Advertisement

Principles of intraoperative neurophysiological monitoring and anaesthetic considerations

Published:December 12, 2019DOI:https://doi.org/10.1016/j.mpaic.2019.10.020

      Abstract

      Surgery to the nervous system poses risks to neural structures be that mechanical, haemodynamical, chemical or thermal. The role of intraoperative neurophysiological monitoring (IONM) is to facilitate the assessment of the functional integrity of neural structures and provide a real time alerting system when changes caused by surgically induced insults are detected, with the goal of reducing the risk of postoperative neurological deficits. Furthermore, it is also used as a guidance system to map eloquent areas within the cortex and to identify specific neuronal structures, particularly when landmarks cannot be easily recognized. In this article, we focus on the various neurophysiological modalities used in intraoperative monitoring, their basic principles, indications and the information that they provide. We also examine the anaesthetic considerations and the checklist for the multidisciplinary team should an intraoperative alert be issued.

      Keywords

      After reading this article, you should be able to:
      • identify the different IONM techniques used, the neurophysiological pathways that they monitor and their value that they play during the surgical procedure
      • compare the different anaesthetic techniques required for the success of the various IONM modalities being monitored
      • formulate an appropriate anaesthetic plan to support the IONM
      • recognize what anaesthetic intervention is required when an IONM alert is issued
      Surgery to the nervous system poses risks to neural structures be that mechanical, haemodynamical, chemical or thermal. The role of intraoperative neurophysiological monitoring (IONM) is to facilitate the assessment of the functional integrity of neural structures and provide a real-time alerting system when changes caused by surgically induced insults are detected, with the goal of reducing the risk of postoperative neurological deficits. Furthermore, it is also used as a guidance system to map eloquent areas within the cortex and to identify specific neuronal structures, particularly when landmarks cannot be easily recognized.
      Prior to the introduction of IONM, the Stagnara wake-up test was used to assess the neurological status of patients. This involved reversal of general anaesthesia and when compliant, the patient was asked to move their lower extremities to assess spinal cord function. The Stagnara wake-up test was only used once during a procedure and was therefore unable to provide continuous assessment of the integrity of the neurological structures. It was a crude indication primarily of motor function. Patient movement on the operating table as well as the risks of accidental extubation of the trachea were concerns.
      • Vauzelle C.
      • Stagnara P.
      • Jouvinroux P.
      Functional monitoring of spinal cord activity during spinal surgery.
      In the first section of this article, we will focus on the different neurophysiological modalities used in intraoperative monitoring, their basic principles, indications and the information that they provide. We will then focus on anaesthetic considerations and the checklist for the multi-disciplinary team should an intraoperative alert be issued.

      Neurophysiological modalities

      Evoked potentials

      Somatosensory evoked potentials (SEPs)

      are a measure of the presynaptic and postsynaptic electrophysiological responses generated within the sensory pathways over the peripheral nerves and propagate rostrally through peripheral and central conduction pathways to the cortex. This technique records the relatively synchronous waves of depolarization arising from myelinated peripheral and central axons and synapses within the grey matter. It is elicited by bipolar transcutaneous non-noxious electrical stimulation to peripheral nerves typically via the median/ulnar and tibial nerves. It involves recording responses from the periphery (popliteal fossa from the lower limb and Erb's point from the upper limb), the spine (at thoracic or cervical level) and the cortical responses from the primary somatosensory cortex.
      Clinically, SEPs are useful in evaluating conduction abnormalities along central pathways (dorsal columns) in conditions such as multiple sclerosis and spinal cord lesions. In some centres it is also used as part of prognosis protocols in anoxic coma.
      Intraoperative SEPs assess changes in conduction in the somatosensory pathways and can be used as guidance for central sulcus localization. SEPs are indicated intraoperatively in scoliosis correction, spinal cord decompression and stabilization procedures, brachial plexus surgeries, tumour resections, clipping of intracranial aneurysms and carotid endarterectomy.
      Intraoperative SEPs alone have limitations. They do not evaluate the motor pathways; therefore injury to the motor tracts can occur with no changes within the SEP parameters. SEPs also require signal averaging (2–3 minutes) and these results in a delay in interpretation and feed-back to the surgeon.
      • Deletis V.
      Basic methodological principles of multimodal intraoperative monitoring during spine surgeries.
      ,
      • MacDonald D.B.
      • Skinner S.
      • Shils J.
      • Yingling C.
      Intraoperative motor evoked potential monitoring – a position statement by the American Society of Neurophysiological Monitoring.

