Introduction

Motor-evoked potentials (MEP) play a pivotal role in spine surgery by facilitating real-time confirmation of preserved nerve functions, thereby improving postoperative motor outcomes and preserving excretory functions. Most anaesthetics suppress neurotransmission across synapses, resulting in suppression of the amplitudes of the signals elicited at the MEP. This inhibitory effect on MEP occurs when inhalational anaesthetics are administered at a minimum alveolar concentration (MAC) greater than 0.50.1 Therefore, the standard anaesthetic management for surgeries with MEP requires minimal suppressive effects on monitoring, where total intravenous anaesthesia (TIVA) with propofol and opioids (eg, fentanyl and remifentanil) without neuromuscular blockade is commonly selected.1

In infants undergoing spinal surgery, MEP serves as an essential tool for monitoring nerve conservation. However, infants require a more limited use of anaesthetics because of their higher sensitivity to anaesthetics due to immature neural structures and delayed metabolism of anaesthetics caused by immature liver and kidney functions. This suppressive effect of anaesthetics on synaptic neurotransmission due to immature neural structures increases the stimulation voltage required to elicit appreciable MEP amplitudes.2 Due to the difficulty in eliciting sufficient signal amplitude monitoring, anaesthesia management with propofol-based TIVA is considered essential for infants. However, immature liver function presents difficulties in metabolising propofol, limiting its intraoperative usage and potentially leading to propofol-related complications such as propofol infusion syndrome and delayed postoperative emergence.

Current Japanese guidelines state that supplementary inhalational anaesthetics of ≤0.50 MAC are acceptable to minimise the usage of propofol in children.3 Moreover, previous studies reported successful anaesthetic regimens including intravenous and inhalational anaesthetics.4–6 There may be a publication bias where successful anaesthetic management tended to be more published. In addition, we previously reported the case of an infant whose MEP was significantly suppressed when sevoflurane, with an age-adjusted MAC of 0.10–0.15, was combined with a propofol-based TIVA.7 Another case series of 19 infants retrospectively evaluated the impact of various anaesthetic regimes on MEP. The study demonstrated that adding even low-dose sevoflurane to propofol-based TIVA suppressed MEP in infants.8

This randomised controlled study aims to evaluate the suppressive effect of low-dose sevoflurane (0.10–0.15 age-adjusted MAC) with propofol-based anaesthesia on MEP. The study hypothesis posited that low-dose supplementary sevoflurane in propofol-based anaesthesia decreases MEP amplitudes.

Methods and analysisStudy design and setting

This randomised controlled study is planned to be conducted at a single tertiary care children’s hospital in Japan between July 2024 and June 2029.

Target population

Inclusion criteria: We will enrol consecutive cases that meet all of the following criteria:

Patients with a postconceptual age of 35–87 weeks (gestational age of 37–89 weeks).

Scheduled spinal surgery requiring intraoperative neuromonitoring IONM(ie, spinal dysraphism, spinal cord tumours, tethered spinal cord).

Exclusion criteria: We will exclude cases that meet at least one of the following criteria:

Contraindications to sevoflurane or propofol for general anaesthesia.

Haemodynamical instability during general anaesthesia.

Signs of insufficient plane of anaesthesia (eg, body movements and detrimental airway reflexes).

Administration of muscle relaxants before the completion of transcranial MEP (TcMEP) data collection.

American Society of Anesthesiologists-Physical Status ≥III.

Presence of preoperative gross motor dysfunction in at least one extremity.

Refusal of participation by the patient’s guardians after receiving detailed medical information.

Duplicate cases during the study period.

Patient and public involvement

None.

Anaesthesia management

No premedication will be administered to patients. Inhalational anaesthesia induction will be performed with sevoflurane (inhalational concentration 5–8%) and 40% nitrous oxide mixed with oxygen. After establishing peripheral intravenous access in the upper extremity, patients will receive boluses of propofol (2.0 mg/kg) and fentanyl (2.0 mcg/kg) of fentanyl (designated as T0). Muscle relaxants will be withheld until the completion of TcMEP data collection.

