Predicting survival in dogs with overt disseminated intravascular

Abstract

OBJECTIVE: The clinical diagnosis of disseminated intravascular coagulation (DIC) in dogs is challenging. It was hypothesized that dogs with laboratory abnormalities consistent with overt DIC have increased mortality and hence survival could be used to evaluate the performance of DIC scoring systems. Four DIC scoring systems were compared in clinically ill dogs: The International Society of Thrombosis and Haemostasis (ISTH) score; a previously published veterinary DIC score; a system using in-house reference intervals (RIs) for coagulation assays; and the published veterinary DIC score modified by use of in-house RI values. DESIGN: Retrospective cohort study. SETTING: University teaching hospital. ANIMALS: Eight hundred four client-owned dogs with underlying disorders associated with DIC. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Medical records were reviewed to identify dogs for which a coagulation panel consisting of coagulation times, D-dimer concentration, antithrombin activity, fibrinogen concentration, and platelet count had been analyzed. Four methods for overt DIC scoring were then applied, and the ability of these scores to predict mortality was evaluated. Use of institution RIs for diagnosis of overt DIC provided the most accurate prognostic assessment. A score based on abnormalities in 3 of 6 parameters to identify overt DIC was 72.7% sensitive and 80.9% specific for mortality and was accurate in 78.4% of cases. The mortality rate of dogs diagnosed with overt DIC via this method was 62.5%, compared to 12.9% in the dogs not in overt DIC, a relative risk of nonsurvival of 4.84 (95% confidence interval 3.80-6.16). CONCLUSIONS: In dogs at-risk for DIC, an in-house RI scoring system accurately predicted mortality. With further prospective validation, this system holds promise as a diagnostic tool for consistent characterization of overt DIC in clinical studies and clinical practice.