Objective This study on a medium-fidelity simulator (SimMan , Laerdal Medical Corpo-ration,Wappingers Falls, NY, USA) examined the management of unanticipated difficult airway by residents of anesthesiology and the effect of training in this context.Methods 30 residents of anesthesiology were devided into two groups (T and E). There were two scenarios investigated:'can't intubate, can oxygenate'(CI) and'can't intubate, can't oxygenate'(CICO). The E group was trained by scene simulation in the second, fourth, sixth month respectively before and after the training, and the T group received the same situation simulation training on the same day after the theoretical training and in the second, sixth month after the training. In con-trast to standard operating procedures, the performance of two groups of trainees was recorded on the basis of pre established evaluation criteria. Data differences between the two groups were analyzed using SPSS 23.0,t test, M-W test andx2 test.ResultIn CI, success rate of placement of astandard and intubating laryngeal mask air-way were high in T group (87％ vs. 55％ ,P=0.037). This was sustained over time. There was no difference in duration and incidence of desaturation between two groups. In CICO, there was a more structured approach following training in T group (P<0.05), which wasn't sustained over time. But this was sustained over 6 months in E group.ConclusionSituational simulation training can significantly improve the anesthesiology residents' ability to take proper response measures to the unexpected difficult airway, and significantly shorten the training time for emergency airway treatment. Repeated situational simulation training should be conducted at intervals of 2 months or less, allowing residents to maintain emergency response to the emergency airway.