The prognostic value of admission point-of-care testing and modified Glasgow Coma Scale score in dogs and cats with traumatic brain injuries (2007-2010): 212 cases.

Abstract

To investigate the association between admission point-of-care parameters and survival to discharge in dogs and cats with traumatic brain injury (TBI). Retrospective study 2007-2010. A rural university teaching hospital and an urban private practice teaching hospital. One hundred thirty-one dogs and 81 cats admitted to the emergency service with evidence of head injury based on history from the owner or physical exam. In dogs, nonsurvivors had significantly higher glucose concentrations at admission than survivors (median=8.49 mmol/L [153mg/dl] vs 6.83mmol/L [123mg/dl], p=0.039). In cats, there was no significant difference in admission glucose between survivors and nonsurvivors (median=10.21mmol/L [184mg/dl] vs 10.93mmol/L [197mg/dl], p=0.17). Modified Glasgow Coma Scale (MGCS) score was available for 105 of the 131 dogs (80%) and was significantly higher in survivors than in nonsurvivors (median=16 vs 11, p<0.0001). MGCS was available for 45 of the 81 cats (56%) and was significantly higher in survivors than in nonsurvivors (median=17vs 14, p=0.0005). The relative prognostic value of the admission point-of-care testing parameters and MGCS were assessed using a stepwise linear regression model, which included admission glucose, pH, base excess, sodium, and MGCS. In dogs, only admission glucose was an independent predictor of survival (odds ratio=1.027, 95% confidence interval, 1.0042-1.05, p=0.019). These results suggest that, as in people with TBI, increased blood glucose concentrations may have prognostic significance in dogs with TBI but not in cats. In addition, MGCS score may be predictive of survival in both dogs and cats with TBI.