经颅电刺激运动诱发电位和皮层体感诱发电位在脊柱侧凸手术中联合监护的观察研究

目的 探讨脊柱侧凸手术中经颅电刺激运动诱发电位(transcranial electrical stimulation motor evoked potential,TES-MEP)和皮层体感诱发电位(cortical somatosensory evoked potential,CSEP)联合监护的可行性和应用价值.方法 2006年7月至2008年4月,在脊柱侧凸手术中同时记录双侧胫前肌、足(足母)短屈肌TES-MEP和双侧胫后神经CSEP 76例.实施全静脉麻醉49例,七氟烷(吸入浓度＜1%)+异丙芬复合麻醉27例.对各麻醉组的TES-MEP检出结果 进行四格表χ2检验,并对术中两种电位的真、假阳性和真、假阴性结果 进行相关的指数统计分析.结果 TES-MEP和CSEP成功检出率均为96.1%,而两种麻醉组的TES-MEP检出率差异无统计学意义.术中诱发电位阳性11例,其中9例为真阳性,均与手术操作不当直接有关.CSEP、TES-MEP、联合监护的灵敏度分别为75.0%、87.5%和100%,特异度分别为98.5%、98.5%和97.0%,约登指数分别为0.74、0.86和0.97.结论 实施异丙芬静脉麻醉为主,辅以七氟烷吸入浓度＜1%的复合麻醉,也是联合监护切实町行的麻醉方案;联合监护对脊髓功能监测的敏感性和准确性明显高于单一的TES-MEP或CSEP监护.     

脊柱手术中TES-MEP和CSEP联合监测脊髓功能的作用评价

目的 探讨脊柱手术中经颅电刺激运动诱发电位(TES-MEP)和皮层体感诱发电位(CSEP)联合监测脊髓功能的临床应用价值.方法 分析2006年7月至2009年4月脊柱手术中同时记录双侧胫前肌、足足母短屈肌(颈椎病变增加鱼际肌)TES-MEP和双侧胫后神经(颈椎病变增加尺神经)CSEP共293例患者的临床资料.术中全静脉麻醉维持,部分患者辅助使用小剂量七氟烷吸入或小剂量肌肉松弛药.将TES-MEP、CSEP、联合监测的结果与术后脊髓运动和感觉功能进行比较及相关分析.结果 术中TES-MEP、CSEP、联合监测的成功检出率分别为90.8%、96.9%和100%.术中TES-MEP、CSEP判断脊髓运动功能的灵敏度分别为100%和89.3%,特异度分别为98.4%和96.9%,约登指数分别为0.984和0.862;而判断脊髓感觉功能的灵敏度分别为76.7%和93.3%,特异度分别为98.7%和98.0%,约登指数分别为0.754和0.913;联合监测的灵敏度为100%,特异度96.9%,约登指数0.969.结论 TES-MEP监测脊髓运动功能的准确性高于CSEP,而CSEP监测脊髓感觉功能的准确性高于TES-MEP;联合监测对脊髓功能监测的灵敏度和准确性又高于单一的TES-MEP或CSEP监测,它是目前脊柱脊髓手术中较为理想的监测方法.     

脊柱侧凸手术中TES-MEP、CSEP阳性与手术操作相关原因分析

目的探讨脊柱侧凸手术中经颅电刺激运动诱发电位（transcranial electrical stimulation motorevoked poten-tial,TES-MEP）和/或皮层体感诱发电位（cortical somatosensory evoked potential,CSEP）阳性与手术操作相关因素。方法 2006年7月～2009年4月,本科行脊柱侧凸手术151例,术中同时监测双侧胫前肌、足拇短屈肌TES-MEP以及双侧胫后肌CSEP。根据术中诱发电位的阳性变化详细检查和分析是否与手术操作不当有关。结果 TES-MEP成功检出率为96.7%,CSEP成功检出率为96%。术中诱发电位阳性27次,除5次不明原因外,22次与手术操作不当相关。其中15次及时采取纠正措施,13次恢复正常,2次明显恢复;7次仅经药物干预和观察处理者,其中6次不能完全恢复,1次无恢复。结论椎弓根钻孔或螺钉置入、内固定器械矫形、截骨过程和截骨面靠拢等手术操作是导致诱发电位阳性的主要原因。     

