Letter to the editor Clinical diagnosis of multiple system atrophy: level of agreement between Quinn’s criteria and the consensus conference guidelines Multiple system atrophy (MSA) is a neurodegen- erative disease characterized clinically by a combi- nation of parkinsonian, autonomic, cerebellar and corticospinal signs (1), and neuropathologically by neuronal loss and presence of glial and neuronal cytoplasmic inclusions in the basal ganglia and ponto-cerebellar system (2). MSA may be recog- nized easily when the full-blown picture is encoun- tered, but some patients presenting with predominantly parkinsonian signs may be misdiag- nosed as having idiopathic Parkinson’s disease (PD) early in the disease phase (3). Quinn first proposed in 1989 a list of diagnostic criteria for MSA, which have since been accepted as the clinical ‘‘gold standard’’ for the identification of this disease (4). He divided cases of MSA into two major categories: those presenting with predominant parkinsonian signs and those with predominant cerebellar involvement. Within each of these groups, patients were labelled as clinically possible or probable according to the extent of multi-system involve- ment, on the basis above all of clinical criteria, though some laboratory criteria were also included, particularly in the version modified in 1994 (5). Cases were classified as definite MSA only when neuropathological confirmation was obtained. Quinn’s criteria were the first to be adopted in clinical practice, representing a tangible effort to understand the complex nosology in this field of neurodegenerative diseases. Nevertheless, in recent years several authors have argued that these criteria have some pitfalls, which are said to consist mainly in the incomplete definition of a few clinical features, in the partial overlapping of probable striatonigral and possible olivopontocerebellar cases, and in the emphasis placed on a neurophy- siological investigation, the urethral and anal sphincter electromyography (EMG), which has been found to have a specificity lower than initially thought (6). Accordingly, the members of the Consensus Conference (CC), who in 1998 convened under the co-sponsorship of the American Autonomic Society and the American Academy of Neurology have proposed new clinical guidelines for the diagnosis of MSA (7). The authors grouped MSA symptoms into four clinical domains – autonomic and urinary dysfunc- tion, parkinsonism, cerebellar dysfunction, and corticospinal tract dysfunction – and described the range of features consistent with MSA within each domain. Criteria were then offered to establish the involvement of each domain as part of the MSA diagnosis. They also defined two clinical diagnostic cate- gories – possible and probable MSA – according to the number of criteria and features present. Definite MSA was defined only histopathologically. The authors suggested replacing the term striatonigral degeneration with MSA-P (for predominance of parkinsonian symptoms) and olivopontocerebellar atrophy with MSA-C (for predominance of cere- bellar symptoms). Exclusion criteria were also offered, including onset <30 years of age, positive family history, hallucinations unrelated to medica- tion, dementia and vertical supranuclear gaze palsy. The main variation from Quinn’s criteria was that a patient has to meet the criterion of ‘‘autonomic dysfunction’’ plus at least another criterion in order to classify as probable MSA. The objective of the present study is to test the level of concordance between Quinn’s criteria and the CC guidelines in a series of clinical cases of MSA. The CC diagnostic guidelines were retrospectively applied to a series of clinical cases of MSA enrolled by the European Study Group on Atypical Parkinsonism (ESGAP) who had previously been diagnosed according to Quinn’s criteria (5). The ESGAP was established in 1994 with the purpose of studying the epidemiology, genetics, and clinical features of atypical parkinsonian syndromes, including MSA (8). Forty-five cases of MSA had been collected in the ESGAP database after their Acta Neurol Scand 2001: 103: 261–264 Printed in UK. All rights reserved Copyright # Munksgaard 2001 ACTA NEUROLOGICA SCANDINAVICA ISSN 0001-6314 261