Interpretation of EN o G

Like electronystagmography, there are normal, expected variations from side to side and from person to person. Therefore, there must be at least a 30% difference between the 2 sides before it can be regarded as evidence of facial nerve dysfunction on the side with the smaller amplitude. Care must be taken in the interpretation of ENoG results, because operator error can result in poor test-retest reliability. In particular, ENoG testing requires training and experience for it to be accurate, including the use of an optimized lead placement strategy instead of standardized placement.

ENoG before surgery is useful for multiple reasons. Patients with clinically normal facial function may have subclinical neuropathy. With this knowledge, the surgeon can better counsel the patient with severe preoperative ENoG abnormalities who may be at higher risk of postoperative facial paralysis. Reasons for severe preoperative ENoG abnormalities include:

• 1.
  

  An acoustic tumor being more infiltrative or compressive than typical tumors

  
  
• 2.
  

  The presence of another type of tumor that may have a higher risk of facial palsy (eg, facial neuroma).

Preoperative counseling in these circumstances can include a greater emphasis on treatment of facial palsy, subtotal resection, or decompression in lieu of resection. Residual tumor can be followed or treated with stereotactic radiosurgery. Preoperative facial nerve dysfunction can influence intraoperative facial nerve monitoring, making such monitoring less effective because the nerve is often in a hyperexcitable state. ENoG allows the surgeon and monitoring neurophysiologist to be prepared for this possibility.

In contrast, ENoG is not typically helpful in the postoperative setting, even in the case of a delayed facial palsy following surgery. ENoG may be helpful in the context of a complete postoperative facial paralysis to determine when facial nerve decompression is indicated, in the same way that it used for patients with Bell palsy, in which edema at the meatal foramen (the narrowest point of the fallopian canal) is posited to be the epicenter of neural injury. However, abnormal postoperative ENoG does not pinpoint the meatal foramen as the epicenter of injury after it has undergone sustained surgical dissection over a broad segment of the nerve during tumor resection. Thus, the usefulness of surgical reexploration to decompress the nerve after an abnormal electroneurogram is unclear, because there may be multiple causes for facial nerve injury. Our practice is to routinely decompress the meatal foramen during the initial translabyrinthine and middle cranial fossa approaches for vestibular schwannoma to minimize this potential mechanism of postoperative nerve injury. In addition, with a postoperative facial palsy, we institute therapy with steroids and antivirals, because viral reactivation is another potential cause of postoperative facial nerve weakness.

Training issues in intraoperative facial nerve monitoring

The successful use of intraoperative facial nerve monitoring requires that the surgeon is trained to understand and interpret the physiologic responses. Furthermore, if the surgeon also takes on the responsibility of the electrode and device setup, specific training in the technical aspects of monitoring is required. Meticulous attention to detail must be paid to anesthesia, the monitoring device, and the electrodes to ensure accurate results. At our institution, the surgeon performs intraoperative monitoring in conjunction with a technologist who has received special training and certification (Certification in Neurophysiologic Intraoperative Monitoring [CNIM]). However, at many institutions, surgeons perform the technical setup. Although this is an appropriate model, many staff physicians have had little or no monitoring training other than a cursory introduction by a sales representative. In academic institutions, monitoring setup is often performed by residents whose only training is cursory, word-of-mouth advice by other residents who themselves typically lack an understanding of the technical aspects of monitoring.

Our review of medicolegal cases, as well as quizzing many residents from around the country, often confirms a lack of monitoring training. Staff surgeons come to rely on the residents, assuming a higher level of knowledge than may be present. To improve monitoring and enhance patient safety, we suggest that a written protocol be established for monitoring at each institution, and that a written curriculum with competency testing be required for all individuals performing monitoring. Furthermore, routine, rather than sporadic, use of facial nerve monitoring during operations in which the nerve potentially is at risk results in staff who are familiar with proper techniques and troubleshooting.

Setup and technique for facial nerve monitoring

Several steps are necessary to reliably perform EMG-based facial nerve monitoring.

Step 1. Ensure that the anesthesiologist avoids the use of long-acting muscle relaxants. Short-acting muscle relaxants such as succinylcholine are typically acceptable for induction, except in the rare case of the patient with pseudocholinesterase deficiency, in whom this drug is broken down slowly and thus experiences prolonged paralysis.

Step 2. Be wary about local anesthesia (eg, lidocaine or marcaine), which can chemically induce a temporary facial paresis, rendering monitoring useless. Most are aware that the injection must avoid the facial nerve at the stylomastoid foramen near the mastoid tip, which is particularly important in children for whom the mastoid tip may be poorly developed. Less well known but of great importance is that care must also be taken to avoid local anesthesia entering the middle ear, which is surprisingly common during chronic ear surgery when injecting the ear canal in the presence of a tympanic membrane perforation.

