Accepted Manuscript Title: Microelectrode recording (MER) findings during sleep-awake anesthesia using dexmedetomidine in deep brain stimulation surgery for Parkinson’s disease Author: Woo-Keun Kwon Jong Hyun Kim Ji-Hye Lee Byung-Gun Lim Il-ok Lee Seong Beom Koh Taek Hyun Kwon PII: DOI: Reference:
S0303-8467(16)30045-2 http://dx.doi.org/doi:10.1016/j.clineuro.2016.02.005 CLINEU 4308
To appear in:
Clinical Neurology and Neurosurgery
Received date: Revised date: Accepted date:
30-12-2015 4-2-2016 5-2-2016
Please cite this article as: Kwon Woo-Keun, Kim Jong Hyun, Lee Ji-Hye, Lim Byung-Gun, Lee Il-ok, Koh Seong Beom, Kwon Taek Hyun.Microelectrode recording (MER) findings during sleep-awake anesthesia using dexmedetomidine in deep brain stimulation surgery for Parkinson’s disease.Clinical Neurology and Neurosurgery http://dx.doi.org/10.1016/j.clineuro.2016.02.005 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Microelectrode recording (MER) findings during sleep-awake anesthesia using dexmedetomidine in deep brain stimulation surgery for Parkinson’s disease Woo-Keun Kwon, M.D.1, Jong Hyun Kim M.D., Ph.D.1* [email protected], Ji-Hye Lee M.D.1, Byung-Gun Lim M.D., Ph.D.2, Il-ok Lee M.D., Ph.D.2, Seong Beom Koh M.D., Ph.D.3, Taek Hyun Kwon, M.D., Ph.D.1 1
Department of Neurosurgery, Korea University Guro Hospital, Korea University College of
Medicine, Seoul, Korea 2
Department of Pain and Anesthesiology, Korea University Guro Hospital, Korea University
College of Medicine, Seoul, Korea 3
Department of Neurology, Korea University Guro Hospital, Korea University College of
Medicine, Seoul, Korea *
Corresponding author at: Department of Neurosurgery, Korea University Guro Hospital, 148
Gurodongro, Gurogu, 08308, Seoul, Korea. Tel: 82) 2-2626-3098; Fax: 82) 2-863-1684.
1
Highlights
GPi and STN DBS for intractable IPD under DEX anesthesia was done
DEX provided excellent surgical conditions, and hemodynamic stability
DEX influences the firing rate of subcortical nuclei during DBS
However, it did not affect the MER interpretation and of target localization
2
Abstract Objective The preferred choice of anesthesia for deep brain stimulation (DBS) has been local anesthesia due to the need of patients’ cooperation during the procedure, and concern on the interference of sedatives on microelectrode recording (MER) results. However, local anesthesia during the whole procedure may be impossible in some patients due to uncontrolled anxiety, fear, delirium or exhaustion. Therefore, sedative drugs have been used for DBS, but findings of MER during the procedures have not been reported in detail, especially in the globus pallidus internus (GPi). We introduce our experience using ‘asleepawake’ technique by dexmedetomidine (DEX) anesthesia with MER findings during DBS in idiopathic Parkinson’s disease (IPD) patients. Patients and Methods Data from 14 different subcortical nuclei from 8 consecutive IPD patients whom had DBS at the GPi (6 patients) and subthalamic nucleus (STN) (2 patients) were retrospectively reviewed. We used continuous DEX and intermittent small boluses of propofol during the painful procedure (‘asleep phase’), accompanied with continuous intraoperative monitorings of bispectral index (BIS) and modified observer’s assessment of sedation (MOAA/S). Then sedatives were discontinued during MER recording (‘awake phase’). Characteristic findings and firing rates of neurons were analyzed and compared to those from other 6 patients who underwent surgery under local anesthesia. Results All patients were satisfactorily sedated using this technique without any respiratory or hemodynamic complications. Characteristics of spike activities of each nucleus were inspected and analyzed quantitatively. We could inspect changes of spike activities according
3
to level of patients’ consciousness in some cases, but the localizing value was good to decide the target in all cases. Firing rates of group whom sedatives were given during asleep phase (‘sedatives’) were significantly lower than those of group under local anesthesia (‘no sedative’). Intraoperative length of target nuclei, postoperative imaging and postoperative changes of UPDRS III score indicated satisfactory outcome. Conclusion We concluded that though MER findings may change during DEX-based monitored ‘sleep-awake’ anesthesia, it did not affect the results of target localization for the clinical purpose. However, it should be considered that use of sedatives before MER could result in changes of firing rate and pattern depending on the patient’s state of consciousness.
KEYWORDS: dexmedetomidine; deep brain stimulation; microelectrode recording; Parkinson’s disease
4
Introduction Deep brain stimulation (DBS) has been widely performed successfully for patients with various intractable movement disorders such as idiopathic Parkinson’s disease (IPD), dystonias or tremors. Previously, DBS has been performed mostly under local anesthesia considering the quality of microelectrode recording (MER) results and the necessity of patient’s cooperation during DBS. However, the duration of DBS intervention, especially in bilateral implantations at the same day, is sometimes too long and some patients suffer from intolerable exhaustion, anxiety or fear. Main concerns of using sedatives during DBS are its possible effect on the quality of MER and respiratory function. Gamma-aminobutyric acidergic (GABAergic) sedatives such as propofol have been used most commonly for DBS, but there are concerns on its interference on MER[1] or its fatal complications such as respiratory depression[2]. Dexmedetomidine (DEX) is a highly selective alpha-2-adrenergic receptor agonist which is a well known alternative sedative, which shows dose-dependent rapid action without respiratory depression[3-5]. However, undesirable effects such as changes of neuronal activities in the subthalamic nucleus (STN) during surgery[6], or prolonged impairment of consciousness[7], have also been reported. There has been a quantitative study of neuronal spikes during the STN DBS using DEX-based anesthesia[8]. In this study, the authors even recommended not to use high-dose DEX during the STN DBS[8] based on their results that spike patterns were changed. In this paper we report our experience of MER during ‘asleep-awake’ anesthesia using DEX-based sedative technique in 8 consecutive IPD patients, mainly from GPi DBS. We stopped the infusion of DEX before the MER and observed the characteristics of neuronal spikes if they are acceptable for the clinical purpose to decide the optimal target. In addition, we analyzed the firing rate of each nucleus and tried to correlate them with the BIS
5
of ‘awake-phase’. To our knowledge, there has been no quantitative analysis using neuronal activities during GPi DBS of IPD patient using DEX-based anesthesia.
Methods Patients A total of eight consecutive patients whom underwent DBS for medically intractable IPD under DEX-based anesthesia between January 2013 and July 2015 in our institute were retrospectively studied. Neurologists and neurosurgeons specializing in movement disorders diagnosed all patients. A multi-department team including neurosurgeons, neurologists and psychiatrists made decision of surgical candidates and target nuclei for DBS, after considering the patients’ main symptoms, unified Parkinson’s disease rating scale (UPDRS) scores and the results of psychiatric evaluations. This study was approved by the Institutional Review Board. Surgical procedure and anesthesia protocol We followed the standard DBS surgical protocol as described in previously reported articles[9]. All operations were performed by a single experienced surgeon. IPD patients for STN DBS had been off all dopaminergic medications for at least 12 hours before the surgery. For patients whom underwent Globus Pallidus internus (GPi) DBS for severe dyskinesia, their dopaminergic medications were administrated until the operation day as usual to verify the anti-dyskinetic effect of test stimulation during the surgery. Magnetic resonance images (MRI) were taken in the operation day morning after placement of the Leksell stereotactic head frame (Elekta Instrument, Stockholm, Sweden). The images were imported into the stereotactic planning software (FramelinkTM, Medtronic, US), and then coordinates of the initial target were decided by the neurosurgeon.
6
We used standard ‘asleep - awake’ anesthetic technique. Patients were sedated during the painful procedure (‘asleep-phase’) using loading and continuous infusion of DEX. Then sedatives were discontinued and patient was awakened for MER and test stimulation (‘awake-phase’).
