BRIEF COMMUNICATION Transcranial magnetic stimulation tracks subminute changes in cortical excitability during propofol anesthesia Roman Gersner1, Carmen Paredes1, Mustafa Q. Hameed1,2 , Sameer C. Dhamne1, Alvaro Pascual-Leone3,4 & Alexander Rotenberg1,5 1 The Neuromodulation Program, Department of Neurology, Boston Children’s Hospital, Harvard Medical School, Boston, Massachusetts 2 Department of Neurosurgery, Boston Children’s Hospital, Harvard Medical School, Boston, Massachusetts 3 Department of Neurology, Hebrew SeniorLife, Harvard Medical School, Boston, Massachusetts 4 Institut Guttmann, Universitat Autonoma, Barcelona, Spain 5 Berenson-Allen Center for Noninvasive Brain Stimulation, Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts Correspondence Abstract Alexander Rotenberg, Department of Neurology, Boston Children’s Hospital, 300 Automated anesthesia systems that continuously monitor cortical excitability Longwood Avenue, Boston, MA 02115. Tel: (CE) changes to govern drug infusion rates, are desirable. Paired-pulse transcra- 617-355-8071; Fax: 617-730-0463; E-mail: nial magnetic stimulation (ppTMS), with electromyography (EMG), provides alexander.rotenberg@childrens.harvard.edu noninvasive CE measures. We tested whether, and with what temporal resolu- tion, ppTMS-EMG detects dose-dependent CE in rats anesthetized with contin- uous intravenous propofol. Motor-evoked potentials (MEPs) were recorded Funding Information This work was supported by the NIH National every 20 seconds as either propofol bolus or change in infusion rate was Institute of Neurological Disorders and Stroke applied. ppTMS-derived measures varied in direct proportion to propofol dose (NINDS) [R01 NS088583 (AR)]; and the with subminute temporal resolution. We conclude that ppTMS-EMG enables Boston Children’s Hospital Translational real-time markers of target engagement by anesthetics that may be incorporated Research Program (AR). into an automated device. Received: 31 October 2019; Revised: 16 December 2019; Accepted: 28 December 2019 Annals of Clinical and Translational Neurology 2020; 7(3): 384–389 doi: 10.1002/acn3.50981 Roman Gersner and Carmen Paredes authors contributed equally to this study. interstimulus interval (ISI) to modulate inhibitory or exci- Introduction tatory circuits. A subset of ppTMS protocols, long interval Transcranial magnetic stimulation (TMS) is a well-toler- ppTMS (LI-ppTMS) in rats, as in humans, activate intra- ated, noninvasive method for focal cortical stimulation,1 cortical inhibitory circuits and can detect changes in long- where small intracranial electrical currents are induced by interval intracortical inhibition (LICI) pursuant to phar- a powerful and fluctuating extracranial magnetic field. macologic intervention or injury.7–9 TMS applied over the motor cortex and coupled with elec- Many general anesthetics shift the excitatory–inhibitory tromyography (TMS-EMG) is widely used to study cortical balance of brain signaling toward greater inhibition, in plasticity and excitability, by measuring changes in motor- part by modulating GABA receptor signaling.10 Propofol, evoked potentials (MEPs).2–6 Paired-pulse TMS (ppTMS) commonly used in clinical practice, is a highly lipophilic involves delivering a conditioning stimulus pulse (CS) fol- agent with a fast onset and short duration of action11 via lowed by a test stimulus (TS) after a predetermined positive modulation of inhibitory function through 384 ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals, Inc on behalf of American Neurological Association. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. R. Gersner et al. TMS in Continuous Propofol Anesthesia GABAA receptors.12 Although propofol also has inhibitory electrical current conducted through the magnetic coil. effects on neuromuscular transmission,13 previous studies Optimal coil position was defined as that with the lowest in rats indicate that the MEP is preserved under propofol stimulation intensity required to elicit lateralized MEPs anesthesia.3 Given the capacity of LI-ppTMS to measure exclusively in the contralateral forelimb.4,14–16 Stimulator cortical inhibition, we tested whether LI-ppTMS can intensity was further adjusted to find motor threshold detect subminute changes in cortical excitability modu- (MT), defined as the lowest stimulator intensity necessary lated by differential dosing of propofol in rats. to elicit MEPs of ≥20 lV peak-to-peak amplitude in ≥5 of 10 consecutive trials.4,14 Materials and Methods Experimental design Animals LI-ppTMS was used to approximate human LICI proto- In total, 52 male Sprague–Dawley rats (250  20 g) were cols.7,8,16 Baseline was obtained for each rat (100 msec used. Experiments were approved by the Institutional ISI, 120% MT, 15 pulse