Objectives: To investigate peripheral and central somatosensory conduction
in patients with diabetes. Methods: The authors recorded sensory nerve acti
on potentials and 5-channel somatosensory evoked potentials (SEPs) with non
cephalic reference after median nerve stimulation in 55 patients with diabe
tes and 41 age- and height-matched normal subjects. The authors determined
onset or peak latencies of the Erb's potential (N9) and the spinal N13-P13
and the cortical N20-P20 components, and obtained the central conduction ti
me (CCT) by onset-to-onset and peak-to-peak measurements. Results: Both ons
et and peak latencies of all SEP components were prolonged in patients with
diabetes. The mean onset CCT in the diabetic group was 6.3 +/- 0.5 msec (m
ean +/- SD)-significantly longer than that in the control group (6.1 +/- 0.
2 msec)-whereas no significant difference was found in the peak CCT. The am
plitudes of N9 and N13-P13 components (but not N20-P20) were significantly
smaller in the diabetic group. The peripheral sensory conduction velocity w
as also decreased in the diabetic group, but there was no significant corre
lation between peripheral conduction slowing and the onset of CCT prolongat
ion. Conclusions: Diabetes affects conductive function in the central as we
ll as peripheral somatosensory pathways. The CCT abnormality does not coinc
ide with lowering of the peripheral sensory conduction. The current results
do not favor a hypothesis that a central-peripheral distal axonopathy play
s an important role in development of diabetic polyneuropathy.