We performed a prospective study to evaluate the prevalence of bifid median nerve (BMN) and if it has a higher prevalence in carpal tunnel syndrome (CTS) patients than symptom-free subjects. We evaluated by ultrasound (US) 162 median nerves at wrist of 89 consecutive patients with symptoms suggestive of CTS and 104 of 52 consecutive symptom-free subjects. Neurophysiological evaluation confirmed CTS diagnosis. All patients underwent ultrasonographic evaluation of median nerve in the elbow-palm segment through high-frequency probe; BMN was arbitrarily classified as "complete" and "incomplete" on the basis of US results. BMN was observed in 16/104 wrists (15.4%) in control group and in 30/162 (18.5%) wrists in CTS group. Our study showed that BMN is frequent, not always bilateral, more often is "incomplete" and it has similar prevalence in CTS and control group. These findings suggest that BMN is not a risk factor of developing CTS.