The suppression of spinal F-waves by propofol does not predict immobility to painful stimuli in humansy

British Journal of Anaesthesia 96 (1): 118 26 (26) doi:1.193/bja/aei283 Advance Access publication November 29, 25 The suppression of spinal F-waves by propofol does not predict immobility to painful stimuli in humansy J. H. Baars*, S. Tas, K. F. Herold, D. A. Hadzidiakos and B. Rehberg Department of Anaesthesiology, Charité Campus Mitte, Berlin, Germany *Corresponding author. Department of Anaesthesiology, Charité Campus Mitte, Schumannstrasse 2/21, D-198 Berlin, Germany. E-mail: jan.baars@charite.de Background. The immobilizing effects of volatile anaesthetics are primarily mediated at the spinal level. A suppression of recurrent spinal responses (F-waves), which reflect spinal excitability, has been shown for propofol. We have assessed the concentration-dependent F-wave suppression by propofol and related it to the logistic regression curve for suppression of movement to noxious stimuli and the effect on the bispectral index TM ( TM ). The predictive power of drug effects on F-waves and for movement responses to noxious stimuli was tested. Methods. In 24 patients anaesthesia was induced and maintained with propofol infused by a target controlled infusion pump at stepwise increasing and decreasing plasma concentrations between.5 and 4.5 mg litre 1. The F-waves of the abductor hallucis muscle were recorded at a frequency of.2 Hz. values were recorded continuously. Calculated propofol concentrations and F-wave amplitude and persistence were analyzed in terms of a pharmacokinetic pharmacodynamic (PK/PD) model with a simple sigmoid concentration response function. Motor responses to tetanic electrical stimulation (5 Hz, 6 ma, 5 s, volar forearm) were tested and the EC 5tetanus was calculated using logistic regression. Results. For slowly increasing propofol concentrations, computer fits of the PK/PD model for the suppression by propofol yielded a median EC 5 of 1.26 (.4 2.3) and 1.9 (1. 2.8) mg litre 1 for the F-wave amplitude and persistence, respectively. These values are far lower than the calculated EC 5 for noxious electrical stimulation of 3.75 mg litre 1. This difference results in a poor prediction probability of movement to noxious stimuli of.59 for the F-wave amplitude. Conclusions. F-waves are almost completely suppressed at subclinical propofol concentrations and they are therefore not suitable for prediction of motor responses to noxious stimuli under propofol mono-anaesthesia. Br J Anaesth 26; 96: 118 26 Keywords: anaesthetics i.v., propofol; monitoring, bispectral index; model, pharmacokinetic pharmacodynamic; responses, F-waves; pain Accepted for publication: September 12, 25 Mechanisms by which volatile anaesthetic suppress motor responses to painful stimuli are predominantly of spinal origin. It has been shown in rats that neither decerebration 1 nor hypothermic transection between the brain and the spinal cord 2 alter the anaesthetic requirements to produce immobility. The spinally-mediated H-reflex and F-waves have been recommended for monitoring surgical immobility in humans. 34 F-waves are low amplitude motor responses to nerve stimulation. After electrical stimulation of peripheral motor fibres, the impulse propagates orthodromically to the corresponding muscle and antidromically to the spinal motoneurones. The orthodromic potential elicits a direct muscle response (M-wave), whereas the antidromic potential results in a recurrent discharge ( backfiring ) of some motoneurones which finally lead to a late muscle response, the F-wave (for reviews on F-wave application see refs 5 and 6). F-waves represent a sensitive measure of changes in motoneurone excitability. 78 Previous studies have shown that propofol depresses F-waves in humans. 9 11 However, there are no reports comparing the concentration- and time-dependent suppression of spinal cord excitability and suppression of the EEG for propofol. In this study we simultaneously examined the effect of propofol on spinal F-waves (amplitude and y Presented in part at the annual meeting of the American Society of Anesthesiologists 24 in Las Vegas. Ó The Board of Management and Trustees of the British Journal of Anaesthesia 25. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

