Objective: To investigate the application of transcranial facial nerve motor evoked potential (FNMEP) and direct nerve electrical stimulation (DNES) for the prediction of facial nerve function after vestibular schwannoma surgery. Methods: The clinical data of 106 patients who underwent vestibular schwannoma surgery under electrophysiological monitoring of facial nerve between 2017 and 2021 were retrospectively examined, and there were 57 males and 49 females, with a mean age of (51±11) years. Neuroelectrophysiological monitoring was performed in all patients during the operation. After the tumor was removed, FNMEP and DNES were used for electrophysiological evaluation of facial nerve function. The amplitude ratios of FNMEP to baseline (M1) and the brainstem segment to the internal auditory canal segment of DNES (M2) were recorded after the tumor was removed, respectively. The correlation between these two ratios and facial nerve function at 1 day, 1 month and 3 months after the operation were compared. According to the House-Brackmann (HB) scale, Grade Ⅰ-Ⅱ refers to good facial nerve function, and grade Ⅲ-Ⅵ refers to moderate and severe facial nerve dysfunction. Non-parameter Spearman correlation coefficient was used to evaluate the correlation between M1 and M2 and facial nerve function at 1 day, 1 month and 3 months after operation, and the receiver operating characteristic (ROC) curves were plotted to verify the diagnostic efficacy of M1 and M2 for predicting the prognosis of postoperative facial nerve function. Results: Among 106 patients, 102 cases (96.3%) underwent total tumor resection, 4 patients (3.7%) underwent subtotal resection, 104 patients (98.1%) had anatomical preservation of facial nerves, and there were no deaths reported. All patients could evoke reliable FNMEP and 2 patients could not evoke DNES in the brainstem segment of the facial nerve. There were 81 (76.4%), 99 (93.4%) and 103 patients (97.2%) with satisfactory function of facial nerve at 1 day, 1 month and 3 months after surgery, respectively. M1 had large absolute values of Spearman correlation coefficient at 1 day (ρ=|-0.648|) and 1 month (ρ=|-0.552|) after surgery (both P<0.001), while M2 showed a greater absolute value of Spearman correlation coefficient at 3 months (ρ=|-0.395|) than that of M1 (ρ=|-0.378|) (P<0.001). Cut-off value of M1 was 0.58 (sensitivity=0.92, specificity=0.64), and that of M2 was 0.36 (sensitivity=0.64, specificity=1.00). Meanwhile, M1<0.58 or M2<0.36 suggested moderate and severe impairment of facial nerve function. Conclusions: Both FNMEP and DNES during vestibular schwannoma surgery can effectively evaluate the postoperative facial nerve function. FNMEP is better than DNES in predicting the early postoperative facial nerve function, but DNES is better for predicting the long-term postoperative facial nerve function.
目的： 探讨经颅面神经运动诱发电位（FNMEP）和直接神经电刺激（DNES）两种神经电生理监测方法在听神经瘤术中对解剖保留的面神经术后功能的预测作用。 方法： 回顾性分析2017年1月至2021年12月在海军军医大学长海医院行面神经功能监测下听神经瘤切除术的106例患者的临床资料，男57例，女49例，年龄（51±11）岁。术中所有患者均全程使用神经电生理监测，肿瘤切除后分别用FNMEP和DNES两种监测方法对面神经功能进行电生理学评估，记录FNMEP在肿瘤切除后与基线的波幅比M1，肿瘤切除后DNES的脑干段与内听道段的波幅比M2，比较两个比值与术后1 d、1个月、3个月面神经功能的相关性，根据HB分级标准，Ⅰ~Ⅱ级为面神经功能良好，Ⅲ~Ⅵ级为面神经中重度障碍。应用非参数Spearman相关系数评估M1和M2与术后1 d、1个月、3个月面神经功能的相关性，绘制M1、M2预测术后面神经功能预后的受试者工作特征（ROC）曲线。 结果： 106例听神经瘤患者中全切102例（96.3%），次全切4例（3.7%），面神经解剖保留104例（98.1%），无死亡病例。术中所有患者均可诱发出FNMEP，2例患者未在脑干段引出DNES。术后1d面神经功能良好81例（76.4%），术后1个月面神经功能良好99例（93.4%），术后3个月面神经功能良好103例（97.2%）。M1在术后1 d和1个月有较大的Spearman相关系数绝对值（ρ=|-0.648|，ρ=|-0.552|，P<0.001），而M2在术后3个月的ρ=|-0.395|大于M1的ρ=|-0.378|（P<0.001）。M1的cut-off值为0.58（灵敏度=0.92，特异度=0.64），M2的cut-off值为0.36（灵敏度=0.64，特异度=1.00），M1<0.58或M2<0.36提示面神经功能中重度障碍。 结论： 听神经瘤切除术中应用FNMEP和DNES两种方法均可以有效评估术后面运动功能，FNMEP对术后早期面神经功能的预测作用优于DNES，对于术后远期的面神经功能预测则DNES更优。.