体感诱发电位在重症动脉瘤性蛛网膜下腔出血患者外科或介入干预术前评估中的应用：一项前瞻性观察性队列研究

Use of somatosensory evoked potentials for preoperative assessment in patients with severe aneurysmal subarachnoid hemorrhage before surgical or interventional treatment： a prospective observational cohort study

目的 探讨短潜伏期体感诱发电位（SLSEP）在重症动脉瘤性蛛网膜下腔出血（aSAH）患者手术或介入治疗前作为术前评估工具的可能性。方法 采用前瞻性观察性队列研究方法，选择2016年11月至2017年4月首都医科大学附属北京天坛医院重症医学科（ICU）收治的重症aSAH患者〔WFNS分级≥Ⅳ级〕作为研究对象。于发病后12 h内对患者进行SLSEP监测，并按Judson标准分级；应用改良Fisher分级评价患者入院时脑CT检查结果；随访患者术后3个月预后，以改良Rankin量表（mRS）评分0~3分为预后良好，4~6分为预后不良。采用单因素分析比较两组患者一般资料及临床、神经影像、SLSEP评估结果，筛选导致预后不良的可能因素，并代入多因素Logistic回归模型，分析影响患者预后的独立危险因素；采用敏感度、特异度、阳性预测值（PPV）和阴性预测值（NPV）评价各项分级结果对预后的预测价值。结果 共入选41例aSAH患者，其中7例因SLSEP监测结果受到干扰而被排除，最终34例患者纳入分析，预后不良组21例，预后良好组13例。两组性别、年龄、体重指数（BMI）、发病至手术和（或）介入治疗时间及干预方式等比较差异均无统计学意义。单因素分析显示，预后不良组WFNS分级、改良Fisher分级和SLSEP Judson分级结果与预后良好组比较差异均有统计学意义。预测价值评估结果显示，WFNS分级Ⅴ级和改良Fisher分级Ⅳ级的敏感度、特异度、PPV、NPV均未超过85%；而SLSEP Judson分级Ⅲ级预测预后不良的敏感度（90.5%比71.4%、71.4%）、特异度（84.6%比69.2%、76.9%）和PPV（90.5%比79.0%、83.3%）均高于WFNS分级Ⅴ级和改良Fisher分级Ⅳ级。多因素Logistic回归分析显示，仅SLSEP Judson分级Ⅲ级是导致预后不良的独立危险因素〔优势比（OR）=45.73，95%可信区间（95%CI）=4.25~499.31，P=0.002〕；而WFNS分级Ⅴ级（OR=1.14，95%CI=0.12~13.06，P=0.912）和改良Fisher分级Ⅳ级（OR=7.22，95%CI=0.51~113.20，P=0.160）虽与预后不良存在一定相关性，但并非其独立危险因素。结论 与WFNS分级和改良Fisher分级相比，SLSEP能在手术或介入干预前更准确地预测重症aSAH患者的远期预后，可考虑应用SLSEP对此类患者进行术前评估。

Objective To explore the application value of short latency somatosensory evoked potentials （SLSEP） as a tool for preoperative assessment of surgical or interventional treatment in patients with severe aneurysmal subarachnoid hemorrhage （aSAH）.Methods A prospective observational cohort study was conducted. The patients with severe aSAH with a WFNS grade of Ⅳ or Ⅴ admitted to intensive care unit （ICU） of Beijing Tiantan Hospital of Capital Medical University from November 2016 to April 2017 were enrolled. The patients received SLSEP monitoring within 12 hours after onset, and the monitoring results were classified according to the Judson scale. Meanwhile, the findings on cerebral CT scans at admission were evaluated by the modified Fisher classification. The follow-up was performed at 3 months after aSAH ictus based on the modified Rankin scale （mRS）, and a mRS score 0-3 was defined as favorable outcome, 4-6 was defined as unfavorable outcome. For statistical evaluation, demographic, clinical, neuroimaging and SLSEP data were evaluated by univariate analysis to identify the risk factors associated with prognosis; afterwards, those factors were analyzed by multivariate Logistic regression; also the validity was assessed by calculating the respective sensitivity, specificity, positive predictive value （PPV） and negative predictive value （NPV）.Results A total of 41 patients with aSAH were selected, of which 7 were excluded because of the interference of the SLSEP monitoring results, 34 patients with aSAH were enrolled finally. Among them, 21 were classified in the unfavorable outcome group, while the rest （n = 13） were allocated into the favorable outcome group. No significant difference was found in gender, age, body mass index （BMI）, time delay from ictus to treatment or the options for therapeutic methods between the two groups. The findings of univariate analysis, however, showed statistically differences in WFNS grade, the modified Fisher scale and Judson scale of SLSEP between the two groups. Yet, the further validity evaluation for these predictors demonstrated that the sensitivity, specificity, PPV and NPV of WFNS grade of Ⅴ and modified Fisher scale of Ⅳ were all less than 85%, whereas the results for SLSEP Judson scale of Ⅲ were much better （sensitivity： 90.5% vs. 71.4% and 71.4%, specificity： 84.6% vs. 69.2% and 76.9%, PPV： 90.5% vs. 79.0% and 83.3%）. In the following multivariate Logistic analysis, only Judson scale of Ⅲ was identified to be the independent risk factor for poor outcome [odds ratio （OR） = 45.73, 95% confidence interval （95%CI） = 4.25-499.31, P = 0.002], while the WFNS grade of Ⅴ （OR = 1.14, 95%CI = 0.12-13.06, P = 0.912） and the modified Fisher scale of Ⅳ （OR = 7.22, 95%CI = 0.51-113.20, P = 0.160） were merely associated with poor outcomes without significant independence.Conclusion In comparison with WFNS grade and the modified Fisher scale, SLSEP seems more accurate in the prediction of long-term outcome of severe aSAH prior to surgical or interventional treatment, and thus may be applied as an effective aid in preoperative assessment.