BACKGROUND Diabetes mellitus (DM) is a common cause of polyneuropathy. The aim of the present study was to evaluate two-point discrimination (TPD) compared with nerve conduction studies in the early stages of DM. METHODS Forty-eight patients with early diagnosed (<5 years) type 2 DM and 17 healthy controls were evaluated. Of the patients with DM, 26 had neuropathic pain and 22 were asymptomatic. TPD and electrophysiological evaluations was obtained for all subjects. RESULTS Nerve conduction studies in patients showed findings related to both demyelination and axonal damage. Patients with neuropathic pain had higher TPD values on the plantar surface of the foot and both groups of DM patients had higher TPD values on the outer lateral malleolus compared with the control group (P < 0.05). There was a correlation between TPD and axonal damage in patients with neuropathic pain (P < 0.05). In patients without neuropathic pain, there was a correlation between TPD values and distal latencies of motor or sensory nerves (P < 0.05). In the control group, only third digit TPD values were related to the distal motor latency of the median nerve (P < 0.05). CONCLUSIONS In conclusion, the TPD method is a less painful, practical, cost-effective, and more easily applicable method that was completed in less time than nerve conduction studies. Higher TPD values in the lower extremities indicate nerve damage in patients. These findings suggest that increased TPD values can easily determine neuropathy starting in the early stages of diabetes in patients with DM.