Évaluation initiale des méningites tuberculeuses : l’IRM encéphalique peut-elle identifier les patients à risque d’aggravation neurologique

Objective: describe early brain MRI alterations in patients with tuberculous meningitis (TBM) and investigate their potential association with neurologic outcome. Material and methods: a restrospective review of all TBM cases was conducted from 2004 to 2014 in a tertiary care hospital of Paris, France. We included patients for whom a brain MRI was performed within the first 90 days after admission. Neurologic outcome was evaluated using the modified Rankin scale (mRS) at 90 days after admission and at the end of follow-up. Results: forty-seven patients were included, among which 29 (62%) had a good outcome (mRS=1) and 18 (38%) a poor outcome at 90 days (mRS>1), and 32 (68%) had a good outome and 15 (32%) a bad outcome at the end of follow-up. Median delay between admission and initial brain MRI was 11 (4-26) days. Tuberculomas (n=29, 62%), focal meningitis (n=16, 34%) and diffuse meningitis (n=11, 23%) had no impact on outcome. Only ischemic strokes (n=11, 23%,), hydrocephalus (n=7, 15%), 44% vs 10%, p=0.01 and 33% vs 3%, p1), et respectivement 32 (68%) et 15 (32%) à la fin du suivi. Le délai médian entre l’admission hospitalière et l’IRM initiale était de 11 (4-26) jours. Les tuberculomes (n=29, 62%), et les méningites focales (n=16, 34%) et diffuses (n=11, 23%) n’étaient pas associées au pronostic. Les AVC ischémiques (n=11, 23%) cas, les hydrocéphalies (n=7, 15%), respectivement 44% vs 10%, p=0.01 et 33% vs 3%, p<0.01, et un délai admission-IRM plus long, 22 (9-31) vs 6 (4-21) jours, p=0.03 étaient associés à un mauvais pronostic à 90 jours. À la fin du suivi, seuls l’hydrocéphalie (33% vs 6%, p=0.03) et un délai admission-IRM plus long [22(14-36) vs 6(4-19) jours, p