Somatosensory-Evoked Potential Phase-Reversal Technique

To indirectly identify the central sulcus, SEPs can be recorded from the exposed cerebral cortex by using the phase-reversal technique. SEPs are elicited by stimulation of the median nerve at the wrist and the posterior tibial nerve at the ankle (40-mA intensity, 0.2-msec duration, 4.3-Hz repetition rate). Recordings are performed from the scalp at CZ′-FZ (for legs) and C3′/C4′-CZ′ (for arms) according to the 10–20 International Electroencephalography System. After craniotomy, a strip electrode is placed across the exposed motor cortex and primary somatosensory cortex, transversing the central sulcus. This technique is based on the principle that an SEP, elicited by median nerve stimulation at the wrist, can be recorded from the primary sensory cortex.¹⁹ Its mirror-image waveform can be identified if some of the contacts of the strip electrode are placed on the opposite side of the central sulcus, over the motor cortex^20–22 (Fig. 4-1). For phase reversal, a strip electrode with four to eight stainless steel contacts with an intercontact distance of 1 cm is used. In the literature, the success rate of the phase reversal technique to indirectly localize the primary motor cortex ranges between 91%^20,21 and 97%.¹⁸ Interestingly, identification of the central sulcus by magnetic resonance imaging provided contradictory results when compared with intraoperative phase reversal.²⁰ Although it is expected that ongoing progress in the field of functional magnetic resonance imaging will eventually replace the need for neurophysiologic tests, ION still retains the highest reliability in mapping of the motor cortex and language areas when compared with functional neuroimaging.^23–26

FIGURE 4-1 Identification of the central sulcus by phase reversal of median nerve cortical somatosensory-evoked potentials. To the right is a schematic drawing of the exposed brain surface with a grid electrode position orthogonally to the central sulcus. On the left are the recorded evoked potentials phase reversed between electrode 6 and 7, showing a mirror image of the evoked potential between the motor and sensory cortices, depicting the central sulcus lying between electrodes 6 and 7.

(From Deletis V. Intraoperative neurophysiological monitoring. In: McLone D, ed. Pediatric Neurosurgery: Surgery of the Developing Nervous System, 4th ed. Philadelphia: WB Saunders; 1999:1204-1213.)

Direct Cortical Stimulation (60-Hz Penfield Technique)

Once the motor strip has indirectly been identified by the phase reversal technique, direct cortical stimulation is needed to confirm the localization of the motor cortex. Most current methods are based on the original Penfield technique. This calls for continuous direct cortical stimulation over a period of a few seconds with a frequency of stimulation of 50 to 60 Hz and observation of muscle movements.^14,16,27 An initial current intensity of 4 mA is used and, if no movements are elicited in contralateral muscles of the limbs and face, stimulation is increased in steps of 2 mA to the point at which movements are elicited.¹⁶ Muscle responses can either be observed visually or documented by multichannel electromyography, which appears to be more sensitive.²⁸ If no response is elicited with an intensity as high as approximately 16 mA, that area of cortex is considered not functional and can therefore be removed.²⁹ It should be emphasized that a negative mapping does not always ensure safety. To increase the chances of obtaining a positive mapping result, technical and anesthesiologic drawbacks have to be carefully ruled out and cortical exposure should be generous.

More in general, a limitation to the reliability of cortical mapping is the large variability of threshold for a positive mapping response across and within individuals.³⁰ A motor response from the same muscular group can be elicited from more than one cortical site, using different stimulation intensities.

Therefore, function localization may vary in different studies as a result of stimulation parameters and mapping strategies. Mapping strategies appear as one of the main variables that may affect the results of stimulation. Two different theories underline the choice of one or the other strategy:

Spreading of the current using the 60-Hz stimulation technique is limited to 2 to 3 mm as detected by optical imaging in monkeys.³⁴ Accordingly, one can assume that using this technique is safe for removal of tumors very close to the motor and sensory pathways as long as stimulation is repeated whenever a 2- to 3-mm section of tumoral tissue is removed.²⁹ Similarly, this technique allows us to map motor pathways subcortically while removing tumors that arise or extend to the insular, subinsular, or thalamic areas.^27,35 At the subcortical level, the stimulation intensity required to elicit a motor response is usually lower than that required for cortical mapping. When performing subcortical mapping, however, we have to keep in mind that a distal muscle response after stimulation of subcortical motor pathways can be misleading. Although this stimulation activates axons distal to the stimulation point, the possibility of damage to the pathways proximal to that point cannot be ruled out. This is a concern, especially when dealing with an insular tumor where there is a risk of cortical or subcortical ischemia induction secondary to manipulation of perforating vessels (Fig. 4-2).