Step 3. Place electrodes carefully. Intramuscular needle electrodes are inserted in a paired manner at the nasolabial groove (orbicularis oris) and near the eyebrow (orbicularis oculi) on the side to be monitored ( Fig. 1 ). Care is taken to direct the electrodes away from the ocular globe to avoid inadvertent trauma to the eye. Our experience over the decades has shown that there is no need to risk ocular injury by placing the needle electrodes inferior to the eyebrow. Two additional electrodes are placed subcutaneously in a distal location such as over the sternum or the ipsilateral shoulder, one serving as a ground for the recording channels and the other as a return for the monopolar nerve stimulator. Fig. 1 shows the ground and anode electrodes placed on the forehand, a location we had used for many years and is used in many centers today. In more recent years, we place these electrodes on the sternum which appears to result in less cautery artifact. Consistency is important to reduce errors. We routinely color code electrodes: blue for eyes, red for lips, green for ground, white for anode, and black for cathode. Our mnemonic for this, which many have adopted internationally, is Blue Eyes, Red Lips, Green Ground. The cathode is used for stimulation rather than the anode because cathodal stimulation is more effective.

EMG recording montage used for 2-channel facial nerve monitoring. Paired electrodes are placed subcutaneously at the nasolabial groove (orbicularis oris) and near the eyebrow (orbicularis oculi) on the side to be monitored. Ground and anodal return electrodes can be placed high on the forehead (as shown) or, more commonly today, near the sternum (not shown).

( From Kartush JM. Electroneurography and intraoperative facial monitoring in contemporary neurotology. Otolaryngol Head Neck Surg 1989;101:501; with permission.)

Step 4. After the electrodes are connected to the nerve monitor, check the impedances of the different electrode pairs. The independent electrode impedance should be less than 5 kohm, whereas interelectrode impedance should be less than 2 kohm. If the impedance is too high, this suggests that the needle electrodes may be poorly positioned or faulty; the electrodes should be repositioned or replaced and then retested.

Step 5. Perform a “tap test” to check the integrity from the electrode to the recording device. Tapping the skin over each pair of facial recording electrodes should elicit a visual signal on the oscilloscope as well as a concurrent acoustic signal from the loudspeaker. It is important that facial nerve monitors represent responses not just visually on an oscilloscope but also acoustically so the surgeon receives real-time feedback. The surgeon cannot continually observe the oscilloscope and operate. For those who only rely on a technician watching the screen, if the technician looks away for a moment during a lengthy case, an important response can be missed. Conversely, devices that only have an audible response with no oscilloscope lose the opportunity to differentiate some forms of artifact from true responses. Multimodality monitoring increases the need to have a technical assistant. Other increasingly common modalities include laryngeal monitoring, auditory brainstem recording, somatosensory recording, and transcranial motor evoked potentials.

One common misunderstanding about the tap test is that hearing a response means that everything is working perfectly. This is not the case. Performing a tap test on a paralyzed face also creates a response, because tapping adjacent to an inserted electrode elicits an electrical artifact, not a true CMAP. The tap test has benefit in that it checks whether the connections from the recording electrodes to the oscilloscope and loudspeaker are intact, but it provides no information regarding the stimulating side of the system or about the functional state of the facial nerve or muscles. Thus, the setup is not finished after obtaining a positive tap test, but should also include the final 2 steps (described later).

Step 6. After incision and soft tissue exposure, check for current flow using a nerve stimulator. Touching the stimulator to soft tissue or wet bone should result in near 100% conduction of current from the stimulator to the monitor. Most devices display the returned current visually. Others may present an audible tone (eg, warble) or digital voice confirmation (eg, stimulus) that is distinct from a true response tone (eg, beep). The volume of the nerve monitor should be adjusted from the start to ensure that the surgeon can hear it over the ambient sounds in the operating room (eg, the drill, suction, laser, babble).

Step 7. Stimulate the nerve at an early point in the surgery before any significant manipulation of the nerve is performed. Placing the nerve stimulator near the porus acousticus, the root entry zone, or the second genu of the facial nerve after entry into the mastoid antrum during surgical exposure triggers a CMAP that is detected by the facial nerve monitor. The distance from the nerve and the amount of intervening tissues determines the current setting. Even though a normal facial nerve responds to 0.05 mA in the cerebellopontine angle, if the surgeon has 0.5 cm of intervening soft tissue or a few millimeters of bone, the current may need to be temporarily turned up to 1 to 2 mA before a baseline, far-field response can be obtained by volume conduction of the current through intervening tissue. Once a confirmatory baseline response is obtained, we reduce the current intensity to lower levels based on how distal we are from the nerve. In 30 years of using such mapping techniques with modern current settings (eg, pulse durations ∼100 microseconds at ∼5 Hz), we have observed no detrimental effect to the nerve. Conversely, those who neglect to routinely obtain a baseline response before nerve dissection risk monitoring with a system that may be faulty due to set up errors, persistent muscle relaxants or chemical paralysis from lidocaine.

See Box 1 for a summary of steps for facial nerve monitoring.

Interpretation of intraoperative facial nerve EMG monitoring

The surgeon must be knowledgeable not only in anatomy but in the pertinent neurophysiology to make proper use of monitoring. Even flawlessly accrued information conveyed to a surgeon is of little value if the data cannot be interpreted and surgical maneuvers implemented based on the information. Electromyographic responses can:

• 1.
  

  Occur because of trauma or other nonelectric stimuli, or

  
  
• 2.
  

  Be electrically triggered.