Background: The A-Line Autoregressive Index (AAI), which is derived from auditory evoked potentials, has been used for determining anaesthetic depth. This study verified the correlation between AAI values and the corresponding end-tidal concentrations of sevoflurane during general anaesthesia induction.Methods: Thirty young male adults undergoing elective minor orthopaedic surgery were sequentially allocated to receive inspiratory 3%, 5%, or 6% sevoflurane for mask induction, followed by mechanical ventilation after tracheal intubation. The inspiratory, end-tidal, and estimated jugular bulb concentrations of sevoflurane were recorded at three target AAI values: below 20, below 10 and at the start of burst suppression.Results: The mean time to loss of consciousness in the 6% sevoflurane group was shorter than that in the 5% and 3% groups; however, the groups had comparable AAI values (range: 16–45). The 6% group had a higher end-tidal concentration (4.5% ± 0.2% vs. 3.8% ± 0.2%, P < 0.05) than did the 5% group, despite having the same target anaesthetic levels by AAI score ≤ 10, whereas the estimated jugular bulb concentrations were comparable (1.9% vs. 1.9%) in both groups. Conclusions: Following mechanical ventilation with inspiratory 3%, 5%, or 6% sevoflurane, the end-tidal concentrations were discrepant at the same end points of AAI levels, despite similar estimated jugular bulb concentrations of sevoflurane. Thus, conventional alveolar concentration may overestimate anaesthesia depth during rapid wash-in of sevoflurane.