FIGURE 4-2 Upper left: Preoperative axial contrast-enhanced, magnetic-resonance, T1-weighted image of a right frontotemporoinsular anaplastic astrocytoma that was removed with the assistance of intraoperative neurophysiologic monitoring. Lower left: Intraoperative view at the end of tumor resection. The internal capsule (C) has been identified using mapping of the subcortical motor pathways with the short train of stimuli technique. The temporal lobe (T) and branches of the middle cerebral artery (MCA) are on view. Right: Motor-evoked potentials (MEPs) recorded intraoperatively from the left abductor brevis pollicis (L-APB) and tibialis anterior (L-TA) muscles. MEP recordings at the end of tumor removal (A); MEP loss during hemostasis (B); transitory MEP reappearance by increasing stimulation to seven stimuli and 130-mA intensity (C); new disappearance of MEPs despite increased stimulation (D); progressive MEP reappearance after papaverine infusion and increased systemic blood pressure (E and F); MEPs at the end of the procedure (G).

(Modified from Sala F, Lanteri P. Brain surgery in motor areas: The invaluable assistance of intraoperative neurophysiological monitoring. J Neurosurg Sci. 2003;47:79-88.)

Despite its popularity in the past, this 60-Hz Penfield technique has some disadvantages. With the exception of speech mapping, it is our opinion that these disadvantages should prevent its use as a motor cortex/pathways mapping technique. First, this technique can induce seizures in as many as 20% of patients, despite therapeutic levels of anticonvulsants and regardless of whether there is a preoperative history of intractable epilepsy.^36,37 Second, in children younger than 5 years old, direct stimulation of the motor cortex for mapping purposes may not yield localizing information because of the relative unexcitability of the motor cortex.^19,38 Third, because this is a mapping and not a monitoring technique, no matter how often cortical or subcortical stimulation is repeated, the functional integrity of the motor pathways cannot be assessed continuously during surgery.

Direct Cortical Stimulation and Motor-Evoked Potential Monitoring (Short Train of Stimuli Technique)

Recently, mapping techniques have integrated monitoring techniques to continuously assess the functional integrity of the motor pathways and therefore increase the safety of these procedures.^20,39–41 The following is a description of the technique that we use at our institutions and have found suitable for both mapping and monitoring.

Muscle MEPs are initially elicited by multipulse transcranial electrical stimulation (TES). Short trains of five to seven square-wave stimuli of 500-μsec duration with an interstimulus interval of 4 msec are applied at a repetition rate of as high as 2 Hz through electrodes placed at C1 and C2 scalp sites, according to the 10–20 International Electroencephalography System. The maximum stimulation intensity should be as high as 200 mA, which is strong enough for most cases. Muscle responses are recorded via needle electrodes inserted into the contralateral upper and lower extremity muscles. We usually monitored the abductor pollicis brevis and the extensor digitorum communis for the upper extremities and the tibialis anterior and the abductor hallucis for the lower extremities. For the face area, the orbicularis oculi and orbicularis oris muscles are typically used.

After exposure of the cortex and once phase reversal has been performed, direct cortical stimulation of the motor cortex can be achieved by using a monopolar-stimulating probe to identify the cortical representation of contralateral facial and limb muscles. The same parameters of stimulation used for TES, except for a much lower intensity (≤20 mA), can be used.³⁹ Sometimes the short train of stimuli technique requires slightly higher current intensities than those required by the Penfield technique. However, by using a very short train, the charge applied to the brain is significantly reduced⁴² and, consequently, the risk of inducing seizures. The number of pulses in the short-train technique is five to seven pulses per second, whereas in the Penfield technique, there are 60 pulses per second. The effect of stimulation on the cerebral cortex, from a neurophysiologic point of view, differs between the Penfield technique and the short train of stimuli technique. The Penfield technique delivers one stimulus every 15 to 20 msec continuously for a couple of seconds. The short train of stimuli technique delivers five to seven stimuli in a period of approximately 30 msec with a long pause between trains (470 to 970 msec, which depends on train repetition rate—1 or 2 Hz). Therefore, the Penfield technique is more prone to produce seizures, activating the cortical circuitry more easily than short-train stimuli do. Furthermore, compared with the Penfield technique, the short-train technique does not induce strong muscle twitches that may interfere with the surgical procedure. Responses are usually recorded from needle electrodes used to record muscle MEPs elicited by TES. However, any combination of recording muscles can be used, according to the tumor location. The larger the number of monitored muscles, the lower the chance of a false-negative mapping result. We suggest that stimulation of the tumoral area should always be performed to rule out the presence of some functional cortex. As already described, this is especially true in the case of low-grade gliomas.^3–5

