INTRODUCTION/AIMS: In traumatic nerve lesions (TNLs), motor unit potentials (MUPs) may be difficult to detect in early injury. Ultrasound-guided electromyography (US-EMG) can aid in identifying areas of muscle activation, but its sensitivity can be improved. In this study we compare the sensitivity of US-EMG alone with US-EMG after peripheral nerve stimulation (NC-US-EMG) to better identify active muscle regions. METHODS: In this prospective study, 32 patients with severe TNLs were evaluated with standard EMG (ST-EMG), US-EMG, and NC-US-EMG at baseline (T0), after 2 to 3 months (T1), and after 5 to 6 months (T2). RESULTS: NC-US-EMG was more sensitive in detecting MUPs compared with US-EMG and ST-EMG at T0 (19 patients vs 14 and 5 patients, respectively). In addition, both US-guided techniques were more sensitive than ST-EMG in detecting MUPs (ST-EMG vs US-EMG: P = .014; ST-EMG vs NC-US-EMG: P = .003). At T1, ST-EMG remained less sensitive NC-US-EMG (P = .019). No significant differences were observed among the three techniques at T2. DISCUSSION: In the evaluation of severe TNLs, the combination of peripheral nerve stimulation and US increases the sensitivity of EMG for MUP detection at baseline and 2 to 3 months postinjury.