Clinical Neurophysiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Transcranial motor evoked potentials during anesthesia with desflurane versus propofol – A prospective randomized trial M.J. Malcharek a,⇑,1, S. Loeffler a,1, D. Schiefer a, M.A. Manceur b, A. Sablotzki a, J. Gille a, S. Pilge c, G. Schneider c a b c
Department of Anesthesiology, Intensive Care and Pain Therapy, Klinikum St. Georg gGmbH, Leipzig, Germany Division of Biostatistics at Computational Landscape Ecology & Community Ecology, Helmholtz-Centre for Environmental Research GmbH – UFZ, Leipzig, Germany Department of Anesthesiology, Witten/Herdecke University, Helios Clinic Wuppertal, Nordrhein-Westfalen, Germany
a r t i c l e
i n f o
Article history: Accepted 30 November 2014 Available online xxxx Keywords: Motor evoked potentials Intraoperative neurophysiologic monitoring Volatile anesthetics Total intravenous anesthesia Propofol Desflurane
h i g h l i g h t s Transcranial electrical motor evoked potential (tcMEP) amplitudes were significantly larger with pro-
pofol anesthesia than with desflurane anesthesia. Intraindividual differences in tcMEP amplitudes for the desflurane–propofol regimen in the effect
group were significantly smaller. Regardless of the quantitative aspects of the tcMEP amplitudes, recordings of tcMEPs were feasible
under desflurane/remifentanil- and propofol/remifentanil-based general anesthesia.
a b s t r a c t Objective: This study aimed to evaluate differences in transcranial electrical motor evoked potential (tcMEP) amplitudes between desflurane/remifentanil and propofol/remifentanil anesthesia treatment plans in patients without preexisting motor deficits (PMDs) undergoing carotid endarterectomy (CEA). Methods: This prospective trial included 21 patients who were randomly assigned to an effect group (GroupDESFLURANE; n = 14) or a control group (GroupSTANDARD-PROPOFOL; n = 7). tcMEP amplitudes were measured 35 min post-induction (T1) either with desflurane or propofol. Treatment was then changed to propofol in GroupDESFLURANE. After an additional 35 min, the tcMEP amplitudes were reevaluated (T2). Differences in amplitudes (DW) between T1 and T2 were calculated for each patient, and the means of these differences were compared between groups. Results: tcMEPs were recorded in all 21 patients. At T1, the mean amplitude was 840.1 (SD 50.3) lV and 358.9 (SD 74) lV for GroupSTANDARD-PROPOFOL and GroupDESFLURANE, respectively. The absolute mean difference (T1 T2) between groups was 496.75 lV (p = 0.0006). Conclusion: Desflurane reduces the tcMEP amplitude significantly more than propofol in patients without PMDs undergoing CEA. Significance: TcMEPs were recorded in all patients regardless of the anesthesia regimen. In patients with initially small amplitudes, desflurane may limit tcMEP recording because it produces a remarkable amplitude reduction, even in patients without PMDs. Ó 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Clinical Neurophysiology.
1. Introduction ⇑ Corresponding author at: Division of Neuroanesthesia & Intraoperative Neuromonitoring, Department of Anesthesiology, Intensive Care and Pain Therapy, Klinikum St. Georg gGmbH, Delitzscher Str. 141, 04129 Leipzig, Germany. Tel.: +49 341 9094009; fax: +49 341 9092568. E-mail address: [email protected] (M.J. Malcharek). 1 Michael J. Malcharek and Silvana Loeffler have contributed equally, each as first author of this work and in preparing the original manuscript.
The intraoperative monitoring of motor function using direct cortical or transcranial electrical stimulated motor evoked potentials (tcMEPs) is widely used in neurosurgery (Kombos et al., 2003; Neuloh et al., 2004), orthopedic surgery (Langeloo et al., 2003; MacDonald et al., 2003) and vascular surgery (MacDonald and Janusz, 2002; Sloan and Lameson, 2007). Since the introduction of
http://dx.doi.org/10.1016/j.clinph.2014.11.025 1388-2457/Ó 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Clinical Neurophysiology.
