Diabetes mellitus is a major chronic disease and is epidemic in many regions of the world. It is associated with long-term complications that affect almost every major part of the body. Neuropathy is one of the main complications of diabetes resulting in loss of peripheral sensory, motor and autonomic nerve function. Often, the sensory neuropathy in the lower legs is often the most apparent. This peripheral nerve damage can lead to a variety of problems, one of which is loss of balance and coordination. Subjects with diabetic neuropathy have been found to be less stable and this has. in the past, been associated exclusively with loss of plantar cutaneous sensation. The postural control system, however, relies on multiple sources of sensory feedback. This raises the question whether loss of plantar cutaneous sensation can sufficiently explain the instability observed in patients with diabetic neuropathy or whether more components of postural feedback are affected when balance becomes a problem.

The aim of this study was two-fold. The first aim was to demonstrate that diabetic neuropathy not only seriously affects plantar cutaneous sensation but that muscle spindle function of the lower legs is also affected.

The second aim was to demonstrate that adaptations in postural strategies occur when loss of peripheral sensation is severe enough. The combined loss of plantar cutaneous sensation and muscle spindle function in the lower leg was predicted to limit the flow of useful kinesthetic information from the lower leg to the CNS. The importance of kinesthetic feedback from the upper leg and hip was therefore expected to be inversely related to the severity of neuropathy in the lower leg. This was predicted to result in a change of control of the lower limb from an ankle strategy to a hip strategy.

An experimental design was chosen to control for important confounding factors while accommodating for the fact that the availability of appropriate subjects with diabetic neuropathy imposed limitations on the number that could be recruited.

Four groups of subjects were selected:​ a non-diabetic control group, a diabetic, non-neuropathic control group, a diabetic, mildly neuropathic group and a diabetic, severely neuropathic group. Before subjects were entered into the study, they had to meet a large number of exclusion criteria. The four groups were successfully matched on marginal distributions for gender, age, height and weight. Other possible confounding factors were measured but not used to match the groups. Fortunately, there were no significant differences between the groups on any of these factors. The results of this study can, therefore, be attributed mainly to the differences between the groups in terms of the presence or absence of diabetic neuropathy and the severity of peripheral neuropathy.

Three tests were used to demonstrate a loss of muscle spindle function in diabetic neuropathy. In all tests, muscle vibration was used to stimulate muscle spindles of the lower leg muscles to bias feedback from those receptors. The assumption was that the main effect of muscle vibration could only occur when the muscle spindles and their nerve supplies were intact. If muscle spindle function was degraded in diabetic neuropathy, the effect of muscle vibration was expected to be reduced. All three tests appear to confirm that there is a loss of muscle spindle function due to diabetic neuropathy. Furthermore, the changes in muscle spindle function described for neuropathic subjects were most pronounced in the group of subjects with severe diabetic neuropathy.

Subjects with significant loss of sensation in the lower legs were expected to adapt their strategies of postural control by controlling lower limb movement at the hip joint (hip strategy). In this situation, the upper leg muscles were predicted to be a more dominant source of afferent feedback resulting in an increased response to upper leg muscle vibration in neuropathic subjects. Also, the relative contribution of ankle joint movement was predicted to be reduced in favor of the relative contribution of hip joint movement. Both of these consequences of a hip strategy were tested in the quiet standing and the voluntary sway tests.

Although there were no significant changes in postural control strategies for the neuropathic subjects compared to the control subjects, there were significant differences for the group of subjects with severe diabetic neuropathy. The severe neuropathic subjects tended to move less at the ankle joint during the voluntary sway test. More importantly, there was a significant difference with respect to the relative contribution of hip movement in the voluntary sway test for the severe neuropathic group. Severe loss of muscle spindle function in combination with loss of cutaneous sensation resulted in adaptations in the postural control strategy in the group of subjects with severe diabetic neuropathy. Therefore, the reduced stability in diabetic neuropathy appears to be the result of a more general loss of peripheral sensory receptor function in the lower legs then has previously been realized.