Later‐Onset Multiple System Atrophy: A Multicenter Asian Study

Consensus guidelines for diagnosis of multiple system atrophy (MSA) stated that onset age can be as early as 31 years and that onset after 75 years is indicated as a feature not supporting diagnosis of MSA.1 Therefore, clinically probable MSA patients whose onset is above the age of 75 years have been diagnostically challenging.2 In this multicenter study, we collected patients who met the MSA diagnostic criteria and whose onset ages were above the age of 75 years to identify clinical characteristics of later-onset MSA group in Asia. Of 1425 MSA patients, data on 39 with clinically probable later-onset MSA (median onset age, 76.8 years) were retrospectively collected from 5 tertiary referral hospitals in Korea and 128 clinically probable MSA patients with mean age of onset (median onset age, 58.8 years) from Severance Hospital in Korea (Supporting Information Table 1, Supporting Information Methods). There were no significant differences in demographic characteristics between the later-onset MSA and MSA with mean-age-at-onset groups, except motor subtypes; the MSA-parkinsonian (MSA-P) subtype (64.1%) was more common than the MSA-cerebellar (MSA-C) subtype (35.9%) in the later-onset MSA group, whereas MSA-C (59.4%) was more common than MSA-P (40.6%) in MSA with the mean-age-at-onset group (P = 0.010). We also collected 10 of 165 pathologically proven MSA patients from Aichi Medical University in Japan who had a disease onset after 75 years (Table 1, Supporting Information Methods). Of those, 8 cases were classified as striatonigral degeneration (SND) dominant type, whereas the other 2 cases exhibited olivopontocerebellar atrophy dominant type and mixed neuronal loss. Compared with pathologically proven MSA with mean-age-at-onset group, SND dominant type was more common in the later-onset MSA group (P < 0.001). In addition, patients with later-onset MSA had shorter survival after disease onset compared with the MSA with mean-age-at-onset group (P < 0.001). Compared with clinical features between 25 clinically probable later-onset MSA-P subtypes from 5 tertiary referral hospitals in Korea and 62 age-matched Parkinson's disease (PD) patients from Severance Hospital in Korea, dysautonomic symptoms, falling, limb ataxia, and speech disturbance were more common in the later-onset MSA group, whereas resting and postural tremor were more common in the PD group (Supporting Information Table 2). The present study identified later-onset MSA patients whose onset ages were above 75 years. Our clinical and pathological data showed a higher proportion of MSA-P or SND type in the later-onset MSA group than the MSA with mean-age-at-onset group. In addition, pathologically proven later-onset MSA tended to have shorter survival after disease onset compared with MSA with mean age at onset. Thus, our data suggest that later-onset MSA may exhibit distinct clinical features and prognosis. It remains controversial whether onset age and MSA phenotypes differentially affect disease progression3-6; therefore, further studies are needed to determine associations among onset age, MSA phenotype, and disease progression in MSA patients. However, our results should be cautiously interpreted given the pathological data only from a single center, the lack of detailed clinical information of pathological data, and the retrospective nature and lack of detailed comorbidities of clinical cohorts. In summary, our study suggests that the second consensus criteria for diagnosis of MSA needs to be revised with respect to the range of onset age of MSA. Yang Hyun Lee: 1C, 2A, 2B, 3A. Takashi Ando: 1C, 2B, 3A. Jae Jung Lee: 1C, 3B. Min Seok Baek: 1C, 3B. Chul Hyoung Lyoo: 1C, 3B. Sang Jin Kim: 1C, 3B. Minkyeong Kim: 1C, 3B. Jin Whan Cho: 1C, 3B. Young H. Sohn: 1B, 3B. Masahisa Katsuno: 1C, 2B, 3A. Hirohisa Watanabe: 1C, 2B, 3A. Phil Hyu Lee: 1A, 1B, 2A, 2C, 3B. None. Ethics approval: Institutional Review Board of the Yonsei University Severance Hospital. Provenance and peer review: Not commissioned; externally peer reviewed. Appendix S1: Supporting Information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.