Prognosis after traumatic brain injury


Update content

Title Prognosis after traumatic brain injury
Period 10 / 1998 - 12 / 2005
Status Completed
Dissertation Yes
Research number OND1286241


- Project summary: Traumatic brain injury (TBI) is a major health and socio-economic problem throughout the world, and is the leading cause of death and disability in younger patients in developed countries. Patients with TBI constitute a heterogeneous population, including patients with varying risk profiles according to baseline clinical characteristics. A consequence of this heterogeneity is that the prediction of individual long-term outcome = e.g. mortality - is complex. Several characteristics have been identified as powerful predictors, e.g. age, Glasgow Coma Scale, motor score, hypoxia, hypotension and Computed Tomography (CT) classification. However, it is still uncertain what characteristics can be included best in a prognostic model predicting outcome after TBI and how this inclusion can be done best. For example, there are still debates about how age can be included best - categorical with an age threshold or continuous - and in what age transformation or with what threshold. Further, it is uncertain whether the frequently used CT classification, constructed of several individual CT parameters, has optimal predictive value or that another combination of CT characteristics may be better from a prognostic point of view. In the past several multivariable models to predict long-term outcome have been developed. However, relatively small patient samples were often used to derive the model and most samples originated from one single place or region, thus restricting generalizability of the model. Also the validity of previously developed models was only seldom examined in a new patient population. The aim of the present project is to gain more knowledge about prognosis after severe or moderated TBI. We distinguish three issues: 1) identification of important predictors of outcome, 2) identification of the optimal way to include predictors into a prognostic model and 3) development and validation of prognostic models. - Progress: When comparing outcome in two trial populations with similar design (n=2176), we identified region (USA versus non-USA) as an important and independent predictor of six-month outcome. We also studied many continuous and categorical transformations of age in four separate large data sets (total n=2664). Currently, we are developing and validating prognostic models to predict six-month mortality and unfavorable outcome according to the Glasgow Outcome Scale.

Related organisations

Related people

Supervisor J.D.F. Habbema
Doctoral/PhD student Dr. C.W.P.M. Hukkelhoven


D21100 Bioinformatics, biomathematics
D23230 Neurology, otorhinolaryngology, opthalmology

Go to page top
Go back to contents
Go back to site navigation