| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The usefulness of intraoperative monitoring of motor-evoked potentials (MEPs) during inhaled anesthesia is limited by the suppressive effects of volatile anesthetics on MEP signals. We investigated the effects of different stimulation patterns and end-tidal concentrations of sevoflurane on intraoperative transcranial electrical MEPs. In 12 patients undergoing craniotomy, stimulation patterns (300500 V, 1001000 Hz, 15 stimuli) and multiples (0.5, 0.75, and 1.0) of minimum alveolar concentration (MAC) of sevoflurane were varied randomly while remifentanil was administered at a constant rate of 0.2 µg · kg1 · min1. MEPs were recorded from thenar and hypothenar muscles and analyzed without knowledge of the respective MAC. Three-way analysis of variance revealed significant main effects for increasing stimulation intensity, frequency, and number of stimuli on MEP amplitude (P < 0.05). Maximum MEP amplitudes and recording success rates were observed during 4 stimuli delivered at 1000 Hz and 300 V. A significant main effect of sevoflurane concentration (0.5 versus 0.75 and 1 MAC multiple) on MEP amplitude was observed at the thenar recording site only (P < 0.05). In conclusion, MEP characteristics varied significantly with changes in stimulation pattern and less so with changes in sevoflurane concentration. The results suggest that high frequency repetitive stimulation allows intraoperative use of MEP monitoring during up to 1 MAC multiple of sevoflurane and constant infusion of remifentanil up to 0.2 µg · kg1 · min1.
Use of motor-evoked potentials (MEP) has been established in intraoperative neurophysiological monitoring of the descending motor pathways because of the capability of detecting impending iatrogenic lesions within the motor system at an early, potentially reversible stage. However, myogenic MEPs are significantly suppressed by general anesthesia, particularly by volatile anesthetics (18). The ability to obtain reproducible responses during intraoperative MEP monitoring has been considerably improved by the introduction of repetitive high-frequency stimulation devices and specifically modified total IV anesthesia protocols (4,914). Whereas the effects of total IV as well as inhaled anesthesia (including isoflurane, halothane, enflurane and desflurane) on intraoperative myogenic MEPs during transcranial electrical stimulation have been reported (13,5,79,15), the effect of sevoflurane has only been studied during single stimulus and paired stimuli (16). When sevoflurane was administered at clinically relevant concentrations, transcranial stimulation of the motor cortex with either single stimulus or paired stimuli did not overcome the depressant effects of sevoflurane on MEPs (16). The present study was designed to further define the effect of different stimulation patterns (induced by changes in stimulus number, stimulation frequency, and intensity) and different end-tidal concentrations of sevoflurane on MEPs evoked by repetitive transcranial electrical stimulation during intracranial neurosurgical procedures. We hypothesized that increasing sevoflurane concentrations produce significant dose-dependent suppressive effects on MEP amplitude, which may be partially overcome by choosing appropriate stimulation patterns.
The study protocol complies with the declaration of Helsinki. After approval by the local Ethics Committee and the patient's informed written consent, the motor pathways were monitored prospectively by transcranial MEP in 12 patients (6 male, 6 female; ASA physical status IIIII: n = 5/7; mean age: 48 yr [range, 2561 yr]; mean weight: 72 kg [range, 6194 kg]; mean height: 171 cm [range, 164187 cm]) during supratentorial intracranial procedures. Pre-existing impairment of the muscle groups under investigation was excluded by preoperative clinical and neurophysiological assessment, including electromyography. All patients were premedicated with 7.5 mg midazolam per os approximately 1 h before induction of anesthesia. No other centrally acting drugs were administered. On arrival in the operating room, catheters were inserted in peripheral veins and the radial artery on the arm opposite of the MEP recording site for administration of fluids and drugs (including IV anesthetics), continuous recording of mean arterial blood pressure (MAP) and regular blood sampling for blood gas analysis (ABL®; Radiometer, Copenhagen, Denmark), respectively. Peripheral oxygen saturation (Spo2) and depth of anesthesia were monitored continuously via pulse oximetry (Siemens, Erlangen, Germany) and electroencephalograph EEG bispectral index (17) (BIS Monitor; Aspect Medical Systems, Newton, MA), respectively.
