| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Intraoperative transcranial motor evoked potential (MEP) monitoring may help prevent neurologic injury during spine surgery. This type of monitoring may be difficult in the pediatric population under general anesthesia. We retrospectively reviewed data from 56 children, aged 2 to 18 yr, who were to undergo surgical correction of idiopathic scoliosis with MEP monitoring. Under combined isoflurane-propofol general anesthesia, before incision, we examined the minimum stimulating threshold voltage required to achieve a 50-microvolt or greater MEP response amplitude. Younger age was associated with an increase in the threshold voltage needed to elicit a sufficient MEP response. In addition, younger age was associated with longer stimulating pulse trains and greater need to adjust stimulating scalp electrodes. Body surface area, height, weight, and body mass index were also significant factors, but they were not independent predictors, after adjusting for age. Younger children received significantly lower levels of isoflurane and comparable doses of propofol, compared with older patients. Stronger stimulation needed to produce MEP responses in younger patients may reflect immaturity of their central nervous system, specifically conduction by the descending corticospinal motor tracts. Greater attention must be given to optimizing physiologic variables, limiting depressant anesthetics, and selecting the most favorable stimulating conditions in children, especially those <10 yr old.
Children undergoing major spine surgery are at risk for developing serious neurologic sequelae. Monitoring transcranial motor evoked potentials (MEP) during these procedures may identify and help prevent injury to motor pathways. MEP responses can usually be detected intraoperatively in neurologically intact children (1,2), but these signals may be missed in 32%61% of children with preexisting neurologic deficits (1,3). The difficulty in obtaining MEP responses during pediatric spine surgery may be attributable in part to the depression of evoked responses caused by volatile inhaled anesthetics or by the sedative-hypnotic propofol. These anesthetics suppress MEP responses in adults in a dose-dependant fashion (4,5), but the effects in children have not been reported. In contrast, depression of somatosensory evoked potentials (SSEP) by these anesthetic drugs has been described for anesthetized pediatric surgical patients (6,7). Age may also influence our ability to elicit MEP responses. In unanesthetized normal children, the stimulating threshold required to obtain transcranial magnetic MEP responses is higher for younger patients (8,9). We are unaware of any studies that have described an association of age with MEP responses in pediatric patients receiving general anesthesia. We retrospectively analyzed data obtained from pediatric patients (218 yr of age) who underwent spine surgery while using MEP monitoring. We specifically evaluated how age affected stimulation variables necessary to elicit myogenic MEP responses to transcranial electric stimulation during general anesthesia but before surgery.
We obtained approval from our institution's Committee for Human Research to review the records of 172 pediatric patients who underwent deformity correction spine-surgery with intraoperative neurophysiologic monitoring at the University of California, San Francisco between January 1998 and January 2004. We restricted our study to patients who had idiopathic scoliosis and excluded patients who had myelopathic, neuromuscular, and congenital spine pathologies. We identified 56 neurologically intact subjects who underwent correction for idiopathic scoliosis and who were monitored using transcranial myogenic MEPs. Age, gender, and weight were recorded for each subject; accurate height was known for only 39 patients. Body Mass Index (BMI; weight [kg]/height2 [cm2]) and Body Surface Area (BSA) (10) were calculated using established formulas. The data from these 56 patients were evaluated and form the basis for this paper. General anesthesia was induced with inhaled sevoflurane or IV propofol. Muscle relaxation was obtained for endotracheal intubation with either succinylcholine or a single dose of a nondepolarizing neuromuscular blocking drug. No additional muscle relaxant was used. In all subjects, anesthesia was maintained with isoflurane plus IV infusions of propofol and fentanyl. After completion of anesthetic induction and tracheal intubation, IV and arterial catheters were inserted, neuromonitoring electrodes were placed, and the patients were positioned for surgery. Complete recovery from neuromuscular blockade was confirmed at this time. The initial doses of administered anesthetic ranged from 0.75% to 1.0% isoflurane and 5075 µg · kg1· min1 of propofol. The anesthesiologist responsible for clinical care of the patient then adjusted anesthetic levels as needed for clinical reasons (e.g., arterial blood pressure). No monitor of cortical activity (e.g., Bispectral Index, BIS) was used on any patient. Multipulse