脊柱外科术中皮层体感诱发电位监护与临床研究

脊柱外科术中脊髓功能的监测方法有：①术中唤醒试验；②皮层体感诱发电位（cortical somatosensory evoked potential，CSEP）监护；③运动诱发电位监测。（moter evoded potedntial，MEP）监护。CSEP监测．近几年在脊柱手术中应用逐渐增多．尽管有许多局限性．但由于操作简单，容易记录、重要性好。不影响手术视野．不需禁用肌松药等优点．在脊柱手术中有一定的实用价值。本院自2004年7月至2005年11月．有选择性对部分重大疑难脊柱手术行术中CSEP监护．大大减少了脊柱外科手术中脊髓损伤的机会．增加手术安全性．并具有法医学上的潜在价值。     

体感诱发电位用于脊柱手术中脊髓功能的监测

目的评估体感诱发电位(SEP)用于脊柱手术中脊髓监测的可行性，以及皮质体感诱发电位(CSEP)与皮质下体感诱发电位(Sub—CSEP)监测的适应证。方法42例脊柱手术患者，年龄6～68岁，咪唑安定、丙泊酚及维库溴铵静脉诱导，安氟醚吸入维持麻醉．术中均应用CSEP和Sub—CSEP监测脊髓功能。结果CSEP受麻醉影响较大，Sub—CSEP波形稳定，全组7例诱发电位出现异常，其中3例与临床相符，4例为假阳性，无假阴性发生，准确率为90．5％。结论CSEP适于麻醉深度的监测，Sub—CSEP是脊柱手术中脊髓功能监测较为理想的方法。     

术中无肌松全静脉麻醉用于联合神经电生理监护脊柱手术的可行性

［摘要］ 目的 比较全静脉麻醉下术中使用或不使用肌松药对脊柱手术中联合神经电生理监测结果的影响,探讨安全有效的神经电生理监测麻醉方案。方法 选择拟行联合神经电生理监测的择期脊柱手术病人 40 例,分为A、B两组。两组病人均采用丙泊酚、瑞芬太尼和右美托咪定全凭静脉麻醉,A组病人术中使用小剂量阿曲库铵维持肌松,B组病人术中不使用肌松药。同时监测体感诱发电位（SEP）和运动诱发电位（MEP）评判脊髓功能。记录术中不同时间点两组病人的生命体征和SEP和MEP的波幅和潜伏期,同时记录经颅电刺激时病人是否出现剧烈体动和自主呼吸。比较两组病人术毕后麻醉苏醒时间和质量。结果 两组病人不同时间点的生命体征差异无统计学意义。两组病人的SEP的波幅和潜伏期差异无统计学意义,MEP的潜伏期差异无统计学意义,MEP的波幅差异有显著性统计学意义。两组病人在电刺激时均无自主呼吸和剧烈体动发生。结论 术中不使用肌松药的全静脉麻醉方案可安全有效地用于行神经电生理监测的脊柱手术,并且在电生理监测信号质量和术后苏醒方面具有明显优势。     

皮层体感诱发电位及经颅电刺激运动诱发电位联合监护在脊柱畸形矫正术中的应用

目的:探讨脊柱畸形矫正术中皮层体感诱发电位(CSEP)及经颅电刺激运动诱发电位(TES-MEP)联合监护在脊柱畸形矫正手术中的应用。方法:对我院收治的脊柱畸形患者根据术中监护方法分为两组:A组37例,其中男25例,女12例,年龄13～42岁,平均26.1岁,术中进行CSEP及TES-MEP联合监护;B组29例,其中男10例,女19例,年龄13～20岁,平均15.8岁,单纯采用CSEP监护作为对照组。术中持续观察CSEP及TES-MEP波幅及潜伏期变化,出现波幅下降50%、潜伏期延长10%或刺激强度高于初始刺激强度100V仍未引出者,停止手术并对症处理,术后给予激素治疗。结果:A组术中出现CSEP异常者4例(10.8%),TES-MEP异常者19例(60%)。术中CSEP及TES-MEP均异常者4例(10.8%),无术中CSEP异常而TES-MEP正常者。2例(陈旧性结核并后凸畸形1例,陈旧性骨折并后凸畸形1例)术中合拢截骨平面时,TES-MEP波形均消失,但仅1例出现CSEP异常,经术中积极处理,至手术结束时波形仍未恢复至术前水平,术后均出现神经功能障碍。余35例患者术后无神经功能损伤(假阴性率0%)。B组9例(21%)术中出现CSEP波形异常,其中2例虽经减少矫形角度及激素冲击治疗,术后仍出现双下肢瘫,余术后未出现神经功能障碍。3例术中监护未见异常者,术后出现重度不可逆性脊髓损伤(假阴性率10%)。结论:CSEP结合TES-MEP联合监护能较可靠、准确的反映术中脊髓功能状态,可降低监护假阴性率,为手术治疗过程提供参考。     

