The current study evaluated the neuromuscular responses following administration of sevoflurane in 14 patients with myasthenia gravis (MG) (I-IIb in Osserman's classification) scheduled for thymectomy and in 11 control patients (ASA I-II) who underwent elective surgery. The electromyographic (EMG) response of the abductor digiti minimi was measured following train-of-four (TOF) stimulation of the ulnar nerve every 20 s. After induction of anesthesia with a combination of 3-4 mg·kg(-1) thiopental and 1-2 μg·kg(-1) fentanyl with 66% N2O and oxygen, an inspired concentration of 4% sevoflurane was administered via a face mask for 7 min. Anesthesia was maintained during surgery with 66% N2O in oxygen and with 1 minimum alveolar concentration (MAC) of end-tidal concentration of sevoflurane. The T1 (the amplitude of the first response) values decreased more profoundly in the MG patients than in the control patients at the end of surgery (P<0.05). Following administration of 4% sevoflurane for 7 min, the TOFR (the ratio of the fourth TOF to the first response) values revealed depressions greater than 10% of preinhalation values in 11 of 14 MG patients with a marked individual variation. This attenuated response was followed by a further depression of the TOFR values with increasing time of 1 MAC sevoflurane anesthesia. On the other hand, no notable changes were observed in patients with normal neuromuscular functions. The most significant factor that correlated with the depression of the TOFR values induced by 1 MAC sevoflurane was the anti-AchR antibody titers (P=0.029). Our results indicate that MG patients have an increased neuromuscular sensitivity to sevoflurane.