Predicting surgical outcome in patients with glioblastoma multiforme using pre-operative magnetic resonance imaging: development and preliminary validation of a grading system.

The lack of a simple, objective and reproducible system to describe glioblastoma multiforme (GBM) represents a major limitation in comparative effectiveness research. The objectives of this study were therefore to develop such a grading system and to validate it on patients who underwent surgical resection. A systematic review of the literature was performed to identify features on pre-operative magnetic resonance imaging (MRI) that predict the surgical outcome of patients with GBM. In all, the five most important features of GBM on pre-operative MRI were as follows: periventricular or deep location, corpus callosum or bilateral location, eloquent location, size and associated oedema. These were then used to develop a grading system. To validate this grading system, a retrospective cohort study of all adult patients with supratentorial GBM who underwent surgical resection between the 1 January 2014 and the 31 June 2015 was performed. There was a substantial agreement between the two neurosurgeons grading GBM (Cohen's κ was 0.625; standard error 0.066). High-complexity lesions were significantly less likely to result in complete resection of contrast-enhancing tumour than low-complexity lesions (50.0 versus 3.4%; p = 0.0007). The proposed grading system may allow for the standardised communication of anatomical features of GBM identified on pre-operative MRI.