In the illustrative case presented in Fig. 4-2,³⁹ an impairment of muscle MEPs occurred at the end of tumor removal when opening and closing mapping procedures had already been done and confirmed the integrity of motor pathways distal to the stimulation point at the level of the internal capsule. However, ischemia of the pyramidal tracts secondary to severe vasospasm of the main perforating branches of the middle cerebral artery occurred during hemostasis and was detected by muscle MEP monitoring. If not detected in time, this event would have likely resulted in an irreversible loss of muscle MEPs and, consequently, a permanent motor deficit. Mapping techniques are unlikely to detect these events because they do not allow a continuous “online” assessment of the functional integrity of neural pathways.

In our experience with using the short-train technique, a threshold lower than 5 mA for eliciting muscle MEPs usually indicated proximity to the motor cortex. When muscle responses are elicited through higher stimulation intensities, activation of the CT is of less localizing value because of the possibility of spreading of the current to adjacent areas.³⁹

Once mapping of the cortex has clarified the relationship between eloquent motor areas and the lesion, continuous MEP monitoring of the contralateral muscles can be sustained throughout the procedure to assist during the surgical manipulation. To do so, one of the same contacts of the strip electrode can be used as an anode for stimulation while the cathode is at Fz. The stimulation point on the motor cortex with the lowest threshold used to elicit muscle MEPs from contralateral limbs or face usually corresponds with the contact from which the largest amplitude of the mirror-image SEPs was obtained. The same stimulation parameters as those used for the short-train mapping technique can be used.

When removing a tumor that extends subcortically, preservation of muscle MEPs during monitoring from the strip electrode will guarantee the functional integrity of motor pathways and avoid the need for periodic remapping of the cortex at known functional sites.³⁹

For insular tumors where the motor cortex is not exposed by the craniotomy, a strip electrode can still be gently inserted into the subdural space to overlap the motor cortex. Phase reversal and/or direct cortical stimulation can be used to identify the electrode with the lowest threshold to elicit muscle MEPs. The use of MEPs during surgery for insular tumors has proved very useful to identify impending vascular derangements to subcortical motor pathways in time for corrective measures to be taken. In spite of the observation that intraoperative MEP changes occurred in nearly half of the procedures, these were reversible in two thirds of the cases.⁴³

Warning Criteria and Correlation with Postoperative Outcome

Still debated are the warning criteria for changes in muscle MEPs that are used to inform the surgeon about an impending injury to the motor system. It should be stressed that although for spinal cord surgery, a “presence/absence” of muscle MEPs criterion has proved to be reliable and strictly correlates with postoperative results,^44,45 there are not definite MEP parameters indicative of significant impairment during supratentorial surgery.⁴⁶ We believe that the predictive value of muscle MEPs is different for supratentorial and spinal cord surgeries. As such, different warning criteria must be employed. This judgment is based on the difference in types of CT fibers in supratentorial portion of the CT as compared with the spinal cord. Different groups with established experience in this field have proposed similar criteria,^40,46 suggesting that a shift in latency between 10% and 15% and a decrease in amplitude of more than 50% to 80% correlate with some degree of postoperative motor deficit. However, a permanent new motor deficit has consistently correlated only with irreversible complete loss of muscle MEPs.⁴⁶

A persistent increase in the threshold to elicit muscle MEPs or a persistent drop in muscle MEP amplitude, despite stable systemic blood pressure, anesthesia, and body temperature, represents a warning sign. However, it should be noted that muscle MEPs are easily affected by muscle relaxants, bolus of intravenous anesthetics and high concentrations of volatile (and other) anesthetics such that wide variation in muscle MEP amplitude and latency can be observed.⁴⁷ Due to this variability, the multisynaptic nature of the pathways involved in the generation of muscle MEPs, and the nonlinear relationship between stimulus intensity and the amplitude of muscle MEPs, the correlation between intraoperative changes in muscle MEPs (amplitude and/or latency) and the motor outcome are not linear. Further clinical investigation is needed to clarify sensitive and specific neurophysiologic warning criteria for brain surgery.