F-wave and propofol persistence) and to the bispectral index TM ( TM ). By using pharmacokinetic pharmacodynamic modelling (PK/PDmodelling) it is possible to compare the dose response curves for different responses. We determined the concentration response curve for the F-wave amplitude and persistence and compared it with a logistic regression curve of response to noxious stimulation. Comparing the equilibration half-times of the different effects may help to disentangle hypnotic effects measured by the from spinal effects measured by the F-wave. In particular, different values of the equilibration half-times would be an argument for different and independent effect sites. Material and methods The protocol for this study was approved by the local ethics committee (Charité, Berlin, Germany), and written informed consent was obtained from all patients. The study was performed before elective gynaecological or urological surgery on 2 female and 4 male patients who were classified as American Society of Anesthesiologists physical Status I or II. Patient exclusion criteria were pregnancy, any neuromuscular disease, use of CNS-acting medication, abuse of alcohol or illicit drugs, and contraindications to the use of propofol. The data evaluation and propofol administration used in this study resembles a previous study we performed on different patients studying the effect of propofol on the H-reflex. 12 Study design Patients fasted at least 6 h before the study and received no premedication. After arrival in the operation room, standard monitoring (non-invasive blood pressure monitoring, ECG and pulse oximetry) and i.v. access via a forearm vein were established. Thereafter baseline recordings of F-waves (Fig. 1) and encephalographic variables were obtained for 1 min before induction of anaesthesia. Propofol was infused i.v. via a computer-controlled infusion pump, programmed using the weight-corrected pharmacokinetic dataset of Marsh and colleagues. 13 Every 3 5 min, the target plasma concentration was increased by.5 mg litre 1 until F-waves were abolished (Fig. 2). The point at which patients lost consciousness (defined as loss of verbal response to repeated loud verbal command) was recorded. The propofol infusion was then stopped until the F-wave amplitude increased again to at least 2% of the baseline value and patients became conscious again. During the period of decreasing propofol concentrations loud verbal commands were repeated every minute to avoid the propofol anaesthesia merging into a natural sleep, which might influence the F-wave. 14 The time of return of consciousness, defined as when there was a verbal response to the loud verbal command, was also recorded. During the entire recording period, patients breathed pure oxygen from the circle system of an anaesthesia machine (Modulus, Ohmeda, Madison, WI, USA) via a tight fitting face mask. End-tidal carbon dioxide was monitored continuously when patients were unconscious and kept constant by assisting ventilation manually, if necessary. The tight fit of the mask was checked by monitoring the capnograph. In 12 patients, the response to a noxious electrical stimulus was investigated. An electrical stimulus (5 Hz, 6 ma, 5 s,.2 ms square wave tetanic stimulus) was applied with a peripheral nerve stimulator (Fischer and Paykel, Auckland, New Zealand) to surface electrodes placed on the volar surface of the forearm at the various target propofol concentrations. A positive response was defined as a gross purposeful movement of the head or extremities, excluding the stimulated arm. After the study period, the propofol concentration was increased to 4.5 mg litre 1 and patients received fentanyl.1 mg and cis-atracurium.1 mg kg 1 before their tracheas were intubated and surgery commenced. Neurophysiologic data acquisition A detailed description of the method has been given previously. 15 In brief, F-waves were recorded over the abductor pollicis muscle after stimulation of the tibial nerve at the ankle. Stimuli were applied continuously throughout the study with a frequency of.2 Hz and duration of.1 ms. Stimulus intensity was kept constant for the entire study period. F-wave persistence, that is, the number of measurable F-waves divided by the number of stimuli, was determined offline from a series of nine successive stimuli. To distinguish F-waves from background noise the recorded EMG-signal was reviewed visually; only appropriately timed deflections, which clearly contrasted with the baseline noise, were accepted as F-waves. Depending on the individual background noise level the smallest accepted F-wave amplitude varied between 3 and 5 mv. The EEG was recorded in a bifrontal montage (Fpz-A1 and Fpz-A2) using an A-1 EEG monitor (Aspect Medical Systems, Natick, USA). The (version B31v2) was used for further analysis. Data from time periods where burst suppression occurred (burst suppression index>) were discarded. Pharmacodynamic analysis Individual concentration response functions were fitted to the data and the calculated plasma concentrations using a spreadsheet program (EXCEL, Microsoft) and a simple sigmoidal model: E=E 1 cl eff EC l 5 +cl eff In this model, E is the baseline effect, c eff is the apparent effect site concentration, EC 5 is the concentration that causes 5% of the maximum effect and l (Hill coefficient) describes the slope of the concentration response relation. The time lag between changes in calculated plasma 119

