PURPOSE: The aim of this study was to compare the clinical and electrodiagnostic efficacy of ultrasound (US)-guided versus landmark (LM)-guided steroid injections in patients with carpal tunnel syndrome (CTS). METHODS: This randomized clinical trial included 47 patients (60 affected hands) with moderate CTS (30 hands in each group). All clinically suspected patients with CTS who met the inclusion criteria confirmed by electrodiagnostic tests were enrolled. The subjects received 40 mg of methylprednisolone either with the US-guide or the LM-guided injection technique. They were evaluated using the Boston Carpal Tunnel Questionnaire (symptom/function/total) and 6 electrodiagnostic findings at the baseline and 12 weeks after injection. RESULTS: Symptom severity scores and functional status scores and electrodiagnostic parameters were significantly improved within each group at week 12 after treatment (P < 0.05), except for compound muscle action potential amplitude and motor nerve conduction velocity (P > 0.05). The improvement in symptom severity scores and functional status scores in the US-guided group was more than in the LM-guided group after 12 weeks, but there was no statistically significant difference (P = 0.79 and 0.64). The mean changes in electrodiagnostic parameters were also not different between groups except for the higher sensory nerve action potential amplitude in LM-guided group (P = 0.003). CONCLUSIONS: Both US-guided and LM-guided steroid injections were effective in reducing the symptoms, improving the function and electrodiagnostic findings of CTS. Although there was better symptomatic improvement with US-guided injections and better increase in sensory nerve action potential amplitude with LM-guided injection, a significant difference was not generally observed between US-guided and LM-guided CTS injections.