Evoked Response Monitoring

General Overview of Monitoring

Using these techniques, a stimulus is applied to the neural tract, and a response is “evoked” that indicates the tract is functioning. This allows probing tissue to locate or “map” the tract, and it allows repeated testing of a tract during procedures to identify conditions of impending neural injury (“monitoring”). This allows improved decision-making during procedures and facilitates improved outcome. It also allows an assessment of the physiological environment of the neural tissue (e.g., perfusion pressure) to facilitate improved outcome.

In general, evoked responses are measured along the neural tract after the stimulus is applied. The response consists of series of peaks, each of which is produced by the neural tissue in the tract. For mapping, the presence of the response indicates that the tract has been located in the vicinity of the stimulator. For monitoring, repeated measurements are taken to watch for significant peak changes. Changes are usually noted from peak measurements such as the time from stimulation to the peak (referred to as “latency”) and the “amplitude” of the peak as measured to an adjacent peak of opposite polarity. Frequently, several monitoring techniques are utilized during procedures recognizing risk in multiple areas (“multimodality monitoring”).

Auditory Brainstem Response

The ABR is produced when the cochlea is stimulated using sound delivered into the external ear canal. It has also been referred to as the brainstem auditory evoked response (BAER) and the brainstem auditory evoked potential. Traditionally, broad-spectrum “clicks” are used with white noise (like a television tuned to an empty channel) delivered to the opposite ear to prevent the other cochlea from being activated by sound vibrations transmitted through the skull.

The response of the eighth cranial nerve can be measured using electrodes placed during surgery on the nerve or near the cochlear nucleus; however, it is usually monitored using electrodes at the external ear and top of head. The response consists of a series of five peaks produced within the 10 ms following stimulation. The latency of peaks I, III, and V and the amplitude of peak V are usually monitored. Wave I is produced by the extracranial portion of CN VIII, wave III by acoustic relay nuclei and tracts deep in the midline of the lower pons, and wave V by the lateral lemniscus and inferior colliculus contralateral to the side of stimulation ( Fig. 6.1 ). The response of the auditory cortex is referred to as the midlatency auditory evoked potential (MLAEP).

Experience has shown that increases in latency of 10% or more of peaks (or between peaks, interwave latency), decrease in peak amplitude of more than 50%, or loss of peaks is associated with significant dysfunction in the pathway, which correlate with changes in useful hearing. Hence the ABR is used to reduce the risk of hearing loss in surgery in the cerebellopontine angle of the brainstem. It has also been used as a general monitor of brainstem viability during surgery on the midbrain and pons. The MLAEP has been used as a measure of anesthetic effect but is rarely used for procedural monitoring.

Monitoring Other Cranial Nerves

The ABR is the only evoked sensory response testing routinely utilizing a sensory cranial nerve. Methods for testing the olfactory nerve (CN 1) have not been adequately developed for monitoring. Methods for monitoring the evoked response of the optic nerve (CN 2) (visual evoked potentials) have been developed using flash stimulation to closed eyes; however, routine use has not found utility similar to ABR. Finally, methods for evoked response testing of the trigeminal nerve (CN 5) have been developed and are occasionally used.

Somatosensory Evoked Potentials

The somatosensory evoked potential (SSEP) is one of the most commonly monitored evoked responses. The response is produced by applying an electrical stimulus to a peripheral nerve (much like that used for assessing train of four for neuromuscular blockade). The nerves (and their component nerve roots) usually utilized for monitoring are major motor and sensory nerves to the hands and feet (median n. (C6-T1), ulnar n. (C8-T1), and posterior tibial n. (L4-S2)). Stimulation activates large-diameter, fast-conducting Ia muscle afferent fibers producing a muscle movement and group II cutaneous nerve sensory fibers that produce cephalad traveling SSEP signals.

The major sensory response travels through the component roots of the nerves in the brachial and lumbar plexus ( Fig. 6.2 ). Entering the spinal cord via the dorsal root, the response ascends the spinal cord via the dorsal columns in the pathway of joint proprioception and vibration. A first synapse occurs near the nucleus cuneatus and nucleus gracilis, and it ascends the brainstem in the medial lemniscus after crossing the midline. After a second synapse in the ventroposterolateral nucleus of the thalamus, it travels to the primary somatosensory cortex contralateral to the side of stimulation.