Please cite this article in press as: Malcharek MJ et al. Transcranial motor evoked potentials during anesthesia with desflurane versus propofol – A prospective randomized trial. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.025
2
M.J. Malcharek et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
multi-pulse tcMEP stimulation, monitoring of the corticospinal tract (CST) has become more successful in patients under general anesthesia (Kalkman et al., 1992; Kawaguchi et al., 1996; Pechstein et al., 1998). Next to the technical development of complex monitoring systems designed for the modern operating room, the adjustment of certain anesthesia regimens might be responsible for the widespread and successful application of tcMEPs. Total intravenous anesthesia using a propofol infusion has been preferred when MEP monitoring was requested during surgery (Scheufler et al., 2005; Hayashi et al., 2008). However, following the synaptic model, which describes anesthetic action on the nervous system, propofol and volatile anesthetics have similar effects on GABA (Wakasugi et al. 1999; Alkire et al., 2008; Eckle et al., 2013) but may have different actions on cortical or spinal glycine receptors (Hales and Lambert, 1991; Mihic et al., 1997; Grasshoff and Antkowiak, 2004). Thus, numerous studies in the literature have investigated differences in MEP recording under the administration of either TIVA or halogenated agents during neurosurgical or orthopedic procedures (e.g., Pechstein et al., 1998; Pelosi et al., 2001; Scheufler et al., 2005). MEP monitoring has only recently been shown to be a useful adjunct to SSEP monitoring during carotid surgery (Alcantara et al., 2013; Malcharek et al., 2013); therefore, to date, there has been no study in the literature evaluating different anesthetics and their effects on tcMEP recordings in this cohort. Thus, our study is unique because it investigates the effects on tcMEP amplitudes using both TIVA (propofol/remifentanil infusion) and desflurane/remifentanil in patients undergoing carotid surgery. The present trial addresses the following objective: the quantitative evaluation of significant differences in tcMEP amplitudes between two anesthetic approaches, desflurane/remifentanil and the standard propofol/remifentanil protocol. 2. Methods Ethical approval for this study (Registration No: EK-BR-75/13-1) was given by the Research Ethical Committee of the State Chamber of Medicine in Saxony. All patients involved in the study signed an informed consent form regarding the application of the two different anesthesia regimens according to the study protocol. 2.1. Patients We prospectively investigated all patients who were scheduled for elective carotid endarterectomy (CEA) at Klinikum St. Georg gGmbH, Leipzig, Germany between August and December 2013 and who did not have preoperative motor deficits on the contralateral side of the surgery. The patients were evaluated to ensure that they had no preoperative motor deficits by a neurologist and a vascular surgeon. Twenty-one patients were allocated to two groups using a block randomization table (GroupDESFLURANE: 14 patients and GroupSTANDARD-PROPOFOL: 7 patients). The anesthesia plan in GroupDESFLURANE was switched after 35 min from desflurane/ remifentanil to propofol/remifentanil, whereas patients in GroupSTANDARD-PROPOFOL exclusively received a propofol/remifentanil infusion. Measurements of the tcMEPs were performed after 35 (T1) and 70 min (T2) in both groups. Therefore, each patient served as his/her own control. Moreover, GroupSTANDARD-PROPOFOL was defined as the control group because TIVA (propofol/remifentanil) has been routinely performed during the intraoperative monitoring of tcMEPs at our institution. To avoid changes in EP recordings because of positioning or the surgical procedure itself, the study was performed in each patient prior to surgical positioning. Patients with existing motor deficits, acute stroke, allergies to propofol, a history of seizures or epilepsy, a history of propofol
infusion syndrome, cardiac insufficiency (NYHA P III) and implanted pacemakers or defibrillators were excluded. We performed an a priori statistical power analysis in a separate cohort to estimate the number of patients needed for statistical analysis. Because the baseline of MEP amplitudes is highly variable, we included 50 patients undergoing CEA with the standardized propofol/remifentanil regimen. The MEP amplitudes (mean out of 5 recordings per patient) were preoperatively registered to estimate the variability of the measurements. Expert opinion provided a range of expected effect sizes. An a priori sample size of 15 patients was calculated to have 90% power to detect a difference between improved (n = 10) and standard treatment (n = 5) with a level of significance set at p = 0.05 (variance: 200; effect size: 405 lV). 2.2. Anesthesia protocol Patients were premedicated with 3.5–7 mg of oral midazolam one hour prior to transfer to the operating room. The heart rate (HR), mean arterial blood pressure from the radial artery (MAP), inspiratory oxygen fraction (FiO2), end-tidal carbon dioxide (etCO2) and ear temperature were measured in all patients. The target oxygen saturation was over 96%. Normothermia was maintained by forced air warming. All patients received etomidate, a continuous infusion of remifentanil, and rocuronium for the induction of general anesthesia (Fig. 1). Anesthesia was maintained either with desflurane in the event group (GroupDESFLURANE) or with TIVA in the control group (GroupSTANDARD-PROPOFOL), as shown in Fig. 1. According to the study protocol in GroupDESFLURANE, inhaled anesthesia was terminated after 35 min and switched to TIVA. An initial bolus of propofol (1 mg/kg) was administered prior to the continuous infusion of propofol. The initial remifentanil infusion rate ranged from 0.2 to 0.6 lg/ kg/min based on clinically relevant bradycardia (target HR > 40/ min). The infusion rate was then not changed unless medical reasons required a dose adjustment (e.g., bradycardia). Given that the dose of the remifentanil infusion was assigned first, the doses of propofol and desflurane were adjusted based on a defined target range (30–45) of the bispectral index (BIS; BIS-Module for Philips M1034A with BISx, frontal montage, Aspect Medical Systems Inc., Framingham, MA, USA). To provide comparable depths of anesthesia, BIS values were assumed to be similar for both regimens within this range. The propofol infusion was adjusted so that it did not exceed 6 mg/kg/h, and a maximum concentration of 4% of desflurane was kept at an age-adjusted minimal alveolar concentration (MAC; Amingo–Avance, Datex-Ohmeda Inc., Madison, WI, USA) of 0.5–0.6. The concentration of desflurane or propofol was reduced when the BIS values decreased 2: 1 mg/kg) in both groups. According to the study protocol, rocuronium was not redosed during the experimental period. The MAP was increased by administering akrinorÒ (1 ampule: cafedrine-HCl 200 mg and theoadrenaline-HCl 10 mg; ASTA Medica AWD GmbH, Dresden, Germany) or norepinephrine in case of a decrease by more than 20% of the preoperative values. The ventilator was set to maintain normocapnia, and the target SpO2 was P96%. 2.3. Intraoperative neurophysiologic monitoring All patients underwent multimodal EP monitoring, which is the standard monitoring technique for CEA at our institution.