The neuronal data from ‘awake-phase’ were used for analysis (‘sedatives’
group). For the quantitative analysis to compare firing rates and patterns, another group of patients (total 6: 2 GPi and 4 STN) whom underwent surgery under local anesthetics without any sedatives (‘no sedative’ group) were included. In addition, we used a data from two patients (patient 1 and 7) whom underwent surgery with one site with sedatives and the other site without, for within-patient analysis. All patients were monitored by an experienced anesthesiologist during the whole procedures. Initially DEX was given by a loading dose of 0.5 − 1 µg/kg over 10 minutes, followed by a maintenance dose of 0.1 − 0.5 µg/kg/hr infusion. If the patient was not sedated enough during the painful procedure, we increased the infusion rate of DEX or used small boluses of propofol (10 - 20 mg) for rapid effect. Low dose of remifentanil (REM) was also given (0.01 − 0.1 µg/kg/min) continuously with an adjustment as needed to control pain in 4 out of 8 cases. Continuous infusion of DEX and REM were stopped after the placement of burr holes. In case of bilateral implantation, DEX infusion was restarted before the intervention of the second side. Detailed control of the amount and doses of sedatives were done by the anesthesiologist, based on the patients’ intraoperative conditions. Additionally, bispectral index (BIS) monitoring system (Aspect Medical System, US) was applied at the patient’s forehead to monitor the depth of sedation. The maintenance dose of DEX was adjusted aiming the target BIS lower than 80. Modified observer’s assessment of alertness/sedation (MOAA/S) scale was also checked by the anesthesiologist to evaluate the level of patients’ consciousness.
7
Intraoperative Microelectrode recordings procedure and postoperative evaluation For the physiological mapping, MER was started from 15 millimeters above the estimated target. The tungsten/stainless steel microelectrode (model FC 1003, FHC Inc.) was advanced carefully, and stopped whenever good and consistent target neuronal activities appeared. Characteristic spike activities representing each target nuclei were displayed on the Leadpoint system (Medtronic). At the same time, the amplified signal was sent out to the Analog Output Board (Medtronic) installed on the system and separately digitized by an analogue-to-digital converter (ADC) (Micro1401, CED, UK). The sampling rate was 50,000 Hz for spike activities and digitized data was stored on a computer for the off-line analysis. After finishing of physiologic mapping according to intraoperative MER findings and evoked responses, testing macro-stimulation effect was performed at depths by more than two experienced movement disorders specialists. Once an agreement was made that the electrode was placed at the best effective target site, the therapeutic electrode (Model 3389 or 3387, Medtronic) was implanted. The final position of the electrode was verified postoperatively once again by co-registration of post-operative thin sliced computed tomography (CT) with preoperative MRI. Postoperative clinical effectiveness was evaluated by comparison of preand postoperative UPDRS III score, which represents the patients’ clinical status related to motor function. We also compared preoperative OFF-medication state and postoperative OFF-medication/DBS-ON state at least 6 months after the surgery. Spike sorting and analyses Once the spontaneous neuronal activity was found during MER procedure, advancement of microelectrode was stopped and stayed at least 20 seconds to record stable spike activities, and then was stored for off-line analyses. Digitally recorded spike activities were sorted using Spike2 software (CED, Cambridge, UK) by template matching of action potential waveforms[9]. Spikes with high signal to noise ratio at least two times of 8
background activities were chosen for analysis. Analyses of spike characteristics of firing rates were carried out using customized software programmed by MATLAB. Firing rate was calculated by number of spikes per duration of recording. Results Patients and target nuclei A total of eight IPD patients (4 males, 4 females) with average age of 62.4 (range ; 54-73) were involved in this study. Six IPD patients underwent GPi DBS (two unilateral, four bilateral) while the other two underwent bilateral STN DBS. GPi was chosen as a target for IPD patients with decreased cognitive function or for those who had psychiatric problems. One patient (patient-4) had major depressive disorder with severe anxiety. Preoperative demographic information and target nuclei of the patients are presented on Table1. Anesthesia After initial loading dose of DEX was given, most patients became sedated efficiently. The state of sedation could be monitored continuously by an anesthesiologist with MOAA/S scale and BIS. BIS monitoring of a representative patient is shown on Figure 1. The BIS score was recorded every minute on the device in 3 patients, while it was recorded intermittently by an anesthesiologists on the chart in other cases. The mean BIS of all patients during the asleep-phase was 71.9 and was 81.9 during the awake-phase. Generally, the BIS value represented the patient’s level of consciousness well, but some fluctuation was found occasionally during the procedure because of some artifact made by the patient’s movement. The mean loading and maintenance doses of DEX were 0.94 µg/kg and 0.46 µg/kg/hr. The mean (+/- standard deviation) infusion time of DEX was 67.5 (+/- 27.1) min. Small boluses of propofol for rapid effect were given intermittently in all patients during the asleep-phase as described in Table 2. Low dose of continuous infusion of REM (0.01 - 0.1 µg/kg /hr; mean 0.045 µg/kg/min) was given to 4 patients (patient 2,4-6). Hemodynamic parameters were 9
stable throughout the procedure among all 8 patients. After the initial loading dose, both blood pressure and heart rate were slightly decreased during the ‘asleep phase’, but gradually increased and recovered baseline levels during the ‘awake phase’. There were no significant changes of respiration rate and end-tidal PCO2 during the whole procedure. More than half of the patients were fully awaken within 20 minutes (10 out of 14 sides) after stopping DEX infusion. The mean awakening time to fully cooperative state (MOAA/S score 5) after stopping DEX was 22.3±16.1 minutes. We observed that not all patients were recovered to MOAA/S score 5 and cooperative during MER. Patient 1 was uncooperative during MER procedure but awakened 40 minutes after the discontinuation of DEX. This patient underwent unilateral surgery previously under local anesthesia, and consequent trial of contralateral implantation at the same day was failed because she was too delirious to continue the procedure. Patient 3 was well cooperative during the first side but was only able to follow simple commands during the 2nd side, though which did not affect the intervention.
Microelectrode recordings (MER) and neuronal activities MER was stably recorded after the discontinuation of DEX and we could observe characteristic patterns of neuronal activities from the globus pallidus and subthalamic nucleus (Fig.2). The mean number of electrode trajectories per side was 1.3. The mean trajectory length was about 5 mm (GPi: 5.06 mm, STN: 5.55 mm). Using the results of characteristic spike activities, each subcortical nucleus could be clearly localized. However, we could observe subtle differences of neuronal activities according to the state of sedation. Because some patient was not fully awake after DEX discontinuation, the state of sedation was different among patient during MER. We observed that background activity and spike amplitudes seemed lower under the deeper sedation than those under light sedation in some cases (Fig.3). A total of 125 single neuronal activities (GPe 27, GPi 79, STN 19 cells) could 10
be sorted and analyzed. The mean firing rate of each nucleus were GPe: 33.5, GPi: 40.0, STN: 22.5 (spike/sec) respectively. These results were compared with those of patients whom underwent surgery with local anesthesia (‘no sedative’ group). A total of 64 neurons were available for the analysis (GPe 23, GPi 18, STN 23 cells). The mean firing rate of ‘no sedative’ group were GPe: 52.9, GPi: 60.75, STN:35.2 (spikes/sec), showed significantly higher firing rates compared to ‘sedatives’ group in all nuclei (Table 3). Because we had two patients (patient 1,7) whom underwent GPi-DBS surgery with two different conditions, within-patient comparison was performed. A total of 18 (7 sedatives, 11 no sedative) GPi cells from patient 1, and 13 (6 sedatives, 7 no sedative) GPi cells from patient 7 were available for the analysis. Patient 1 was deeply sedated after discontinuation of DEX, while patient 7 was quickly awakened. The mean BIS during awake-phase of patient 1 and 7 were 57.3 and 93 respectively. We observed GPi cells of patient 1 at ‘awake-phase’, however this patient was still deeply sedated even after DEX discontinuation. The results showed more pauses with lower firing rate compared with ‘no sedative’ condition in this patient (Fig.4 A and B). On the other hand, GPi cells of patient 7 of ‘awake-phase’ showed similar firing rate compared with ‘no sedative’ state (Fig.4 B). Next, we investigated the correlation between firing rate and BIS. As conducting a retrospective study, BIS data was recorded continuously in only three patients, so mean firing rate and mean BIS during the procedure was used for the analysis in other patients. We could observe a significant correlation between firing rate and BIS in one patient but not in other two patients from continuous data. However, there was a certain tendency of positive correlation between mean firing rate and mean BIS using available data from all patients (Fig. 5). Clinical improvements comparing UPDRS III score before and after the surgery could be evaluated in 5 patients. Three patients could not be evaluated as two were referred back to another center, and one patient was not tested postoperatively with OFF-medication. These 11
three patients reported their state as ‘much better’ after the surgery. Evaluation of other 5 patients showed significant improvement with a mean 49.8% decrease of UPDRS III score.