pairs, one pair/ 20 sec). Animal Care and Use Committee at Boston Children’s Experiment 1 (propofol bolus): After baseline, rats were Hospital (Boston, MA), and in accordance with the NIH randomized into (1) no bolus (control), (2) 10 mg/kg, or Guide for the Care and Use of Laboratory Animals. (3) 20 mg/kg; followed by LI-ppTMS for 15 min. Experiment 2 (varied propofol infusion): After baseline, rats were randomized into (1) no change in infusion rate Anesthesia (1 mg/kg per min; control) or (2) increase to 2 mg/kg Rats were anesthetized with isoflurane and placed into a per min propofol for 15 min followed by return to 1 mg/ stereotaxic frame. A venous catheter (27G) was inserted kg per min. Each group was monitored by LI-ppTMS for into the lateral tail vein. Following intravenous propofol 30 min. load (10 mg/kg), isoflurane was gradually decreased over 4 min until completely off, while oxygen supplementation Data analysis continued (1 l/min). Anesthesia was maintained using continuous propofol infusion (1 mg/kg per min unless All data are presented as mean  SEM. Peak-to-peak otherwise specified, see “Experimental Design” section). voltage of the response to the CS (ResponseCS) and TS (ResponseTS) was measured. Paired-pulse inhibition was computed as ResponseTS:ResponseCS. All measures were Electromyography (EMG) normalized and log transformed. Differences between MEPs were recorded with monopolar 27G stainless-steel groups were determined by repeated-measures ANOVA needle electrodes inserted into the brachioradialis muscle, and Bonferroni post hoc tests, with group as the and a reference electrode positioned distally in the fore- between-subject factor and time as the within-subject fac- limb contralateral to the side stimulated by the TMS tor. coil.4,14 EMG was acquired at 40 kHz, band-pass filtered at 0.3–20 kHz, and amplified 1000x for post hoc analy- Results sis.4,14 Experiment 1: MEP amplitude decreases Transcranial Magnetic Stimulation (TMS) with increasing doses of propofol bolus Focal TMS was delivered to the motor cortex of anes- Repeated-measures ANOVA of ResponseCS amplitudes thetized rats through a figure-of-eight coil (diameter: out- (Fig. 1A) revealed no significant main effect of group, but side = 66 mm, inside = 15 mm; Magstim, Wales, UK).7,14 a significant main effect of time (F (2,44) = 7.63; TMS intensity was documented as percent machine out- P < 0.001) and interaction between the two (F (2, put (% MO) with 100% corresponding to maximal 88) = 2.36; P < 0.001). Post hoc analysis revealed that Figure 1. Dose-dependent decrease in MEP amplitude after propofol bolus. Data are presented as mean  SEM of MEP amplitude recorded at baseline and at follow-up timepoints. Following a short baseline, animals received either 10 mg/kg, 20 mg/kg or no bolus of IV propofol. The dotted line indicates time of propofol bolus administration. (A) Propofol significantly reduced the response to the conditioning stimulus in a dose- dependent manner. (B) However, minimal MEP amplitude in response to the test stimulus was similar following either bolus, with 20 mg/kg resulting in a prolonged decrement. (C) No changes in the inhibition ratio were observed. ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals, Inc on behalf of American Neurological Association. 385 TMS in Continuous Propofol Anesthesia R. Gersner et al. 386 ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals, Inc on behalf of American Neurological Association. R. Gersner et al. TMS in Continuous Propofol Anesthesia ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals, Inc on behalf of American Neurological Association. 387 TMS in Continuous Propofol Anesthesia R. Gersner et al. Figure 2. Dose-dependent decrease in MEP amplitude after propofol bolus. Data are presented as mean  SEM of MEP amplitude recorded at baseline and at follow-up timepoints. Following a short baseline, animals received an increased infusion rate of 2 mg/kg per min during 15 min, before returning to 1 mg/kg per min for another 15 min. A control group received a continuous infusion rate of 1 mg/kg per min for the entire 30-min period. The shaded region indicates the timing of propofol infusion rate changes. (A, B) Increasing the infusion rate from 1 to 2 mg/kg per min resulted in a progressive decrease in the MEP amplitudes in response to the conditioning and test stimuli, respectively. When the infusion rate was restored to 1 mg/kg per min, a progressive increase in both responses was observed. (C) No changes in the inhibition ratio were observed. propofol boluses of both 10 mg/kg (P < 0.001) and No significant change in the inhibition ratio was 20 mg/kg (P < 0.001) significantly reduced ResponseCS. observed (Fig. 2C). Moreover, the identified MEP decrease was dose-depen- dent (P = 0.01), was evident within 20 sec, and signifi- Discussion cant difference from control lasted for ~7 min after 10 mg/kg and ~12 min after 20 mg/kg boluses. We demonstrate for the first time that propofol modulates In ResponseTS measures (Fig. 1B), significant main a LI-ppTMS-EMG measure of cortical excitability parame- effects of group (F(2,44) = 36.03; P < 0.001), time (F ters in a dose-dependent manner, and that dose-dependent (2,44) = 3.40; P < 0.001), and interaction (F(2,88) = 2.41; fluctuation in inhibition is detectable with subminute tem- P < 0.001) between the two factors were observed. Changes poral resolution. We identified rapid decreases in MEP in amplitude were evident within 20 sec of the bolus. In rats amplitudes following propofol bolus, and gradual dose-de- that received a bolus of 10 mg/kg, the effect lasted for pendent changes in MEP amplitudes following changes in ~5 min before trending toward recovery to baseline. Nota- the propofol infusion rate. Notably, a conspicuous change bly, while minimal ResponseTS amplitude was similar fol- in the MEP was detected following both CS and TS. These lowing either the 10 mg/kg or 20 mg/kg propofol bolus, data indicate that individual MEP amplitudes may be valid rats receiving 20 mg/kg showed a prolonged decrement, markers of change in cortical excitability in LI-ppTMS pro- lasting ~12 min and remaining practically unchanged dur- tocols, in addition to the ratio of MEPs produced by the CS ing the first 8 min of this period. These findings indicate a and TS. While the MEP ratio is the typical primary measure floor effect for ResponseTS after the 20 mg/kg bolus. As in in LI-ppTMS studies, ResponseTS patterns were like the first group, a complete return to baseline was not ResponseCS changes in our experiments, resulting in an achieved by the end of the follow-up. essentially stable MEP ratio. No significant change in the inhibition ratio was Propofol’s mechanism of action includes both potentia- observed (Fig. 1C). tion of GABAA17 and voltage-gated sodium channel block.18 Standalone GABAA potentiation decreases the ResponseTS:ResponseCS ratio, while sodium channel block- Experiment 2: MEP amplitude is inversely ers reduce MEP amplitude.18 Our results indicate that LI- affected by changing continuous propofol ppTMS is more sensitive to the latter propofol effect where infusion rate increasing propofol doses decrease corticospinal excitability In ResponseCS measures (Fig. 2A), significant main effects and, correspondingly, the MEP amplitude. These alter- of group (F(1,89) = 6.27; P = 0.026), time (F(1,89)=2.83; ations of brain excitability were identified within 20 sec fol- P < 0.001) and interaction between the factors (F(1,89) lowing changes in propofol dose, predictably due to the =2.83; p < 0.001) were found. Increasing the infusion rate rapid penetration of propofol through the blood-brain bar- (from 1 to 2 mg/kg/min) provoked a progressive decrease rier and its fast distribution to the central nervous system. in the MEP amplitude, reaching a minimum value 13 min Recovery to baseline took ~10–12 min from maximal after the rate was increased. With the infusion rate restored ResponseCS suppression, which mimics the clinical timeline to 1 mg/kg/min, a progressive increase in amplitude was of recovery from propofol anesthesia.19 Together with a observed, reaching baseline levels after 10 min. decrease in MEP amplitude within 20 sec of bolus adminis- In ResponseTS measures (Fig. 2B), significant main tration, this suggests a high temporal resolution of LI- effects of group (F(1,89) = 6.62; P = 0.023), time (F ppTMS in measuring depth of anesthesia. Our results also (1,89) = 1.5; P = 0.002), and interaction between the fac- suggest that propofol-induced changes in voltage-gated tors (F(1,89) = 1.5; P = 0.002) were observed. A mini- sodium channel-dependent excitability are more readily mum value was reached 10 min following the increase in measured by single-pulse TMS than GABAergic inhibition rate. The amplitude returned to baseline levels 6 min after assessed by LI-ppTMS. Moreover, since a floor effect is restoration of infusion rate to 1mg/kg per min. identified in ResponseTS in both experiments, the 388 ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals, Inc on behalf of American Neurological Association. R. Gersner et al. TMS in Continuous Propofol Anesthesia ResponseCS is a better measure of propofol-induced anes- 3. Luft AR, Kaelin-Lang A, Hauser TK, et al. Transcranial thesia level. This floor effect also points to follow-up studies magnetic stimulation in the rat. Exp Brain Res where stimulus strength and inter-pulse interval may be 2001;140:112–121. modified to mitigate the complete ResponseCS inhibition 4. Rotenberg A, Muller PA, Vahabzadeh-Hagh