Factors Affecting the Diagnostic Yield of Frame-Based Stereotactic Intracranial Biopsies

Abstract

Frame-based stereotactic biopsy (FSB) remains the "gold standard" for obtaining diagnostic samples of intracranial lesions to guide therapy. Nevertheless, diagnostic yield is highly variable. This study aims to provide an analysis of diagnostic yield, surgical complications, and factors associated with obtaining nondiagnostic samples in a contemporary FSB series. A retrospective cohort study was conducted of all adult patients with imaging-documented lesions undergoing FSB at our institution between 2013 and 2018. Diagnostic accuracy, lesion characteristics associated with nondiagnostic biopsy, and surgical complications were evaluated. A biopsy was considered nondiagnostic if all frozen samples and the final pathology yielded normal brain tissue or nonspecific reactive tissue unless the "reactive" pathology was consistent with radiation injury from prior therapy. Our search identified 198 FSB patients. Mean (standard deviation) age was 62 ± 17 years, and 44.2% were female. Median procedure time was 32 minutes. A definitive histologic diagnosis was established in 187cases (94.4% diagnostic yield). Mean lesion diameter was 31.9 ± 16.8 mm. Multivariable logistic regression revealed only lesion diameter to be significantly associated with diagnostic result (odds ratio for nondiagnostic result: 0.94 per mm diameter decrease, 95% confidence interval 0.87-0.99, P= 0.028). On univariable analysis, diagnosis of central nervous system lymphoma appeared to increase the risk of a nondiagnostic biopsy (P= 0.025), but this association disappeared when controlling for lesion size and steroid administration before biopsy. Eight patients (4.0%) developed postoperative hemorrhagic complications, 3 of whom required reoperation, and another expired. This study demonstrates that diagnostic yield from contemporary FSB is high and depends predominantly on lesion size.