      Motor evoked potentials (MEPs)

      Transcranial electrical stimulation of the motor cortex (TcMEP) produces a descending volley that transverses through the corticospinal tracts and alpha motor neurons, producing a measurable synaptic response in the anterior horn cell (recorded epidurally in the form of a D-wave response) or a measurable compound muscle action potential (CMAP) in the periphery (myogenic MEP). MEPs are used to evaluate the integrity of the corticospinal tracts and anterior spinal cord pathways.
      The generation of myogenic MEPs is dependent on factors such as the excitability of the motor cortex, the corticospinal tracts, the alpha motor neurons, and the integrity of the motor nerves and motor endplates.
      D-wave recordings are a variant of MEPs. They are a neurographic representation obtained directly from the corticospinal tracts from levels of T8 and above. D-waves are corticospinal whereas myogenic MEPs are corticosynaptic and this explains why myogenic MEPs are more susceptible to influences of general anaesthesia.
      Intraoperative MEPs are indicated in any surgery where spinal cord parenchyma is at risk and motor deficit is a concern: anterior spinal procedures, extensive decompression in spinal stenosis with features of myelopathy, functional disturbance of cauda equina and nerve roots, spinal neurovascular surgeries where compromise of spinal cord perfusion is a concern, resection of all spinal cord tumours especially intradural intramedullary, spinal deformities and corrective scoliosis and procedures involving the cerebellopontine angle and brainstem.
      The combination of repetitive motor cortex stimulation, spinal epidural recording, patient movement and adjusted anaesthetic techniques raises concerns with TcMEP: tongue bite injuries, seizures (under general anaesthesia, seizures are very rare), kindling (a process in which repeated stimuli sensitize the brain lowering the threshold to react when the stimulus is reapplied), epidural complications, accidental injury from movement, scalp burns, and jaw pain.
      • Deletis V.
      Basic methodological principles of multimodal intraoperative monitoring during spine surgeries.
      ,
      • MacDonald D.B.
      • Skinner S.
      • Shils J.
      • Yingling C.
      Intraoperative motor evoked potential monitoring – a position statement by the American Society of Neurophysiological Monitoring.
      Tongue injuries are the most frequent complication from TcMEPs (four times more likely than lip lacerations). Tamkus et al. reported an incidence of 0.63% and, although this is low, the use of bilateral soft bite blocks and checking for displacement throughout surgery is strongly recommended.
      • Tamkus A.
      • Rice K.
      The incidence of bite injuries associated withTranscranial motor-evoked potential monitoring.
      ,
      • Pisklakov S.
      • Le V.
      • Sandoval P.
      Tongue laceration during neurophysiologic monitoring with motor evoked potentials.

      Intraoperative EMG techniques

      Free-run EMG (FrEMG)

      is a technique that evaluates the function of motor root axons by recording electromyographic (EMG) activity from the muscle fibres innervated by the axons of the root under exploration. It is used to detect motor nerve root mechanical activation and compromise during surgical procedures involving placement of pedicle screws, decompression for spinal stenosis where roots are at risk, correction for scoliosis deformity, radiculopathy secondary to disc herniation and/or spondylosis and removal of tumours involving the nerve roots.
      Single non-repetitive and asynchronous discharges (phasic bursts) are seen when direct nerve irritation occurs from tugging, displacement, vigorous irrigation, direct bumping or sacrificing the root (Figure 1).
      Persistent neurotonic, train activity is of greater concern. This activity consists of prolonged multiple or repetitive synchronously grouped discharges that last up to several minutes. It is usually associated with a more serious neural insult and it is typically seen during sustained traction, compression and stretching of the nerve root (Figure 2).
      Figure 2
      Figure 2Persistent neurotonic trains localised to the specific root under exploration.
      EMG activity over a widespread distribution, often indicative of increased muscle tone, can be a consequence of lightened general anaesthesia.
      The benefits of FrEMG is that is provides real-time feedback to the surgical team. From the anaesthetic perspective, this technique cannot be done in the presence of neuromuscular blockade.

      Triggered EMG (TrEMG)

      is an intraoperative technique involving direct electrical stimulation to either a pedicle screw or directly onto a neuronal structure with the subsequent measurement of the compound muscle action potentials (CMAPs) from muscles innervated by the nerve roots near to the stimulated screw, or directly from stimulation of the neuronal structure itself.
      This technique is used to test the integrity of the motor nerve root, cranial nerves, and to assist in the identification of nerve structures during resection of tumours or untethering spinal procedures. TrEMG requires the absence of neuromuscular blocking agents.