For mapping purposes, stimulation of the posterior surface of the spinal cord can be used to locate the midline, and recording on the surface of the brain can be used to locate the gyrus between the sensory and motor cortex. For procedure monitoring, the response can be monitored along the peripheral nerve (verifying stimulation similar to seeing the motor response), over the brachial plexus (Erb point), epidurally near the spinal cord, and, most commonly, over the cervical spine (referred to as the subcortical response) and over the sensory cortex.

Monitoring has been used for surgery on peripheral nerves and plexus, monitoring for unfavorable positioning of the arm (brachial plexus), spine procedures, general brainstem viability, and procedures placing the subcortical and cortical pathways at risk. Experience indicates that more than 50% decrease in amplitude of response peaks or increase in latency of 10% or more are associated with concerns of unfavorable neural environment.

Motor Evoked Potentials

MEP are the most recent addition to the techniques used for monitoring. Of the various methods that have been used to stimulate the motor pathways, electrical stimulation of the motor cortex has emerged as the most commonly used technique that unquestionably allows motor tract assessment. The pyramidal cells of the motor cortex can be stimulated directly through a craniotomy (direct cortical stimulation); however the most commonly employed technique utilizes multipulse electrical stimuli applied via scalp electrodes (transcranial motor evoked potentials, TcMEP). Enhanced stimulation techniques have also been developed using a priming stimulus prior to the main stimulus to assist in young children, adults with neural pathology, and with partial neuromuscular blockade. Transcranial stimulation utilizing a brief magnetic stimulation has largely been abandoned due to depression by anesthesia.

This technique activates 4%–5% of the fast-conducting fibers of the cortico-spinal tract (CST), which descends the brain, brainstem (crossing in the brainstem), and spinal cord until it synapses on anterior horn cells ( Fig. 6.3 ). The descending response in the spinal cord consists of a D wave from the direct stimulation of the motor cortex and is considered an approximation of the volume of motor pathway activated. Its amplitude varies less than 10%. I waves accompany the D wave and represent the contribution of the response from transsynaptic activation of internuncial pathways in the motor cortex. I waves are sensitive to anesthesia.

Temporal summation of the D wave and I waves activates the anterior horn cells to produce a response in motor fibers of lower motor neurons. This response travels to the neuromuscular junction and produces a compound muscle action potential (CMAP) in about 4%–5% of muscle fibers. Not all of the descending spinal fibers directly innervate anterior horn cells; many have intervening synapses. In addition, several other motor pathways are either activated or influence the sensitivity of the anterior horn cell to activation. These additional synapses and other pathways give rise to wide variability of the CMAP produced under anesthesia. Of note, individual muscles are often innervated by muscle fibers from more than one nerve root.

MEP stimulation techniques can be utilized for mapping of the location of the motor cortex, and for the location of the CST in the subcortical tissue, brainstem, and spinal cord. The MEP response can be monitored utilizing an epidural electrode (monitoring the D wave) or using electrodes placed in peripheral muscles (monitoring the CMAP). The most commonly monitored muscles include those of the distal upper (e.g., abductor or flexor pollicis brevis) and lower extremity (e.g., abductor hallucis brevis and tibialis anterior), although muscles related to specific spinal cord regions can be used. Since the MEP has good correlation with motor outcome, it has been used to monitor motor cortex and general brainstem viability and procedures on the spine and spinal cord. It is frequently monitored along with SSEP, where it is thought to be more sensitive than the SSEP to ischemia in areas of the cortex, brainstem, and spinal cord.

Because of the stability of the D wave, a 50% reduction in amplitude has been used as criteria for concern. However, the wide variability of the CMAP response has created controversy about warning criteria. Suggestions have included changes in the configuration of the multipeaked CMAP response (indicating changes in the motor units responding), the need for increased cortical stimulation voltage (50–100 V) or current, and very substantial declines in the CMAP amplitude (i.e., >70%). One of the most commonly utilized MEP criteria is simply the loss of a CMAP response.

Electromyography

Electromyography (EMG) is the recording of activity using needle pairs placed in muscles. This differs from MEP in that the muscle responses are spontaneous or evoked by stimulation of cranial or peripheral nerves rather than central motor stimulation of the cerebral cortex. Two basic types of recording are used. Electrical stimulation of nerves leading to a muscle response is referred to as triggered EMG and is used to map the location or assess the continuity of nerve structures. The second type of recording is referred to as spontaneous or “free-run” EMG. In this case, muscle activity is seen in a usually silent background caused by other forms of stimulation such as mechanical or thermal irritations. Some of these stimuli are usually innocuous (e.g., inadvertent mechanical irritation); however, some are potentially injurious and signal the need to reassess the procedure (e.g., nerve stretching). In addition to seeing the response on the recording device, the response may be played on a loudspeaker.

