A study of normal nerve action potentials using an averaging technique (barrier grid storage tube)

Abstract

Nerve action potentials from the median and ulnar nerves in healthy subjects were studied by means of a response averager incorporating a barrier grid storage tube. Instead of photographically superimposing the traces, 100 sweeps were stored and the average of these, photographed on the cathode ray tube. In the re-investigation of the effect of positioning and separating electrodes it was shown that as the distance between electrodes was reduced, the peaks of the main upward deflection and of the 2nd downward deflection fell progressively earlier and the amplitude of the initial downward deflection and of the main upward deflection decreased. The waveform obtained from a pair of recording electrodes over a nerve is actually the algebraic sum of the absolute values of potential changes occurring at each electrode. Latency measurements to the onset of the negative deflection were seldom found to vary with the exact arrangement of the electrode. In measurements of conduction velocity, results indicated that nerve impulses were initiated close to the cathode and not at some point between cathode and anode. When both anode and cathode are close to the nerve trunk, excitation will occur at both electrodes. A study of the effect of stimulus intensity on action potential latency showed that the latency was shorter with a supra-maximal than with a near-threshold stimIllus. The difference in latency was 0.05 to 0.15 msec, with a mean of 0.11 msec suggesting some current spread. Recording accurate latency measurements in spite of small random fluctuations with each trace is one advantage of superimposing several averaged traces. From a comparative study of wrist and finger stimulation of the ulnar nerve it was concluded that if there are fibers in the muscle branches of the ulnar nerve with a markedly higher conduction velocity than digital fibers they are not of sufficient number to affect the action potentials that were recorded. Vhen recording from peripheral nerves the averaging process is advantageous in that it eliminates noise and randm activity without obscuring the details of individual action potentials and permits latency measurements with a degree of accuracy not possible to obtain from photographic records of the primary trace.