A QUANTITATIVE TEST OF MUSCLE SPINDLE FUNCTION IN DIABETIC NEUROPATHY

Robert W. M. van Deursen, (1) Maria Matilde Sanchez, Jan S. Ulbrecht, and Peter R. Cavanagh
Center for Locomotion Studies, Department of Kinesiology and (1) Statistical Consulting Center
Penn State University, University Park, PA 16802

Presented at the 20th Annual Meeting of the American Society of Biomechanics
Atlanta, Georgia. October 17-19, 1996

INTRODUCTION

Diabetic neuropathy is a very common complication of diabetes mellitus and involves progressive loss of peripheral nerve fibers secondary to prolonged hyperglycemia. Loss of plantar cutaneous sensation has been related to impaired postural control that is observed in diabetic neuropathy (Simoneau et al., 1994). Additionally, the muscle receptors, in particular the muscle spindles, play an important role in postural control. For example, muscles around the ankle joint play a critical role in the maintenance of balance (Pyykkö et. al., 1989). However, there is presently no quantitative test of muscle spindle function available in the literature. In this study, such a test was developed using muscle vibration to provide potentially confusing signals to intact muscle spindles (Gilhodes et al., 1986). Muscle vibration at a frequency of 80 Hz and with a small (0.5 mm) amplitude is known to specifically stimulate the muscle spindle primary endings. The effect of vibration is that a bias is introduced into the muscle spindle output, provided subjects are unable to see the area that is vibrated. The vibrated muscle is perceived to be longer than it actually is, leading to a perception of motion in excess of the actual movement (Sittig et al., 1987). The effect of muscle vibration only occurs when the muscle spindles and their nerve supplies are intact. Therefore, the illusory effects of muscle vibration would be reduced in individuals suffering damage to muscle spindles secondary to diabetic neuropathy. We hypothesized that the adverse effects of vibration would be inversely related to the degree of nerve damage in diabetic neuropathy.

PROCEDURES

A total of 40 subjects, 10 young and healthy, 15 with diabetic neuropathy, and 15 age matched non-diabetic controls performed an ankle movement matching task with and without muscle vibration. The feet were placed in clamping devices which allowed actuator controlled movement of the right ankle (the "driven" side) and "free" movement of the left ankle (the "tracking" side) without stimulation of the plantar cutaneous mechanoreceptors. Ankle movement was measured by Penny & Giles goniometers attached to both ankles. To ensure muscle relaxation, electromyographic signals were measured in the lower leg on the driven side.

In each trial, subjects were instructed to continuously match the position of the tracking ankle to that of the driven ankle while it was moved through 3 cycles of dorsiflexion and plantar flexion at a velocity of 5°/s over a range of ± 25° (30 seconds total). The following conditions were presented in three (young, healthy) or four (matched groups) trials per condition:

RESULTS AND DISCUSSION

The one-factor ANOVA demonstrated that muscle vibration had a highly significant effect on the gain for the young, healthy subjects (p<0.001). In the two-factor ANOVA the interaction between group and vibration was highly significant (p<0.001), indicating that the control subjects responded to a greater degree in the presence of vibration than the subjects with diabetic neuropathy. The reliability of the test measured by the Intraclass Correlation was approximately 0.9 both within and between days.

Figure 1: Mean gain of the tracking movement with and without muscle vibration for the neuropathic and control groups and the young, healthy group. The black arrows indicate the increase in gain (D gain) as a result of muscle vibration for each group.

The mean gains for the three groups with and without muscle vibration are shown in Figure 1. The young subjects (average age: 30.4) tracked the movement very well in the absence of vibration (average gain = 0.93). Muscle vibration caused a dramatic and significant increase in their movement amplitude (D gain: 0.42). The older control subjects (average age: 62.3) had some difficulty tracking the movement even without vibration (average gain = 0.71). However, muscle vibration caused a substantial increase in their movement amplitude as it had done in the young subjects (D gain: 0.32). The neuropathic subjects (average age: 62.1) had the most difficulty tracking the ankle movement in the absence of vibration (average gain = 0.62). More importantly, muscle vibration had much less effect on their performance (D gain: 0.16). The reduced effect of vibration indicates that diabetic neuropathy results in damage to the muscle spindles. We conclude that: a) the change in tracking performance when vibration is added during an ankle movement matching task provides a test of muscle spindle function, and b) Diabetic neuropathy degrades muscle sensory function, presumably through the loss of either afferent or gamma efferent nerves to muscle spindles in the lower leg. This may contribute to the impaired balance and unsteadiness of gait that has been observed in diabetic neuropathy.

REFERENCES

Gilhodes, J. C. et al., Exp. Brain Research, 61: 395-402, 1986.

Pyykkö, I. et al., Acta Otolaryngologia, 468: 175-180, 1989.

Simoneau, G. G. et al. Diabetes Care, 17: 1411-1421, 1994.

Sittig, A. C. et al. Exp. Brain Research, 67: 35-40, 1987.

ACKNOWLEDGMENTS

This study was supported by NIH Grant 1R01 AG09345. The authors wish to thank Mary B. Becker, David R. Lemmon, and Doug J. Tubbs for their contributions.