The effect of haemolysis on antithrombin concentration as determined by a chromogenic method

Abstract

The presence of free haemoglobin in serum or plasma can markedly affect the outcome of laboratory tests. Normal concentrations of plasma haemoglobin in carefully obtained specimens are less than 0.025g/l. The presence of free haemoglobin in a sample increases the spectrophotometric absorbance of tests run at wavelengths within the absorbance range of haemoglobin (400 – 440nm). Little is known about the effects of haemolysis on the determination of antithrombin levels in canine plasma samples. Two plasma pools, designated AT 100 and AT 70 were prepared. The AT 70 pool was prepared by diluting pooled plasma with 0.9% saline in a ratio of 7:3. A unit of whole blood was collected from a healthy donor animal. The erythrocytes were lysed by freezing and thawing. The solution was centrifuged, the supernatant collected and filtered using 1.2 um and 0.22 um filters sequentially to remove residual red cell stroma. The haemoglobin concentration of the solution was determined using a modification of the automated Drabkin method. Intermediate haemoglobin solutions of decreasing known concentrations were prepared by the addition of saline. The intermediate solutions were added to the plasma pools in a 1+9 manner and a series of samples were prepared with final calculated and measured haemoglobin concentrations of 0.0; 0.5; 1.5; 2.5; 3.5; 4.5 and 5.5 g/l. The AT determinations were performed using a functional chromogenic assay and the spectrophotometric absorbances were read using a 405 nm filter, as specified. Increasing concentrations of haemoglobin resulted in a decrease in the AT value measured. A simple linear regression analysis was performed on both AT70 and AT100 using a two-step regression analysis. The slopes up to [Hb] 1.5g/l were not significant whereas the slopes at between [Hb] 1.5 – 5.5 g/l were significant ( p< 0.001). The slope equation for AT 100 was y= -5.742X+ 115.24 with R 2 = 0.794 and for AT 70 was y = - 4.2037X + 66.821, with R 2 = 0.936. A conversion table was created by interpolation of data between these two lines. These results show that it is possible, using a conversion equation, to perform the AT assay in haemoglobinaenic serum, thus opening the way to further evaluation of the coagulation status in patients with haemolytic disease processes.