Please cite this article in press as: Malcharek MJ et al. Transcranial motor evoked potentials during anesthesia with desflurane versus propofol – A prospective randomized trial. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.025
3
M.J. Malcharek et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
Remifentanil infusion rate: 0.2-0.6 µg/kg/min G
DESFLURANE
Propofol
(n=14)
MAC : 0.5-0.6 Target BIS2: 30-45
Initial Bolus: 1 mg/kg Infusion rate: 4-6 mg/kg/h Target BIS2: 30-45
Propofol
Propofol
Initial Bolus: 1 mg/kg Infusion rate: 4-6 mg/kg/h Target BIS2: 30-45
Infusion rate: 4-6 mg/kg/h Target BIS2: 30-45
1
G STANDARD-PROPOFOL (n=7)
35 min
70 min
T1
T2 3
Induction
Reversal of NMB
(etomidate 0.2 mg/kg, rocuronium 0.6 mg/kg)
(sugammadex 1-2 mg/kg)
1
Minimal alveolar concentration Bispectral index 3 Neuromuscular blockade 2
Fig. 1. Scheme of the study protocol to time the administration of anesthetic agents.
Table 1 Modalities for the assessment of tcMEP monitoring. Standardized institutional setting during the investigation. tcMEPa stimulation Monitoring system Electrodes Montage of electrodes Specifications
a
tcMEPa recording
ISIS IOM Neuromonitoring system (Inomed Medizintechnik GmbH) Corkscrew electrodes (MNCS0610DS, Spes Medica S.r.l., Italy) Needle electrodes (SDN 530620, Inomed Medizintechnik GmbH, Germany) C4 + 1 cm C3 + 1 cm/C3 + 1 cm C4 + 1 cm Bilateral abductor digiti minimi muscle (ADM) (according to the 10/20-system) Band pass filter: 5–2500 - Constant current (upper voltage limit: 300 V) - Range of intensity needed for suprathreshold stimulation: 35–100 mA - Train: 5; Interstimulus interval: 2 ms - Impulse: 0.5 ms, monophasic - Stimulation rate: 2 Hz
Transcranially electrical stimulated motor evoked potential.
Multimodal EP monitoring includes median nerve somatosensory evoked potentials (mSSEPs), tibial nerve somatosensory evoked potentials (tSSEPs) and tcMEPs from the upper extremities. The monitoring of SSEPs was not subject to investigation in this trial. There was no pre-anesthesia referential assessment of tcMEPs; therefore, an initial stimulation threshold was set at 5 mA above the first intensity that produced stable and reproducible MEP recordings (just before the 1st test interval: T1). This threshold intensity was not changed during the experimental period. Concerning the recording of tcMEPs during the two test intervals (T1 and T2), Table 1 shows the parameters of the tcMEP measurement. The mean amplitude was analyzed out of five single measurements (inter train interval: 15 s) per test. According to the study protocol, only responses from the side contralateral to the surgical site were analyzed because the results might be potentially more important at the surgically relevant side. The tcMEP monitoring was performed by one of the three neuroanesthesiologists who were involved in the study. All were trained and certified in the assessment of intraoperative neurophysiologic monitoring (‘‘Curriculum Neuromonitoring’’: educational program of the German Society of Anaesthesiology and Intensive Care Medicine).
temperature, SpO2, FiO2 and etCO2) and the BIS values at T1 and T2 within each patient was evaluated using a permutation t-test (via a Monte Carlo simulation of size 10,000, which does not require normality) with a two-sided null hypothesis. We investigated significant changes in the tcMEP amplitudes with desflurane administration. There were constraints on the experimental design, as propofol remains in the body for a long time, and the preoperative time period usually lasts 70 min. Thus, it was impossible to use a classical cross-over design (Wellek and Blettner, 2012) starting with propofol. Given the physiologic details of the medication and local logistics, we used a control group (GROUPSTAN-PROPOFOL). The control group was also used to evaluate potential time-related changes (‘‘anesthetic fade’’) in the MEP amplitudes (Lyon et al., 2005). The differences in mean amplitudes within patients between experimental groups at T1 and T2 (Dw) were analyzed with an unpaired t-test; to evaluate differences between the absolute mean amplitudes, a paired t-test was also used. A type I error rate of 0.05 was used for all tests.