Discussion The present study showed that monitored ‘asleep-awake’ technique using Dex-based anesthetics was safe and reliable method for DBS for different subcortical nuclei. The quality of MER was good enough to decide the target nuclei and test stimulation was successfully performed with patient’s cooperation. Permanent electrodes were implanted at the sites where macrostimulation showed significant improvements without side effects. Proper location of implantation were also later confirmed by UPDRS motor scores comparing preoperative off medication state vs postoperative off-medication/on-DBS state. Mean length of target nucleus about 5 mm also suggested satisfactory target exploration. Most centers withhold anti-parkinsonism medications at least 12 hours before the surgery to render the PD patients in the ‘off’ state for intraoperative neurological testing. Without medication, patients may have pain, anxiety or exhaustion during the surgery if they are not properly managed. We have experienced that some PD patients were too exhausted to be cooperative especially during the second side operation of bilateral surgery under the local anesthesia. Patient-1 of the current study had developed delirium during second side operation after the first side. It was impossible to implant permanent electrode on the second side because the patient was not cooperative. Another patient (patient-4) had severe anxiety and painful off-dystonia symptom without anti-parkinsonism medications. Both patients underwent the surgery successfully using the current protocol without any difficulty. ‘Asleep-awake-asleep’ or ‘asleep-awake’ anesthetic method allows enough sedation during painful procedure followed by rapid emergence for intraoperative brain mapping procedures. Propofol has been used for craniotomy requiring intraoperative awake functional 12
brain mapping [10], and MER for therapeutic electrode insertion during DBS on different target sites [5, 11, 12]. The median interval wakeup time of propofol and remifentanil was 9 minutes in brain tumor, vascular malformation and epilepsy patients [10], but more delayed up to 36 minutes when used to dystonia patients undergoing DBS surgery[13]. We also observed that wakeup time was various among patients. Respiratory complications are the most feared during DBS surgery because when it happens it may hard to keep the airway due to fixed patient’s head via the stereotactic frame to the operating table [14]. DEX is a highly selective alpha-2-adrenoreceptor agonist, and it produces dose-dependent sedation without respiratory depression [15]. Its sedative and anxiolytic effects seem to through activation of locus ceruleus that modulates arousal, sleep and anxiety[15]. This non-cortical site of action may result in a state of ‘cooperative-sedation’[15]. Continuous infusion of DEX has been used for DBS in several studies successfully [3, 4, 16]. Previous report also showed that adequate sedation using DEX or propofol were equally satisfactory after DBS [17]. The protocols of these studies were different from ours. We used continuous infusion of DEX with intermittent small dose of propofol in case the patient was awake and complained discomfort during the painful procedure (asleep-phase). Because propofol achieves sedation more rapidly than DEX [17], it was faster using propofol to adjust the level of sedation than changing the infusion rate of DEX. Using DEX plus small dose of propofol has been successfully used in awake surgery for cognitive testing in other study[18]. We used both BIS and MOAA/S scale to monitor and control the level of sedation during the ‘asleep-phase’. As shown in Fig 1, BIS values were generally decreased during the ‘asleep-phase’ and increased in ‘awake- phase’ as expected. However, using only BIS may not correctly represent the patient’s level of sedation when using continuous DEX infusion. There was a report that BIS monitoring during sedation with DEX and REM showed variable
13
and fluctuating results[19]. Therefore, the combined use of both BIS and sedative scales should be used for complementary data for evaluating the patient’s response to sedation[20]. There are concerns about the changes of neuronal activities while using propofol or DEX[1, 6, 12]. Infusion of sedatives during the MER procedure showed various effects on neuronal activities of the STN. Administration of propofol decreased STN neuronal activity significantly, but returned to the baseline shortly after the discontinuation [1]. Another report showed decrease of activities in the STN under DEX infusion, and recommended the discontinuation after placement of burr holes to keep the patient awake during MER[21]. One recent paper showed increased firing rate but decreased burst index of the STN during DEX infusion and concluded that high-dose DEX infusion should be avoided during the STN DBS [8]. These reports might raise the idea of using ‘asleep-awake’ technique, used as in this study for better MER results. Interestingly, our results also showed even after the discontinuation of DEX infusion, firing rate and patterns could be changed according to the state of patient’s consciousness. This change of neuronal activities after discontinuation of sedatives has not been reported before. However, these changes do not necessarily mean poor localization of target nuclei for the clinical purpose as excellent positioning has been achieved as shown in this study. Our study has several limitations. Because we used ‘asleep-awake’ technique and discontinued DEX before MER, it is not clear that neuronal activities were actually affected by DEX infusion. Observed differences of neuronal activity could be due to difference in individual patients and not related to DEX at all. To study the direct effect of DEX or other sedatives on neurons, one should exam the neuronal change before and after the infusion in the same cell, which is not easy during the surgical procedure, then pair-wise comparison should be performed [1]. Previous retrospective studies presented observational data [6] or used different cells with different conditions [8, 12] as in this study. Within-patient 14
observation of this study suggest possible effects of patient’s sedation state (Fig.4). However, due to the limitation of retrospective study, we could not present exact correlation between BIS and neuronal firing because many BIS data were recorded intermittently rather than continuously. Our results may seem somewhat different compared with the recent paper of Krishna et al. [8], which showed increased firing rate of the STN after DEX infusion. The reason might be due to the different condition of neurons. This study used neurons after discontinuation of sedatives, but previous study used neurons under DEX infusion continued or titrated depending on patients’ condition. In addition, previous study investigated STN neurons, but this study used mainly GPi neurons and relatively small numbers of STN neurons. Results of firing rates of ‘no sedative’ group of this study were consistent with previous studies [8, 12, 22, 23], considering STN was recorded under OFF-medication and GPi under ON-medication state. We observed the tendency of more pauses in GPi and smaller amplitudes in STN when the patient was in the deeper sedation state (Fig 2.). Within patient analysis also showed similar results (Fig 4). Those changes might have been because of the remaining effect of sedatives or sleep induced changes[24]. We observed patients often became lightly sedated again (see Fig.1) even after discontinuation of DEX if there was no external stimulation. This prolonged anxiolytic effect was noticed in other study as well [6]. Changes of EEG pattern similar to stage II sleep was noted after DEX infusion [25], which showed increased low frequency activity as seen in our cases (Fig 3, left column). Subcortical nuclei are directly or indirectly connected to the cortex, thus sedation- or sleep-induced changes of neuronal activities may not be surprising. However, it is not possible to conclude whether the changes of neuronal activities were due to remaining effect of DEX or sleepinduced in this retrospective study.
15
Conclusion In this clinical research, we found that the ‘sleep-awake’ anesthetic technique using DEX does have a tendency to decrease the firing rate of subcortical nucleus in DBS for IPD patients even after discontinuation of the sedative. Those findings may be related to the degree of sedation due to the remaining effect of sedatives or sleep-induced changes. However, even though there were some changes neuronal activities of MER, we concluded that this anesthetic technique provided safe and reliable results enough to decide the optimal target for DBS.