transcranial electrical stimulation was generated with a Digitimer D-185 constant-voltage stimulator (Digitimer LTD, Welwyn Garden City, UK). Stimuli were delivered through two "corkscrew" stimulating electrodes (Nicolet Biomedical, Inc., Madison, WI) placed at C3 and C4 (International 1020 system). The D-185 stimulator was interfaced with an electrophysiological recording platform (Viking IV P [Nicolet Biomedical Inc., Madison, WI] or Cadwell Cascade [Cadwell Laboratories Inc., Kennewick, WA]) so that a "triggered" electromyelographic response was elicited with each stimulus. Electromyelographic (EMG) activity evoked by transcranial stimulation was recorded using subdermal needle electrodes (29-gauge, 1.5 cm length, Medtronic-Xomed, Rochester, MN) placed bilaterally, approximately 46 cm apart, in the thenar-hypothenar muscles of the hands and in the tibialis anterior, extensor hallucis longus, and abductor hallucis muscles of the lower extremities. Recording and filtering parameters were typically 301000 Hz, with a time base of 100 ms. MEP responses were elicited contralateral to anodal stimulation (e.g., right-sided MEPs were triggered after left anodal stimulation). Our typical practice was to begin with a train of 5 pulses, each of 50-µs duration, with an Interstimulus Interval (ISI) set at 2.0 ms. Stimulation intensity started at 150 Volts (V) and was increased by 25 V increments up to a maximum of 400 V. MEP responses were deemed "adequate" when stimulation yielded reproducible response waveforms of sufficient duration and complexity, with amplitudes of at least 50 µV in all muscle groups. If there were no replicable MEP responses of sufficient amplitude after increments of 5075 V, an additional pulse was usually added. In patients in whom voltage had exceeded 300 V, and there was a continued absence of an adequate response, additional stimulating electrodes were usually placed 12 cm posterior to C34, less often 1 cm anterior to C12 or elsewhere. ISI parameters were adjusted to produce the most robust waveform whenever stimulation voltage or pulse train number was changed. The ISI ranged from 1.8 to 4.0 ms. If MEP responses were unobtainable after all technical adjustments had been exhausted, the anesthesiologist was then asked to reduce the dose of the volatile anesthetic drug. All recordings were done before surgical incision. All baseline MEP tracings were reviewed, and it was confirmed by the authors that the amplitudes exceeded 50 µV. The neurophysiologist recorded anesthetic doses at the time baseline MEP recordings were successfully obtained.
A power analysis was not performed, as there are no previous data quantifying the observed changes that we measured. Statistical analysis was performed using JMP 4.0 (SAS Institute, Cary, NC). Within the patient cohort, data were analyzed using analysis of variance with Tukey-Kramer correction for multiple comparisons. We performed multiple linear regression to develop a model of the change in threshold voltage and to investigate the contribution of multiple factors, including age, height, weight, gender, BMI, BSA, and anesthetic dose.
Of the 56 children enrolled in this study, 39 were female and 17 were male. The age range was 2 to 18 yr of age. Adequate baseline MEP responses (amplitude more than 50 µV) were obtained in all 56 subjects. There were no significant differences in body temperature or hypotensive episodes (alteration from baseline) related to age at the time baseline MEP responses were recorded. MEP stimulation threshold voltage was higher in younger patients and decreased with increasing age (Fig. 1; P < 0.0001). There was a similar, but weaker relationship between threshold voltage and increasing BSA (Fig. 2; [R2 = 0.42 versus 0.53 for age], P < 0.0001), weight (data not shown; [R2 = 0.36], P < 0.0001) and height (data not shown, [R2 = 0.27], P < 0.0001). BMI was weakly associated with threshold voltage (data not shown; [R2 = 0.12], P < 0.05). Gender was not a significant factor. Age highly correlated to height, weight, BSA, and BMI (P < 0.001).
Younger patients also required more stimulating pulses, as compared with older patients, to elicit reproducible MEP responses (Fig. 3; [R2 = 0.53], P < 0.0001). We placed additional stimulating electrodes in 48% (n = 26) of the patients; extra electrodes were required only in patients younger than 13 yr old. Younger patients received smaller isoflurane doses, measured as both absolute end-tidal isoflurane concentration or as age-adjusted minimum alveolar concentration (MAC) (11) (Fig. 4; P < 0.0001). MEP stimulation threshold was higher in subjects receiving smaller isoflurane doses (Fig. 5; [R2 = 0.24], P < 0.001). All subjects received similar amounts of IV propofol infusion (50125 µg · kg1 · min1). The dose of propofol showed a barely significant correlation to age (Fig. 6, [R2 = 0.1], P = 0.045) but no relationship to MEP stimulating threshold voltage (P = 0.98). Patients receiving larger doses of propofol received smaller isoflurane doses, by end-tidal concentration and age-adjusted MAC (P < 0.01).
