Validity of electromyograms and tension as a means of motor unit number estimation

The purpose of this study was to validate three different techniques for obtaining motor unit number estimates of the rat medial gastrocnemius muscle. These consisted of two electromyographic techniques using unprocessed and digitally averaged unitary muscle action potentials, and one mechanical technique. We also injected subunit B of cholera toxin into this muscle and counted the number of spinal motor neurons labeled by the toxin. Our results revealed that a motor unit number estimate obtained by using the unprocessed unitary muscle action potential was statistically different from the actual number of motor neurons. The other two motor unit number estimates, however, were not statistically different from the actual motor neuron number. These two methods thus seem more appropriate than the first electromyographic method for obtaining an accurate motor unit number estimate.