Mapping Techniques

Neurophysiologic mapping techniques have been increasingly used to localize CT and cranial nerve motor nuclei on the lateral aspect of the midbrain and on the floor of the fourth ventricle.^50–54

Mapping of the Corticospinal Tract at the Level of the Cerebral Peduncle

This is a recently described technique used to map the CT tract within the brain stem at the level of the cerebral peduncle.^54,55 To identify the CT, we use a hand-held monopolar-stimulating probe (0.75-mm tip diameter) as a cathode, with a needle electrode inserted in a nearby muscle as an anode. If the response (D wave) is recorded from an epidural electrode, a single stimulus is used. Conversely, if the response is recorded as a compound muscle action potential from one or more muscles of contralateral limbs, a short train of stimuli should be used.

We usually increase stimulation intensity to 2 mA. When a motor response is recorded, the probe is then moved in small increments of 1 mm to find the lowest threshold to elicit that response.

This technique is particularly useful for midbrain tumors that have displaced the CT tract from its original position. Usually, the so-called midbrain lateral vein described by Rhoton⁵⁶ represents a useful anatomic landmark because it allows an indirect identification of the CT tract, located anterior to the vein. However, when an expansive lesion distorts anatomy, only neurophysiologic mapping allows the identification of the CT and, consequently, a safe entry zone to the lateral midbrain.

In the case of a cystic midbrain lesion, sometimes mapping of the CT is negative at the beginning of the procedure, but a positive response can be recorded when mapping from within the cystic cavity toward the anterolateral cystic wall.⁵⁷

Mapping of Motor Nuclei of Cranial Nerves on the Floor of the Fourth Ventricle

This technique is based on intraoperative electrical stimulation of the motor nuclei of the cranial nerves on the floor of the fourth ventricle, using a hand-held monopolar stimulating probe. Compound muscle action potentials are then elicited in the muscles innervated by the cranial motor nerves. A single stimulus of 0.2-msec duration is delivered at a repetitive rate of 2.0 Hz. Stimulation intensity starts at approximately 1 mA and is then reduced to determine the point with the lowest threshold that elicits muscle responses corresponding with the mapped nucleus (Fig. 4-3). No stimulation intensity higher than 2 mA should be used.^50,51 To record the responses from cranial motor nerves VII, IX/X, and XII, wire electrodes are inserted into the orbicularis oculi and orbicularis oris muscles, the posterior wall of the pharynx, and the lateral aspect of the tongue muscles, respectively. Based on mapping studies, characteristic patterns of motor cranial nerve displacement, secondary to tumor growth, have been described (Fig. 4-4).⁵⁸ The case described in Fig. 4-5 is consistent with this observation.

FIGURE 4-3 Mapping of the brain stem cranial nerve motor nuclei. Upper left: Drawing of the exposed floor of the fourth ventricle with the surgeon’s handheld stimulating probe in view. Upper middle: The sites of insertion of wire hook electrodes for recording the muscle responses are depicted. Far upper right: Compound muscle action potentials recorded from the orbicularis oculi and oris muscles after stimulation of the upper and lower facial nuclei (upper two traces) and from the pharyngeal wall and tongue muscles after stimulation of the motor nuclei of cranial nerves IX/X and XII (lower two traces). Lower left: Photograph obtained from the operating microscope shows the hand-held stimulating probe placed on the floor of the fourth ventricle (F). A, aqueduct.

(Reproduced from Deletis V, Sala F, Morota N. Intraoperative neurophysiological monitoring and mapping during brain stem surgery: A modern approach. Oper Tech Neurosurg. 2000;3:109-113.)

