Neurophysiologic Intraoperative Monitoring

32713

Aatif M. Husain

Neurophysiologic intraoperative monitoring (NIOM) is used to reduce neurologic morbidity during surgeries in which the nervous system is at risk. NIOM is a useful adjunct to surgeries in which the brain, spinal cord, and peripheral nerves are at risk. NIOM allows assessment of neurologic function when the patient cannot be examined. Monitoring waveforms is performed by the neurophysiologist during the active portion of the surgical procedure. Several techniques can be used to monitor the nervous system during surgery, and these are chosen depending on the part of the nervous system at risk and type of surgery. Common modalities used for monitoring include brainstem auditory evoked potentials (BAEP), somatosensory evoked potentials, transcranial electrical motor evoked potentials, electromyography, and electroencephalography. BAEP monitoring has been shown to reduce the incidence of hearing loss associated with microvascular decompression surgeries.

brainstem auditory evoked potentials, electroencephalography, electromyography, hearing loss, microvascular decompression surgeries, neurophysiologic intraoperative monitoring, somatosensory evoked potentials, transcranial electrical motor evoked potentials

Electroencephalography, Electromyography, Evoked Potentials, Auditory, Brain Stem, Evoked Potentials, Motor, Evoked Potentials, Somatosensory, Hearing Loss, Intraoperative Neurophysiological Monitoring, Microvascular Decompression Surgery

Neurophysiologic intraoperative monitoring (NIOM) is used to reduce neurologic morbidity during surgeries in which the nervous system is at risk. NIOM allows assessment of neurologic function when the patient cannot be examined. Monitoring waveforms is performed by the neurophysiologist during the active portion of the surgical procedure. Often, the neurophysiologist is able to alert the surgeon of impending injury, allowing the surgeon to modify or reverse the procedure. Several techniques can be used to monitor the nervous system during surgery, and these are chosen depending on the part of the nervous system at risk and type of surgery. Common modalities used for monitoring include brainstem auditory evoked potentials (BAEP), somatosensory evoked potentials (SEP), transcranial electrical motor evoked potentials (MEP), electromyography (EMG), and EEG. Often, more than one modality will be used; this is known as multimodality monitoring. In this chapter, each modality is shown separately for illustration purposes; in practice, several modalities are monitored simultaneously.

BRAINSTEM AUDITORY EVOKED POTENTIALS

BAEP monitoring is used whenever there is potential for injury to the vestibulocochlear nerve or its pathways. Common procedures during which BAEP monitoring is used include microvascular decompression (MVD) surgery for trigeminal neuralgia and hemifacial spasm and cerebellopontine angle (CPA) tumor surgery. It may also be used during other types of brainstem surgery. BAEP monitoring has been shown to reduce the incidence of hearing loss associated with MVD surgeries.

During surgery the BAEP ipsilateral to the side of surgery is monitored closely. Changes in the latencies and amplitudes of the wave I and wave V in relation to the baseline are noted. The contralateral median nerve SEP is also periodically monitored. This evaluates conduction in the dorsal column pathways in the brainstem which lie close to the vestibulocochlear pathway. This type of multimodality monitoring is particularly useful in 329CPA tumor surgery. At times, the contralateral BAEP and ipsilateral median SEP are also checked for comparison purposes. The example (Figure 13.1) is from a 63-year-old patient undergoing MVD surgery for right-sided trigeminal neuralgia. BAEPs from right ear stimulation are shown. Stimulation parameters are noted at the bottom of the graph, as is the montage (thin arrows). Notice that the latency and amplitude of the wave V did not change significantly during the procedure (thick arrows). The vertical line is drawn on the wave V; notice the consistency with which the wave V falls on this line, indicating no significant change in latency. This monitoring did not suggest permanent damage to the vestibulocochlear pathway ipsilateral to the side of surgery.

**FIGURE 13.2.** This is intraoperative brainstem auditory evoked potentials (BAEP) monitoring data showing a small increase in wave V latency and a 50% decrease in amplitude.

