Using phonomyography, a new monitoring technique of neuromuscular blockade (NMB), we compared NMB after mivacurium 0.1 mg/kg at the lateral cricoarytenoid muscle (LCA) with a possible external monitoring site of the larynx. In 12 patients, data were obtained at both sites using phonomyography. Anesthesia was induced with remifentanil 0.25-0.5 microg . kg(-1) . min(-1) followed by propofol 2-3 mg/kg. A small piezo-electric microphone was positioned beside the vocal cords into the muscular process at the base of the arytenoid cartilage to record acoustic signals from the contraction of the LCA. A second microphone was positioned at an external site, lateral to the trachea, just below the thyroid notch. The recurrent laryngeal nerve was stimulated supramaximally using train-of-four (TOF) stimulation every 12 s. Onset, maximum effect, and offset of NMB were measured and compared. Peak effect, time to reach (T) 25%, 75%, and 90% of control twitch response, and TOF recovery to TOF ratios 0.5-0.8 were significantly longer at the external site. The onset time was not significantly different between the two sites. We used phonomyography with a microphone placed at the neck to evaluate the possibility to externally monitor NMB at the larynx. When compared with LCA, we found a more pronounced peak effect and longer offset of NMB. The acoustic signals recorded at this external site are unlikely to stem from laryngeal muscle contraction but are rather a result of contraction of the strap muscles of the neck.