Baars et al. A 14 12 F-wave amplitude (1 µv) 1 8 6 4 2 2 2 4 6 8 1 12 14 Time (ms) B 14 12 F-wave amplitude (1 µv) 1 8 6 4 2 2 2 4 6 8 1 12 14 Time (ms) Fig 1 Original tracings of M-wave and F-wave. Original tracings of M-wave and F-wave (A) before anaesthesia and (B) under 2 mg litre 1 propofol plasma concentration in a male patient (characteristics: age, 49; body height, 2 cm; weight, 11 kg). (Note: as F-wave latency is related to the body height it is far above the average in this patient.) concentration and observed effect was modelled by an effect compartment and a first-order rate-constant determining the efflux from the effect compartment k e : dc eff =dt=ðc p C eff Þ k eo where C p is the predicted plasma propofol concentration, C eff is the effect compartment concentration of propofol and k e is the first-order rate-constant determining the efflux from the effect compartment. The effect site equilibration half-life t 1/2 k e was calculated as=ln 2/k e. The fitted parameters of the different electrophysiological measures were compared using one way ANOVA with Tukey s multiple comparison post hoc test (Prism 3.; Graph pad Software, San Diego, USA). The logistic regression model developed by Waud 16 was used to determine the EC 5 -value for both the loss of movement response to noxious electrical stimulus and the loss of consciousness. For the latter, the propofol effect site as predicted by the pump concentration was averaged more than 2 min before and after loss of and return of consciousness; for the former, the predicted plasma concentration was averaged more than 1 min before the tetanic stimulation. 12

F-wave and propofol A 2. B 2. F-wave amplitude 1.5 1..5 F-wave amplitude 1.5 1..5. 1 2 3 4 5 C plasma propofol (mg litre 1 ). 1 2 3 4 5 C eff propofol (mg litre 1 ) C 1. D 1. F-wave persistence.8.6.4.2 F-wave persistence.8.6.4.2. 1 2 3 4 5 C plasma propofol (mg litre 1 ). 1 2 3 4 5 C effect propofol (mg litre 1 ) E 1 F 1 8 8 6 6 4 4 2 2 1 2 3 4 5 C plasma propofol (mg litre 1 ) 1 2 3 4 5 C effect propofol (mg litre 1 ) Fig 2 Individual hysteresis loops (effect vs propofol plasma concentration) and collapsed loops (effect vs effect compartment concentrations) of the different drug effect measures. Original data were collected from all patients. The collapsed curves were generated from the individual equilibration time constants. To estimate the goodness of fit the grey line in the collapsed loops shows the mean concentration response curve of all patients (see Fig. 6). Loops for F-wave amplitude are shown in (A) for plasma concentration and (B) for effect site concentration, for F-wave persistence in (C) for plasma concentration and (D) for effect site concentration and for the in (E) for plasma concentration and (F) for effect site concentration. For the F-wave amplitude the data were averaged within a time-window of 2 s. To estimate and compare the predictive value of the different variables, we calculated the prediction probability (P k ) introduced by Smith and colleagues. 17 P k is a nonparametric correlation measure that indicates the probability that a variable correctly predicts anaesthetic depth, that is, in this case, a movement response. A P k -value of 1. indicates perfect prediction, whereas a value of.5 indicates that the predictive value of the variable is no better than chance 121

Baars et al. alone. The calculation of the P k was based on values of each variable that were averaged more than 1 min before the noxious electrical stimulation. Results In all 24 patients included in the study, F-waves were reliably elicited. The F-wave data of three patients could not be further evaluated as the F-wave amplitude did not recover after reducing the propofol concentration which was probably caused by electrode displacement after uncontrolled movement during propofol wash-out. In four other patients the EEG data could not be evaluated because of a data transmission error that led to a complete loss of the EEG data. The characteristics (mean SD) of the 24 patients (2 female) were an age