. In some plasma samples from patie nts with hyperfibrinogenaemia the factor XIII determination actually s hows decreased values for factor XIII. During the reaction, a fibrin c lot is formed. The resulting turbidity simulates an increase in absorb ance so that NADH consumption is apparently decreased. In six patients with hyperfibrinogenaemia (8.1 -9.4 g/1), a factor XIII activity of 2 6 U/I or less was determined. Using 50 mul instead of 100 mul sample v olume, 50% (3/6) of the patients showed a normal factor XIII activity (80-96 U/l), whereas 50% (3/6) values of 6 - 15 U/l were found. In our modified assay we measured normal factor XIII activities (72 - 151 U/ l) in all 6 patients. The procedure is optimized by reducing the sampl e volume from 100 mul to 50 pl. The concentration of the fibrin aggreg ation inhibitor is raised from 0.5 g/l to 1.0 g/l in the test solution , and the period of constant reaction rate is used for the determinati on. Enzyme activity is expressed in U/l (37-degrees-C). Optimized in t his way, the assay shows a linear relationship between the decrease of absorbance and factor XIII activity up to more than 128 U/l, correspo nding to 140%. Plasma samples with high concentrations of fibrinogen c an be analysed without problems. Precision was 2.3% in series and 2.8% from day to day, using a control plasma with 81 U/l; for a control pl asma with 35 U/1, the respective values were 4.9% and 5.8%. In apparen tly healthy 100 persons a reference interval of factor XIII activity w as established from 66 to 142 U/l or 73 to 155%.