Multivariate analysis showed that age was the dominant predictor of threshold voltage (P < 0.0001). Weight, height, BMI, and BSA were not statistically significant, after accounting for age, in the multivariate model. Isoflurane dose (either absolute end-tidal value or age-adjusted MAC) was not a statistically significant factor in the multivariate analysis. Propofol dose was just significant at P = 0.02.
These data show a strong association between age and MEP threshold requirements. The younger the patient, the more stimulating voltage is required to elicit a motor response. The association of height, weight, BMI, and BSA to MEP threshold can be explained by the close relationship of these variables to age. We observed that younger patients also received more stimulating pulses in each train than older patients. However, this association is not independent, as pulses were only added when patients required a high stimulation threshold voltage. The age-related difference in stimulating voltage cannot be attributed to larger concentrations of suppressive anesthetics administered to younger subjects. In fact, significantly less isoflurane was administered to the younger subjects. Propofol, as dosed by weight, showed a barely significant relationship to age, but the doses administered did not differ substantially within the study population (50 to 125 µg · kg1 · min1 in all subjects). This range of propofol dosage has not been shown to substantially depress MEP responses. Several studies have shown that younger children require larger doses of propofol for induction of anesthesia and more rapid infusion rates of this drug for maintenance of anesthesia than older subjects (12,13). Total anesthetic depth is not known. We lacked sufficient detail of propofol dosing to reasonably estimate plasma or brain levels. "Anesthetic depth" measurements, as recorded by BIS or entropy monitors, were not available. However, these indices would be difficult to interpret as suppression of MEPs by isoflurane and propofol differ significantly at comparable BIS levels (14). Younger subjects received less isoflurane, at the discretion of their anesthesiologist. We suspect that this was done to maintain patient arterial blood pressure close to baseline levels or in response to high stimulation voltage needs reported by the neurophysiologist. The multivariate analysis suggests that larger propofol dosing was weakly associated with lower MEP threshold requirements. We attribute this to the anesthesiologist choosing to increase propofol infusion rates in subjects whose isoflurane was reduced secondary to high threshold voltage needs. The highly significant interrelationships between the predictor variables make conclusions from the multivariate analysis uncertain. We were primarily interested in demonstrating that age was a dominant variable, and that the relationship between age and threshold voltage was not explained by a confounding variable, such as anesthetic regimen. We believe that the analysis has demonstrated this. In a retrospective study, we could not control for all physiologic variables that might affect MEP responses. Arterial blood pressure was not consistently recorded at the time we obtained baseline MEP responses, but no hypotensive episodes were noted on the anesthetic record. This is notable, as hypotension may diminish MEP responses (15,16). The normal arterial blood pressure of younger subjects tends to be lower. Our data were collected before surgical incision and an acceptable arterial blood pressure was maintained per the anesthesiologist. Temperature, another physiologic factor that affects MEP responses (17), did not differ by age at the time MEP measurements were made. Variation in the hematocrit of our subjects might affect MEP signals. However, isovolemic anemia does not alter somatosensory evoked potentials signals (18), but the effect on MEPs is less clear. Intraoperative decreases of MEP signals have improved when anemia was corrected (19), but other physiologic factors (e.g., arterial blood pressure) were simultaneously augmented. Preoperative hematocrit values were rarely available in our patients, but our selection of otherwise healthy children minimized their likelihood for occult anemia. Furthermore, baseline MEP responses were obtained before skin incision and before any blood loss. "Anesthetic fade" was not a factor, as the data were collected early in the anesthetic course (20). Perhaps the key limitation of this study is that it was not designed to test maturational effects of MEP responses. Our age range was limited to children aged 2 years and older. This reflects the absence of corrective surgery for asymptomatic idiopathic scoliosis offered to children younger than 2 years of age at our institution. Our analysis focused on defining the minimum parameters needed to generate interpretable responses rather than on examining the characteristics of the responses. We might better identify age-related differences by stimulating all subjects in the same fashion, then comparing response amplitudes, latencies, and morphology. Optimal stimulating electrode placement might differ in younger versus older subjects, contributing to higher threshold voltage requirements. However, our strategy of mapping out more optimal electrode sites when threshold voltage exceeded 300 V may ameliorate this as a factor. Another issue not explored was current direction. We used traditional anodal stimulation to elicit contralateral muscle responses (21). Anodal transcranial magnetic stimulation can produce more attainable ipsilateral MEP responses than contralateral ones in some children younger than 10 years of age (22) and