gms | German Medical Science

58. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie e. V. (DGNC)

Deutsche Gesellschaft für Neurochirurgie (DGNC) e. V.

26. bis 29.04.2007, Leipzig

The predictive ability of the Glasgow Coma Score

Die Vorhersagekraft des Glasgow Coma Scores

Meeting Abstract

  • corresponding author D. Woischneck - Neurochirurgische Abteilung, Universitätsklinikum Ulm
  • E. Rickels - Neurochirurgische Abteilung, Universitätsklinikum Ulm
  • M. Skalaj - Neuroradiologische Abteilung, Universitätsklinikum Magdeburg
  • R. Firsching - Klinik für Neurochirurgie, Universitätsklinikum Magdeburg

Deutsche Gesellschaft für Neurochirurgie. 58. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie e.V. (DGNC). Leipzig, 26.-29.04.2007. Düsseldorf: German Medical Science GMS Publishing House; 2007. DocFR.07.01

The electronic version of this article is the complete one and can be found online at:

Published: April 11, 2007

© 2007 Woischneck et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Objective: The study analyzes the predictive ability of the Glasgow Coma Score (GCS) in correlation with the distribution of cerebral lesions after severe brain injuries (BI).

Methods: In 200 patients with BI, GCS was recorded continuously. In all patients, magnetic resonance (MRI) data were acquired within 8 days after the injury. The outcome was measured 6 months after injury by use of the Glasgow Outcome Score (GOS). GCS and MRI data were correlated with the GOS in a multivariate analysis. Statistics were performed by use of the software package SPSS 13.0 for Windows XP. Significance was assumed for p<0,01.

Results: GCS correlated significantly with outcome on the day of MRI examination (mean: 3 days after BI). GCS on and 24 hours after admittance did not correlate with GOS nor with the location of any intracerebral lesion. GCS at the day of the MRI examination revealed a significant relation to theMRI data: In patients with GCS£5, the number of lesions in the pons and medulla oblongata was increased. In patients with GCS>5, lesions of the mesencephalon and cases without any infratentorial damage were statistically more frequent. A peculiarity was a significant accumulation of patients with basal ganglia lesions among those cases with GCS 5-6, associated with severe or slight disability.

Conclusions: The minor predictive potency of the GCS after BI is caused by an inadequate correlation with those cerebral lesions, which decide the clinical outcome. Several studies demonstrated the low predictive value of the GCS. Nevertheless, this might also be explained by deficits in outcome measurement. Consideration of the location of cerebral lesions is a measuring rod based on pathophysiology: it helps to improve income and outcome research.