Clinical usefulness of measuring prothrombin time as a routine admission test

A questionnaire indicating the presence of a history or physical findings consistent with liver desease or bleeding disorders was completed by house officers on 301 admissions to a Veterans Administration medical service. Each patient had prothrombin time determined on admission. Only two of 107 patients for whom the prothrombin time was measured as a screening test had an abnormal results and one of these was normal when repeated. Of 73 patients with a history of alcoholism but no other pertient case history or physical examination results, only one had an abnormal prothrombin time. Of the remaining 121 patients, who had a pertinent history or physical examination, 41 had an abnormal admission prothrombin time. We conclude that measurement of prothrombin time adds little additional information to that obtained by history and physical examination in screening for liver disease and coagulation defects unless the patient has specific clinical evidence of liver disease, anticoagulation, or other conditions predisposing to bleeding disorders.