of 4 12 yr (range 21 5), a body weight of 65 14 kg and a height of 17 1 cm. The increasing propofol concentration caused a clear reduction of the F-wave amplitude and persistence (Fig. 1). When propofol concentration was decreased there was recovery of the F-waves. The sigmoidal model adequately described the concentration-dependent suppression of the F-wave amplitude and persistence. To evaluate the goodness of fit all individual hysteresis loops and the corresponding collapsed loops are presented in Figure 2. The individual EC 5 -values and equilibration half-time values are displayed in Figure 3. The other fitted parameters are presented in Table 1. The average F-wave minimal latency (shortest latency on a series of 1 stimuli) in our study was 49.5 6.3 ms (mean SD) and is in the range of results reported previously (for references see ref. 6). The noxious electrical stimulation at the wrist led to an increase in F-wave amplitude in the abductor pollicis muscle only in those patients that moved after the noxious electrical stimulation (Fig. 4A). After 15 s, the F-wave amplitude reached pretetanic levels again. The time-course of the F-wave amplitude and the around the tetanic stimulation are presented in Figure 4B and C. Figure 5 displays the different sigmoidal concentration response curves constructed with the mean values for the EC 5 and slope parameters together with the logistic regression curves for the movement response to noxious electrical stimulation and for the loss of consciousness. The logistic regression curve of the tetanic electrical stimulation, based on a total of 19 tetanic electrical stimuli (9 positive and 1 negative responses), yielded an EC 5 of 3.75 mg litre 1 (SE 1.1) and a steepness coefficient of 1.86 (SE 1.9). The prediction probabilities of the different parameters for the movement response to painful electrical stimulation are given in Table 2. The logistic regression curve of loss and return of consciousness is based on 4 changes of consciousness (2 loss of consciousness and 2 return of consciousness). Loss and return of consciousness (note: return of consciousness was determined during the study period, i.e. before surgery) mostly occurred during periods of A Equilibration half time (min) B EC 5 (mg litre 1 ) 15 1 5 15 1 5 F-amp F-pers F-amp F-pers Fig 3 EC 5 -values and equilibration half-time derived from the individual sigmoidal pharmacokinetic pharmacodynamic modelling of the examined drug effects. Horizontal bar=mean value., bispectral index (n=21). Table 1 Parameters derived from the sigmoidal PK/PD-modelling of the examined effects. Mean values and standard deviation of the individual fit results from all patients included in the study., bispectral index; E, baseline effect; l, slope parameter of the concentration response relation; r 2, correlation coefficient E l r 2 F-wave amplitude 1.83 (1.72) 2.18 (1.72).78 (.15) F-wave persistence.96 (.6) 4.76 (2.82).73 (.13) 96.3 (2.57) 2.79 (1.74).87 (.11) quickly changing plasma concentration, which resulted in a high interindividual variability of the plasma concentration values. Therefore, the calculated effect site concentration derived from the individual fits, rather than the plasma concentration, was used for this regression calculation. It should be noted that neither loss of consciousness nor return of consciousness occurred at steady state concentrations. The shape of the observed concentration response curve for the F-wave amplitude resembles the logistic regression curve for the change of consciousness. To eliminate the possibility that change of consciousness leads to an abrupt chance of the F-wave amplitude, the time-course of the F-wave amplitude around the time point of changing 122

A 15 F-wave and propofol 1 F-wave amplitude 5 Time Noxious simulus 5 2 4 6 8 1 12 14 Time (ms) B 2.5 2. Movers Non-movers F-wave amplitude (mv) 1.5 1..5. 8 6 4 2 2 4 6 8 Time relative to tetanic stimulation (s) C 7 65 6 Non-mover Mover 55 5 45 4 35 3 8 6 4 2 2 4 6 8 Time relative to tetanic stimulation (s) Fig 4 Response of F-wave amplitude to noxious electrical stimulation. (A) Consecutive F-wave tracings around noxious stimulation in one patient. The noxious stimulation was performed during a calculated propofol plasma concentrations of 4.5 mg litre 1, which equals a calculated brain concentration or 3.5 mg litre 1 (k e.25 min 1 ). The patient did move after stimulation. Note: background noise remained stable during the tetanic stimulation at the wrist which did not cause any recordable artefacts. Response of F-wave amplitude (B) and (C) to noxious electrical stimulation (at t= s; stimulus intensity: 5 Hz, 6 ma, 5 s,.2 ms square wave) for all patients who did not move after stimulation (non-movers) and those patients who did move (movers). Data are means and standard error of the mean. 123