2.4. Statistical analysis
Twenty-one patients were enrolled in the study. Table 2 shows demographic and epidemiologic details of the cohort and some relevant clinical data. Furthermore, Table 3 illustrates the analysis of significant differences of hemodynamic parameters between T1 and T2 in both cohorts. The BIS values were kept in the target range of 30–45 but were significantly lower during propofol administration in GroupDESFLURANE. The mean concentrations of propofol
The results are presented in box plots that show the median, quartiles, minimum and maximum of the data. A two-sided Fisher’s exact test (without distributional assumptions because it was simulation-based) was performed to analyze potential differences between both groups with regard to ASA rating and gender. Additionally, the comparability of physiological variables (MAP,
3. Results
Please cite this article in press as: Malcharek MJ et al. Transcranial motor evoked potentials during anesthesia with desflurane versus propofol – A prospective randomized trial. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.025
4
M.J. Malcharek et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
and remifentanil were not significantly different between GroupDESFLURANE and GroupSTANDARD-PROPOFOL (p = 0.35 and p = 0.079). Fig. 2 clearly illustrates the changes in MEP amplitudes during the switch from desflurane to propofol anesthesia. With regard to the primary goal of the study, we observed that, as shown in Fig. 3a, there were visibly larger mean differences between T1 and T2 within each patient (Dw) in GroupDESFLURANE and GroupSTANDARD-PROPOFOL. The absolute difference in Dw between the experimental groups was 496.75 lV (p = 0.0006; 95% confidence interval: 925.8, 253.4). Although the wide confidence interval suggests that the effect is variable, the magnitude of the effect is large enough that the difference is significant. Note that although the mean amplitude at T1 was different between the two experimental groups (840.1 vs 358.9 lV), the analysis is based on the mean difference of the differences within each patient between T1 and T2 (i.e., ‘each patient serves as his own control’); therefore, differences between groups at the initial time do not influence the statistical analysis. Additionally, the mean amplitudes at T1 were 840.1 and 358.9 lV for GroupSTANDARD-PROPOFOL and GroupDESFLURANE, respectively, and the values at T2 were 793.7 and 902.1 lV, respectively (Fig. 3b). Table 4 provides information about the additional results of the intra- and interindividual comparisons. Because we did not change the suprathreshold level within each patient, we found significant higher mean stimulation intensities at T2 in GroupDESFLURANE (Table 4). 4. Discussion The analysis confirmed results from studies in patients undergoing spinal surgery that indicated that tcMEPs are recordable independent of whether TIVA or inhalational agents were administered (Pechstein et al., 1998; Lo et al., 2004, 2006; Hayashi et al.,
Table 2 Patient characteristics. Values are mean (range) or numbers.
Age (yr) Sex (F/M) ASAa status (II/ III) BMIb (kg m 2) a b
GroupDESFLURANE n = 14
GroupSTANDARD-PROPOFOL n=7
Pvalue
66.3 (54–81) 5/9 5/9
64.4 (50–77) 2/5 1/6
0.681 1.000 0.613
27.4 (23–39)
28.4 (22–39)
0.611
American Society of Anesthesiology. Body mass index.
2008, 2009). However, the primary goal of this study was to investigate the differences in tcMEP amplitudes depending on the choice of the anesthetic in CEA patients. However, according to the literature, different interpretation techniques are described to evaluate tcMEP recordings based on the following factors: (1) alteration in MEP morphology (Quinones-Hinojosa et al., 2005), (2) changes in threshold-level parameters (Calancie et al., 1998), (3) changes in MEP amplitude (Szelényi et al., 2006; Neuloh and Schramm, 2009) and (4) the ‘‘all-or-none’’ interpretation criteria (Zentner et al., 1989; Kothbauer et al., 1998; Dong et al., 2002; Sala et al., 2006). We were analyzing quantitative differences in tcMEPs; therefore, the last technique was unsuitable for this study. Thus, we assessed tcMEPs by analyzing peak-to-peak amplitudes. Because this interpretation criterion has been commonly described in the literature to evaluate the alteration of MEPs by anesthetic agents (Scheufler et al., 2005; Lo et al., 2006; Reinacher et al., 2006; Hayashi et al., 2008, 2009), it seemed to be the most reliable method for quantitative evaluation. However, the use of another criterion to interpret alterations in the tcMEP responses (e.g., changes in threshold level parameters) could provide different results. With regard to the classification of our results in the literature, evaluating whether our study may be comparable to other trials is important. A review of the literature (Wang et al., 2009) showed that groups investigating the anesthesia effects on MEPs predominately used a study protocol in which patients receiving either one or the other agent were compared. In addition, sample sizes in most of the prospective investigations were small, thereby increasing the risk of coincidental results. Our study design was based on an intraindividual comparison; therefore, we could, for the most part, avoid both problems. In addition, the quality of the statistical analysis was improved by calculating the number of patients using an a priori power analysis. Another fundamental problem in comparing various anesthetics is the use of a reliable parameter to evaluate the depth of anesthesia to administer nearly equipotent doses of the different anesthetic agents. For example, Pelosi et al. (2001) found propofol infusion to be superior to isoflurane according to successful MEP recordings (97% versus 61%) during orthopedic spinal surgery. However, whether or how they defined comparable depth of anesthesia in both groups is unclear. In contrast, Chen (2004) investigated the alteration of tcMEPs by comparing propofol and isoflurane anesthesia using different graduations of BIS values. In a prospective trial of 35 patients during spinal surgery, they described a higher rate of successful tcMEP recordings during the administration of propofol anesthesia than isoflurane (i.e., BIS 55–65: 100% versus 58%; BIS 35–45: 100% versus 11.8%). Similarly,
Table 3 Preoperative hemodynamic parameters under general anesthesia during experimental testing. Difference within each patient between T1 and T2 (Dw) by experimental group. Differences were considered statistically significant at a P-value
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Transcranial motor evoked potentials during anesthesia with desflurane versus propofol – A prospective randomized trial M.J. Malcharek a,⇑,1, S. Loeffler a,1, D. Schiefer a, M.A. Manceur b, A. Sablotzki a, J. Gille a, S. Pilge c, G. Schneider c a b c
Department of Anesthesiology, Intensive Care and Pain Therapy, Klinikum St. Georg gGmbH, Leipzig, Germany Division of Biostatistics at Computational Landscape Ecology & Community Ecology, Helmholtz-Centre for Environmental Research GmbH – UFZ, Leipzig, Germany Department of Anesthesiology, Witten/Herdecke University, Helios Clinic Wuppertal, Nordrhein-Westfalen, Germany
a r t i c l e
i n f o
Article history: Accepted 30 November 2014 Available online xxxx Keywords: Motor evoked potentials Intraoperative neurophysiologic monitoring Volatile anesthetics Total intravenous anesthesia Propofol Desflurane
h i g h l i g h t s Transcranial electrical motor evoked potential (tcMEP) amplitudes were significantly larger with pro-
pofol anesthesia than with desflurane anesthesia. Intraindividual differences in tcMEP amplitudes for the desflurane–propofol regimen in the effect
group were significantly smaller. Regardless of the quantitative aspects of the tcMEP amplitudes, recordings of tcMEPs were feasible
under desflurane/remifentanil- and propofol/remifentanil-based general anesthesia.
