Study on biological variation of haemostatic parameters in clinically healthy dogs.

Thromboelastography (TEG) may be a valuable supplement to the coagulation assays activated partial thromboplastin time (aPTT), prothrombin time (PT), thrombin time (TT), fibrinogen, antithrombin (AT) and D-Dimer currently used in most clinical pathology laboratories. Allowable imprecision and bias reference limits for analytical tests can be calculated based on measurements of biological variation. No studies to date have examined the effect of biological variation on these haemostasis parameters in the same group of dogs. Plasma samples were collected after a set protocol once weekly for five consecutive weeks from eight healthy dogs (four males and four females) and stored at -80 degrees C until analysis. Randomized duplicate coagulation tests and TEG analyses were performed on all plasma samples within one run. The data were analyzed for outliers and subsequently subjected to nested analysis of variance to obtain the coefficient of analytical, intra-individual and inter-individual variation. From these objective analytical performance standards for imprecision, critical difference, total error and the index of individuality were calculated to assess the utility of conventional population-based reference ranges. All the clotting times (aPTT, PT and TT), fibrinogen, AT and D-Dimer showed a degree of individuality, which may make the use of population-based reference ranges alone an insensitive interpretation criterion, whereas a population-based reference interval seems to be sensitive for interpreting all TEG parameters. Analytical performance standards for imprecision were only met for one of the coagulation assays, whereas all TEG parameters except the alpha angle, alpha achieved this analytical goal.