The Somatosensory Evoked Potential

Three decades have elapsed since Dawson (1947) recorded the first somatosensory evoked potential (SEP). Simple superimposition of individual responses was possible because the patient had progressive myoclonic epilepsy. In this disease the SEP amplitude is much enhanced (Shibasaki et al, 1978; Kelly et al, 1981). Subsequently Dawson (1951, 1954) presented his averager to the Physiological Society, thereby initiating the present-day explosive growth of evoked potentials. SEPs are made up of components with varying latencies. The components are best identified by latency and polarity as recorded at the scalp (P = positive and N = negative). Nevertheless, the nomenclature of somatosensory evoked potentials can be extremely confusing, mainly because the same component can have a different polarity depending on the electrode montage used. Generally speaking (but this is not a firm rule), far-field (subcortical) potentials are positive in polarity when a non-cephalic reference is used, whereas these same components have a negative polarity when the reference is on the scalp. It is therefore useful to always indicate the recording montage being employed. In addition, use of absolute latencies in the terminology can cause confusion because they are dependent upon length and body height. For example, the brachial plexus component usually occurs at about 9 msec, but may extend to as long as 11 or more msec in a very tall individual. Subsequent components then become difficult to identify in relation to normal means.