BACKGROUND: For the early detection and prevention of diabetic neuropathy, it is important to identify subclinical diabetic neuropathies. A routine nerve conduction study often fails to detect the early stages of neuropathy. The purpose of this study is to evaluate the clinical usefulness of electrophysiological indexes including the residual latency(RL), terminal latency index (TLI) and modified F ratio (MFR) in detecting early diabetic neuropathy with no objective clinical or electrophysiological abnormalities. METHODS: A nerve conduction study of the upper/lower limbs was investigated in 38 subclinical diabetic neuropathy patients with normal nerve conduction studies (group I), 35 clinical diabetic neuropathy patients with normal nerve conduction studies (group II) and 31 normal controls. RL, TLI and MFR were calculated and compared among the groups. RESULTS: Compared with the control group, the MFR of the lower limbs and TLI of both the upper/lower limbs were significantly decreased in both group I and II (p<0.05). RL was increased in both groups, but the difference was not statistically significant. Comparing the indexes between group I and II, there was no significant difference. CONCLUSIONS: RL, TLI and MFR are useful indexes for reflecting distal conduction slowing especially in slowly progressing polyneuropathies such as diabetic neuropathy. The results also suggest that electrophysiological changes veiled in a routine nerve conduction study were present before the clinical manifestations.