Conflict of interest This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) founded by the Ministry of Education (NRF2010-0005420)
16
References [1] Raz A, Eimerl D, Zaidel A, Bergman H, Israel Z. Propofol decreases neuronal population spiking activity in the subthalamic nucleus of Parkinsonian patients. Anesth Analg 2010;111:1285-1289. [2] Mirski MA, Hemstreet MK. Critical care sedation for neuroscience patients. J Neurol Sci 2007;261:16-34. [3] Rozet I, Muangman S, Vavilala MS, Lee LA, Souter MJ, Domino KJ, Slimp JC, Goodkin R, Lam AM. Clinical experience with dexmedetomidine for implantation of deep brain stimulators in Parkinson's disease. Anesth Analg 2006;103:1224-1228. [4] Sassi M, Zekaj E, Grotta A, Pollini A, Pellanda A, Borroni M, Pacchetti C, Menghetti C, Porta M, Servello D. Safety in the use of dexmedetomidine (precedex) for deep brain stimulation surgery: our experience in 23 randomized patients. Neuromodulation 2013;16:401-406; discussion 406. [5] Venkatraghavan L, Luciano M, Manninen P. Review article: anesthetic management of patients undergoing deep brain stimulator insertion. Anesth Analg 2010;110:1138-1145. [6] Elias WJ, Durieux ME, Huss D, Frysinger RC. Dexmedetomidine and arousal affect subthalamic neurons. Mov Disord 2008;23:1317-1320. [7] Bustillo MA, Lazar RM, Finck AD, Fitzsimmons B, Berman MF, Pile-Spellman J, Heyer EJ. Dexmedetomidine may impair cognitive testing during endovascular embolization of cerebral arteriovenous malformations: a retrospective case report series. Journal of neurosurgical anesthesiology 2002;14:209-212. [8] Krishna V, Elias G, Sammartino F, Basha D, King NK, Fasano A, Munhoz R, Kalia SK, Hodaie M, Venkatraghavan L, Lozano AM, Hutchison WD. The effect of dexmedetomidine on the firing properties of STN neurons in Parkinson's disease. The European journal of neuroscience 2015;42:2070-2077. 17
[9] Kim JH, Ohara S, Lenz FA. Mental Arithmetic Leads to Multiple Discrete Changes From Baseline in the Firing Patterns of Human Thalamic Neurons. Journal of Neurophysiology 2009;101:2107-2119. [10] Keifer JC, Dentchev D, Little K, Warner DS, Friedman AH, Borel CO. A retrospective analysis of a remifentanil/propofol general anesthetic for craniotomy before awake functional brain mapping. Anesthesia & Analgesia 2005;101:502-508- table of contents. [11] Maltete D, Navarro S, Welter ML, Roche S, Bonnet AM, Houeto JL, Mesnage V, Pidoux B, Dormont D, Cornu P, Agid Y. Subthalamic stimulation in Parkinson disease: with or without anesthesia? Arch Neurol 2004;61:390-392. [12] Hutchison WD, Lang AE, Dostrovsky JO, Lozano AM. Pallidal neuronal activity: implications for models of dystonia. Ann Neurol 2003;53:480-488. [13] Trombetta C, Deogaonkar A, Deogaonkar M, Ebrahim Z, Rezai A, Machado A, Farag E. Delayed awakening in dystonia patients undergoing deep brain stimulation surgery. J Clin Neurosci 2010;17:865-868. [14] Venkatraghavan L, Manninen P, Mak P, Lukitto K, Hodaie M, Lozano A. Anesthesia for functional neurosurgery: review of complications. Journal of neurosurgical anesthesiology 2006;18:64-67. [15] Bekker A, Sturaitis MK. Dexmedetomidine for Neurological Surgery. Neurosurgery 2005;57:1-10. [16] Maurtua MA, Cata JP, Martirena M, Deogaonkar M, Rezai A, Sung W, Lotto M, Niezgoda J, Schubert A. Dexmedetomidine for deep brain stimulator placement in a child with primary generalized dystonia: case report and literature review. J Clin Anesth 2009;21:213-216.
18
[17] Arain SR, Ebert TJ. The efficacy, side effects, and recovery characteristics of dexmedetomidine versus propofol when used for intraoperative sedation. Anesthesia and Analgesia 2002;95:461-466. [18] Mack PF, Perrine K, Kobylarz E, Schwartz TH, Lien CA. Dexmedetomidine and neurocognitive testing in awake craniotomy. Journal of neurosurgical anesthesiology 2004;16:20-25. [19] Haenggi M, Ypparila-Wolters H, Hauser K, Caviezel C, Takala J, Korhonen I, Jakob SM. Intra- and inter-individual variation of BIS-index and Entropy during controlled sedation with midazolam/remifentanil and dexmedetomidine/remifentanil in healthy volunteers: an interventional study. Crit Care 2009;13:R20. [20] Kasuya Y, Govinda R, Rauch S, Mascha EJ, Sessler DI, Turan A. The correlation between bispectral index and observational sedation scale in volunteers sedated with dexmedetomidine and propofol. Anesth Analg 2009;109:1811-1815. [21] Elias S, Joshua M, Goldberg JA, Heimer G, Arkadir D, Morris G, Bergman H. Statistical properties of pauses of the high-frequency discharge neurons in the external segment of the globus pallidus. Journal of Neuroscience 2007;27:2525-2538. [22] Hutchison WD, Allan RJ, Opitz H, Levy R, Dostrovsky JO, Lang AE, Lozano AM. Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson's disease. Annals of Neurology 1998;44:622-628. [23] Lee JI, Verhagen Metman L, Ohara S, Dougherty PM, Kim JH, Lenz FA. Internal pallidal neuronal activity during mild drug-related dyskinesias in Parkinson's disease: decreased firing rates and altered firing patterns. Journal of Neurophysiology 2007;97:26272641.
19
[24] Stefani A, Galati S, Peppe A, Bassi A, Pierantozzi M, Hainsworth AH, Bernardi G, Orlacchio A, Stanzione P, Mazzone P. Spontaneous sleep modulates the firing pattern of parkinsonian subthalamic nucleus. Experimental brain research 2006;168:277-280. [25] Mason KP, O'Mahony E, Zurakowski D, Libenson MH. Effects of dexmedetomidine sedation on the EEG in children. Paediatr Anaesth 2009;19:1175-1183.
20
Figure Captions Fig 1. A representative example of continuous monitoring of sedation during ‘asleepawake’ phase. BIS was decreased soon after starting DEX infusion and was maintained around a score of 60 with some fluctuations. After discontinuation of DEX, BIS increased rapidly above a score of 80 indicating rapid regain of conciousness. BIS ; bispectral index, DEX ; dexmedetomidine, MER ; microelectrode recording
Fig 2. . Representative MER results during ‘awake phase’ at two different target nuclei. MER spikes of a representative patients, during the DBS at the Globus pallidus(A), and the STN(B). MER spike images are shown with the schematic image for each subcortical nuclei during insertion of the implant. Increased background activity (indicated *) is noted during advancement of electrode in the zona incerta (ZI), which means electrode is close to the STN. GPe ; globus pallidus externus, lat lami. ; lateral medullary lamina GPi ; Globus pallidus internus, med lami ; medial medullary lamina, ZI ; zona inserta, STN ; subthalamic nucleus, SNR ; substantia nigra
Fig 3. Representative MER changes of globus pallidus and subthalamic nucleus during DBS with different state of sedation. Results on the left column (patient 1 & 5) were recorded during awake-phase but patient was still in deep sedation, while those on the right column (patient 2 & 6) were recorded under light sedation. Simultaneously recorded EEG shows predominant lower frequency activities on the left column. It was observed that more pauses in the GPi cells and lower amplitude in the STN under deep sedated state. Although some difference of the amplitude was seen, results of MER from all cases showed excellent characteristic patterns. GPi ; Globus pallidus internus, STN ; subthalamic nucleus
21
Fig 4. Within-patient analysis of two different conditions. A) Neuronal activities from same patient under ‘no sedative’ condition and awake phase of ‘sedatives’ condition. More pauses are seen in ‘sedatives’ condition (indicated dots). Small figures on the right side are sorted spikes used for the analysis. B) Firing rates of GPi neurons from same patient under different conditions. Patient 1 was under deep sedation (mean BIS 57.3) during the MER after discontinuation of sedatives. GPi cells of patient 1 shows significantly lower firing rate compared with those ‘no sedative’ condition (P0.05 Mean HY 2.75 (0.37) 2.5 (0.44) >0.05 Disease duration 12.5 (4.34) 9.6 (5.39) >0.05 FR GPe 33.5(15.42) 52.9(21.17)
S0303-8467(16)30045-2 http://dx.doi.org/doi:10.1016/j.clineuro.2016.02.005 CLINEU 4308
To appear in:
Clinical Neurology and Neurosurgery
Received date: Revised date: Accepted date:
30-12-2015 4-2-2016 5-2-2016
Please cite this article as: Kwon Woo-Keun, Kim Jong Hyun, Lee Ji-Hye, Lim Byung-Gun, Lee Il-ok, Koh Seong Beom, Kwon Taek Hyun.Microelectrode recording (MER) findings during sleep-awake anesthesia using dexmedetomidine in deep brain stimulation surgery for Parkinson’s disease.Clinical Neurology and Neurosurgery http://dx.doi.org/10.1016/j.clineuro.2016.02.005 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Microelectrode recording (MER) findings during sleep-awake anesthesia using dexmedetomidine in deep brain stimulation surgery for Parkinson’s disease Woo-Keun Kwon, M.D.1, Jong Hyun Kim M.D., Ph.D.1* [email protected], Ji-Hye Lee M.D.1, Byung-Gun Lim M.D., Ph.D.2, Il-ok Lee M.D., Ph.D.2, Seong Beom Koh M.D., Ph.D.3, Taek Hyun Kwon, M.D., Ph.D.1 1
Department of Neurosurgery, Korea University Guro Hospital, Korea University College of
Medicine, Seoul, Korea 2
Department of Pain and Anesthesiology, Korea University Guro Hospital, Korea University
College of Medicine, Seoul, Korea 3
Department of Neurology, Korea University Guro Hospital, Korea University College of
Medicine, Seoul, Korea *
Corresponding author at: Department of Neurosurgery, Korea University Guro Hospital, 148
Gurodongro, Gurogu, 08308, Seoul, Korea. Tel: 82) 2-2626-3098; Fax: 82) 2-863-1684.
