Factors contributing to variability in neurocognitive performance before glioma neurosurgery

Abstract

Background Cognitive impairment following anti-tumor treatment is a common concern for brain tumor patients. However, growing evidence indicates that significant impairments can be present even before treatment. The purpose of this study was to identify factors that explain variability in pre-treatment test performance, beyond that of tumor burden. Methods Using multi-step linear regression, we retrospectively probed the contribution of clinical-, tumor-, patient-, and self-reported factors to variance in performance among 96 treatment-naïve brain tumor patients across 13 objective neurocognitive tests. Agreement between subjective and objective reports of cognitive impairment was also examined. Results Clinically significant pre-operative impairments were observed in both objective and subjective domains. Estimated premorbid intelligence quotient (IQ), tumor volume, diagnosis of an astrocytoma, self-reported Depression and perceived Cognitive Functioning Scores were the most common predictors of objective neurocognitive performance prior to treatment, explaining 12.3%-58.3% of the variance. No association was identified between objective and subjective reports of cognitive impairment. Conclusions Glioma patients frequently exhibit objective and subjective impairments prior to treatment. Both tumor- and self-reported factors were identified as predictors of performance, after correcting for estimated premorbid IQ. Nevertheless, more than 41.7% of the variance in cognitive performance remained unexplained, indicating a substantial role for additional, as yet unaccounted for, clinical factors. Notable disparity between objective- and subjective cognitive impairment status re-emphasizes the importance of assessing both domains to ascertain a patient’s overall functioning in the context of treatment outcomes.