AM, et al. and thus enable use of the ResponseTS:ResponseCS ratio. Lateralization of forelimb motor evoked potentials by We aimed to evaluate the feasibility of using LI- transcranial magnetic stimulation in rats. Clin ppTMS-EMG to detect changes in cortical excitability Neurophysiol 2010;121:104–108. modulated by propofol anesthesia. However, we note that 5. Kobayashi M, Pascual-Leone A. Transcranial magnetic we assessed only single bolus doses and infusion rate stimulation in neurology. Lancet Neurol 2003;2:145–156. changes per experiment. Other regimens, including multi- 6. Pascual-Leone A, Tormos JM, Keenan J, et al. Study and ple consecutive bolus doses, additional infusion rate modulation of human cortical excitability with transcranial changes, and tests of other anesthetics warrant future magnetic stimulation. J Clin Neurophysiol 1998;15:333–343. studies. Further, beyond the scope of this report, our 7. Vahabzadeh-Hagh AM, Muller PA, Pascual-Leone A, et al. results raise prospects for coupling LI-ppTMS with EEG Measures of cortical inhibition by paired-pulse transcranial to measure pharmacologically induced cortical changes magnetic stimulation in anesthetized rats. J Neurophysiol excitability outside of the motor cortex. 2011;105:615–624. 8. Hsieh TH, Dhamne SC, Chen JJ, et al. A new measure of cortical inhibition by mechanomyography and paired- Conclusion pulse transcranial magnetic stimulation in unanesthetized rats. J Neurophysiol 2012;107:966–972. We demonstrate for the first time that protocols employing 9. Hsieh TH, Lee HHC, Hameed MQ, et al. Trajectory of motor cortex TMS coupled with EMG can detect changes parvalbumin cell impairment and loss of cortical in cortical excitability with subminute resolution, which inhibition in traumatic brain injury. Cereb Cortex raises prospects for the use of TMS in closed-loop anesthe- 2017;27:5509–5524. sia systems. While the MEP ratio is the typical primary 10. Franks NP. Molecular targets underlying general measure in LI-ppTMS studies, our data also indicate that anaesthesia. Br J Pharmacol 2006;147(Suppl 1):S72–S81. the MEP amplitudes pursuant to both conditioning and 11. Kanto J, Gepts E. Pharmacokinetic implications for the test stimuli may be valid markers of changes in intracortical clinical use of propofol. Clin Pharmacokinet 1989;17:308– inhibition in LI-ppTMS protocols, and that the ratio may 326. remain misleadingly stable in the case of similar modula- 12. Hardman J, Limbird L, Gilman A.General Anesthetics. In: tion of individual MEP responses. Beyond the scope of this L. Brunton, B. Chabner, B. Knollman, eds. Goodman and report, our results raise prospects for coupling LI-ppTMS Gilman’s The Pharmacological Basis of Therapeutics, 12 with EEG or other readouts to measure cortical excitability ed. 2011. New York: McGraw-Hill Health. outside of the motor cortex. 13. Abdel-Zaher AO, Askar FG. The myoneural effects of propofol emulsion (Diprivan) on the nerve-muscle preparations of rats. Pharmacol Res 1997;36:323–332. Conflict of Interest 14. Muller PA, Dhamne SC, Vahabzadeh-Hagh AM, et al. Suppression of motor cortical excitability in anesthetized A.P.L. serves on the scientific advisory boards for Starlab rats by low frequency repetitive transcranial magnetic Neuroscience and Neuroelectrics. A.R. is a founder and stimulation. PLoS ONE 2014;9:e91065. advisor to Neuromotion; serves on the scientific advisory 15. Whishaw IQ, Cioe JD, Previsich N, Kolb B. The variability board of EpiHunter, Gamify and NeuroRex; and has of the interaural line vs the stability of bregma in rat received consulting fees or research support from Brain- stereotaxic surgery. Physiol Behav 1977;19:719–722. sway, Cavion, Neuroelectrics, Roche, Sage, and Takeda, 16. Gersner R, Ekstein D, Dhamne SC, et al. Huperzine A outside the submitted work. The other authors declare no prophylaxis against pentylenetetrazole-induced seizures in competing interests. rats is associated with increased cortical inhibition. Epilepsy Res 2015;117:97–103. References 17. Yip GM, Chen ZW, Edge CJ, et al. A propofol binding 1. Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic site on mammalian GABAA receptors identified by stimulation of human motor cortex. Lancet photolabeling. Nat Chem Biol 2013;9:715–720. 1985;325:1106–1107. 18. Ziemann U, Reis J, Schwenkreis P, et al. TMS and drugs 2. Gersner R, Shachter S, Rotenberg A. Huperzine A revisited 2014. Clin Neurophysiol 2015;126:1847–1868. increases cortical inhibition and decreases PTZ-induced 19. White PF. Propofol: its role in changing the practice of seizures in rats. Work in progress2013. anesthesia. Anesthesiology 2008;109:1132–1136. ª 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals, Inc on behalf of American Neurological Association. 389