      Recurrent laryngeal nerve (RLN) monitoring

      Erwood et al. report that the incidence of RLN injury during anterior cervical decompression/fixation (ACDF) surgeries is 1–11% and may lead to dysphonia, hoarseness, speaking and singing difficulties. The suggested causes of this neural insult are traumatic division, entrapment by retractors and overstretching during lateral retraction. The RLN ascends along the tracheoesophageal groove, bifurcating before entering the larynx where the RLN innervates vocalis and arytenoid muscles. The EMG impregnated endotracheal tube allows the monitoring of these muscles during surgery (Figure 3).
      Figure 3
      Figure 3EMG impregnated tube (Courtesy of NuVasive).

      Cortical stimulation and motor mapping

      Electrical stimulation is either applied cortically directly on the surface of the brain or subcortically to map eloquent motor, sensory, language or visual cortex.
      This technique can be performed on awake patients or under general anaesthesia. It is important to remember that the depth of the anaesthesia can affect the outcome of the IONM due to suppressed cortical excitability or the presence of neuromuscular blockade.
      During motor mapping, stimulation is applied to the cortex with CMAP recording from peripheral muscles and therefore requires intravenous anaesthetic agents and absence of neuromuscular blockade.
      Language mapping involves cortical stimulation while various language tasks are being performed by the patient. It is used to evaluate eloquent language brain functions such as reading, auditory comprehension and spontaneous speech. Cortical stimulation in language areas typically induces inhibition of language functions.

      Central sulcus localization

      The use of a specialized SEP technique known as phase-reversal can be employed to localize the central sulcus. Following the craniotomy, a strip electrode is placed across the exposed motor and primary somatosensory cortex, traversing the central sulcus. The SEP, elicited by median nerve stimulation at the wrist, is then recorded from the primary sensory cortex. Electrophysiologically, one would expect a phase reversal in the cortical responses indicating the area of the central sulcus.

      Electroencephalography (EEG)

      EEG is commonly known for its clinical use in the epilepsy field and in the hospital setting to aid with diagnoses of cortical dysfunction. EEG is also used intraoperatively to monitor the degree of cortical perfusion and oxygenation during a variety of vascular, cardiac, and neurosurgical procedures. The primary advantages of intraoperative EEG monitoring include: (i) recognizing and/or preventing perioperative ischaemic insults; and (ii) monitoring of brain function in order to determine depth of anaesthesia (and level of consciousness), thereby assisting with tailoring of drug administration to achieve a predefined neural effect (e.g. burst suppression).
      General anaesthetic drugs cause an increase in slow activity, but can also result in an increase in fast activity especially over the anterior cortical regions. Deeper levels of anaesthesia can eventually cause a burst-suppression pattern or even complete disappearance of all EEG activity (i.e. isoelectric EEG).
      EEG can aid both the anaesthetist and the neurophysiologist in examining the level of cortical suppression during total intravenous anaesthetic techniques which can be adjusted accordingly to maintain sufficient anaesthesia while ensuring sufficient cortical excitability required for the generation of neurophysiological intraoperative variables.

      Anaesthetic considerations

      Inhalational anaesthesia

      Inhalational anaesthetic agents (IHA) are known to produce a dose-related increase in latency and reduction in the amplitude of cortically recorded SEPs. MEPs are more sensitive to interference and are easily abolished by IHA (Figure 4).
      Figure 4
      Figure 4Waveform stimulus. Waveform stimulus is delivered (stimulus onset), the response is then recorded from the associated neural structure. Latency, de_ned as the time in milliseconds (ms) from stimulation to the onset of the action potential. Amplitude, peak-to-peak measure of waveform in milli- or micro-volts. Shown as a baseline recording in example “a”. Example “b” shows dose related anaesthetic changes with an increase in latency and decrease in amplitude.
      Generally speaking, keeping the concentration of the volatile anaesthetic to less than 0.5 MAC should allow for the acquisition of MEPs; however, they are exceptionally unstable and the recommended approach to monitoring MEPs is to avoid volatile anaesthetic agents and instead rely on a propofol-based total intravenous anaesthesia.
      Nitrous oxide produces more profound changes in cortical SEPs and MEPs whereas its effects on subcortical and peripheral sensory responses are minimal. Regardless of whether inhalational anaesthetics or intravenous anaesthetics are used it is important to maintain constant (steady-state) concentrations of anaesthetics. Bolus administration or acute changes in inhalational anaesthetics can result in marked alterations in evoked potentials which must be distinguished from potential neuronal injury.
      • Erb T.O.
      • Ryhult S.E.
      Improvement of motor-evoked potentials by ketamine and spatial facilitation during spinal surgery in a young child.
      ,
      • Sloan T.
      • Jantti V.