Baars et al. Probability of movement /normalized parameters 1..8.6.4.2 LOC/ROC Response to tetanus F-amp. 1 2 3 4 5 6 c(propofol) (mg litre 1 ) Fig 5 Comparison of the sigmoidal concentration response curves. Comparison of the sigmoidal concentration response curves for suppression of F-wave amplitude (continuous line), F-wave persistence (dashed dotted line) and (dotted line). These curves were generated from the mean values of the computer fits of all individual concentration response curves at each concentration. Dashed and dashed dotted lines show the logistic regression curve for the suppression of the motor response to noxious electrical stimulation and the logistic regression curve for loss and return of consciousness, respectively. Horizontal error bars=standard errors of the EC 5 -values. Note: for the, F-wave amplitude and loss of consciousness/return of consciousness the x-axis displays the effect site concentrations whereas the logistic regression curve for the response to tetanus is based on calculated plasma concentrations. Table 2 Prediction probability P k for movement to painful electrical stimulation for different predictors. Prediction probabilities calculated from 19 reactions (9 positive; 1 negative) to tetanic electrical stimuli. Standard errors were calculated by the jack-knife method., bispectral index P k -value F-wave amplitude.59.13 F-wave persistence.71.12.89.7 consciousness was examined. In order to pool the data from loss of consciousness and return of consciousness the time-course for the return of consciousness was inversed. The 4 min time-course around the change of consciousness averaged for all 2 patients shows a gradual attenuation of the F-wave amplitude (Fig. 6) but no sudden decrease at loss of consciousness. Discussion This study investigated the relationship between propofol concentration and suppression of the spinal F-wave in comparison to an EEG effect, the. Furthermore, the different concentration response curves were related to the suppression of movement to noxious stimulation. The results demonstrate that the same sigmoidal model used for EEG variables can describe the suppression of the F-wave by propofol. The propofol concentration at which the F-wave amplitude and persistence is suppressed by 5% in this study is in accordance with the results published by other groups. 9 11 SE However these in vivo results are at variance with the results obtained by in vitro studies. Matute and colleagues 18 have shown in isolated rat spinal cord preparations that propofol did not suppress a monosynaptic reflex, which reflects motoneurone excitability, even at supraanaesthetic propofol concentrations. This difference may be because of the absence of supraspinal input in the isolated spinal cord preparation. Suppression of F-waves might be partly caused by the reduction of facilitating or enhancement of inhibitory supraspinal input. Our method, however, does not differentiate between spinal or supraspinal contributions to the reduced motoneurone excitability. To distinguish between spinal and supraspinal effects, recordings from spinal motoneurones deprived of any supraspinal input would be necessary, but this is impossible in man. We found that the F-wave amplitude is almost completely suppressed at propofol concentrations below the EC 5 -value for the motor response to tetanic electrical stimulation. Taking into account that the EC 5 -value for the motor response to tetanic electrical stimulation in this study is below the EC 5 -value of 1 mg litre 1 reported for skin incision 19 leads us to conclude that F-wave amplitude or persistence is unsuitable for monitoring immobility under propofol anaesthesia. During sevoflurane anaesthesia the spinal H-reflex, another measure of spinal excitability, but not the F-wave predicted motor responses to painful stimuli better than the. 2 During sevoflurane anaesthesia F-waves were also completely suppressed around the MAC value of sevoflurane. 15 It is therefore likely that F-waves are generally too sensitive to anaesthetic drug effects in humans. A possible explanation for a high sensitivity to anaesthetics 124