肌电图监测面神经功能用于预测标准麻醉深度患者的体动反应

背景面神经（facialnerve，FN）监测已成为颅面部手术、中耳手术以及颅底手术的标准监护手段，以减少医源性损伤，进行面神经定位，并预测术后神经功能。以往的研究也表明面神经肌电图监测（facialnerve electromyography，FNEMG）具有评估麻醉深度和预测患者体动的临床价值。本研究旨在探讨脑电双频指数（bispectral index，BIS）和FNEMG在两种不同的麻醉方式中，是否能用于预测颅骨手术患者的术中体动。方法采用单盲、随机、对照的临床试验方法，将FNEMG监测和BIS监测用于特殊颅面部手术和颅骨手术的全麻患者，评价这两种监测手段预测术中患者体动的相关性。另外采用异丙酚联合瑞芬太尼的全凭静脉麻醉（total IV anesthetic，TIVA）与地氟烷吸入麻醉进行比较，确定何种麻醉方式既能够提供最佳的充分麻醉状态，也能够预防需要FNEMG监测的非肌松患者术中运动。结果TIVA组血流动力学状态更好。两组间BIS值无显著性差异，而TIVA组FNEMG活性更低。地氟烷组出现运动的患者多于TTVA组，10例运动患者中有8例存在显著FNEMG活性，FNEMG对术中可见运动的阳性预测价值为38％，阴性预测值为95％。FNEMG表现活性的运动患者与无运动患者相比，血流动力学、BIS值与基础值相比均无显著变化。结论本研究表明FNEMG可能成为预测颅面部手术和颅底手术患者术中运动的有效监测。但是，BIS监测不能充分预测这类患者的运动。DES组更常见血流动力学变化和FNEMG活性，而TTVA更有效的防止患者运动，可调整剂量为临床操作提供稳定的状态。FNEMG是辅助预测并阻止颅面部手术和颅底手术患者运动的一种有效的临床工具。     

无肌松药全凭静脉麻醉对脊柱侧凸手术患者运动诱发电位监测的影响

目的 评价无肌松药全凭静脉麻醉对脊柱侧凸手术患者运动诱发电位(MEP)监测的影响.方法 择期行脊柱侧凸后路融合术患者50例,性别不限,年龄18～ 25岁,ASA分级Ⅰ或Ⅱ级,采用随机数字表法,将患者随机分为2组(n=25):肌松药复合全凭静脉麻醉组(Ⅰ组)和无肌松药全凭静脉麻醉组(Ⅱ组).2组均经20 min静脉输注右美托咪定负荷量1 μg/kg,随后以0.2μg· kg-1 ·h-1速率维持.给予右美托咪定负荷量后进行麻醉诱导,气管插管,机械通气.麻醉维持:静脉输注瑞芬太尼0.2 μg·kg-1·min-1和丙泊酚80～ 120 μg·kg-1·min-1,维持脑电双频谱指数值40 ～ 60;静脉输注顺阿曲库铵0.5～1.5 μg·kg-1 ·min-1,采用四个成串刺激监测肌松程度,维持T1 45％ ～ 55％,Ⅱ组于椎旁肌肉分离完毕前30 min停用顺阿曲库铵.记录停用肌松药期间MEP监测情况,并由外科医生评价肌松程度.记录成功唤醒情况.结果 与Ⅰ组比较,Ⅱ组MEP监测成功率升高(P＜0.05),肌松程度差异无统计学意义(P＞0.05).2组患者均成功唤醒.结论 无肌松药全凭静脉麻醉用于脊柱侧凸手术患者可在满足手术所需肌松程度的同时,提高MEP监测的成功机率.     

