Comparison of the reliability of brain lesion localization when using traditional and stereotactic image-guided techniques: a prospective study.

Object: Accurate localization of brain lesions is of utmost importance. Traditional methods of localization that involve the use of neuroimaging and surface anatomy have been replaced in certain cases by using frameless stereotactic neuronavigational systems. Even though these systems have been found to be accurate, no studies have been conducted to investigate whether the systems provide improved localization accuracy compared with traditional methods.

Methods: Twenty-two patients undergoing image-guided surgery with the aid of the Stealth Neuro-Station were prospectively enrolled in this study. All patients underwent standard magnetic resonance or computerized tomography imaging, as well as special Stealth-sequenced imaging acquired using scalp fiducial markers. Traditional and Stealth estimates of the surface projection of lesions were determined, digitally photographed, and later compared. The mean (+/- standard deviation) error associated with traditional localization of lesions was 1.1 +/- 0.7 cm in the mediolateral plane and 1.3 +/- 1.1 cm in the anteroposterior plane. This error was not significantly affected by the size or location of the lesion.

Conclusions: Findings of this study indicate that the conventional localization technique used to demarcate brain cortical and subcortical lesions has an error of approximately 1 to 1.5 cm in both the mediolateral and anteroposterior directions. This error can be reduced by judicious use of image-guided techniques.