Association of Traumatic Meningeal Enhancement on MRI With Clinical Recovery in Patients With Traumatic Brain Injury.

Background And Objectives: Traumatic meningeal enhancement (TME) can be observed on MRI of patients with acute traumatic brain injury (TBI) and reflects abnormal contrast extravasation into the meninges. Resolution of TME occurs over time, but TME can persist for weeks, suggesting incomplete meningeal repair. This study's objectives were to describe TME's prevalence, severity, and evolution over time and to investigate TME's association with other imaging findings, blood-based biomarkers commonly associated with TBI, and incomplete recovery.

Methods: Patients with suspected TBI presenting within 48 hours of injury to Suburban Hospital or Washington Hospital Center were prospectively enrolled between 2010 and 2019, received MRI, and underwent optional blood collection at baseline and follow-up visits at 1 week, 30 days, 90 days, and 1 year. Independent trained raters blinded to patient details scored for the presence and severity of TME on postcontrast MRI. Neuroimaging findings on CT and MRI, other than TME, were extracted from neuroradiology reports. Plasma biomarker levels (total tau [t-tau]; neurofilament light chain [NfL]; glial fibrillary acidic protein [GFAP]; ubiquitin C-terminal hydrolase-1 [UCH-L1]) were assessed with single-molecule array kits. Incomplete recovery was defined as a Glasgow Outcome Scale-Extended (GOSE) score <7 at 30-90-day follow-up. Factors associated with recovery were assessed through multivariable logistic regression analysis controlled for confounding variables.

Results: Of 675 patients (male/female/neither 68%/31%/1%; median [interquartile range] age: 45 [28-58] years; Glasgow Coma Scale score 15 [15-15]), 359 (53%) were positive for TME at baseline (16 [6-25] hours after injury). At 30-90-day follow-up, TME remained absent in 117 (37%), resolved in 139 (45%), and persisted in 56 (18%). Acute TME had a high positive predictive value (PPV) for acute TBI-related findings on CT (87.7%) and MRI (86.1%). One-way analysis of covariance demonstrated significant associations between baseline TME and CT for plasma biomarker levels ((): t-tau = 19.328 (2); NfL = 20.458 (2); GFAP = 78.662 (2); UCH-LI = 46.680 (2)). Patients with persistent TME were more likely (odds ratio 3.809; 95% CI 1.703-8.519; = 0.001) to have GOSE score <7.

Discussion: TME was prevalent at baseline, with high PPV for other neuroimaging findings, but was not associated with recovery. TME's persistence was independently associated with incomplete recovery, suggesting that the meninges' failure to repair may be a mediator of recovery after TBI.

Trial Registration Information: ClinicalTrials.gov NCT01132937.