Erschienen in:
23.03.2019 | Original Communication
Relapse rates and long-term outcome in primary angiitis of the central nervous system
verfasst von:
Simon Schuster, Ann-Kathrin Ozga, Jan-Patrick Stellmann, Milani Deb-Chatterji, Vivien Häußler, Jakob Matschke, Christian Gerloff, Götz Thomalla, Tim Magnus
Erschienen in:
Journal of Neurology
|
Ausgabe 6/2019
Einloggen, um Zugang zu erhalten
Abstract
Objective
To analyze the treatment response in patients with primary angiitis of the central nervous system (PACNS).
Methods
In a single-center retrospective observational study, we assessed relapses, remission, and long-term outcome by use of the modified Rankin Scale (mRS) under different immunotherapies. Eligible patients had CNS biopsy in favor of PACNS or neuroimaging compatible with PACNS after exclusion of an alternative diagnosis. Regression models, recurrent event, and linear mixed-effects models were used to estimate the annual relapse rate, relapse and outcome predictors. Favorable outcome was defined as mRS < 3.
Results
Of 44 patients, 26 (59%) were female, median age at diagnosis was 43.5 (range 14–83) years, and 25 (57%) had biopsy-proven diagnosis. Median follow-up was 5.1 years. Glucocorticoids were administered in 30 patients at diagnosis (68%), 33 patients (75%) received cyclophosphamide, and 86% of patients had maintenance therapy > 24 months. Overall, 201 treatment episodes with 104 relapses and 4 (9%) deaths occurred. 26 patients had relapses (59.1%). The annual relapse rate was 1.4 (CI 1.1–1.8). Male sex was the only significant predictor of relapse (HR = 3.27, 95% CI 1.57–6.82). Remission occurred in 30 patients (68%). Favorable outcome was evident in 80% of patients after 2 years and 66% of patients at last follow-up.
Conclusions
PACNS is a relapsing-remitting disease with a heterogeneous disease course and mostly favorable outcome under immunotherapy. Male patients have a higher relapse risk; no other relapse or outcome predictor could be identified. PACNS subtype stratification is needed to further evaluate predictors of response.