F-wave and propofol 1. 8 6 4 2 F-wave amplitude.8.6.4.2 F-wave amplitude normalized to control. 15 1 5 5 1 15 Time relative to LOC/ROC (s) Fig 6 Time-course of F-wave amplitude and relative to loss and return of consciousness. Filled circles, ; open circles, F-wave amplitude normalized to individual control values. Grey lines show the linear regression curves. Data are means and standard errors of the mean. might be the fact that full F-wave production requires only 1 2% of the entire motoneurone pool of the tested muscle. 21 The likelihood that motoneurones will produce F-waves depends upon the summation of multiple excitatory and inhibitory inputs from various sources of the central and peripheral nervous system and so even a small increase in inhibitory input by propofol could lead to a complete suppression of F-wave production. The marked increase of the F-wave amplitude after the noxious stimulation only in those patients that moved indicates that the movement response is associated with an increase in spinal excitability, which can be detected by the F-wave. It is therefore possible that under surgical stimulation the F-wave might be less suppressed by propofol than under our experimental conditions. Nevertheless our data imply that F-waves will not be suitable for monitoring immobility during propofol anaesthesia. In contrast to sevoflurane anaesthesia, in this study the predicted movement to tetanic stimulation during propofol anaesthesia quite well. There is controversy in the literature concerning the prediction of movement in response to noxious stimuli by the during propofol anaesthesia. While earlier studies 22 24 reported a good predictive power a recent study has called these findings into question. 25 In contrast, during sevoflurane anaesthesia the 2 26 probably does not predict movement responses. This difference may either reflect the better correlation of propofol concentration with values or may be because of a greater supraspinal component in the immobilizing effect of propofol compared with sevoflurane. This study is the first to analyse the equilibration halftime for F-wave suppression during propofol anaesthesia. This allows generation of complete concentration response curves and effect site models, and a separation of different pharmacokinetic or pharmodynamic effects. The equilibration half-time found for suppression of the F-wave amplitude was 1.6 times longer than that of the. The phenomenon that effects on spinal cord indices occur slower than effects on EEG variables is in line with the findings in previous studies. Comparing the equilibration half-times of the spinal effects expressed as a multiple of the equilibration half-time yields 1.53, 2.38 and 1.74 for the H-reflex suppression under sevoflurane, 27 for the H-reflex suppression under propofol 12 and F-wave suppression under sevoflurane, 15 respectively. The different equilibration half-time of spinal cord and forebrain effects may originate either from differences in the anaesthetic wash-in and wash-out of two different effect compartments or from different neuronal effects at possibly identical anatomical sites. Evidence for the former is given by studies in rats that indicate substantially less blood flow to the spinal cord than to the brain. 28 In the latter case, the suppression of F-waves by propofol would be only secondary to forebrain effects. Such a possible mechanism seems rather unlikely, as F-waves persist independently of cortical influence. F-wave amplitudes in spastic patients with upper spinal cord injury affecting corticospinal pathways are not different from healthy volunteers. 29 The delay in the spinal action of propofol could explain why reflex movements can be still observed after induction of anaesthesia even though low values indicate deep sedation. In summary we have shown that F-wave persistence and amplitude are abolished at propofol concentrations much lower than those at which immobility to noxious stimulation 125

Baars et al. occurs. This suppression at subclinical propofol concentrations indicates that the F-wave is not a useful tool to measure anaesthetic induced immobility. The PK/PD analysis of our data indicate that there may be different effect sites for suppression of spinal excitability as measured by F-waves and the hypnotic effect of propofol measured by EEG variables such as the. However it remains unclear whether the immobilizing effect of propofol is indeed caused by a direct spinal effect. Acknowledgement Supported by DFG-grant Re1534/2. References 1 Rampil IJ, Mason P, Singh H. Anesthetic potency (MAC) is independent of forebrain structures in the rat. Anesthesiology 1993; 78: 77 12 2 Rampil IJ. Anesthetic potency is not altered after hypothermic spinal cord transection in rats. Anesthesiology 1994; 8: 66 1 3 Antognini JF, Carstens E. In vivo characterization of clinical anaesthesia and its components. Br J Anaesth 22; 89: 156 66 4 Zhou HH, Jin TT, Qin B, Turndorf H. Suppression of spinal cord motoneuron excitability correlates with surgical immobility during isoflurane anesthesia. Anesthesiology 1998; 88: 955 61 5 Fisher MA. AAEM Minimonograph #13: H reflexes and F waves: physiology and clinical indications. Muscle Nerve 1992; 15: 1223 33 6 Mesrati F, Vecchierini MF. F-waves: neurophysiology and clinical value. Neurophysiol Clin 24; 34: 217 43 7 Eisen A, Fisher M. The F wave. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 1999; 52: 255 7 8 Panayiotopoulos CP, Chroni E. F-waves in clinical neurophysiology: a review, methodological issues and overall value in peripheral neuropathies. Electroencephalogr Clin Neurophysiol 1996; 11: 365 74 9 Dueck MH, Oberthuer A, Wedekind C, Paul M, Boerner U. Propofol impairs the central but not the peripheral part of the motor system. Anesth Analg 23; 96: 449 55, table 1 Kakinohana M, Fuchigami T, Nakamura S, Kawabata T, Sugahara K. Propofol reduces spinal motor neuron excitability in humans. Anesth Analg 22; 94: 1586 8, table 11 Kammer T, Rehberg B, Menne D, Wartenberg HC, Wenningmann I, Urban BW. Propofol and sevoflurane in subanesthetic concentrations act preferentially on the spinal cord: evidence from multimodal electrophysiological assessment. Anesthesiology 22; 97: 1416 25 12 Rehberg, B, Dangel C, Baars J. Pharmacodynamic analysis of H-reflex suppression by propofol. Anesthesiology 24; A-3 13 Marsh B, White M, Morton N, Kenny GN. Pharmacokinetic model driven infusion of propofol in children. Br J Anaesth 1991; 67: 41 8 14 Ichikawa T, Yokota T. F wave change by decreased motoneuronal excitability: a sleep study. Bull Tokyo Med Dent Univ 1994; 41: 15 22 15 Baars JH, Kalisch D, Herold KF, Hadzidiakos DA, Rehberg B. The concentration-dependent suppression of F-waves by sevoflurane does not predict immobility to painful stimuli in humans. Br J Anaesth 25; 95: 789 97 16 Waud DR. On biological assays involving quantal responses. J Pharmacol Exp Ther 1972; 183: 577 67 17 Smith WD, Dutton RC, Smith NT. Measuring the performance of anesthetic depth indicators. Anesthesiology 1996; 84: 38 51 18 Matute E, Rivera-Arconada I, Lopez-Garcia JA. Effects of propofol and sevoflurane on the excitability of rat spinal motoneurones and nociceptive reflexes in vitro. Br J Anaesth 24; 93: 422 7 19 Kazama T, Ikeda K, Morita K. Reduction by fentanyl of the Cp5 values of propofol and hemodynamic responses to various noxious stimuli. Anesthesiology 1997; 87: 213 27 2 Rehberg B, Grunewald M, Baars J, Fuegener K, Urban BW, Kox WJ. Monitoring of immobility to noxious stimulation during sevoflurane anesthesia using the spinal H-reflex. Anesthesiology 24; 1: 44 5 21 Dumitru D, Zwarts MJ. Special nerve conduction techniques. In: Dumituru D, Amato AA, Zwards MJ, eds. Electrodiagnostic Medicine, 2nd edn. Philadelphia: Hanley and Belfus, 22; 225 56 22 Leslie K, Sessler DI, Smith WD, et al. Prediction of movement during propofol/nitrous oxide anesthesia. Performance of concentration, electroencephalographic, pupillary, and hemodynamic indicators. Anesthesiology 1996; 84: 52 63 23 Kearse LA Jr, Manberg P, Chamoun N, debros F, Zaslavsky A. Bispectral analysis of the electroencephalogram correlates with patient movement to skin incision during propofol/nitrous oxide anesthesia. Anesthesiology 1994; 81: 1365 7 24 Vernon JM, Lang E, Sebel PS, Manberg P. Prediction of movement using bispectral electroencephalographic analysis during propofol/alfentanil or isoflurane/alfentanil anesthesia. Anesth Analg 1995; 8: 78 5 25 Struys MM, Jensen EW, Smith W, et al. Performance of the ARX-derived auditory evoked potential index as an indicator of anesthetic depth: a comparison with bispectral index and hemodynamic measures during propofol administration. Anesthesiology 22; 96: 83 16 26 Katoh T, Suzuki A, Ikeda K. Electroencephalographic derivatives as a tool for predicting the depth of sedation and anesthesia induced by sevoflurane. Anesthesiology 1998; 88: 642 5 27 Rehberg B, Bouillon T, Gruenewald M et al. Comparison of the concentration-dependent effect of sevoflurane on the spinal H-reflex and the EEG in humans. Acta Anaesthesiol Scand 24; 48: 569 76 28 Archer DP, Ewen A, Roth SH, Samanani N. Plasma, brain, and spinal cord concentrations of thiopental associated with hyperalgesia in the rat. Anesthesiology 1994; 8: 168 76 29 Tsai CT, Chen HW, Chang CW. Assessments of chronodispersion and tacheodispersion of F waves in patients with spinal cord injury. Am J Phys Med Rehabil 23; 82: 498 53 126