Dependence of the linear model for the nerve compound action potential on the single fibre action potential waveform

Abstract

The linear model of the nerve compound action potential (CAP) depends on the assumed waveform for the single fibre action potential (SFAP). A general method has been developed to investigate the influence of the unknown features of the SFAP on the estimation of nerve fibre conduction velocity (CV) distribution. A SFAP waveform is considered consistent with the model and the experimental data if recorded and reconstructed CAPS fit and the distribution is physically meaningful. Experimental CAPS were monopolarly recorded using surface electrodes over the median nerve at the wrist. To fit the model, SFAP waveforms must satisfy some internal relationship. The most important feature is that the ratio between positive and negative areas of the SFAP is almost one and does not vary in different subjects and recording sites. Many SFAP waveforms fit the model, and the relative conduction velocity distributions may be very different. These must be regarded as conventional distributions. As for inter-subject comparison, the dependence of the method on the recording site has been reduced by choosing the place where stimulus intensity and relative motor response amplitude have given values. In this recording environment CV distributions of normal subjects can be properly compared using the same SFAP and deviations from normality evidenced.