小儿硫喷妥钠／七氟烷麻醉及复合使用酮洛芬和芬太尼对刺激正中神经的体感诱发电位的影响

背景体感诱发电位（somatosensory evoked potentials,SEPs）常被用于外科手术中判断脊髓和脑功能。总的来说,SEPs对吸入性麻醉药物是敏感的,但有关小儿用七氟烷维持麻醉对sEPs的影响了解甚少。麻醉中常会使用镇痛药物,辅助用药也会影响SEPs。在这项前瞻性的临床试验中,共计27例3。8岁的健康儿童接受研究,对其静脉给予苯二氮革和巴比妥类药物麻醉诱导后,用七氟烷维持麻醉,并观察它们对正中神经SF2s的影响。此外,也观察了两种镇痛药物,酮洛芬和芬太尼对SEPs的影响。方法麻醉前静脉给予0．1mg／kg咪达唑仑,用呼气末浓度为2％的七氟烷维持麻醉,分别在给予咪达唑仑前,吸入七氟烷后15分钟、25分钟（给予或不给予酮洛芬）、35分钟（给予芬太尼）记录正中神经SEPs。结果与基础水平相比,应用七氟烷维持麻醉期间N20潜伏期延长（P=0．015）,中枢传导时间延长（P=0．001）;使用镇痛药物对N20潜伏期和中枢传导时间并没有影响。5-8岁儿童的平均皮层N20-P25的振幅显著降低（P=0．008）。此外,对于年长的儿童,复合用酮洛芬和芬太尼后N20-P25振幅比使用前明显减低（P=0．03）,而在年幼的儿童则没有这样的变化。讨论本研究发现,儿童使用2％七氟烷维持麻醉会延长正中神经的SEPs,与用其他吸入性麻醉药的报道相似。然而,吸入七氟烷时可以进行SEP的监测,但吸入的剂量应根据个体差异调整。复合使用酮洛芬和芬太尼不影响SEP的潜伏期,但posthoc分析结果提示,较年长的儿童皮层波幅会减低。     

Evaluation of the applicability of sevoflurane during post-tetanic myogenic motor evoked potential monitoring in patients undergoing spinal surgery

Purpose  Recent evidence has indicated that post-tetanic motor evoked potentials (p-MEPs) can be used to improve the reliability of the monitoring of motor function during spinal surgery. However, data on p-MEP monitoring are limited to those in subjects under propofol anesthesia. The present study was conducted to assess the applicability of sevoflurane during p-MEP monitoring in patients undergoing spinal surgery. Methods  Thirty-five patients undergoing spinal surgery under sevoflurane anesthesia were enrolled in the study and classified as being without preoperative motor deficits (n = 25) or with preoperative motor deficits (n = 10). For conventional MEP (c-MEP), transcranial train-pulse stimulation was delivered and the compound muscle action potentials were bilaterally recorded from the abductor pollicis brevis, abductor hallucis, tibialis anterior, and soleus muscles. For p-MEP, tetanic stimulation (50 Hz, 50 mA stimulus intensity) for 5 s was applied to the bilateral median and left tibial nerves 1 s prior to transcranial stimulation. Results  The amplitudes of p-MEP were significantly higher in all muscle recording sites than those of c-MEP in patients without motor deficits, whereas these amplitudes were significantly higher in only four of the eight muscles in patients with motor deficits (P < 0.05). The success rates of c-MEP and p-MEP recording were 48% and 64%, respectively, in patients without motor deficits and 30% and 60%, respectively, in patients with motor deficits. There were no statistically significant differences in success rates between c-MEP and p-MEP recording. Conclusion  Although the application of tetanic stimulation prior to transcranial stimulation did not significantly increase the success rates of MEP recording, it significantly enlarged MEP amplitude under sevoflurane anesthesia in patients without preoperative motor deficits.     