The nerves innervating the muscles involved in EMG recording include cranial and peripheral nerves. For cranial nerves, any nerve with a motor component can potentially be monitored using the muscles innervated by the nerve ( Table 6.1 ). Mapping techniques have been developed to locate cranial nerve nuclei and the cranial nerve as it traverses the brainstem to the muscles recorded. Mapping of cranial nerve nuclei on the brainstem surface allows surgeons to approach deeper structures through designated safe entry zones.

Monitoring usually involves repeated intermittent stimulation in the operative field. Techniques which are not dependent on the surgeon providing the stimulation have been developed that are similar to MEP where transcranial stimuli evoke the muscle activity, thereby allowing monitoring independent of surgical stimulation (corticobulbar techniques). Such techniques have been described for facial nerve and vagus/recurrent laryngeal nerve monitoring. This latter technique also allows assessment of the cranial nerve function proximal to the surgical site.

For peripheral nerves a large number of muscles can be used depending on the specific nerve root desired ( Table 6.2 ). For these nerves, mapping can be utilized in injured peripheral nerves (neuroma in continuity) or to assess locations of injury in brachial plexus lesions. Mapping is also used to identify component rootlets of nerve roots to be sacrificed in dorsal rhizotomy for spasticity. Mapping is also used in surgery on the cauda equina (e.g., release of tethered cord) to identify functional nerve tissue to be preserved and nonfunctional tissue for sacrifice. Finally, as discussed subsequently, pedicle screw testing is a form of mapping the proximity of screws to nerve roots.

Monitoring of peripheral nerves using EMG is frequently done to identify the injurious stimuli mentioned earlier during procedures where nerve roots, the cauda equina, or peripheral nerves are at risk. Of note, when a particular nerve root is at risk, monitoring using intermittent triggered EMG of that root is more frequently utilized than free-run EMG since muscles used with MEP are usually innervated by more than one nerve root.

Several forms of evoked response can be seen in EMG. Triggered responses are usually brief CMAP responses linked in time to the stimulus as mentioned before. Free-run EMG has two basic forms of response ( Fig. 6.4 ). High-frequency CMAP bursts can identify blunt mechanical trauma or irritation to motor nerves. Causes of irritation include mechanical stimulation (e.g., nearby dissection; ultrasonic aspiration or drilling), nerve retraction, thermal irritation (e.g., heating from irrigation, lasers, drilling, or electrocautery), and chemical or metabolic insults.

More continuous EMG activity is referred to as “trains”or “neurotonic discharges,” and may be associated with impending nerve injury (nerve compression, traction, or ischemia of the nerve). When played on a loudspeaker, they have musical qualities like the sound of an outboard motor boat engine, swarming bees, popping corn (“popcorn”), or an aircraft engine (“bomber”) potentials. These neurotonic discharge trains raise concern for nerve injury as does very high amplitude bust responses. Unfortunately, nerve injury can occur without EMG activity such as when the nerve is severed.

Other Specialized EMG Responses

A specialized type of triggered EMG response is reflex testing. In this case the stimulus is applied to a peripheral sensory nerve that sends a volley of activity to the spinal cord via the dorsal root. It triggers reflex pathways in the spinal cord that activates motor fibers in the ventral root and peripheral nerve resulting in a CMAP. Three CMAP responses can be seen. The first “M” response is from direct activation of motor fibers in the stimulated nerve (like used for train of four assessment). The second “Hoffmann” (H) response is the reflex response. The last is an “F” response that is produced by the reflection of the incoming sensory volley at the spinal cord that sends a motor response back to the muscle via the stimulated nerve. Of note, the H reflex pathway in the spinal cord is dependent on normal cephalad spinal cord function. As such, a spinal cord injury will depress the H reflex activity caudal to the injury, even when the reflex occurs in uninjured spinal cord. As such the H reflex has been used in spinal surgery as a measure of spinal integrity cephalad to the region of reflex.

Another commonly utilized reflex is the bulbocavernosus reflex. This reflex results from stimulation of the pudendal nerve at the genitalia, reflex activity in the S2–S4 spinal cord segments, and efferent activity also via the pudendal nerve resulting in a CMAP in the external anal sphincter. This has found utility for testing during cauda equina surgery to preserve neural innervation of the bowel and bladder, which travel in the pudendal nerve.