Use of a test for proteins induced by vitamin K absence or antagonism in diagnosis of anticoagulant poisoning in dogs: 325 cases (1987-1997).

OBJECTIVE: To determine usefulness of the test for proteins induced by vitamin K absence or antagonism (PIVKA) to identify anticoagulant-poisoned dogs, compared with one-stage prothrombin time (OSPT) and activated partial thromboplastin time (APTT) tests.
DESIGN: Retrospective study. ANIMALS: 325 dogs. PROCEDURE: Comparisons of results of PIVKA, OSPT, and APTT measurements in dogs with anticoagulant poisoning, hepatic disease, disseminated intravascular coagulation, other blood-related disorders, immune-mediated diseases, or other chronic and acute diseases were performed. Median, quartile, and range values were determined.
RESULTS: PIVKA tests with a 150-second critical value had > 98% specificity and > 90% sensitivity for diagnosis of anticoagulant poisoning versus > 99% specificity and > 79% sensitivity with a 300-second critical value. Comparison of PIVKA values among diagnostic groups revealed significant differences between dogs with anticoagulant poisoning and all other groups. CONCLUSIONS AND CLINICAL RELEVANCE: The PIVKA test with a 150-second critical value is diagnostically useful for distinguishing anticoagulant poisoning from other coagulopathies. Severe liver disease can cause false-positive results. Administration of vitamin K1 or early evaluation (within a few hours of ingesting anticoagulant) may cause false-negative results. Dogs with PIVKA test values > 150 seconds and clinical signs of anticoagulant poisoning can confidently be considered to have anticoagulant poisoning because of the high test sensitivity and specificity.