Somatosensory- and motor-evoked potential monitoring during spine and spinal cord surgery

STUDY DESIGN: Prospective, observational study. SETTING: Regional Trauma Center, Torino, Italy. OBJECTIVES: Complex spinal surgery carries a significant risk of neurological damage. The aim of this study is to determine the reliability and applicability of multimodality motor-evoked potentials (MEPs) and somatosensory-evoked potentials (SEPs) monitoring during spine and spinal cord surgery in our institute. METHODS: Recordings of MEPs to multipulse transcranial electrical stimulation (TES) and cortical SEPs were made on 52 patients during spine and spinal cord surgery under propofol/fentanyl anaesthesia, without neuromuscular blockade. RESULTS: Combined MEPs and SEPs monitoring was successful in 38/52 patients (73.1%), whereas only MEPs from at least one of the target muscles were obtained in 12 patients (23.1%); both MEPs and SEPs were absent in two (3.8%). Significant intraoperative-evoked potential changes occurred in one or both modalities in five (10%) patients. Transitory changes were noted in two patients, whereas three had persistent changes, associated with new deficits or a worsening of the pre-existing neurological disabilities. When no postoperative changes in MEP or MEP/SEP modalities occurred, it was predictive of the absence of new motor deficits in all cases. CONCLUSION: Intraoperative combined SEP and MEP monitoring is a safe, reliable and sensitive method to detect and reduce intraoperative injury to the spinal cord. Therefore, the authors suggest that a combination of SEP/MEP techniques could be used routinely during complex spine and/or spinal cord surgery.     

经颅刺激运动诱发电位联合体感诱发电位监测在特发性脊柱侧凸矫形手术中应用的研究

背景：特发性脊柱侧凸手术治疗中神经功能障碍是最应受到关注的并发症,术中神经功能监测可帮助早期发现可能的神经功能损伤。目的：分析联合应用经颅刺激运动诱发电位（TcMEP）和体感诱发电位（SEP）的多模式术中神经功能监测在特发性脊柱侧凸矫形手术中对预测医源性神经功能损害的作用。方法：在特发性脊柱侧凸矫形手术中,运用TcMEP和（或）SEP进行神经功能监测。MEP监测采用经颅刺激C3、C4,记录外周肌源性MEP,SEP监测采用刺激双侧胫后神经,记录电极采用Cz—CPz。阳性诊断标准：与基线相比,MEP波幅下降75％,SEP波幅下降50％。结果：112例特发性脊柱侧凸患者中,联合MEP、SEP监测的检出率为100％。MEP监测阳性6例,假阳性1例。1例出现一过性神经功能障碍。MEP的监测敏感性为100％,特异性为99％;SEP监测敏感性50％,特异性为100％;联合MEP、SEP监测的敏感性、特异性均为100％。结论：特发性脊柱侧凸矫形手术中MEP＋SEP的术中神经功能监测可提高监测敏感性及特异性,可预测术中神经功能损伤事件的发生。MEP是多模式监测的基础,SEP是重要补充。单独应用MEP监测在特发性脊柱侧凸患者手术中有应用前景。     

Recovery from rocuronium by sugammadex does not affect motor evoked potentials

Motor evoked potential (MEP) monitoring has been employed to detect the spinal cord injury during spinal, neurosurgical and cardiovascular operations. Muscle relaxants diminish the amplitude of MEP because MEP is the picture of electromyogram. In 5 cases undergoing MEP monitoring, we examined the effect of rocuronium followed by the administration of sugammadex on MEP Anesthesia was induced with propofol (target controlled infusion 3.0-3.5 microg x ml(-1)) and remifentanil 0.15-0.3 microg x kg(-1) x min(-1), and the trachea was intubated with the use of rocuronium 0.6 mg x kg(-1) without any muscle rigidity, bucking and laryngospasm. General anesthesia was maintained by total intravenous anesthesia using propofol and remifentanil with no muscle relaxants. Immediately after the tracheal intubation, sugammadex 4 mg x kg(-1) was intravenously given. The amplitude of MEP was measured just before the administration of rocuronium, immediately after the tracheal intubation, and 1, 2, 3, 5 min following the administration of sugammadex. Sugammadex restored the MEP amplitude, deteriorated by rocuronium, in 3 to 5 min to the level of non-paralytic muscles. In one case, it took 8 min to restore the MEP of hemiparetic leg. Taking these findings into consideration, it is likely that rocuronium might not affect the MEP when reversed by sugammadex, and should be safe for smooth tracheal intubation in patients who need MEP monitoring.     