1
Highlights
GPi and STN DBS for intractable IPD under DEX anesthesia was done
DEX provided excellent surgical conditions, and hemodynamic stability
DEX influences the firing rate of subcortical nuclei during DBS
However, it did not affect the MER interpretation and of target localization
2
Abstract Objective The preferred choice of anesthesia for deep brain stimulation (DBS) has been local anesthesia due to the need of patients’ cooperation during the procedure, and concern on the interference of sedatives on microelectrode recording (MER) results. However, local anesthesia during the whole procedure may be impossible in some patients due to uncontrolled anxiety, fear, delirium or exhaustion. Therefore, sedative drugs have been used for DBS, but findings of MER during the procedures have not been reported in detail, especially in the globus pallidus internus (GPi). We introduce our experience using ‘asleepawake’ technique by dexmedetomidine (DEX) anesthesia with MER findings during DBS in idiopathic Parkinson’s disease (IPD) patients. Patients and Methods Data from 14 different subcortical nuclei from 8 consecutive IPD patients whom had DBS at the GPi (6 patients) and subthalamic nucleus (STN) (2 patients) were retrospectively reviewed. We used continuous DEX and intermittent small boluses of propofol during the painful procedure (‘asleep phase’), accompanied with continuous intraoperative monitorings of bispectral index (BIS) and modified observer’s assessment of sedation (MOAA/S). Then sedatives were discontinued during MER recording (‘awake phase’). Characteristic findings and firing rates of neurons were analyzed and compared to those from other 6 patients who underwent surgery under local anesthesia. Results All patients were satisfactorily sedated using this technique without any respiratory or hemodynamic complications. Characteristics of spike activities of each nucleus were inspected and analyzed quantitatively. We could inspect changes of spike activities according
3
to level of patients’ consciousness in some cases, but the localizing value was good to decide the target in all cases. Firing rates of group whom sedatives were given during asleep phase (‘sedatives’) were significantly lower than those of group under local anesthesia (‘no sedative’). Intraoperative length of target nuclei, postoperative imaging and postoperative changes of UPDRS III score indicated satisfactory outcome. Conclusion We concluded that though MER findings may change during DEX-based monitored ‘sleep-awake’ anesthesia, it did not affect the results of target localization for the clinical purpose. However, it should be considered that use of sedatives before MER could result in changes of firing rate and pattern depending on the patient’s state of consciousness.
KEYWORDS: dexmedetomidine; deep brain stimulation; microelectrode recording; Parkinson’s disease
4
Introduction Deep brain stimulation (DBS) has been widely performed successfully for patients with various intractable movement disorders such as idiopathic Parkinson’s disease (IPD), dystonias or tremors. Previously, DBS has been performed mostly under local anesthesia considering the quality of microelectrode recording (MER) results and the necessity of patient’s cooperation during DBS. However, the duration of DBS intervention, especially in bilateral implantations at the same day, is sometimes too long and some patients suffer from intolerable exhaustion, anxiety or fear. Main concerns of using sedatives during DBS are its possible effect on the quality of MER and respiratory function. Gamma-aminobutyric acidergic (GABAergic) sedatives such as propofol have been used most commonly for DBS, but there are concerns on its interference on MER[1] or its fatal complications such as respiratory depression[2]. Dexmedetomidine (DEX) is a highly selective alpha-2-adrenergic receptor agonist which is a well known alternative sedative, which shows dose-dependent rapid action without respiratory depression[3-5]. However, undesirable effects such as changes of neuronal activities in the subthalamic nucleus (STN) during surgery[6], or prolonged impairment of consciousness[7], have also been reported. There has been a quantitative study of neuronal spikes during the STN DBS using DEX-based anesthesia[8]. In this study, the authors even recommended not to use high-dose DEX during the STN DBS[8] based on their results that spike patterns were changed. In this paper we report our experience of MER during ‘asleep-awake’ anesthesia using DEX-based sedative technique in 8 consecutive IPD patients, mainly from GPi DBS. We stopped the infusion of DEX before the MER and observed the characteristics of neuronal spikes if they are acceptable for the clinical purpose to decide the optimal target. In addition, we analyzed the firing rate of each nucleus and tried to correlate them with the BIS
5
of ‘awake-phase’. To our knowledge, there has been no quantitative analysis using neuronal activities during GPi DBS of IPD patient using DEX-based anesthesia.
Methods Patients A total of eight consecutive patients whom underwent DBS for medically intractable IPD under DEX-based anesthesia between January 2013 and July 2015 in our institute were retrospectively studied. Neurologists and neurosurgeons specializing in movement disorders diagnosed all patients. A multi-department team including neurosurgeons, neurologists and psychiatrists made decision of surgical candidates and target nuclei for DBS, after considering the patients’ main symptoms, unified Parkinson’s disease rating scale (UPDRS) scores and the results of psychiatric evaluations. This study was approved by the Institutional Review Board. Surgical procedure and anesthesia protocol We followed the standard DBS surgical protocol as described in previously reported articles[9]. All operations were performed by a single experienced surgeon. IPD patients for STN DBS had been off all dopaminergic medications for at least 12 hours before the surgery. For patients whom underwent Globus Pallidus internus (GPi) DBS for severe dyskinesia, their dopaminergic medications were administrated until the operation day as usual to verify the anti-dyskinetic effect of test stimulation during the surgery. Magnetic resonance images (MRI) were taken in the operation day morning after placement of the Leksell stereotactic head frame (Elekta Instrument, Stockholm, Sweden). The images were imported into the stereotactic planning software (FramelinkTM, Medtronic, US), and then coordinates of the initial target were decided by the neurosurgeon.
6
We used standard ‘asleep - awake’ anesthetic technique. Patients were sedated during the painful procedure (‘asleep-phase’) using loading and continuous infusion of DEX. Then sedatives were discontinued and patient was awakened for MER and test stimulation (‘awake-phase’).
