Role of Steroids in the Diagnosis of Primary Central Nervous System Lymphomas: A Single-Center Experience

The effect of preoperative steroids on primary central nervous system lymphoma (PCNSL) diagnosis is debated, because of concerns about lesion regression and false-negative biopsies. Steroids are commonly administered to manage the mass effect and neurological symptoms, but may obscure differentiation from steroid-responsive inflammatory syndromes like chronic lymphocytic inflammation with pontine perivascular enhancement responsive to steroids and supratentorial lymphocytic inflammation with parenchymal perivascular enhancement responsive to steroids. The objective is to evaluate diagnostic yield in patients with PCNSL receiving preoperative steroids and assess whether steroids compromise histopathological confirmation in the context of inflammatory mimics.

We retrospectively reviewed 50 patients with histologically confirmed PCNSL between 2014 and 2024 and stratified them by preoperative steroid exposure. Demographics, steroid dosing, surgical approach, imaging, and need for repeat biopsy were compared.

Of 50 patients, 34 (68%) received steroids before biopsy. The mean age was 63.96 years. All cases were diffuse large B-cell lymphoma except 1 T-cell lymphoma. In the steroid group, 17 underwent stereotactic biopsy, 1 underwent open biopsy, and 16 had craniotomy; 1 required repeat biopsy. In the nonsteroid group, 11 underwent stereotactic biopsy, 3 open biopsy, and 2 craniotomy; none required repeat biopsies. Cumulative steroid doses varied widely. One patient with prolonged steroid exposure had a nondiagnostic biopsy and required a rebiopsy. MRI typically showed diffuse enhancement (42/50), with lesions mainly in the frontal (n = 31), temporal (n = 17), and parietal (n = 10) lobes. The average tumor size was 34.0 mm in the steroid group and 29.5 mm in the nonsteroid group. There were no statistical differences between groups.

Short-course preoperative steroids did not substantially compromise the diagnostic yield in this PCNSL cohort. Prolonged exposure may obscure the diagnosis, particularly in cases mimicking chronic lymphocytic inflammation with pontine perivascular enhancement responsive to steroids/supratentorial lymphocytic inflammation with parenchymal perivascular enhancement responsive to steroids. Timely biopsy with adequate sampling remains effective. These findings support a pragmatic approach to steroid use, emphasizing diagnostic vigilance and individualized management.