Propofol and sevoflurane depress spinal neurons in vitro via different molecular targets

BACKGROUND: The capacity of general anesthetics to produce immobility is primarily spinally mediated. Recently, compelling evidence has been provided that the spinal actions of propofol involve gamma-aminobutyric acid type A (GABAA) receptors, whereas the contribution of glycine receptors remains uncertain. The relevant molecular targets of the commonly used volatile anesthetic sevoflurane in the spinal cord are largely unknown, but indirect evidence suggests a mechanism of action distinct from propofol. METHODS: The effects of sevoflurane and propofol on spontaneous action potential firing were investigated by extracellular voltage recordings from ventral horn interneurons in cultured spinal cord tissue slices obtained from embryonic rats (embryonic days 14-15). RESULTS: Propofol and sevoflurane reduced spontaneous action potential firing of neurons. Concentrations causing half-maximal effects (0.11 microm propofol, 0.11 mm sevoflurane) were lower than the median effective concentration immobility (1-1.5 microm propofol, 0.35 mm sevoflurane). At higher concentrations, complete inhibition of action potential activity was observed with sevoflurane but not with propofol. Effects of sevoflurane were mediated predominantly by glycine receptors (45%) and GABAA receptors (38%), whereas propofol acted almost exclusively via GABAA receptors (96%). CONCLUSIONS: The authors' results suggest that glycine and GABAA receptors are the most important molecular targets mediating depressant effects of sevoflurane in the spinal cord. They provide evidence that sevoflurane causes immobility by a mechanism distinct from the actions of the intravenous anesthetic propofol. The finding that propofol acts exclusively via GABAA receptors can explain its limited capacity to depress spinal neurons in the authors' study.     

A case report of a patient who developed hemiparaplegia with multiple cerebral infarction during thoracoabdominal aortic aneurysm repair

To protect the spinal cord during thoracoabdominal aortic aneurysm repair, motor evoked potentials (MEP) monitoring and cerebrospinal fluid drainage are often employed. Herein, we report a case, where intraoperative diminishment of motor evoked potentials was accompanied by multiple cerebral infarction. A 63-year-old man underwent elective surgery for both thoracoabdominal aortic aneurysm and abdominal aortic aneurysm. He had a past history of cerebral infarction, resulting in Wernicke aphasia but no paralysis. Preoperative magnetic resonance angiography and echocardiography revealed occlusion of the intercostal and lumbar arteries, mild aortic regurgitation, and atherosclerotic lesions at the aortic arch as well as descending aorta. Anesthesia and muscular relaxation were maintained with fentanyl, propofol, and continuous administration of vecuronium at 0.5 mg x kg(-1) x h(-1). The thoracoabdominal aortic aneurysm was repaired under distal aortic perfusion with femorofemoral bypass. After terminating the bypass, we found that the MEP at the lower limb had disappeared. Although we reconstructed intercostal arteries under mild hypothermia and partial bypass, the amplitude of MEP remained very low. Suspecting spinal cord ischemia, we performed cerebrospinal fluid drainage immediately after the operation. On the postoperative day 4, when we stopped the cerebrospinal fluid drainage and propofol administration, his level of consciousness was poor and brain CT revealed multiple cerebral infarction. On the postoperative day 30, he was discharged from an intensive care unit with complications of hemiplagia and paraplegia. Although cerebrospinal fluid drainage may be recommended to protect spinal cord during thoracoabdominal aortic aneurysm repair, we should consider performing brain CT to exclude a risk of brain herniation secondary to cerebrospinal fluid drainage if there is a possibility of cerebral incidents.     

Effect of bolus propofol administration on muscle evoked potential (MsEP) during spine surgery

Intraoperative monitoring of descending pathways by means of muscle evoked potential (MsEP) is a reliable method to monitor spinal cord motor function, but MsEP is readily affected by anesthetics. We monitored MsEP evoked by repetitive transcranial electrical stimulation of the motor cortex in 30 patients receiving spine surgery. Total intravenous anesthesia was maintained with propofol and fentanyl without any muscle relaxant. Onset latencies and peak to peak amplitudes of MsEP were evaluated before and after the bolus propofol administration. The concentrations of propofol in blood and the effect-site during MsEP monitoring were predicted by computer simulation software. The amplitude of MsEP decreased slightly by bolus propofol administration, but the latencies showed no significant change with propofol under the same condition. We consider that total intravenous anesthesia with propofol and fentanyl without muscle relaxants is compatible with the recording of MsEP evoked by high frequency repetitive electrical transcranial stimulations. When MsEP is monitored during spine surgery, anesthetic condition should be controlled carefully in order to maintain a stable blood concentration of propofol and thus to assure the reliability of MsEP measurements.