Anesthesia was induced by continuous IV infusion of remifentanil (0.20.5 µg · kg1 · min1) and propofol (11.5 mg/kg). Cisatracurium (0.1 mg/kg) was administered to facilitate endotracheal intubation. To assure systemic arterial oxygen partial pressures (Pao2) >100 mm Hg at all times, fractional inspired oxygen concentration (Fio2) was administered at a minimum of 0.5 in air and adjusted to maintain Spo2 at Hypotension (MAP <60 mm Hg in normotensive or <70 mm Hg in hypertensive patients, respectively) and bradycardia (heart rate <45 bpm) were treated with IV administration of Akrinor® (12 mL IV of a 2:8 Cafedrin-HCl/Theodrenalin-HCl normal saline mixture) and 510 mg etilefrin (Effortil®; Boehringer Ingelheim, Basel, Switzerland), respectively. All vital signs were recorded continuously (SC 9000®; Siemens). Fluid administration was guided by central venous pressure and urine output. All anesthetics were administered by two of the investigators (W. B., H.-J. P.). In each patient, the end-tidal concentration of sevoflurane was varied randomly (sealed envelope technique) corresponding to 0.5, 0.75 and 1.0 multiples of minimum alveolar concentration (MAC) of sevoflurane. The reported age-adjusted sevoflurane MAC values vary considerably. For the age brackets 1825 yr, 2640 yr, and >40 yr we defined 1 MAC as 2.6%, 2.2%, and 1.8%, respectively (18,19). After each change in sevoflurane concentration, the end-tidal concentration of sevoflurane had to have remained unchanged at the selected concentration, and the continuously monitored MAP and heart rate values had to have remained within a 5% range for 15 min before MEPs were recorded. Throughout the investigation, remifentanil was administered unchanged at 0.2 µg · kg1 · min1.
MEPs were recorded by standard neurophysiological equipment (Spirit® evoked potential system; Nicolet Biomedical, Madison, WI). Compound muscle action potentials were derived from subdermal needle electrodes placed in the abductor pollicis brevis (thenar) and abductor digiti minimi (hypothenar) muscles using a belly-tendon montage. Before commencing with MEP recordings, complete recovery from neuromuscular blockade was verified by train-of-four and double-burst stimulation. Electrode impedances below 5 k Each recording cycle assessed MEPs in response to a) variation in stimulation current (100, 150, 200, 250, 300 V) at constant stimulation frequency (500 Hz) and constant number of stimuli (n = 4), b) variation in stimulation frequency (100, 200, 500, 1000 Hz) at constant stimulation intensity (300 V) and constant number of stimuli (n = 4), or c) variation in number of stimuli (n = 15) at constant stimulation intensity (300 V) and constant stimulation frequency (500 Hz). Thus, 12 different stimulation patterns were investigated at each of the 3 sevoflurane concentrations (Table 1). Two consecutive MEP recordings (separated by 30 s) were performed after each change in stimulation pattern (i.e., 12 x 2 measurements per MAC). To exclude conditioning effects of repetitive stimulation on MEP characteristics, the stimulation pattern was varied randomly.
Amplitudes were measured from peak to peak. Latencies were defined as the interval between onset of stimulation artifact and onset of the MEP. MEP amplitudes and latencies were assessed individually by two investigators blinded for the sevoflurane concentrations (P. R., K-M.S). Data from the two investigators were averaged. Because assessment of each possible combination among stimulation variables (voltage, frequency, number of stimuli) and end-tidal sevoflurane concentrations (0.5, 0.75, and 1 MAC multiples) in terms of a complete factorial design was not feasible, our study protocol comprised a reduced factorial design. The individual effects of variation in stimulation pattern and MAC multiple on MEP characteristics were evaluated independently. Each recording cycle consisted of 12 different stimulation patterns applied twice in random order after each change in end-tidal sevoflurane concentration. Thus, 24 measurements were done at each MAC multiple in each patient. With 3 changes in end-tidal concentration of sevoflurane 72 measurements were performed in each individual patient. A three-way within-subjects (repeated measures) analysis of variance was conducted. MAC multiples, stimulation variables (voltage, frequency, and number of stimuli) and measurement replication were treated as dependent variables (repeated measures). The 3 repeated measures factors consisted of a) MAC multiple (3 levels: 0.5, 0.75, and 1), b) stimulation pattern (5 levels for voltage: 100-150-200-250-300 V and number of stimuli: 1-2-3-4-5 stimuli; 4 levels for stimulation frequency: 100-200-500-1000 Hz), and c) replication of each measurement within the randomized stimulation protocol at each MAC multiple (i.e., 2 levels). This resulted in a reduced factorial 3 x 5 (4) x 2 design adjusting for repeated measures. In addition to the assessment of each factor's main effect (i.e., collapsing over the remaining factors) as well as two-way and three-way interactions between the repeated measures factors (analysis of variance), the specific effect of each change in stimulation intensity, frequency, and stimulus number on MEP characteristics was evaluated at each MAC multiple and measurement replication by assigning contrast coefficients to the various levels of each factor (contrast analysis). Data derived from thenar and hypothenar muscles are presented separately as required by statistical methodology. Physiological data are expressed as mean ± sd or as median and range for data that were not normally distributed as shown by Shapiro-Wilk test. Friedman's analysis of variance was used to detect differences between corresponding data during variation of MAC multiples. Statistical significance was assumed for P < 0.05. Statistical analysis was performed using the SYSTAT 11 package (Systat Software, Inc., Point Richmond, CA).