During BAEP monitoring, a wave V latency prolongation of 1 msec or an amplitude decrement of 50% is considered significant. Three possible mechanisms can account for a change in the BAEP. First, one must consider technical issues, then global physiological changes such as anesthesia or blood pressure variation, and finally surgically induced change. The pattern of change of the BAEP (i.e., which waves are affected and how quickly) helps determine the etiology. If the change is thought to be surgery induced, the surgeon must be alerted immediately. During MVD surgery, the cerebellum may be retracted to expose the CPA. This can cause stretch injury to the vestibulocochlear nerve, which if severe enough can lead to hearing loss. BAEP are commonly performed in MVD to prevent this complication. In the sample above, the patient is undergoing MVD for right trigeminal neuralgia. At the start of the case, waves I (thin arrow) and V (thick arrow) are clearly identified. Soon after placement of the cerebellar retractor, there is prolongation of the wave V latency (notice the dot placed on the peak of wave V at baseline). The maximum latency prolongation is 0.6 msec, which does not reach the critical 1 msec point (dashed arrow). However, there is a significant decrease in the amplitude (greater than 50%; dotted arrow). The surgeon is alerted, and he repositions the cerebellar retractor. After completing the decompression, the retractor is removed and the wave V gradually returns to baseline (dash and dot arrow). There is change in wave I latency and morphology as well, but not as much as with wave V. The return of the BAEP to near baseline suggests that permanent damage to the vestibulocochlear pathway ipsilateral to the side of surgery did not occur (Figure 13.2).

**FIGURE 13.3.** This is intraoperative brainstem auditory evoked potentials (BAEP) monitoring data showing a wave V latency prolongation of 1.5 msec and amplitude reduction of more than 50%.

A 1.0 msec prolongation of wave V latency is considered significant, and the surgeon should be alerted. A persistent 1.0 msec or worsening latency shift is more likely to be associated with postoperative hearing loss. Latency shifts should be interpreted based on the underlying disease state. For example, in patients with CPA tumors, even smaller latency shifts may be clinically significant. In this sample from a patient undergoing resection of a left acoustic neuroma, there is gradual prolongation of the wave V latency, with the maximum shift being 1.5 msec (thin arrow; Figure 13.3). Notice that the vertical line is over the wave V at baseline; at the time of tumor dissection, there is maximal shift of the wave V. In this example, there is no significant change in wave V amplitude. By the end of the surgery, the latency of wave V is close to baseline (thick arrow; notice vertical line). Presence of wave I at the time of maximal wave V shift implies adequacy of stimulation (dashed arrow).

**FIGURE 13.4.** This is intraoperative brainstem auditory evoked potentials (BAEP) monitoring data showing loss of wave V during cerebellopontine angle (CPA) tumor dissection without return by the end of the surgery.

Loss of the wave V waveform is the most severe type of change that can occur with intraoperative BAEP monitoring. If it does not return by the end of the surgery, the patient is more likely to have postoperative hearing loss. However, loss of the wave V is not incompatible with preserved hearing (i.e., false-positive). When complete loss of wave V occurs suddenly, it is usually due to interruption of vascular supply of the vestibulocochlear nerve. If the loss is gradual, the etiology is more likely to be either mechanical or thermal trauma to the nerve. In the above sample of BAEP monitoring during left acoustic neuroma surgery, there is a robust wave V at the start of the case (thin arrow). However, as tumor dissection progresses, there is gradual loss of amplitude (thick arrow) and eventually there is complete loss of the wave V (dashed arrow; Figure 13.4). It does not return by the end of the case. The preserved wave I (dotted arrow) confirms that this change is not due to technical reasons.

**FIGURE 13.5.** This is intraoperative brainstem auditory evoked potentials (BAEP) monitoring data showing loss of wave V during microvascular decompression (MVD) with return by the end of the surgery.

Loss of wave V followed by recovery before the end of the surgery suggests that hearing will be preserved. This is especially true in patients undergoing MVD surgery. In the sample above, the patient is undergoing MVD for right trigeminal neuralgia. A wave V is noted at baseline (thin arrow), however, with cerebellar retraction there is gradual latency prolongation up to 0.7 msec and amplitude loss (thick arrow) and finally disappearance (dashed arrow). Once the retractor is removed, there is gradual recovery of the wave V (dotted arrow; Figure 13.5).

**FIGURE 13.6.** This is intraoperative brainstem auditory evoked potentials (BAEP) monitoring data showing loss of wave V during drilling of bone.

As noted previously, with surgery induced changes, BAEP changes can occur due to technical issues and systemic physiological changes. One such technical issue arises with bone drilling during exposure. The drill makes a loud noise which can mask the acoustic stimulus of the BAEP. This may cause loss of the BAEP waveforms. It is recommended that BAEP averaging be suspended during drilling. In the sample above, BAEP averaging is continued during drilling (Figure 13.6). Notice the lower amplitude wave V waveform (thin arrow). Before and after drilling, the wave V waveform is robust (thick arrows).