Following the propofol and fentanyl boluses, nitrous oxide and sevoflurane will be terminated. Continuous infusion of propofol (4.0–6.0 mg/kg/hour) and remifentanil (0.50–1.0 mcg/kg/min) will be initiated for maintenance of general anaesthesia. After securing the airway with a size-appropriate tracheal tube (3.0 or 3.5 mm), mechanical ventilation will be initiated with 6.0 L/min oxygen to eliminate nitrous oxide and sevoflurane. After establishing mechanical ventilation, the patients will be positioned prone, and the neurosurgeons and clinical engineers will then attach the TcMEP. At least 20 min after initiating the mechanical ventilation with 6.0 L/min of 40% air mixed with oxygen, the TcMEPs will be recorded as control data (T1) after confirming that the end-tidal sevoflurane concentration is at undetectable level by the case-assigned anaesthesiologists.

In the intervention group, the assigned anaesthesiologists will administer 0.10–0.15 age-adjusted MAC of end-tidal sevoflurane concentration under mechanical ventilation with 3.0 L/min of 40% air mixed with oxygen under continuous infusion of propofol (4.0–6.0 mg/kg/hour) and remifentanil (0.50–1.0 mcg/kg/min) at the exact dosages (figure 1). After 20 min of sevoflurane administration, if the assigned anaesthesiologist confirmed that the end-tidal sevoflurane concentration was 0.10–0.15 age-adjusted MAC, the second TcMEP (T2) recording will be performed (figure 1). In contrast, in the control group, general anaesthesia will be maintained with propofol (4.0–6.0 mg/kg/hour) and remifentanil (0.50–1.0 mcg/kg/min) without sevoflurane for 20 min after the first TcMEP recording (T1). All recordings will be completed before the incision of the epidural membrane in the surgical field. Our pilot study in 10 infants demonstrated the feasibility of obtaining T1 and T2 measurements under this time constraint without hindering surgical proceedings. Anaesthesiologists are restricted from administering anaesthetics that can affect MEP—such as ketamine, dexmedetomidine and methadone—until data measurement. The intervention will be discontinued in the following potential situations: unstable haemodynamics, participant family’s request and other situations that may require discontinuing the intervention.

Figure 1

Protocol of anaesthesia regimens and timing of transcranial motor-evoked potential (TcMEP) recordings. T0: time at anaesthesia induction facilitated with boluses of 2.0 mg/kg of propofol and 2.0 mcg/kg of fentanyl . T1: time to first TcMEP recording after at least 20 min from T0 with an undetectable level of end-tidal sevoflurane. T2: time at the second TcMEP recording after at least 20 min from T1 with a 0.10–0.15 age-adjusted minimum alveolar concentration (MAC) of the end-tidal level of sevoflurane.

Procedure of TcMEP

An intraoperative MEP measurement system (Neuromaster-MEE1232; Nihonkoden, Tokyo, Japan) will be used to record TcMEP. The stimulating electrodes of a pair of 20 mm plated disks will be placed with conductive paste at 1–2 cm anterior sites of C3 (cathode) and C4 (anode) on the international 10-20 system, targeting the motor cortex area in infants.9 A train of five to eight square-wave pulses will be delivered at an interval of 2.0 m/s at a stimulation frequency of 0.9 Hz. The intensity of stimulation will be determined during the first TcMEP recording for control data, where the intensity was set supramaximal to each stimulus (ie, 300–600 V). TcMEP detectors will be placed on both the upper extremities (bilateral adductor pollicis muscles) and lower extremities (quadriceps femoris, tibialis anterior and gastrocnemius muscles).

Data collection

The patient’s underlying characteristics will be reviewed using electronic medical and surgical records (HAPPY ACTIS, Canon Medical System, Tochigi, Japan). Additionally, electronic anaesthetic records (Fortec ORSYS, Phillips, Amsterdam, Netherlands) will be reviewed to determine the patient’s underlying characteristics (table 1).

