Comparison of the prothrombin time-derived and Clauss assays for the measurement of plasma fibrinogen in hospitalized dogs.

ABSTRACT

BACKGROUND: Accurate measurement of fibrinogen is necessary for detecting bleeding tendencies and inflammation. The Clauss assay determines fibrinogen concentration from its inverse relationship with thrombin-induced clot times. PT-derived assays determine fibrinogen concentrations from changes in the optical density during a routine prothrombin assay and allow determination of fibrinogen without additional reagents. This method has not been assessed in clinically ill dogs. OBJECTIVES: We aimed to determine the agreement between the Clauss and PT-derived fibrinogen assays and compare the ability of the assays to predict surgery-associated transfusions and discriminate between dogs with and without bleeding. METHODS: Retrospective medical record review identified 200 dogs with a variety of underlying diseases with results from both assays. The two assays were compared using Passing-Bablok regression, and the ability of the assays to identify bleeding and predict the need for transfusions was assessed with receiver operating characteristic (ROC) curve analyses. RESULTS: The PT-derived assay displayed constant (y-intercept, 32 mg/dL; 95% CI 18-41) and proportional (slope, 0.79; 95% CI 0.75-0.82) bias compared with the Clauss assay. The Clauss assay reported lower values than the PT-derived assay at lower fibrinogen concentrations and higher values at higher concentrations. Comparing the area under the ROC curve did not detect significant differences in the ability of the two assays to discriminate between dogs with and without bleeding or predict the need for surgery-associated transfusions. CONCLUSIONS: The PT-derived and Clauss assays are not interchangeable, and the Clauss assay could be more sensitive to hypofibrinogenemia in dogs.