a b s t r a c t Objective: This study aimed to evaluate differences in transcranial electrical motor evoked potential (tcMEP) amplitudes between desflurane/remifentanil and propofol/remifentanil anesthesia treatment plans in patients without preexisting motor deficits (PMDs) undergoing carotid endarterectomy (CEA). Methods: This prospective trial included 21 patients who were randomly assigned to an effect group (GroupDESFLURANE; n = 14) or a control group (GroupSTANDARD-PROPOFOL; n = 7). tcMEP amplitudes were measured 35 min post-induction (T1) either with desflurane or propofol. Treatment was then changed to propofol in GroupDESFLURANE. After an additional 35 min, the tcMEP amplitudes were reevaluated (T2). Differences in amplitudes (DW) between T1 and T2 were calculated for each patient, and the means of these differences were compared between groups. Results: tcMEPs were recorded in all 21 patients. At T1, the mean amplitude was 840.1 (SD 50.3) lV and 358.9 (SD 74) lV for GroupSTANDARD-PROPOFOL and GroupDESFLURANE, respectively. The absolute mean difference (T1 T2) between groups was 496.75 lV (p = 0.0006). Conclusion: Desflurane reduces the tcMEP amplitude significantly more than propofol in patients without PMDs undergoing CEA. Significance: TcMEPs were recorded in all patients regardless of the anesthesia regimen. In patients with initially small amplitudes, desflurane may limit tcMEP recording because it produces a remarkable amplitude reduction, even in patients without PMDs. Ó 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Clinical Neurophysiology.
1. Introduction ⇑ Corresponding author at: Division of Neuroanesthesia & Intraoperative Neuromonitoring, Department of Anesthesiology, Intensive Care and Pain Therapy, Klinikum St. Georg gGmbH, Delitzscher Str. 141, 04129 Leipzig, Germany. Tel.: +49 341 9094009; fax: +49 341 9092568. E-mail address: [email protected] (M.J. Malcharek). 1 Michael J. Malcharek and Silvana Loeffler have contributed equally, each as first author of this work and in preparing the original manuscript.
The intraoperative monitoring of motor function using direct cortical or transcranial electrical stimulated motor evoked potentials (tcMEPs) is widely used in neurosurgery (Kombos et al., 2003; Neuloh et al., 2004), orthopedic surgery (Langeloo et al., 2003; MacDonald et al., 2003) and vascular surgery (MacDonald and Janusz, 2002; Sloan and Lameson, 2007). Since the introduction of
http://dx.doi.org/10.1016/j.clinph.2014.11.025 1388-2457/Ó 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Clinical Neurophysiology.