      Intravenous anaesthesia

      Propofol, a short-acting intravenous anaesthetic agent, is used for the induction and maintenance of general anaesthesia. It produces a dose-dependent reduction in the amplitude of MEPs, but has no effect on the latency. A comparison of the effects of propofol with the inhaled anaesthetics isoflurane plus nitrous oxide on MEPs revealed that with multipulse stimulation propofol provided better recording conditions. These findings support the current position of Propofol being the standard anaesthetic approach for IONM recording of MEPs.
      Ketamine, a dissociative anaesthetic with intense analgesic properties, has the ability to enhance the signals acquired when monitoring evoked potentials. This can be achieved with clinically relevant doses. While some authors have advocated doses of ketamine that are on the high side in order to enhance the evoked potentials, the downside of this approach is the greater likelihood that the patient will experience psychomimetic effects upon emergence from anaesthesia.
      • Erb T.O.
      • Ryhult S.E.
      Improvement of motor-evoked potentials by ketamine and spatial facilitation during spinal surgery in a young child.
      ,
      • Sloan T.
      • Jantti V.

      Neuromuscular blockade (NMB)

      MEP, TrEMG, DNS and FrEMG, which require recording of CMAPs, are impacted by the extent of neuromuscular blockade.
      Partial blockade, to reduce the potential for movement, has been advocated for spine surgery, but must be maintained in a narrow range if used. If NMB is required for intubation a short-acting agent that can be rapidly reversed is advisable. The ability to obtain reliable and true neurophysiological baseline responses pre-positioning, and before knife to skin, are highly reliant on the absence of NMB.

      Effects of temperature

      Marked temperature-related drops in amplitudes of evoked potentials may occur after exposure of the spine.
      Hyperthermia reduces the latency and increases the conduction velocity of SEPs and MEPs. Spinal SEP amplitudes are unchanged, whereas cortical SEPs and spinal MEPs show attenuation above 42°C.
      Hypothermia increases latency and decreases the conduction velocities of SEPs and MEPs. Below 28°C, the amplitude of SEPs and MEPs attenuate or disappear. It has been suggested that SEP and MEP measurements are performed within a constant range of 2–2.5°C above or below the baseline temperature throughout the procedure
      • Oro J.
      • Haghighi S.S.
      Effects of altering core body temperature on somatosensory and motor evoked potentials in rats.
      .

      Oxygen and ventilation

      Both SEP and MEP deterioration can be seen in hypoxemia before any other clinical parameters change. Arterial carbon dioxide (PaCO2) alterations alter spinal cord and cortical blood flow with the most notable changes seen in cortical SEP when the arterial carbon dioxide tension is extremely low. This suggests that excessive vasoconstriction may produce ischemia (PaCO2 20 mmHg). The effects of this have been suggested to contribute to alterations in SEP during spinal surgery and may be expected to produce some degree of alterations in MEPs.
      Hypocapnia may produce small SEP changes, and possibly MEP; it is recommended that intraoperative neurophysiological baseline recordings should be acquired prior to initiation of any hyperventilation.
      • Gravenstein M.A.
      • Sasse F.
      Effects of hypocapnia on canine spinal, subcortical, and cortical somatosensory-evoked potentials during isoflurane anaesthesia.

      Blood rheology and haemodynamic effects

      Haematocrit changes can alter both oxygen carrying capacity and blood viscosity and the maximum oxygen delivery is often thought to occur in a midrange haematocrit of 30–32%.
      The effects of these variables are seen in the SEPs demonstrating an increase in amplitude with mild anaemia, followed by an increase in latency at haematocrits of 10–15%. Further latency and amplitude reduction can be observed with haematocrit levels less than 10%. It has been reported that these changes are partially restored by an increase in the haematocrit. There are no comparable studies investigating the effects on MEPs.
      Numerous studies have demonstrated a threshold relationship between regional cerebral blood flow and cortical evoked responses. Decreases in blood pressure may therefore result in substantial recording changes within both the SEP and MEP modalities. These changes typically reverse when perfusion pressure is increased and restored to baseline. Cortical SEPs remain normal until blood flow is reduced to ∼20 ml/min/100 g and at more restricted blood flows of between 15 and 18 ml/min/100 g of tissue; SEPs are significantly altered or abolished.
      As with general anaesthetic effects, the subcortical SEPs are less sensitive than cortical responses to effects of reductions in blood flow.
      • Nagao S.
      • Roccaforte P.
      The effects of isovolemic hemodilution and reinfusion of packed erythrocytes on somatosensory and visual evoked potentials.

