New forms of therapy for diabetic and other neuropathies may prevent, stabilize, or ameliorate loss of nerve fibers. Clinically meaningful changes in mean Neurological Disability Score (NDS), and the associated mean change of electrophysiologic attributes have been described in diabetic polyneuropathy. It is unknown what magnitude of myelinated fiber (MF) density change is associated with these meaningful changes of clinical and electrophysiologic alterations. In 18 diabetics and 5 normal controls associations between the mean NDS, summated (ulnar, peroneal and tibial) compound muscle action potential (sigma CMAP), summated (ulnar and sural) sensory nerve action potential (sigma SNAP), sural SNAP, and MF density in the sural nerve, were assessed using linear regression analyses. Values were corrected for age and sex. For a decrease of: 2 points in the mean NDS (minimum clinically detectable change), MF density decreased by approximately 200 fibers/mm2 (p < 0.001) 1 mV in the mean sigma CMAP (sum of the ulnar, peroneal and tibial CMAP amplitudes), MF density decreased by 160 fibers/mm2 (p < 0.01) 1 microV in the mean sigma SNAP (sum of ulnar and sural SNAP amplitudes), MF density decreased by approximately 70 fibers/mm2 (p < 0.001) 1 microV in the mean sural SNAP, MF density decreased by approximately 150 fibers/mm2 (p < 0.01). Changes in sensory detection thresholds were also associated with a measurable change in the MF density. A quantifiable association exists between the magnitude of change in density of MF, and a meaningful alteration in mean NDS and various electrophysiologic parameters. Knowledge of this is needed to assess the statistical power of a clinical trial in which density of myelinated fibers is an outcome measurement.