Many technical problems can lead to loss of BAEP waveforms. Inadequacy of stimulation may occur for many different reasons. A relatively common cause is kinking or clamping of the tubing used to transmit the acoustic stimulus from the sound generator to the ear. Obstruction of this tubing prevents the clicks from reaching the auditory system, and consequently a BAEP is not produced. The sample is from a patient undergoing MVD for right trigeminal neuralgia. Soon after positioning, the baseline response was obtained and revealed a robust wave V waveform (thin arrow). However, soon after draping the 335patient, there was sudden loss of the wave V (thick arrow). Notice also that the wave I also disappeared (dashed arrow). The absence of all BAEP waveforms suggested inadequacy of stimulation. After the clamps of the drape were removed, the BAEP response returned (dotted arrow; Figure 13.7).

**FIGURE 13.7.** This is intraoperative brainstem auditory evoked potentials (BAEP) monitoring data showing loss of wave V soon after draping the patient.

**FIGURE 13.8.** This is intraoperative brainstem auditory evoked potentials (BAEP) monitoring data showing latency prolongation and amplitude decrement of the wave V toward the end of the surgery.

The auditory stimulator which is secured in the external auditory canal can be inadvertently pulled out, especially towards the end of a long surgery. With dislodgement of the stimulator, the stimulus intensity reaching the ear is reduced. Consequently, BAEP may be misinterpreted as demonstrating prolonged latency and reduced amplitude. The technologist should confirm that the change is due to stimulator problems. Once the problem is found, it should be corrected; if possible another BAEP response is obtained to confirm no interval change. In this patient undergoing MVD for right trigeminal pain, there is gradual prolongation of latency and drop in amplitude of the wave V waveform towards the end of the case (thin arrow; Figure 13.8). At the end of the case, the technologist confirmed that the stimulator tubing had been dislodged. Notice that as the wave V disappears, so does the wave I, indicating a peripheral etiology for this change (thick arrow).

**FIGURE 13.9.** This is intraoperative monitoring data obtained from stimulation of the left ear. The surgery being performed was microvascular decompression for trigeminal neuralgia for the right ear. Notice the drop in amplitude of the wave V of the brainstem auditory evoked potentials (BAEP) obtained after stimulation of the left ear (thick arrow). The BAEP from the operative side (right) was unchanged.

As noted above, technical issues must be considered when changes in BAEP are noted. When right-sided MVD is being performed, it would be very unlikely for the left-sided BAEP to change at the time of closing, as was the case in this example (thick arrow). Without any changes in the right-sided BAEP, physiological changes, such as a drop in blood pressure, are unlikely to account for the change (Figure 13.9). A clue pointing to the cause of the change is the loss of amplitude of the wave I (thin arrow). This suggests that the stimulation probe may have been dislodged. As suspected, at the time of undraping, the ear insert on the left side had partially come out of the ear canal.

338SOMATOSENSORY EVOKED POTENTIALS

SEP monitoring is most often used for monitoring the spinal cord during surgeries in which it may be at risk. In scoliosis surgery, SEP has been shown to reduce neurologic morbidity. SEP can also be used during surgeries on the spinal cord, brainstem, and thoracic aorta. Additionally, median SEP have been used for cortical localization (discussed later). Tibial SEP are used when surgery involves risk to the spinal cord below the lower cervical level. In such cases, median or ulnar SEP can be used as controls to evaluate for systemic changes. For surgeries involving risk to the upper to mid cervical spinal cord, median or ulnar SEP are used (ulnar preferentially used if C6-7 region is at risk). Subcortical (P14/N18 for upper, P31/N34 for lower) and cortical (N20 for upper, P37 for lower) waveforms are followed during surgery.

**FIGURE 13.10.** This is intraoperative median and tibial somatosensory evoked potentials (SEP) monitoring data showing no significant change during the procedure.

As with BAEP, SEP changes can be induced by technical problems, general physiological changes, and surgery-induced neural tissue damage. The pattern of change helps localize the problem. It should be remembered that SEP monitors the dorsal column pathway (posterior aspect of spinal cord) and serves only as a surrogate marker for the corticospinal pathways. In many instances, SEP monitoring is done in conjunction with other types of monitoring. In the sample shown here, the patient is undergoing posterior spinal fusion for scoliosis. The median SEP are being used as a control. Both subcortical (P14, thin arrows) (first and third columns) and cortical (N20, thick arrows; second and fourth columns) responses are displayed. Tibial SEP are followed closely. Subcortical (P31, dashed arrows; first and third columns) and cortical (P37, dotted arrows; second and fourth columns) are displayed. No significant change in the responses is noted, and neurologic morbidity is not anticipated (Figure 13.10).