The neuronal data from ‘awake-phase’ were used for analysis (‘sedatives’
group). For the quantitative analysis to compare firing rates and patterns, another group of patients (total 6: 2 GPi and 4 STN) whom underwent surgery under local anesthetics without any sedatives (‘no sedative’ group) were included. In addition, we used a data from two patients (patient 1 and 7) whom underwent surgery with one site with sedatives and the other site without, for within-patient analysis. All patients were monitored by an experienced anesthesiologist during the whole procedures. Initially DEX was given by a loading dose of 0.5 − 1 µg/kg over 10 minutes, followed by a maintenance dose of 0.1 − 0.5 µg/kg/hr infusion. If the patient was not sedated enough during the painful procedure, we increased the infusion rate of DEX or used small boluses of propofol (10 - 20 mg) for rapid effect. Low dose of remifentanil (REM) was also given (0.01 − 0.1 µg/kg/min) continuously with an adjustment as needed to control pain in 4 out of 8 cases. Continuous infusion of DEX and REM were stopped after the placement of burr holes. In case of bilateral implantation, DEX infusion was restarted before the intervention of the second side. Detailed control of the amount and doses of sedatives were done by the anesthesiologist, based on the patients’ intraoperative conditions. Additionally, bispectral index (BIS) monitoring system (Aspect Medical System, US) was applied at the patient’s forehead to monitor the depth of sedation. The maintenance dose of DEX was adjusted aiming the target BIS lower than 80. Modified observer’s assessment of alertness/sedation (MOAA/S) scale was also checked by the anesthesiologist to evaluate the level of patients’ consciousness.
7
Intraoperative Microelectrode recordings procedure and postoperative evaluation For the physiological mapping, MER was started from 15 millimeters above the estimated target. The tungsten/stainless steel microelectrode (model FC 1003, FHC Inc.) was advanced carefully, and stopped whenever good and consistent target neuronal activities appeared. Characteristic spike activities representing each target nuclei were displayed on the Leadpoint system (Medtronic). At the same time, the amplified signal was sent out to the Analog Output Board (Medtronic) installed on the system and separately digitized by an analogue-to-digital converter (ADC) (Micro1401, CED, UK). The sampling rate was 50,000 Hz for spike activities and digitized data was stored on a computer for the off-line analysis. After finishing of physiologic mapping according to intraoperative MER findings and evoked responses, testing macro-stimulation effect was performed at depths by more than two experienced movement disorders specialists. Once an agreement was made that the electrode was placed at the best effective target site, the therapeutic electrode (Model 3389 or 3387, Medtronic) was implanted. The final position of the electrode was verified postoperatively once again by co-registration of post-operative thin sliced computed tomography (CT) with preoperative MRI. Postoperative clinical effectiveness was evaluated by comparison of preand postoperative UPDRS III score, which represents the patients’ clinical status related to motor function. We also compared preoperative OFF-medication state and postoperative OFF-medication/DBS-ON state at least 6 months after the surgery. Spike sorting and analyses Once the spontaneous neuronal activity was found during MER procedure, advancement of microelectrode was stopped and stayed at least 20 seconds to record stable spike activities, and then was stored for off-line analyses. Digitally recorded spike activities were sorted using Spike2 software (CED, Cambridge, UK) by template matching of action potential waveforms[9]. Spikes with high signal to noise ratio at least two times of 8
background activities were chosen for analysis. Analyses of spike characteristics of firing rates were carried out using customized software programmed by MATLAB. Firing rate was calculated by number of spikes per duration of recording. Results Patients and target nuclei A total of eight IPD patients (4 males, 4 females) with average age of 62.4 (range ; 54-73) were involved in this study. Six IPD patients underwent GPi DBS (two unilateral, four bilateral) while the other two underwent bilateral STN DBS. GPi was chosen as a target for IPD patients with decreased cognitive function or for those who had psychiatric problems. One patient (patient-4) had major depressive disorder with severe anxiety. Preoperative demographic information and target nuclei of the patients are presented on Table1. Anesthesia After initial loading dose of DEX was given, most patients became sedated efficiently. The state of sedation could be monitored continuously by an anesthesiologist with MOAA/S scale and BIS. BIS monitoring of a representative patient is shown on Figure 1. The BIS score was recorded every minute on the device in 3 patients, while it was recorded intermittently by an anesthesiologists on the chart in other cases. The mean BIS of all patients during the asleep-phase was 71.9 and was 81.9 during the awake-phase. Generally, the BIS value represented the patient’s level of consciousness well, but some fluctuation was found occasionally during the procedure because of some artifact made by the patient’s movement. The mean loading and maintenance doses of DEX were 0.94 µg/kg and 0.46 µg/kg/hr. The mean (+/- standard deviation) infusion time of DEX was 67.5 (+/- 27.1) min. Small boluses of propofol for rapid effect were given intermittently in all patients during the asleep-phase as described in Table 2. Low dose of continuous infusion of REM (0.01 - 0.1 µg/kg /hr; mean 0.045 µg/kg/min) was given to 4 patients (patient 2,4-6). Hemodynamic parameters were 9
stable throughout the procedure among all 8 patients. After the initial loading dose, both blood pressure and heart rate were slightly decreased during the ‘asleep phase’, but gradually increased and recovered baseline levels during the ‘awake phase’. There were no significant changes of respiration rate and end-tidal PCO2 during the whole procedure. More than half of the patients were fully awaken within 20 minutes (10 out of 14 sides) after stopping DEX infusion. The mean awakening time to fully cooperative state (MOAA/S score 5) after stopping DEX was 22.3±16.1 minutes. We observed that not all patients were recovered to MOAA/S score 5 and cooperative during MER. Patient 1 was uncooperative during MER procedure but awakened 40 minutes after the discontinuation of DEX. This patient underwent unilateral surgery previously under local anesthesia, and consequent trial of contralateral implantation at the same day was failed because she was too delirious to continue the procedure. Patient 3 was well cooperative during the first side but was only able to follow simple commands during the 2nd side, though which did not affect the intervention.
Microelectrode recordings (MER) and neuronal activities MER was stably recorded after the discontinuation of DEX and we could observe characteristic patterns of neuronal activities from the globus pallidus and subthalamic nucleus (Fig.2). The mean number of electrode trajectories per side was 1.3. The mean trajectory length was about 5 mm (GPi: 5.06 mm, STN: 5.55 mm). Using the results of characteristic spike activities, each subcortical nucleus could be clearly localized. However, we could observe subtle differences of neuronal activities according to the state of sedation. Because some patient was not fully awake after DEX discontinuation, the state of sedation was different among patient during MER. We observed that background activity and spike amplitudes seemed lower under the deeper sedation than those under light sedation in some cases (Fig.3). A total of 125 single neuronal activities (GPe 27, GPi 79, STN 19 cells) could 10
be sorted and analyzed. The mean firing rate of each nucleus were GPe: 33.5, GPi: 40.0, STN: 22.5 (spike/sec) respectively. These results were compared with those of patients whom underwent surgery with local anesthesia (‘no sedative’ group). A total of 64 neurons were available for the analysis (GPe 23, GPi 18, STN 23 cells). The mean firing rate of ‘no sedative’ group were GPe: 52.9, GPi: 60.75, STN:35.2 (spikes/sec), showed significantly higher firing rates compared to ‘sedatives’ group in all nuclei (Table 3). Because we had two patients (patient 1,7) whom underwent GPi-DBS surgery with two different conditions, within-patient comparison was performed. A total of 18 (7 sedatives, 11 no sedative) GPi cells from patient 1, and 13 (6 sedatives, 7 no sedative) GPi cells from patient 7 were available for the analysis. Patient 1 was deeply sedated after discontinuation of DEX, while patient 7 was quickly awakened. The mean BIS during awake-phase of patient 1 and 7 were 57.3 and 93 respectively. We observed GPi cells of patient 1 at ‘awake-phase’, however this patient was still deeply sedated even after DEX discontinuation. The results showed more pauses with lower firing rate compared with ‘no sedative’ condition in this patient (Fig.4 A and B). On the other hand, GPi cells of patient 7 of ‘awake-phase’ showed similar firing rate compared with ‘no sedative’ state (Fig.4 B). Next, we investigated the correlation between firing rate and BIS. As conducting a retrospective study, BIS data was recorded continuously in only three patients, so mean firing rate and mean BIS during the procedure was used for the analysis in other patients. We could observe a significant correlation between firing rate and BIS in one patient but not in other two patients from continuous data. However, there was a certain tendency of positive correlation between mean firing rate and mean BIS using available data from all patients (Fig. 5). Clinical improvements comparing UPDRS III score before and after the surgery could be evaluated in 5 patients. Three patients could not be evaluated as two were referred back to another center, and one patient was not tested postoperatively with OFF-medication. These 11
three patients reported their state as ‘much better’ after the surgery. Evaluation of other 5 patients showed significant improvement with a mean 49.8% decrease of UPDRS III score.