Please cite this article in press as: Malcharek MJ et al. Transcranial motor evoked potentials during anesthesia with desflurane versus propofol – A prospective randomized trial. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.025
2
M.J. Malcharek et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
multi-pulse tcMEP stimulation, monitoring of the corticospinal tract (CST) has become more successful in patients under general anesthesia (Kalkman et al., 1992; Kawaguchi et al., 1996; Pechstein et al., 1998). Next to the technical development of complex monitoring systems designed for the modern operating room, the adjustment of certain anesthesia regimens might be responsible for the widespread and successful application of tcMEPs. Total intravenous anesthesia using a propofol infusion has been preferred when MEP monitoring was requested during surgery (Scheufler et al., 2005; Hayashi et al., 2008). However, following the synaptic model, which describes anesthetic action on the nervous system, propofol and volatile anesthetics have similar effects on GABA (Wakasugi et al. 1999; Alkire et al., 2008; Eckle et al., 2013) but may have different actions on cortical or spinal glycine receptors (Hales and Lambert, 1991; Mihic et al., 1997; Grasshoff and Antkowiak, 2004). Thus, numerous studies in the literature have investigated differences in MEP recording under the administration of either TIVA or halogenated agents during neurosurgical or orthopedic procedures (e.g., Pechstein et al., 1998; Pelosi et al., 2001; Scheufler et al., 2005). MEP monitoring has only recently been shown to be a useful adjunct to SSEP monitoring during carotid surgery (Alcantara et al., 2013; Malcharek et al., 2013); therefore, to date, there has been no study in the literature evaluating different anesthetics and their effects on tcMEP recordings in this cohort. Thus, our study is unique because it investigates the effects on tcMEP amplitudes using both TIVA (propofol/remifentanil infusion) and desflurane/remifentanil in patients undergoing carotid surgery. The present trial addresses the following objective: the quantitative evaluation of significant differences in tcMEP amplitudes between two anesthetic approaches, desflurane/remifentanil and the standard propofol/remifentanil protocol. 2. Methods Ethical approval for this study (Registration No: EK-BR-75/13-1) was given by the Research Ethical Committee of the State Chamber of Medicine in Saxony. All patients involved in the study signed an informed consent form regarding the application of the two different anesthesia regimens according to the study protocol. 2.1. Patients We prospectively investigated all patients who were scheduled for elective carotid endarterectomy (CEA) at Klinikum St. Georg gGmbH, Leipzig, Germany between August and December 2013 and who did not have preoperative motor deficits on the contralateral side of the surgery. The patients were evaluated to ensure that they had no preoperative motor deficits by a neurologist and a vascular surgeon. Twenty-one patients were allocated to two groups using a block randomization table (GroupDESFLURANE: 14 patients and GroupSTANDARD-PROPOFOL: 7 patients). The anesthesia plan in GroupDESFLURANE was switched after 35 min from desflurane/ remifentanil to propofol/remifentanil, whereas patients in GroupSTANDARD-PROPOFOL exclusively received a propofol/remifentanil infusion. Measurements of the tcMEPs were performed after 35 (T1) and 70 min (T2) in both groups. Therefore, each patient served as his/her own control. Moreover, GroupSTANDARD-PROPOFOL was defined as the control group because TIVA (propofol/remifentanil) has been routinely performed during the intraoperative monitoring of tcMEPs at our institution. To avoid changes in EP recordings because of positioning or the surgical procedure itself, the study was performed in each patient prior to surgical positioning. Patients with existing motor deficits, acute stroke, allergies to propofol, a history of seizures or epilepsy, a history of propofol
infusion syndrome, cardiac insufficiency (NYHA P III) and implanted pacemakers or defibrillators were excluded. We performed an a priori statistical power analysis in a separate cohort to estimate the number of patients needed for statistical analysis. Because the baseline of MEP amplitudes is highly variable, we included 50 patients undergoing CEA with the standardized propofol/remifentanil regimen. The MEP amplitudes (mean out of 5 recordings per patient) were preoperatively registered to estimate the variability of the measurements. Expert opinion provided a range of expected effect sizes. An a priori sample size of 15 patients was calculated to have 90% power to detect a difference between improved (n = 10) and standard treatment (n = 5) with a level of significance set at p = 0.05 (variance: 200; effect size: 405 lV). 2.2. Anesthesia protocol Patients were premedicated with 3.5–7 mg of oral midazolam one hour prior to transfer to the operating room. The heart rate (HR), mean arterial blood pressure from the radial artery (MAP), inspiratory oxygen fraction (FiO2), end-tidal carbon dioxide (etCO2) and ear temperature were measured in all patients. The target oxygen saturation was over 96%. Normothermia was maintained by forced air warming. All patients received etomidate, a continuous infusion of remifentanil, and rocuronium for the induction of general anesthesia (Fig. 1). Anesthesia was maintained either with desflurane in the event group (GroupDESFLURANE) or with TIVA in the control group (GroupSTANDARD-PROPOFOL), as shown in Fig. 1. According to the study protocol in GroupDESFLURANE, inhaled anesthesia was terminated after 35 min and switched to TIVA. An initial bolus of propofol (1 mg/kg) was administered prior to the continuous infusion of propofol. The initial remifentanil infusion rate ranged from 0.2 to 0.6 lg/ kg/min based on clinically relevant bradycardia (target HR > 40/ min). The infusion rate was then not changed unless medical reasons required a dose adjustment (e.g., bradycardia). Given that the dose of the remifentanil infusion was assigned first, the doses of propofol and desflurane were adjusted based on a defined target range (30–45) of the bispectral index (BIS; BIS-Module for Philips M1034A with BISx, frontal montage, Aspect Medical Systems Inc., Framingham, MA, USA). To provide comparable depths of anesthesia, BIS values were assumed to be similar for both regimens within this range. The propofol infusion was adjusted so that it did not exceed 6 mg/kg/h, and a maximum concentration of 4% of desflurane was kept at an age-adjusted minimal alveolar concentration (MAC; Amingo–Avance, Datex-Ohmeda Inc., Madison, WI, USA) of 0.5–0.6. The concentration of desflurane or propofol was reduced when the BIS values decreased 2: 1 mg/kg) in both groups. According to the study protocol, rocuronium was not redosed during the experimental period. The MAP was increased by administering akrinorÒ (1 ampule: cafedrine-HCl 200 mg and theoadrenaline-HCl 10 mg; ASTA Medica AWD GmbH, Dresden, Germany) or norepinephrine in case of a decrease by more than 20% of the preoperative values. The ventilator was set to maintain normocapnia, and the target SpO2 was P96%. 2.3. Intraoperative neurophysiologic monitoring All patients underwent multimodal EP monitoring, which is the standard monitoring technique for CEA at our institution.