As required by protocol, end-tidal sevoflurane concentrations differed significantly among the three groups (Table 2). Bispectral index values were less than 65 at all times and decreased dose-dependently (P < 0.05). Heart rate, Pao2, Paco2, arterial pH and rectal temperature were not significantly different (P > 0.05) among the three end-tidal concentrations (Table 2). At 1 MAC equivalent, MAP was approximately 15% less than during 0.5 MAC equivalent. In 5 patients, 6 bolus administrations of 12 mL of the cafedrin-HCl/theodrenalin-HCl/saline mixture were required (2 during each of the 3 sevoflurane concentrations) to stabilize MAP. One patient received 2 bolus administrations of 5 mg etilefrin (at 0.5 and 0.75 MAC multiples) to maintain MAP >60 mm Hg and heart rate >45 bpm, respectively. No MEP measurements were made at MAP <60 mm Hg.
MEP amplitudes (range 5 µV4.97 mV) and latencies (20.853.6 ms) varied primarily with stimulation pattern (intensity, frequency, number of stimuli), MAC multiple, and (to a lesser degree) between thenar and hypothenar muscles, i.e., the average MEP amplitudes at hypothenar being slightly (not significant) lower than at thenar. The average duration of intraoperative data acquisition was 52 min. Measurement repetition (measurement 1 versus 2; Table 3) conducted during each stimulation pattern did not cause statistically significant effects.
The results of three-way repeated-measures analysis of variance (fixed effects) for MAC multiple and stimulation intensity (voltage) were as follows. Stimulation intensity (thenar and hypothenar; Fig. 1A), but not end-tidal sevoflurane concentration had a significant main effect (P < 0.001) on MEP amplitude. Contrast analysis revealed significant two-way interaction between the first (MAC multiple) and the second (voltage) factor (P < 0.001; thenar only).
Both end-tidal sevoflurane concentration and voltage displayed significant main effects on MEP latency (Fig. 1B). Contrast analysis revealed no significant two- or three-way interaction between factors. Whereas stimulation with 100 V evoked a reproducible myogenic response in only 1 of 72 attempts (1.4%), incremental increases in stimulation intensity increased MEP amplitudes and recording success rates. At 300 V, delivery of 4 stimuli with a stimulation frequency of 500 Hz resulted in recording success rates of 81% (hypothenar) and 82% (thenar) in all patients irrespective of end-tidal sevoflurane concentration. The results of three-way repeated-measures analysis of variance (fixed effects) for MAC multiple and number of stimuli were as follows. MAC multiple (factor 1) and number of stimuli (factor 2) showed apparent main effects on MEP amplitude (P < 0.01; Fig. 2A). However, significant interactions between MAC multiple and number of stimuli (thenar; P < 0.001) render the effects of end-tidal sevoflurane concentration and stimulus number on MEP amplitude statistically insignificant. Neither significant three-way interactions nor significant main effects on MEP latency were observed (Fig. 2B, Table 3).
An increase in the number of stimuli The results of three-way repeated-measures analysis of variance (fixed effects) for MAC multiple and stimulation frequency were as follows. MAC multiple (factor 1) and stimulation frequency (factor 2) showed apparent main effects (P < 0.001) on MEP amplitude (Fig. 3A) and latency (Fig. 3B). However, interaction between factor 1 and factor 2 renders these findings statistically insignificant at thenar, but not hypothenar recording sites.
Incremental increases in stimulation frequency yielded progressive increases in MEP amplitudes and recording success rates at each MAC multiple. Maximum recording success rates of up to 90% (hypothenar) and 97% (thenar), respectively, were achieved with stimulation trains delivered at 1000 Hz (4 stimuli, 300 V) in all patients at each MAC multiple. Regarding the combinations of stimulation variables under investigation, maximum MEP amplitudes and recording success rates were achieved by a train of four repetitive stimuli delivered at 300 V and 1000 Hz. At 0.5 and 0.75 MAC multiples, the recording success rates were 100% (thenar) and 88% (hypothenar), respectively. At 1 MAC multiple, the recording success rates decreased slightly to 92% (thenar) and 82% (hypothenar), respectively.