Table 1

Details of the data collection forms regarding the characteristics of patients

Intraoperative vital signs at the moment of the first (T1) and second (T2) TcMEP recordings (ie, blood pressure, heart rate, peripheral capillary oxygen saturation and body temperature TcMEPs) will be performed using Neuromaster-MEE1232 (Nihonkoden). The amplitudes (mA) between the top and bottom and waveforms of TcMEPs obtained from both upper and lower extremities will be recorded in Neuromaster-MEE1232 (Nihonkoden).

Random allocation and masking

Patients will be randomly allocated to one of the control or intervention groups using simple randomisation. The primary investigator (TK) will write codes to create a 1-to-1 randomisation table by computer randomisation (Stata V.18.0, StataCorp, College Station, Texas, USA). A research assistant (YS) will run the codes to create the randomisation table and consecutively allocate the patients based on the random allocation table. The research assistant (YS) will write the allocated group on a small piece of paper wrapped in aluminium foil, which will be concealed in an envelope. Case-assigned anaesthesiologists will open this envelope and follow the anaesthesia regimen outlined in the study protocol based on the allocated group of patients. The research assistant (YS) will confirm that the case-assigned anaesthesiologists strictly followed the anaesthetic regimens as specified in the protocol during the recording.

The collected data of voltage of TcMEP at T1 and T2 (ie, endpoint data) for each measurement site (ie, bilateral adductor pollicis, quadriceps femoris, tibialis anterior and gastrocnemius muscles) will be recorded in a spreadsheet (Excel, Microsoft, Washington, USA) by the research assistant (YS) to mask the assigned allocation for the primary investigator (TK) who will perform the data analysis. The primary investigator (TK) will have access to the final trial dataset for data analysis without knowledge of group allocation.

Endpoints

The primary endpoint is the reduction percentage in the mean value of the TcMEP top-to-bottom amplitudes in the upper extremity (adductor pollicis muscle) in the first (T1) and second (T2) recordings.

The secondary endpoints are the reduction percentage of the mean value of the TcMEP top-to-bottom amplitudes in the lower extremities (quadriceps femoris, tibialis anterior and gastrocnemius muscles) in the first (T1) and second (T2) recordings.

Statistical considerationsStatistical analysis

For summary statistics, normally and non-normally distributed continuous variables will be described as means and SDs or medians and IQRs. Categorical variables will be described as numbers and percentages. The normality of the data distribution will be evaluated graphically (ie, quantile-quantile plot, histogram) and using analytical methods (ie, Shapiro-Wilk test). To evaluate the primary and secondary endpoints, the reduction percentages of the mean values of TcMEP amplitudes (mean values at T1)−(mean values at T2)/(mean values at T1) will be compared between the control and intervention groups using the χ2 test. For both the primary and secondary endpoints, the reduction percentage will be calculated as zero if the TcMEP voltages at T2 are larger than those at T1.

Data will be analysed using Stata V.18.0 (StataCorp), with a two-sided p value <0.05, serving as the criterion for assessing the null hypothesis for each analysis. We predetermined that the recorded data of the right upper and lower extremities (adductor pollicis, quadriceps femoris, tibialis anterior and gastrocnemius muscles) will be used for the primary analysis. The TcMEP amplitude at the left upper and lower extremities will be used for secondary analysis. Bonferroni correction will be applied for the three pairwise comparisons among the lower extremity muscles (quadriceps femoris, anterior tibialis and gastrocnemius muscles), and a two-tailed p value <0.016 will be used to assess the null hypothesis for each analysis. Intention-to-treat analysis will be applied to non-adherence participants.

Sample size estimation

Based on our pilot study, the mean value of the reduction percentage in TcMEP voltage in the right upper extremity was approximately 35% (SD=31): this pilot data are unpublished. The estimated sample size was calculated as 36 (each arm, n=18), assuming a type 1 error of 5%, a type 2 error of 20% and a two-sided test. Adjusting for a dropout percentage of 10%, the final estimation of the sample size was 40 (each arm, n=20).

