Defining the limits of survivorship after very severe head injury

Reliable prediction of outcome after head injury is a daunting task. Although previous reports have highlighted the difficulties of determining outcome in the cohort of severe head injury Glasgow Coma Scale (GCS) score < or = 8), we wondered within the very severely injured population (GCS score 3-5) if a simple combination of clinical parameters may be predictive of poor outcome.

All patients admitted to a Level 1 trauma center with a GCS score of 3 to 5 from 1986 to 1991 inclusive (380 patients) were retrospectively reviewed and outcome a minimum of 6 months after injury was determined by chart review or telephone.

Follow-up was accomplished in all but five patients (1.3%). Functional survival (nonvegetative) was correlated to admission GCS score, pupillary abnormalities, and age. As anticipated, overall functional survival was poor (12.5%), and even worse among those evidencing pupillary abnormalities (6.6%). Interestingly, there was an absence of survivors in the advanced age decades, with the oldest functional survivor of any GCS increasing in a stepwise fashion with increasing coma score. This translated into the oldest survivor of a GCS score of 3 being in their chronologic 30s, a score of 4 in their 40s, and a score of 5 in their 50s. Among patients older than these age decades, that is beyond this simple age/GCS cut-off (32.8% of cohort), there were no functional survivors (95% confidence interval 0, 2.4).

Within the population of very severely head injured patients (GCS score 3-5), the simple combination of age and admission GCS score appears to predict accurately non-functional outcome in almost one third of patients. If confirmed at other centers, this may have wide-ranging implications regarding counseling of families, utilization of resources, and the design of head injury studies.