Discussion The present study showed that monitored ‘asleep-awake’ technique using Dex-based anesthetics was safe and reliable method for DBS for different subcortical nuclei. The quality of MER was good enough to decide the target nuclei and test stimulation was successfully performed with patient’s cooperation. Permanent electrodes were implanted at the sites where macrostimulation showed significant improvements without side effects. Proper location of implantation were also later confirmed by UPDRS motor scores comparing preoperative off medication state vs postoperative off-medication/on-DBS state. Mean length of target nucleus about 5 mm also suggested satisfactory target exploration. Most centers withhold anti-parkinsonism medications at least 12 hours before the surgery to render the PD patients in the ‘off’ state for intraoperative neurological testing. Without medication, patients may have pain, anxiety or exhaustion during the surgery if they are not properly managed. We have experienced that some PD patients were too exhausted to be cooperative especially during the second side operation of bilateral surgery under the local anesthesia. Patient-1 of the current study had developed delirium during second side operation after the first side. It was impossible to implant permanent electrode on the second side because the patient was not cooperative. Another patient (patient-4) had severe anxiety and painful off-dystonia symptom without anti-parkinsonism medications. Both patients underwent the surgery successfully using the current protocol without any difficulty. ‘Asleep-awake-asleep’ or ‘asleep-awake’ anesthetic method allows enough sedation during painful procedure followed by rapid emergence for intraoperative brain mapping procedures. Propofol has been used for craniotomy requiring intraoperative awake functional 12
brain mapping [10], and MER for therapeutic electrode insertion during DBS on different target sites [5, 11, 12]. The median interval wakeup time of propofol and remifentanil was 9 minutes in brain tumor, vascular malformation and epilepsy patients [10], but more delayed up to 36 minutes when used to dystonia patients undergoing DBS surgery[13]. We also observed that wakeup time was various among patients. Respiratory complications are the most feared during DBS surgery because when it happens it may hard to keep the airway due to fixed patient’s head via the stereotactic frame to the operating table [14]. DEX is a highly selective alpha-2-adrenoreceptor agonist, and it produces dose-dependent sedation without respiratory depression [15]. Its sedative and anxiolytic effects seem to through activation of locus ceruleus that modulates arousal, sleep and anxiety[15]. This non-cortical site of action may result in a state of ‘cooperative-sedation’[15]. Continuous infusion of DEX has been used for DBS in several studies successfully [3, 4, 16]. Previous report also showed that adequate sedation using DEX or propofol were equally satisfactory after DBS [17]. The protocols of these studies were different from ours. We used continuous infusion of DEX with intermittent small dose of propofol in case the patient was awake and complained discomfort during the painful procedure (asleep-phase). Because propofol achieves sedation more rapidly than DEX [17], it was faster using propofol to adjust the level of sedation than changing the infusion rate of DEX. Using DEX plus small dose of propofol has been successfully used in awake surgery for cognitive testing in other study[18]. We used both BIS and MOAA/S scale to monitor and control the level of sedation during the ‘asleep-phase’. As shown in Fig 1, BIS values were generally decreased during the ‘asleep-phase’ and increased in ‘awake- phase’ as expected. However, using only BIS may not correctly represent the patient’s level of sedation when using continuous DEX infusion. There was a report that BIS monitoring during sedation with DEX and REM showed variable
13
and fluctuating results[19]. Therefore, the combined use of both BIS and sedative scales should be used for complementary data for evaluating the patient’s response to sedation[20]. There are concerns about the changes of neuronal activities while using propofol or DEX[1, 6, 12]. Infusion of sedatives during the MER procedure showed various effects on neuronal activities of the STN. Administration of propofol decreased STN neuronal activity significantly, but returned to the baseline shortly after the discontinuation [1]. Another report showed decrease of activities in the STN under DEX infusion, and recommended the discontinuation after placement of burr holes to keep the patient awake during MER[21]. One recent paper showed increased firing rate but decreased burst index of the STN during DEX infusion and concluded that high-dose DEX infusion should be avoided during the STN DBS [8]. These reports might raise the idea of using ‘asleep-awake’ technique, used as in this study for better MER results. Interestingly, our results also showed even after the discontinuation of DEX infusion, firing rate and patterns could be changed according to the state of patient’s consciousness. This change of neuronal activities after discontinuation of sedatives has not been reported before. However, these changes do not necessarily mean poor localization of target nuclei for the clinical purpose as excellent positioning has been achieved as shown in this study. Our study has several limitations. Because we used ‘asleep-awake’ technique and discontinued DEX before MER, it is not clear that neuronal activities were actually affected by DEX infusion. Observed differences of neuronal activity could be due to difference in individual patients and not related to DEX at all. To study the direct effect of DEX or other sedatives on neurons, one should exam the neuronal change before and after the infusion in the same cell, which is not easy during the surgical procedure, then pair-wise comparison should be performed [1]. Previous retrospective studies presented observational data [6] or used different cells with different conditions [8, 12] as in this study. Within-patient 14
observation of this study suggest possible effects of patient’s sedation state (Fig.4). However, due to the limitation of retrospective study, we could not present exact correlation between BIS and neuronal firing because many BIS data were recorded intermittently rather than continuously. Our results may seem somewhat different compared with the recent paper of Krishna et al. [8], which showed increased firing rate of the STN after DEX infusion. The reason might be due to the different condition of neurons. This study used neurons after discontinuation of sedatives, but previous study used neurons under DEX infusion continued or titrated depending on patients’ condition. In addition, previous study investigated STN neurons, but this study used mainly GPi neurons and relatively small numbers of STN neurons. Results of firing rates of ‘no sedative’ group of this study were consistent with previous studies [8, 12, 22, 23], considering STN was recorded under OFF-medication and GPi under ON-medication state. We observed the tendency of more pauses in GPi and smaller amplitudes in STN when the patient was in the deeper sedation state (Fig 2.). Within patient analysis also showed similar results (Fig 4). Those changes might have been because of the remaining effect of sedatives or sleep induced changes[24]. We observed patients often became lightly sedated again (see Fig.1) even after discontinuation of DEX if there was no external stimulation. This prolonged anxiolytic effect was noticed in other study as well [6]. Changes of EEG pattern similar to stage II sleep was noted after DEX infusion [25], which showed increased low frequency activity as seen in our cases (Fig 3, left column). Subcortical nuclei are directly or indirectly connected to the cortex, thus sedation- or sleep-induced changes of neuronal activities may not be surprising. However, it is not possible to conclude whether the changes of neuronal activities were due to remaining effect of DEX or sleepinduced in this retrospective study.
15
Conclusion In this clinical research, we found that the ‘sleep-awake’ anesthetic technique using DEX does have a tendency to decrease the firing rate of subcortical nucleus in DBS for IPD patients even after discontinuation of the sedative. Those findings may be related to the degree of sedation due to the remaining effect of sedatives or sleep-induced changes. However, even though there were some changes neuronal activities of MER, we concluded that this anesthetic technique provided safe and reliable results enough to decide the optimal target for DBS.