Harms

No harm occurred in our preliminary study involving 10 infants, which received approval from the local ethics committee (approval number: 2022058; 17 November 2022). However, if any detrimental events arise during the current study, the researchers will report the details to the local ethics committee in accordance with ethical protocols.

Data monitoring

The local institutional data monitoring committee will periodically review the progress and occurrence of harm and decide the appropriateness to continue this study. The committee comprises specialists in biostatistics, well-experienced clinical researchers and several non-medical professionals. Interim analyses to evaluate the safety and efficacy of the intervention are not planned because both intervention and non-intervention anaesthesia regimens are applied as regular anaesthesia regimens in our institution. However, if there is concerning harm related to this study, the incident will be immediately reported to the hospital director to determine the study’s discontinuation. Auditing trial conduct is held monthly. The reviewers are the same members as the data monitoring committee members and are independent of investigators and conflicts of interest. If important protocol modifications are needed, the investigators will report the changes and reasons to the local institutional ethics committee.

Ethics and dissemination

This study will be conducted in accordance with the principles outlined in the Declaration of Helsinki. The study protocol was reviewed and approved by the Institutional Review Board of Aichi Children’s Health and Medical Center (approval number: 2023036; 12 July 2023). Written consent will be obtained from all study participant guardians. The results of this study will be reported in a peer-reviewed journal and relevant academic conferences. This study was registered in the Japan Registry of Clinical Trials (jRCT1041230094; 17 October 2023, version 1). The trial information, registered in the Japan Registry of Clinical Trials, is in accordance with the WHO registration dataset (version 1.3.1).10 All personal information will be collected and kept confidential by storing the data in hospital computers where outside access is limited. All personal data will be discarded 1 year after study completion. Authorship of the manuscript will be decided based on the recommendations of International Committee of Medical Journal Editors.11

Discussion

This single-centre randomised controlled study will investigate the impact of low-dose sevoflurane (0.10–0.15 age-adjusted MAC) combined with propofol-based anaesthesia on intraoperative MEP monitoring in infants. Currently, there is a lack of evidence regarding the influence of the interaction between low-dose sevoflurane and propofol-based TIVA on MEP. Understanding the influence of low-dose sevoflurane with propofol-based anaesthesia can provide clinicians with essential information to discuss optimal anaesthetic regimens for improving the efficiency of MEP, resulting in reduced false-positive (suppressive) MEP results during surgery.

In infants, anaesthesia providers must consider potential complications associated with propofol, such as propofol infusion syndrome, prolonged awakening time, lipidaemia and alteration of platelet function. Additionally, propofol accumulation can lead to MEP suppression. Therefore, the coadministration of low-dose sevoflurane with propofol-based anaesthesia seems rational for minimising the total dose of intraoperative propofol infusion. Furthermore, this combination of anaesthesia (ie, low-dose sevoflurane, low-dose propofol and remifentanil) is widely used for anaesthesia maintenance in children requiring MEP. However, there is limited evidence regarding the impact of the coadministration of sevoflurane and propofol on IONM. A previous prospective observational study reported successful anaesthesia in infants with IONM under the coadministration of sevoflurane and propofol.4Additionally, current Japanese guidelines suggest that low-dose inhalational anaesthetics (ie, ≤0.50 MAC) can be coadministered with low-dose propofol during anaesthesia maintenance to avoid suppression of IONM.3 In contrast, we reported a case of an infant that showed the eminent suppression of intraoperative MEPs induced by administering low-dose sevoflurane (0.10–0.15 age-adjusted MAC) with propofol-based anaesthesia.7 Another study suggested that younger children (less than 3 years of age) require greater stimulating voltage during IONM due to the immaturity of the central nervous system, which may be more sensitive to anaesthetics compared with adults. Disappearance of IONM can occur with low-dose anaesthetics in this paediatric population, potentially leading to misleading surgical strategies, which is detrimental to patients. Considering the current contradictory evidence and the unique neurological anatomy and physiology of infants, there is a need for an interventional study with minimal bias in this paediatric population.