Please cite this article in press as: Malcharek MJ et al. Transcranial motor evoked potentials during anesthesia with desflurane versus propofol – A prospective randomized trial. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.025
3
M.J. Malcharek et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
Remifentanil infusion rate: 0.2-0.6 µg/kg/min G
DESFLURANE
Propofol
(n=14)
MAC : 0.5-0.6 Target BIS2: 30-45
Initial Bolus: 1 mg/kg Infusion rate: 4-6 mg/kg/h Target BIS2: 30-45
Propofol
Propofol
Initial Bolus: 1 mg/kg Infusion rate: 4-6 mg/kg/h Target BIS2: 30-45
Infusion rate: 4-6 mg/kg/h Target BIS2: 30-45
1
G STANDARD-PROPOFOL (n=7)
35 min
70 min
T1
T2 3
Induction
Reversal of NMB
(etomidate 0.2 mg/kg, rocuronium 0.6 mg/kg)
(sugammadex 1-2 mg/kg)
1
Minimal alveolar concentration Bispectral index 3 Neuromuscular blockade 2
Fig. 1. Scheme of the study protocol to time the administration of anesthetic agents.
Table 1 Modalities for the assessment of tcMEP monitoring. Standardized institutional setting during the investigation. tcMEPa stimulation Monitoring system Electrodes Montage of electrodes Specifications
a
tcMEPa recording
ISIS IOM Neuromonitoring system (Inomed Medizintechnik GmbH) Corkscrew electrodes (MNCS0610DS, Spes Medica S.r.l., Italy) Needle electrodes (SDN 530620, Inomed Medizintechnik GmbH, Germany) C4 + 1 cm C3 + 1 cm/C3 + 1 cm C4 + 1 cm Bilateral abductor digiti minimi muscle (ADM) (according to the 10/20-system) Band pass filter: 5–2500 - Constant current (upper voltage limit: 300 V) - Range of intensity needed for suprathreshold stimulation: 35–100 mA - Train: 5; Interstimulus interval: 2 ms - Impulse: 0.5 ms, monophasic - Stimulation rate: 2 Hz
Transcranially electrical stimulated motor evoked potential.
Multimodal EP monitoring includes median nerve somatosensory evoked potentials (mSSEPs), tibial nerve somatosensory evoked potentials (tSSEPs) and tcMEPs from the upper extremities. The monitoring of SSEPs was not subject to investigation in this trial. There was no pre-anesthesia referential assessment of tcMEPs; therefore, an initial stimulation threshold was set at 5 mA above the first intensity that produced stable and reproducible MEP recordings (just before the 1st test interval: T1). This threshold intensity was not changed during the experimental period. Concerning the recording of tcMEPs during the two test intervals (T1 and T2), Table 1 shows the parameters of the tcMEP measurement. The mean amplitude was analyzed out of five single measurements (inter train interval: 15 s) per test. According to the study protocol, only responses from the side contralateral to the surgical site were analyzed because the results might be potentially more important at the surgically relevant side. The tcMEP monitoring was performed by one of the three neuroanesthesiologists who were involved in the study. All were trained and certified in the assessment of intraoperative neurophysiologic monitoring (‘‘Curriculum Neuromonitoring’’: educational program of the German Society of Anaesthesiology and Intensive Care Medicine).
temperature, SpO2, FiO2 and etCO2) and the BIS values at T1 and T2 within each patient was evaluated using a permutation t-test (via a Monte Carlo simulation of size 10,000, which does not require normality) with a two-sided null hypothesis. We investigated significant changes in the tcMEP amplitudes with desflurane administration. There were constraints on the experimental design, as propofol remains in the body for a long time, and the preoperative time period usually lasts 70 min. Thus, it was impossible to use a classical cross-over design (Wellek and Blettner, 2012) starting with propofol. Given the physiologic details of the medication and local logistics, we used a control group (GROUPSTAN-PROPOFOL). The control group was also used to evaluate potential time-related changes (‘‘anesthetic fade’’) in the MEP amplitudes (Lyon et al., 2005). The differences in mean amplitudes within patients between experimental groups at T1 and T2 (Dw) were analyzed with an unpaired t-test; to evaluate differences between the absolute mean amplitudes, a paired t-test was also used. A type I error rate of 0.05 was used for all tests.