The two main findings of this study are 1) sevoflurane modifies MEP amplitudes in a dose-dependent manner and 2) recording quality and success rate depend mainly on stimulation pattern. In all patients and at all MAC multiples, maximum recording success rates were achieved after stimulation with 4 stimuli, delivered at 1000 Hz and 300 V. Each stimulation variable (stimulation intensity, number of stimuli, stimulation frequency) significantly affected MEP amplitude at both muscle groups (thenar and hypothenar). Generally, the variability of absolute MEP amplitudes observed between different recording sites during MEP monitoring remains a critical factor during interpretation of intraoperative signal changes. However, the minor differences in stimulus-response characteristics observed between thenar and hypothenar muscles in this study may be primarily attributed to the limited raw data entering multivariate analysis. Several advances in electrophysiological and anesthetic technique have significantly improved the clinical utility of intraoperative MEP monitoring (2,3,9,15,20). Confidence in the safety of repetitive transcranial electrical stimulation is growing (21). We have not observed any adverse event (e.g., induction of seizures, cardiac arrhythmias) related to repetitive high frequency electrical stimulation, even in patients surgically treated for seizure disorders. This impression is supported by the lack of seizure induction during transcranial high frequency repetitive magneto-electrical stimulation in awake patients with epileptic seizure disorders (22). Transcranial high-frequency repetitive electrical stimulation has emerged as the preferred technique for the recording of myogenic MEPs using various anesthetic techniques (4,1014). Continuing improvement in intraoperative MEP acquisition warrants further systematic evaluation of stimulation pattern-dependent and anesthetic-dependent modulation of MEPs recorded under standardized clinical conditions. Considering the absolute refractory period of axonal conduction of different types of motor units (0.58 and 0.88 ms), stimulation frequency is limited to approximately 1000 Hz. Under certain conditions (e.g., myelopathy), use of up to 6 repetitive stimuli applied at lower frequencies (200500 Hz) may yield superior recording success rates, whereas increases in stimulation intensity beyond 250 V appear not to increase recording success. Consequently, the optimal stimulation pattern likely varies with preoperative neurological status and will need to be tailored individually. The effects of total IV (combination of propofol and opioids) and inhaled anesthesia with isoflurane, halothane, enflurane, and desflurane on myogenic MEPs have been reported (13,5,79,15). There are only few experimental (6,23,24) and clinical data (16,25) on the effect of sevoflurane on MEPs. MEPs have been successfully recorded intraoperatively during inhaled anesthesia with nitrous oxide in oxygen, fentanyl, and 0.751.5 MAC sevoflurane during stimulation of the exposed motor cortex with 5 rectangular pulses at 500 Hz (24). However, when delivering single transcranial electrical stimuli, MEPs could only be obtained during administration of small sevoflurane concentrations (0.25 and 0.5 MAC) with recording success rates of merely 67% and 22%, respectively (16). The use of paired stimuli delivered at 500 Hz increased the recording success rates up to 90%, 60%, and 15% during administration of 0.25, 0.5, and 0.75 MAC sevoflurane, respectively. Further increases in sevoflurane concentration up to 1 MAC completely abolished the MEPs. Thus, paired stimuli appear to be insufficient to overcome the depressant effects of clinically relevant sevoflurane concentrations. Improvements in intraoperative MEP signal quality are likely to require modifications in stimulation technique, particularly using trains of multiple repetitive stimuli. Using a train of four stimuli applied at 300 V and 1000 Hz, we were able to achieve MEP recording success rates between 82% and 100% under balanced anesthesia with up to 1 MAC multiple of sevoflurane during continuous infusion of remifentanil. These recording success rates were achieved in intact motor tracts. It is likely that preexisting motor deficits will reduce monitoring success regardless of the use of optimized stimulation pattern and anesthetic technique (13). In conclusion, the findings obtained during this study suggest that constant-current transcranial electrical stimulation using multiple stimuli delivered at high frequency and high intensity yields clinically useful monitoring results under inhaled anesthesia with an end-tidal sevoflurane concentration equivalent to 1 MAC multiple during remifentanil infusion of up to 0.2 µg · kg1 · min1. We gratefully acknowledge the considerable contribution of Dr. J. Schulte-Mönting, Professor of Biometry and Statistics, University of Freiburg to the statistical data analysis.
Supported, in part, by a grant from the German Research Foundation (Ze 267/3-2). Accepted for publication September 27, 2005.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|