This study has several potential limitations. First, this study will be conducted in a single institution. Therefore, the results of this study may not be applicable to other institutions. However, this study uses a widely applied anaesthesia regimen (ie, propofol and remifentanil infusion with or without coadministered sevoflurane), and the procedure of TcMEP attachment and stimulation followed an internationally standardised methodology (ie, the international 10-20 system). This standardised procedure may increase the applicability of the results to clinical practice in other institutions. In our pilot study, we observed that the TcMEP in the lower extremities was suppressed more than that in the upper extremities by adding low-dose sevoflurane. Although the exact mechanism remains unclear, a possible explanation is the proximity of the TcMEP-stimulating electrode to the primary motor cortex of the upper extremity compared with the lower extremity. To address this, we will apply a larger electrode size (20 mm in diameter) in this study compared with the pilot study (10 mm in diameter) to provide sufficient electrical stimulation in both the primary motor cortex areas of the upper and lower extremities.

Second, there can be a wide variance in the measured values of TcMEP amplitudes for each individual, potentially influenced by mechanical factors (small differences in the procedure of TcMEP attachment and stimulation) and/or patient factors (differences in sensitivity to anaesthetics). In our pilot study, several patients exhibited no reduction in TcMEP voltage when low-dose sevoflurane was coadministered with propofol-based anaesthesia. Such measurement variation among different personnel could result in null findings (no significant difference in the reduction percentage of the TcMEP voltage between the control and intervention arms). To mitigate this, this study will apply a cross-over design to compare the reduction percentages of TcMEP voltages between T1 and T2 in the same patient to avoid measurement variations among different individuals.

Third, the residual sevoflurane in the body from anaesthesia induction might influence the TcMEP voltage at T1. In our pilot study, prone positioning of the patient and TcMEP preparation required approximately 30 min after terminating sevoflurane use and securing the airway during anaesthesia induction. We propose that mechanical ventilation with 6.0 L/min of oxygen and air for 30 min will be sufficient to eliminate the residual sevoflurane from the body. Additionally, the TcMEP voltage at T1 will not be recorded until the end-tidal sevoflurane concentration becomes undetectable.

Finally, this randomised controlled study will be unable to apply a double-masking design, potentially introducing bias. This single-masking randomised study will mask the patients for the random allocation. However, the researcher (MI) who will record the TcMEP voltage cannot be masked for the random allocation because the sevoflurane concentration will appear on a screen of vital signs in the operating suite. However, the primary investigator (TK) will be masked for the random allocation until the entire data analysis process is completed.

This randomised controlled study in infants has the potential to offer anaesthesia providers significant insights into re-evaluating the current anaesthesia regimen involving a combination of low-dose inhalational anaesthetics and propofol infusion in infants with IONM. Suppression of MEP due to interactions of anaesthetics can mislead the surgeon’s decisions and can be interpreted as the occurrence of a surgical manipulation-related nerve injury. This can also mislead anaesthesiologists’ decisions, where the difficulty in obtaining the baseline MEP might be recognised as an issue based on the mechanical (monitoring device) aspect; however, it can be caused by the addition of a small dose of sevoflurane. Therefore, if this study finds that adding even a small dose of sevoflurane on propofol-based TIVA can inhibit MEP amplitudes, our recommendation is to avoid using sevoflurane, even in small doses, when used with propofol-based TIVA in infant MEP cases. If anaesthesiologists require the use of sevoflurane, all surgical team members must be vigilant to interpret intraoperative MEP when its suppression or disappearance is observed.

Furthermore, comparing the MEP voltages between propofol-based anaesthesia with and without low-dose inhalational anaesthetics can be a stepstone towards investigating the impact of the interaction of several anaesthetics on IONM responses. Such investigations hold promise for enhancing the success rate of IONM detection in infants, thereby improving surgical outcomes and patient safety.