FIGURE 4-4 Typical patterns of cranial nerve motor nuclei displacement by brain stem tumors in different locations. Upper and lower pontine tumors: Pontine tumors typically grow to push the facial nuclei around the edge of the tumor, suggesting that precise localization of the facial nuclei before tumor resection is necessary to avoid their damage during surgery. Medullary tumors: Medullary tumors typically grow more exophytically and compress the lower cranial nerve motor nuclei ventrally; these nuclei may be located on the ventral edge of the tumor cavity. Because of the interposed tumor, in these cases mapping before tumor resection usually does not allow identification of cranial nerve IX/X and XII motor nuclei. Responses, however, could be obtained close to the end of the tumor resection when most of the tumoral tissue between the stimulating probe and the motor nuclei has been removed. At this point, repeat mapping is recommended because the risk of damaging motor nuclei is significantly higher than at the beginning of tumor debulking. Cervicomedullary junction spinal cord tumors: These tumors simply push the lower cranial nerve motor nuclei rostrally when extending into the fourth ventricle.

(Reproduced from Morota N, Deletis V, Lee M, et al. Functional anatomic relationship between brain stem tumors and cranial motor nuclei. Neurosurgery. 1996;39:787-794.)

FIGURE 4-5 Upper panel: (Top): Preoperative contrast-enhanced, T1-weighted magnetic resonance imaging (MRI) of an upper left pontine low-grade astrocytoma in a 16 year old female. Bottom: Postoperative MRI study showing complete tumor removal. Surgery was performed under neurophysiologic guidance. Middle panel: Direct mapping of the facial nerve motor nuclei on the floor of the fourth ventricle. The tumor was approached through a median suboccipital craniectomy. When the floor of the fourth ventricle was exposed, the median sulcus appeared dislocated to the right and the left median eminence was expanded. Electromyographic wire electrodes were inserted bilaterally in the left (LU) and right (RU) orbicularis oculi and left (LL) and right (RL) orbicularis oris muscles for mapping and monitoring of the seventh nerve, and in the abductor pollicis brevis (LA and RA) for continuous monitoring of the corticospinal tract integrity. We initially stimulated on the left side, approximately 1.5 cm rostral to the striae medullares, where the motor nuclei were expected to be according to normal brain stem functional anatomy. A response was obtained from the left orbicularis oculi (LU) at a stimulation intensity of 1.5 mA (A). By moving the stimulating probe caudally and to the right, a consistent response from the left orbicularis oris (LL) was recorded at a stimulation intensity of 0.5 mA (B). At this point, we moved the stimulation probe more laterally to the right side, approximately 1 cm above the striae medullares, and a clear response was recorded from the right orbicularis oris (RL) at the lowest threshold intensity of 0.2 mA (C). Finally, by moving the stimulating probe paramedially to the left side, a few millimeters above the striae medullares, a consistent response was recorded from both the left orbicularis oris (LL) and orbicularis oculi (LU), using the same low threshold (0.2 mA) (D). The conclusion was drawn that the tumor displaced caudally the facial nerve motor nuclei, especially on the left side. Based on mapping results, the surgeon decided to enter the brain stem on the left side in correspondence with the higher threshold stimulating point (A).Lower left: Schematic summary of mapping results. A and B represent the original position of the left and right facial colliculi, as expected according to brain stem anatomy. A, B, C, and D correspond to the stimulating point illustrated in the upper panel. C and D also correspond to the lower threshold to elicit a consistent response from, respectively, the right and left muscles innervated by the facial nerve. The conclusion was made that real location of facial nerve motor nuclei (C and D) was more caudal than expected, especially on the left side, due to the tumor mass effect. Accordingly, initial incision (I) was carried on transversely in correspondence with stimulating point A Lower right: Continuous neurophysiologic monitoring of muscle motor-evoked potentials during tumor removal. Electromyographic wire electrodes were inserted in the left orbicularis oris (LL) and abductor pollicis brevis (LA) muscles for continuous monitoring of, respectively, the corticobulbar and corticospinal tract integrity, after transcranial electrical stimulation (electrode montage C4/Cz; short train of four stimuli; intensity 50 mA).

(Modified from Sala F, Lanteri P, Bricolo A. Intraoperative neurophysiological monitoring of motor evoked potentials during brain stem and spinal cord surgery. Adv Tech Stand Neurosurg. 2004;29:133-169.)