• Bram Jacobs, Pieter E Vos, Tjemme Beems, Ton M van der Vliet, Ramon R Diaz-Arrastia, and George F Borm.
• Department of Neurology, Biostatistics and HTA, Radboud University Nijmegen Medical Centre, Nijmegen, the Netherlands.
• J. Neurotrauma. 2011 Feb 1;28(2):203-15.

AbstractIntracranial lesion volume and midline shift are powerful outcome predictors in moderate and severe traumatic brain injury (TBI), and therefore they are used in TBI and computed tomography (CT) classification schemes, like the Traumatic Coma Data Bank (TCDB) classification. In this study we aimed to explore the prognostic value of lesion volume and midline shift in moderate and severe TBI as measured from acute cranial CT scans. Also, we wanted to determine interrater reliability for the evaluation of these CT abnormalities. We included all consecutive moderate and severe TBI patients admitted to our hospital who were aged ≥16 years, over an 8-year period, as part of the prospective Radboud University Brain Injury Cohort Study. Six months post-trauma we assessed outcomes using the Glasgow Outcome Scale-Extended (GOS-E). We analyzed 605 patients and found an association of both lesion volume and midline shift with outcome; increases were associated with a higher frequency of patients with an unfavorable outcome or death. A cut-off value, such as that used in the TCDB CT classification (lesion volume 25 mL and midline shift 5 mm), was not found. The average interrater difference in volume measurement was 6.8 mL, and it was 0.2 mm for the determination of degree of shift. Using lesion volume and midline shift as continuous variables in prognostic models might be preferable over the use of threshold values, although an association of these variables with outcome in relation to other CT abnormalities was not tested. The data provided here will be useful for stratification of patients enrolled in clinical trials of neuroprotective therapies.

25 keywords...

hide…