Impact of Postoperative Diffuse-Weighted Imaging Rim Restriction on Neurological Outcomes After Resection of Newly Diagnosed Isocitrate Dehydrogenase-Wildtype Glioblastoma

BACKGROUND/OBJECTIVES: Postoperative diffusion-weighted imaging (DWI) abnormalities, appearing as a rim of restricted diffusion around the surgical cavity, are frequently observed on early postoperative MRI in patients with glioma. This study aims to explore the evolution and impact of these DWI abnormalities on neurological outcomes. METHODS: A retrospective cohort analysis was conducted on 303 patients who underwent initial resection for newly diagnosed isocitrate dehydrogenase-wildtype glioblastoma at the University of California, San Francisco, between 2017 and 2021. DWI abnormalities were classified according to severity: 0 (no rim restriction), 1 (thin/minimal rim restriction), 2 (moderate rim restriction), 3 (minimal region/sector [<1 cm3] of restriction), or 4 (moderate/large region or sector [>1 cm3] of restriction). DWI volumes were segmented and used to calculate median lesional apparent diffusion coefficient. Maximum perpendicular diameters of the restriction abnormality were measured manually. Multivariable regression models assessed associations with neurological outcomes. RESULTS: Of the 303 patients analyzed, 11 (3.6%) patients exhibited no rim restriction (Grade 0), 247 (81.5%) patients had some contiguous, nonbulky DWI rim restriction (Grades 1-2), and 45 patients (14.9%) had thick contiguous, bulky (Grades 3-4) DWI restriction. Overall, all postoperative DWI abnormalities resolved within 6 months. Among the 17 patients (5.6%) who developed new or worsening deficits immediately postoperation or at discharge, only 3 (17.6%) had contiguous, bulky DWI restriction and 9 (52.9%) fully recovered at 6 months. Among the patients who did not recover (n = 8), 6 (75%) had DWI Grades 1 to 2 and 2 (25%) had DWI Grades 3 to 4 (P = .29). DWI volume and maximum DWI diameter were significantly associated with new/worsening deficits at all time points; however, neither DWI grade nor apparent diffusion coefficient were significantly associated with new/worsening deficits at any time point. CONCLUSION: DWI rim restriction is common after glioblastoma resection, and these changes uniformly resolve within 6 months of surgery. The severity of DWI abnormalities does not reliably predict postoperative deficit recovery.