Conflict of interest This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) founded by the Ministry of Education (NRF2010-0005420)
16
References [1] Raz A, Eimerl D, Zaidel A, Bergman H, Israel Z. Propofol decreases neuronal population spiking activity in the subthalamic nucleus of Parkinsonian patients. Anesth Analg 2010;111:1285-1289. [2] Mirski MA, Hemstreet MK. Critical care sedation for neuroscience patients. J Neurol Sci 2007;261:16-34. [3] Rozet I, Muangman S, Vavilala MS, Lee LA, Souter MJ, Domino KJ, Slimp JC, Goodkin R, Lam AM. Clinical experience with dexmedetomidine for implantation of deep brain stimulators in Parkinson's disease. Anesth Analg 2006;103:1224-1228. [4] Sassi M, Zekaj E, Grotta A, Pollini A, Pellanda A, Borroni M, Pacchetti C, Menghetti C, Porta M, Servello D. Safety in the use of dexmedetomidine (precedex) for deep brain stimulation surgery: our experience in 23 randomized patients. Neuromodulation 2013;16:401-406; discussion 406. [5] Venkatraghavan L, Luciano M, Manninen P. Review article: anesthetic management of patients undergoing deep brain stimulator insertion. Anesth Analg 2010;110:1138-1145. [6] Elias WJ, Durieux ME, Huss D, Frysinger RC. Dexmedetomidine and arousal affect subthalamic neurons. Mov Disord 2008;23:1317-1320. [7] Bustillo MA, Lazar RM, Finck AD, Fitzsimmons B, Berman MF, Pile-Spellman J, Heyer EJ. Dexmedetomidine may impair cognitive testing during endovascular embolization of cerebral arteriovenous malformations: a retrospective case report series. Journal of neurosurgical anesthesiology 2002;14:209-212. [8] Krishna V, Elias G, Sammartino F, Basha D, King NK, Fasano A, Munhoz R, Kalia SK, Hodaie M, Venkatraghavan L, Lozano AM, Hutchison WD. The effect of dexmedetomidine on the firing properties of STN neurons in Parkinson's disease. The European journal of neuroscience 2015;42:2070-2077. 17
[9] Kim JH, Ohara S, Lenz FA. Mental Arithmetic Leads to Multiple Discrete Changes From Baseline in the Firing Patterns of Human Thalamic Neurons. Journal of Neurophysiology 2009;101:2107-2119. [10] Keifer JC, Dentchev D, Little K, Warner DS, Friedman AH, Borel CO. A retrospective analysis of a remifentanil/propofol general anesthetic for craniotomy before awake functional brain mapping. Anesthesia & Analgesia 2005;101:502-508- table of contents. [11] Maltete D, Navarro S, Welter ML, Roche S, Bonnet AM, Houeto JL, Mesnage V, Pidoux B, Dormont D, Cornu P, Agid Y. Subthalamic stimulation in Parkinson disease: with or without anesthesia? Arch Neurol 2004;61:390-392. [12] Hutchison WD, Lang AE, Dostrovsky JO, Lozano AM. Pallidal neuronal activity: implications for models of dystonia. Ann Neurol 2003;53:480-488. [13] Trombetta C, Deogaonkar A, Deogaonkar M, Ebrahim Z, Rezai A, Machado A, Farag E. Delayed awakening in dystonia patients undergoing deep brain stimulation surgery. J Clin Neurosci 2010;17:865-868. [14] Venkatraghavan L, Manninen P, Mak P, Lukitto K, Hodaie M, Lozano A. Anesthesia for functional neurosurgery: review of complications. Journal of neurosurgical anesthesiology 2006;18:64-67. [15] Bekker A, Sturaitis MK. Dexmedetomidine for Neurological Surgery. Neurosurgery 2005;57:1-10. [16] Maurtua MA, Cata JP, Martirena M, Deogaonkar M, Rezai A, Sung W, Lotto M, Niezgoda J, Schubert A. Dexmedetomidine for deep brain stimulator placement in a child with primary generalized dystonia: case report and literature review. J Clin Anesth 2009;21:213-216.
18
[17] Arain SR, Ebert TJ. The efficacy, side effects, and recovery characteristics of dexmedetomidine versus propofol when used for intraoperative sedation. Anesthesia and Analgesia 2002;95:461-466. [18] Mack PF, Perrine K, Kobylarz E, Schwartz TH, Lien CA. Dexmedetomidine and neurocognitive testing in awake craniotomy. Journal of neurosurgical anesthesiology 2004;16:20-25. [19] Haenggi M, Ypparila-Wolters H, Hauser K, Caviezel C, Takala J, Korhonen I, Jakob SM. Intra- and inter-individual variation of BIS-index and Entropy during controlled sedation with midazolam/remifentanil and dexmedetomidine/remifentanil in healthy volunteers: an interventional study. Crit Care 2009;13:R20. [20] Kasuya Y, Govinda R, Rauch S, Mascha EJ, Sessler DI, Turan A. The correlation between bispectral index and observational sedation scale in volunteers sedated with dexmedetomidine and propofol. Anesth Analg 2009;109:1811-1815. [21] Elias S, Joshua M, Goldberg JA, Heimer G, Arkadir D, Morris G, Bergman H. Statistical properties of pauses of the high-frequency discharge neurons in the external segment of the globus pallidus. Journal of Neuroscience 2007;27:2525-2538. [22] Hutchison WD, Allan RJ, Opitz H, Levy R, Dostrovsky JO, Lang AE, Lozano AM. Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson's disease. Annals of Neurology 1998;44:622-628. [23] Lee JI, Verhagen Metman L, Ohara S, Dougherty PM, Kim JH, Lenz FA. Internal pallidal neuronal activity during mild drug-related dyskinesias in Parkinson's disease: decreased firing rates and altered firing patterns. Journal of Neurophysiology 2007;97:26272641.
19
[24] Stefani A, Galati S, Peppe A, Bassi A, Pierantozzi M, Hainsworth AH, Bernardi G, Orlacchio A, Stanzione P, Mazzone P. Spontaneous sleep modulates the firing pattern of parkinsonian subthalamic nucleus. Experimental brain research 2006;168:277-280. [25] Mason KP, O'Mahony E, Zurakowski D, Libenson MH. Effects of dexmedetomidine sedation on the EEG in children. Paediatr Anaesth 2009;19:1175-1183.
20
Figure Captions Fig 1. A representative example of continuous monitoring of sedation during ‘asleepawake’ phase. BIS was decreased soon after starting DEX infusion and was maintained around a score of 60 with some fluctuations. After discontinuation of DEX, BIS increased rapidly above a score of 80 indicating rapid regain of conciousness. BIS ; bispectral index, DEX ; dexmedetomidine, MER ; microelectrode recording
Fig 2. . Representative MER results during ‘awake phase’ at two different target nuclei. MER spikes of a representative patients, during the DBS at the Globus pallidus(A), and the STN(B). MER spike images are shown with the schematic image for each subcortical nuclei during insertion of the implant. Increased background activity (indicated *) is noted during advancement of electrode in the zona incerta (ZI), which means electrode is close to the STN. GPe ; globus pallidus externus, lat lami. ; lateral medullary lamina GPi ; Globus pallidus internus, med lami ; medial medullary lamina, ZI ; zona inserta, STN ; subthalamic nucleus, SNR ; substantia nigra
Fig 3. Representative MER changes of globus pallidus and subthalamic nucleus during DBS with different state of sedation. Results on the left column (patient 1 & 5) were recorded during awake-phase but patient was still in deep sedation, while those on the right column (patient 2 & 6) were recorded under light sedation. Simultaneously recorded EEG shows predominant lower frequency activities on the left column. It was observed that more pauses in the GPi cells and lower amplitude in the STN under deep sedated state. Although some difference of the amplitude was seen, results of MER from all cases showed excellent characteristic patterns. GPi ; Globus pallidus internus, STN ; subthalamic nucleus
21
Fig 4. Within-patient analysis of two different conditions. A) Neuronal activities from same patient under ‘no sedative’ condition and awake phase of ‘sedatives’ condition. More pauses are seen in ‘sedatives’ condition (indicated dots). Small figures on the right side are sorted spikes used for the analysis. B) Firing rates of GPi neurons from same patient under different conditions. Patient 1 was under deep sedation (mean BIS 57.3) during the MER after discontinuation of sedatives. GPi cells of patient 1 shows significantly lower firing rate compared with those ‘no sedative’ condition (P0.05 Mean HY 2.75 (0.37) 2.5 (0.44) >0.05 Disease duration 12.5 (4.34) 9.6 (5.39) >0.05 FR GPe 33.5(15.42) 52.9(21.17)
