Functional MRI (fMRI)–based estimations of language dominance may be distorted by brain tumors near the language cortex. The cause of this distortion is uncertain, as is whether it is an artifact of multiple confounders. The authors hypothesized tumor bias based on confounder-free laterality estimates and that the effects are largest for tumors close to Broca’s area. The researchers looked at 1113 patients who had preoperative fMRI. There were 60 patients in total. The authors matched patients with malignancies in the left and right hemispheres and anterior and posterior tumors. They used whole-brain and region of interest (ROI) techniques (Broca’s and Wernicke’s regions) to calculate standard fMRI laterality indices (LIs).
Tumors in the left hemisphere near Broca’s region reduced LIs regardless of known confounders. It indicates a drop in LI values in patients with left anterior tumors compared to right anterior tumors at the whole-brain level. When compared to patients with right anterior tumors, Broca’s area LIs were considerably lower (p = 0.02) in patients with left anterior tumors (mean LI 0.28) (mean LI 0.70). The authors warn against assuming that there is no or little risk of language function after tumor removal in the left inferior frontal gyrus when language laterality is reduced. Their findings address how to interpret functional magnetic resonance imaging (fMRI) data for neurosurgical purposes, as well as theoretical problems of contralesional active compensation and disinhibition.
Reference:thejns.org/view/journals/j-neurosurg/135/6/article-p1674.xml
