Modification of the radiosurgery-based arteriovenous malformation grading system

Area of Science:

• Neurosurgery
• Radiation Oncology
• Medical Imaging

Background:

• The radiosurgery-based arteriovenous malformation (AVM) grading scale predicts outcomes after AVM radiosurgery.
• Simplifying the scale using location as a two-tiered variable was proposed to maintain predictive accuracy.

Purpose of the Study:

• To evaluate if a simplified AVM grading scale, using location as a two-tiered variable, maintains accuracy in predicting outcomes after radiosurgery.
• To assess the impact of scale simplification on predicting AVM obliteration and neurological deficits.

Main Methods:

• Regression analysis modeling was used on 220 patients who underwent AVM radiosurgery (1987-1992) to develop a modified scale.
• The modified scale equation incorporates volume, age, and a two-tiered location variable (hemispheric/corpus callosum/cerebellar vs. basal ganglia/thalamus/brainstem).
• The modified scale's accuracy was tested on 247 patients who underwent AVM radiosurgery (1990-2001).

Main Results:

• No significant difference was found between the original and modified scales regarding AVM obliteration without new neurological deficits (P = 0.53).
• No significant difference was observed in the decline of the Modified Rankin Scale between the two scales (P = 0.56).
• The modified scale showed strong correlations with AVM obliteration rates (r = -0.98, P < 0.01) and Modified Rankin Scale decline (r = 0.99, P < 0.01).

Conclusions:

• Simplifying the radiosurgery-based AVM grading scale using location as a two-tiered variable does not detract from its accuracy.
• This validated, simplified scale is recommended for future AVM radiosurgery studies to stratify patients for comparative analyses.
• The modified scale has been validated across multiple centers using different radiosurgery modalities (gamma knife and linear accelerator).