Natural history in proximal spinal muscular atrophy. Clinical analysis of 445 patients and suggestions for a modification of existing classifications

Abstract

Objectives: To describe the natural history in all types of proximal spinal muscular atrophy (SMA) and to propose a modified classification scheme that takes the long-term course of SMA into account.

Design: Patients with proximal SMA were studied prospectively and retrospectively in a genetic study that was based on clinical and family data.

Patients: Four hundred forty-five patients with SMA were ascertained since 1985 through various departments of neurology and neuropediatrics, institutes of human genetics, and the German muscular dystrophy association (Deutsche Gesellschaft für Muskelkranke, Freiburg, Germany).

Results: The study group was subdivided into patients with four types of SMA (ie, SMA types I, II, III, and IV) on the basis of achieved motor development and age at onset. Survival probabilities at 2, 4, 10, and 20 years of age were 32%, 18%, 8%, and 0%, respectively, in patients with SMA type I (those who were never able to sit) and 100%, 100%, 98%, and 77%, respectively, in patients with SMA type II (those who were able to sit but were unable to walk). Nineteen of 104 patients with SMA type II lost the ability to sit; this inability to sit was not of prognostic relevance. Patients with SMA type III (those who were able to walk [age at onset, younger than 30 years]) were subdivided into those with an age at onset before (SMA type IIIa) and after (SMA type IIIb) 3 years. The probabilities of being ambulatory at 10, 20, and 40 years after onset were 73%, 44%, and 34%, respectively, in patients with SMA type IIIa, and they were 97%, 89%, and 67%, respectively, in patients with SMA type IIIb.

Conclusions: The definition of long-term characteristics of SMA is helpful in providing medical care to families with members who have SMA and also in providing important information for future genotype-phenotype studies and therapeutic trials of patients with SMA. Our data indicate that the widely used classification schemes did not consider the broad spectrum of SMA so a practical modification was suggested.