      The spinal cord at risk checklist

      It is imperative that when a neurophysiological alert is issued, protocols are in place to ensure that real-time communication does not get lost in translation. An established ‘cord at risk’ or ‘neuro alert’ protocol is recommended so that everyone across the multi-disciplinary team is aware of their role and responsibilities. When an alert is issued all surgery should be stopped and your protocol set into motion (Figure 5).
      Figure 5
      Figure 5Alert workflow illustrating the workflow of the anaesthetic, surgical and neurophysiological teams.
      The intraoperative neurophysiologist must look for a possible technical problem and assess all the different monitoring modalities with the understanding that most technical problems are observed in an area unrelated to surgery.
      The anaesthetist must determine whether the blood pressure is adequate. In addition, oxygenation, ventilation, and haematocrit should be verified to be in an acceptable range. Finally, any alterations in the anaesthetic, particularly bolus administration of medications, need to be reconciled with the alterations in IONM signals (Table 1).
      Table 1Checklist for the anaesthetist during an IONM alert
      1Cross-check no new drugs have been administered
      2Correlate EEG with TIVA levels
      3Consider propofol reduction
      4Check BP, O2, ventilation and haematocrit
      5Optimize haemodynamics – raise MAP and oxygenation
      6Dexamethasone or methyl-prednisone for cord at risk
      7Administer ketamine for enhancement of waveforms

      Keys to success

      Understanding the proposed operative procedure to determine what forms of IONM will be employed is imperative to ensuring that all neuronal structures at risk are assessed and this ultimately involves multimodal IONM.
      The decision of the anaesthetic approach to maximize signal acquisition while maintaining, supporting and, if necessary, correcting the physiologic state of the patient is key to the success of the IONM.
      It is crucial that the maintenance of a constant concentration of the inhalational and/or intravenous anaesthetics is adopted, since rapid alterations in anaesthetic concentrations may make interpretation of the IONM signals challenging, especially during critical portions of the operative procedure.
      The ultimate success of IONM is highly reliant on multi-disciplinary teamwork comprised of anaesthetists, surgeons and neurophysiologists. It requires excellent communication and protocols in place should adverse events occur during the procedure.

      References

        • Vauzelle C.
        • Stagnara P.
        • Jouvinroux P.
        Functional monitoring of spinal cord activity during spinal surgery.
        Clin Orthop Relat Res. 1973; 93: 173-178
        • Deletis V.
        Basic methodological principles of multimodal intraoperative monitoring during spine surgeries.
        Eur Spine J. 2007; 16: 147-152
        • MacDonald D.B.
        • Skinner S.
        • Shils J.
        • Yingling C.
        Intraoperative motor evoked potential monitoring – a position statement by the American Society of Neurophysiological Monitoring.
        Clin Neurophysiol. 2013; 124: 2291-2316
        • Tamkus A.
        • Rice K.
        The incidence of bite injuries associated withTranscranial motor-evoked potential monitoring.
        Neurosci Anesthesiol Periope Med. 2012; 115: 663-667
        • Pisklakov S.
        • Le V.
        • Sandoval P.
        Tongue laceration during neurophysiologic monitoring with motor evoked potentials.
        Open J Anesthesiology. 2012; 2: 226-227
        • Erb T.O.
        • Ryhult S.E.
        Improvement of motor-evoked potentials by ketamine and spatial facilitation during spinal surgery in a young child.
        Anesth Analg. 2005; 100: 1634-1636
        • Sloan T.
        • Jantti V.
        Anesthetic effects on evoked potentials. vol. 8. Elsevier, 2008: 94-126
        • Oro J.
        • Haghighi S.S.
        Effects of altering core body temperature on somatosensory and motor evoked potentials in rats.
        Spine. 1992; 17: 498-503
        • Gravenstein M.A.
        • Sasse F.
        Effects of hypocapnia on canine spinal, subcortical, and cortical somatosensory-evoked potentials during isoflurane anaesthesia.
        J Clin Monit. 1992; 8: 126-130
        • Nagao S.
        • Roccaforte P.
        The effects of isovolemic hemodilution and reinfusion of packed erythrocytes on somatosensory and visual evoked potentials.
        J Surg Res. 1978; 25: 530-537