in hemiplegic cerebral palsy patients (23). The association of younger age with higher MEP stimulus requirements is supported by other studies. Higher thresholds are needed when using transcranial magnetic MEPs in awake children versus adults (8,9,24). Parano et al. (25) observed diminished evoked response amplitudes in infancy and childhood compared to adults; this difference was more pronounced in the first 2 years of life. Reliable MEP responses may be unobtainable in children younger than 6 years, even when they are awake (8,9). The minimum threshold needed to elicit a motor response during transcranial magnetic stimulation does not decrease to adult levels until the age of 16 years (24). The higher MEP threshold we observed, and the increased variability of MEP responses seen by others, may be related to immaturity of the motor pathways in younger children. Maturational changes with age include improved synaptogenesis, greater nerve integration, and increased myelination (26). Our study results can best be explained by immaturity of the central nervous system. Maturation of the peripheral nervous system proceeds faster than central pathways and peripheral nerves conduct at adult velocities by the age of 3 years (27). Cortical changes with aging may affect MEP responses. Hagelthorn et al. (28), observed decreased evoked potential inter-hemispheric transmission time with increasing age (717 years), suggesting increased corpus callosal myelination and integration during childhood. A linear increase in axonal density also occurs with increasing age (8 to 18 years) in both the temporal and the frontal regions of the brain (29). These increases correlated with the development of specific cognitive functions. Klingberg et al. (30), comparing fractional anisotropy measurements in children (mean age of 10 years) to adults (mean age of 27 years), found significantly less myelination of the frontal white matter of the children. Spinal cord motor pathways also undergo a prolonged period of maturation. Nezu et al. (8) estimated that electrophysiologic maturity of the corticospinal tracts (CST) innervating the hand muscles was complete by 13 years of age. Unlike other pathways, the CSTs appear to undergo a protracted period of myelogenesis and synaptogenesis, which continues well into the second decade of life. Myelination in the spinal cord begins in the second trimester of pregnancy, yet the CST is the only pathway of the spinal cord not myelinated by birth (31,32). It has been estimated from the diameter of corticospinal axons that they do not reach full myelination until 16 years of age (33). At birth, the conduction velocity of central motor fibers of the spinal cord are approximately 10 m/s, whereas adult values are in the range of 5070 m/s (34). This is the first study to show an association between age and MEP stimulation requirements under general anesthesia. Heightened threshold voltage may overcome the lack of myelination and synaptogenesis by depolarizing more cortical volume or by activating the CST at deeper levels, such as the cerebral peduncle and pyramids (35). The longer train of stimulating pulses in younger patients may help overcome the temporal dispersion (lack of synchrony) of signals traveling down the CST. A longer train of high-frequency, repetitive stimuli fosters better temporal summation at the anterior horns (36). Comparisons of our results with other investigators are difficult because of different anesthetic regimens, transcranial stimulation techniques, and response variables. We cannot conclude that these observations are attributable solely to maturation effects. Younger patients may have enhanced sensitivity to suppression by volatile anesthetics or propofol. Technical challenges with performing MEP may also be relevant in the young. We reasonably conclude that it is more difficult to generate MEP responses in younger children under combined propofol/isoflurane anesthesia. We note that more than 90% of our patients 2 to 10 years old needed baseline threshold voltages at or above 300V, but only 30% of children older than 10 years required voltages this high. Stimulating threshold voltage requirements may further increase during the course of surgery (20,37), especially with changes in physiologic variables, such as arterial blood pressure, temperature, and hematocrit. This suggests that younger children are at greater risk for diminished or lost MEP responses during surgery when using this anesthetic regimen. The selection of the anesthetic regimen for any child must consider all of the desired anesthetic and surgical goals, including the effects on MEPs. Volatile anesthetics are potent hypnotics whose levels can be monitored and adjusted quickly. However, they depress cortical activity and are potent muscle relaxants. To improve MEP monitoring, one may consider reducing volatile anesthetics or using total IV anesthesia. Combined propofol and fentanyl anesthesia has been used successfully for obtaining both myogenic and epidurally recorded MEPs in a pediatric population as young as 8 to 12 months (36). In addition, minimally suppressive drugs, such as ketamine (38) or etomidate, may help to reduce the need for volatile anesthetics. Improved MEP stimulation techniques may be especially useful for obtaining MEP responses in very young subjects (38,39). A prospective study is warranted to test these observations, including in children younger than 2 years old.
Accepted for publication April 17, 2006. Reprints will not be available from the author. Supported by the Department of Anesthesia & Perioperative Care, University of California.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|