The different techniques to measure and analyze surface EMG are summarized
with an emphasis on the clinician's point of view. The application of surfa
ce EMG in neurological disease is hampered by many inherent problems, espec
ially the difficulties in extracting features of single motor units. Howeve
r, the evolution of surface EMG from single bipolar recordings via a linear
array of multiple electrodes to densely packed, multi-channel electrode ar
rays could in principle solve this problem. The added value of using multip
le channels (up to 128) with an interelectrode distance of a few millimetre
s to obtain more spatial information is emphasized. At least for some muscl
es it is now possible to extract information from the surface EMG, conventi
onally thought to belong to the domain of needle EMG (for example the "elec
trical size" of motor units). The use of analysis techniques such as the es
timation of muscle fiber conduction velocity has already proven to be of di
agnostic value in several myopathies characterized by a disturbed membrane
function and in metabolic myopathies with abnormal fatigue profiles. Future
research should be directed at the development of analysis techniques enab
ling the extraction of more relevant motor unit variables from surface EMG
signals. (C) 2000 Elsevier Science Ltd. All rights reserved.