2.4. Statistical analysis
Twenty-one patients were enrolled in the study. Table 2 shows demographic and epidemiologic details of the cohort and some relevant clinical data. Furthermore, Table 3 illustrates the analysis of significant differences of hemodynamic parameters between T1 and T2 in both cohorts. The BIS values were kept in the target range of 30–45 but were significantly lower during propofol administration in GroupDESFLURANE. The mean concentrations of propofol
The results are presented in box plots that show the median, quartiles, minimum and maximum of the data. A two-sided Fisher’s exact test (without distributional assumptions because it was simulation-based) was performed to analyze potential differences between both groups with regard to ASA rating and gender. Additionally, the comparability of physiological variables (MAP,
3. Results
Please cite this article in press as: Malcharek MJ et al. Transcranial motor evoked potentials during anesthesia with desflurane versus propofol – A prospective randomized trial. Clin Neurophysiol (2014), http://dx.doi.org/10.1016/j.clinph.2014.11.025
4
M.J. Malcharek et al. / Clinical Neurophysiology xxx (2014) xxx–xxx
and remifentanil were not significantly different between GroupDESFLURANE and GroupSTANDARD-PROPOFOL (p = 0.35 and p = 0.079). Fig. 2 clearly illustrates the changes in MEP amplitudes during the switch from desflurane to propofol anesthesia. With regard to the primary goal of the study, we observed that, as shown in Fig. 3a, there were visibly larger mean differences between T1 and T2 within each patient (Dw) in GroupDESFLURANE and GroupSTANDARD-PROPOFOL. The absolute difference in Dw between the experimental groups was 496.75 lV (p = 0.0006; 95% confidence interval: 925.8, 253.4). Although the wide confidence interval suggests that the effect is variable, the magnitude of the effect is large enough that the difference is significant. Note that although the mean amplitude at T1 was different between the two experimental groups (840.1 vs 358.9 lV), the analysis is based on the mean difference of the differences within each patient between T1 and T2 (i.e., ‘each patient serves as his own control’); therefore, differences between groups at the initial time do not influence the statistical analysis. Additionally, the mean amplitudes at T1 were 840.1 and 358.9 lV for GroupSTANDARD-PROPOFOL and GroupDESFLURANE, respectively, and the values at T2 were 793.7 and 902.1 lV, respectively (Fig. 3b). Table 4 provides information about the additional results of the intra- and interindividual comparisons. Because we did not change the suprathreshold level within each patient, we found significant higher mean stimulation intensities at T2 in GroupDESFLURANE (Table 4). 4. Discussion The analysis confirmed results from studies in patients undergoing spinal surgery that indicated that tcMEPs are recordable independent of whether TIVA or inhalational agents were administered (Pechstein et al., 1998; Lo et al., 2004, 2006; Hayashi et al.,
Table 2 Patient characteristics. Values are mean (range) or numbers.
Age (yr) Sex (F/M) ASAa status (II/ III) BMIb (kg m 2) a b
GroupDESFLURANE n = 14
GroupSTANDARD-PROPOFOL n=7
Pvalue
66.3 (54–81) 5/9 5/9
64.4 (50–77) 2/5 1/6
0.681 1.000 0.613
27.4 (23–39)
28.4 (22–39)
0.611
American Society of Anesthesiology. Body mass index.
2008, 2009). However, the primary goal of this study was to investigate the differences in tcMEP amplitudes depending on the choice of the anesthetic in CEA patients. However, according to the literature, different interpretation techniques are described to evaluate tcMEP recordings based on the following factors: (1) alteration in MEP morphology (Quinones-Hinojosa et al., 2005), (2) changes in threshold-level parameters (Calancie et al., 1998), (3) changes in MEP amplitude (Szelényi et al., 2006; Neuloh and Schramm, 2009) and (4) the ‘‘all-or-none’’ interpretation criteria (Zentner et al., 1989; Kothbauer et al., 1998; Dong et al., 2002; Sala et al., 2006). We were analyzing quantitative differences in tcMEPs; therefore, the last technique was unsuitable for this study. Thus, we assessed tcMEPs by analyzing peak-to-peak amplitudes. Because this interpretation criterion has been commonly described in the literature to evaluate the alteration of MEPs by anesthetic agents (Scheufler et al., 2005; Lo et al., 2006; Reinacher et al., 2006; Hayashi et al., 2008, 2009), it seemed to be the most reliable method for quantitative evaluation. However, the use of another criterion to interpret alterations in the tcMEP responses (e.g., changes in threshold level parameters) could provide different results. With regard to the classification of our results in the literature, evaluating whether our study may be comparable to other trials is important. A review of the literature (Wang et al., 2009) showed that groups investigating the anesthesia effects on MEPs predominately used a study protocol in which patients receiving either one or the other agent were compared. In addition, sample sizes in most of the prospective investigations were small, thereby increasing the risk of coincidental results. Our study design was based on an intraindividual comparison; therefore, we could, for the most part, avoid both problems. In addition, the quality of the statistical analysis was improved by calculating the number of patients using an a priori power analysis. Another fundamental problem in comparing various anesthetics is the use of a reliable parameter to evaluate the depth of anesthesia to administer nearly equipotent doses of the different anesthetic agents. For example, Pelosi et al. (2001) found propofol infusion to be superior to isoflurane according to successful MEP recordings (97% versus 61%) during orthopedic spinal surgery. However, whether or how they defined comparable depth of anesthesia in both groups is unclear. In contrast, Chen (2004) investigated the alteration of tcMEPs by comparing propofol and isoflurane anesthesia using different graduations of BIS values. In a prospective trial of 35 patients during spinal surgery, they described a higher rate of successful tcMEP recordings during the administration of propofol anesthesia than isoflurane (i.e., BIS 55–65: 100% versus 58%; BIS 35–45: 100% versus 11.8%). Similarly,
Table 3 Preoperative hemodynamic parameters under general anesthesia during experimental testing. Difference within each patient between T1 and T2 (Dw) by experimental group. Differences were considered statistically significant at a P-value
