In order to identify variables obtained at admission that could be used to predict survival in septicemic foals, medical records of 65 foals diagnosed with septicemia were reviewed. Initially, variables were analyzed independently (univariate analysis) for association with survival. Of the physical examination and historical data examined using univariate analysis, the ability to stand at admission, respiratory rate greater than or equal to 60 breaths per minute (bpm), and normal-appearing mucous membranes were significantly associated with survival. Foals with history of induced parturition were significantly less likely to survive. The following hematologic and serum biochemical variables determined at admission were significantly associated with survival: white blood cell count greater than or equal to 6,000 cells/mu L, neutrophil count > 4,000 cells/mu L, serum albumin concentration > 2.2 g/dL, serum glucose concentration > 120 mg/dL. blood pH greater than or equal to 7.35, and positive base excess. The administration of plasma at admission was significantly associated with survival. Stepwise multiple logistic regression analysis was performed to evaluate the association between survival and variables identified as significantly associated with survival in bivariate analysis. The final multivariate model selected included the variables standing, duration of clinical signs (24-hour intervals) prior to admission, respiratory rate greater than or equal to 60 bpm, neutropenia (less than or equal to 4,000 cells/ mu L), and neonatal age category. The probability of survival was significantly increased for foals that were standing, had a respiratory rate Ir 60 bpm, and that had a neutrophil count > 4,000 cells/mu L at admission. Probability of survival was significantly decreased for foals that had a longer duration of clinical signs prior to admission. For each 24 hours of duration, the estimated risk of death was increased by 5.8-fold.
