26.2.1 Acoustic Nerve Monitoring

The acoustic nerve (CN VIII) is commonly monitored with brainstem auditory evoked potentials when hearing preservation surgery is attempted. Anesthetics and muscle relaxants have little effect on these potentials, so they do not need to be considered when selecting anesthetic agents.s. Literatur

26.2.2 EMG Monitoring

In the context of VS surgery, “EMG monitoring” utilizes two different types of stimulus at the nerves. Literatur:

• 
  

  Spontaneous muscle activity, also referred to as passive or free-run EMG. Baseline muscle activity reflects spontaneous nerve fascicle discharges. The low frequency of spontaneous muscle activity will increase with irritation or injury to the nerve. When total spike frequency exceeds 30 Hz, it is considered a neurotonic discharge and a sign of possible nerve injury.s. Literatur In addition to the graphic display of discharges, they can be translated into a frequency-dependent audio output, thus providing immediate real-time warning to the surgeon.

  
  
• 
  

  Stimulus-evoked EMG, also referred to as active or triggered EMG. Application of a stimulating electrode to a motor nerve produces a compound muscle action potential in its target muscle.s. Literatur This can assist in identification of the nerve, particularly if its anatomy has been distorted by the tumor. Conversely, it can be used to exclude unknown structures encountered during surgical dissection as the motor nerve is being monitored. It can also be used to evaluate functional status of the nerve, using stimulus threshold and EMG amplitude.

26.2.3 EMG Monitoring Targets

The cranial nerves most frequently monitored during VS surgery, and their target muscles, are listed in Table 26‑1 . Because of the close anatomic course of the facial nerve alongside the acoustic nerve, the facial nerve is at particular risk of damage during VS resection. EMG monitoring of the facial nerve has been shown to improve outcome,s. Literatur and should always be utilized. Larger VS tumors may place other cranial nerves at risk and make EMG monitoring of them prudent.s. Literatur ^,​ s. Literatur The rostral border of a large tumor may threaten the trigeminal nerve (V). The caudal border of a large tumor may threaten the lower cranial nerves (IX–XII). There is no compelling evidence that EMG monitoring of cranial nerves other than the facial improves outcome.s. Literatur In any event, EMG can assist in identification of cranial nerves where the tumor has distorted the nearby anatomy and thereby aid the surgical approach.

26.3.1 Timely Emergence from Anesthesia at End of Long Surgery

As in most craniotomies, it is desirable when VS surgery is concluded to have the patient awaken from anesthesia quickly and be able to undergo a rudimentary neurologic exam, so as to exclude hemorrhage or ischemia at the operative site. In other neurosurgical anesthetics, this is facilitated by using the minimal adequate dose of general anesthetic to prevent awareness and blunt hemodynamic responses, together with a dense muscle paralysis. However, dense paralysis is not compatible with EMG monitoring.

Furthermore, the characteristics of VS surgery act to increase the time to emergence from anesthesia. It is prolonged by the dose of anesthetic, which must be increased for EMG monitoring without muscle relaxant to ensure immobility, and the length of time under anesthesia.s. Literatur VS surgery requires meticulous exposure and dissection to remove the tumor without damage to the facial and other cranial nerves. Operative time will vary with tumor size and other anatomic variables, but it is one of the longer neurosurgeries. An older study reported an average operating room time of 6.7 ± 2.0 hours for tumors ≤ 1 cm, and 10.4 ± 2.8 hours for tumors ≥ 2.5 cm.s. Literatur

26.3.2 Immobility with Minimal Paralytic Muscle Relaxant

There are two major approaches to providing immobility without a paralytic agent preventing the muscle response necessary for EMG monitoring (except for initial, short-acting dose at intubation). The classical approach uses infusion of muscle relaxant to achieve a submaximal paralysis allowing EMG, but preventing patient movement in combination with an anesthetic level that by itself would not ensure immobility. Given the limitations of anesthetic agents that were available previously, primarily prolonged emergence after high doses, it was reasonable to conclude that “muscle relaxation is an essential component of balanced anesthesia.”s. Literatur Most studies have been done with atracurium or a derivative at <75% neuromuscular blockade. Its half-life is determined solely by its rate of spontaneous breakdown in the blood, independent of hepatic or renal function, giving the most stable submaximal paralytic level with an infusion.

The classical approach has two advantages related to decreasing the anesthetic depth required. Older anesthetic agents such as isoflurane and fentanyl can be used at lower levels without overly delaying time to awakening at the end of surgery. Also, lower levels of anesthetic will cause less hemodynamic depression. In a patient with severe cardiovascular disease, this may be a compelling consideration. The major disadvantage of the classical approach is that even partial paralysis can inhibit EMG. Although this is not clinically significant in all patients, it is most problematic in patients with some degree of preexisting nerve injury, or surgical anatomy putting the nerve at heightened risk of injury. If an EMG response is altered under these circumstances, it may be difficult to confidently distinguish between surgical nerve injury and drug paralytic effect. Furthermore, the effect of partial neuromuscular blockade on spontaneous EMG is poorly validated.s. Literatur The more recent, nonrelaxant approach uses high levels of anesthetic to provide complete immobility without administering any paralytic agent except for intubation. Its major advantage is that it removes anesthetic paralysis completely as an influence on EMG. With few exceptions, this is the author’s preferred approach for VS surgery. Facial nerve paresis represents a significant morbidity and all efforts should be made to remove anesthesia as a confounding variable in EMG monitoring.

Unfortunately, there is no completely reliable predictor of movement under anesthesia. Neither hemodynamics nor the bispectral index (BIS) processed EEG monitor is a reliable predictor of movement during anesthesia.s. Literatur ^,​ s. Literatur BIS was originally designed to predict movement but was not reliable for that purpose, particularly with a narcotic as a significant component of the general anesthetic, and its algorithms were modified to provide an awareness monitor instead.s. Literatur Interestingly, there is one study reporting that spontaneous EMG of the facial nerve may have utility as a predictor of movement under anesthesia in craniotomies, with a negative predictive value of 95%.s. Literatur

Without a monitor that reliably predicts movement under anesthesia, it is necessary to use a supramaximal dose of anesthetic. The effect of a volatile anesthetic on movement is defined by its minimum alveolar concentration (MAC), the gas concentration in the lungs at steady state at which 50% of patients will not move in response to surgical incision. The standard deviation of volatile anesthetic dose response curves for movement is approximately 10%. To achieve no movement in 95% of patients (ED95) will then require a dose 2 SDs above the mean, or 1.2 MAC.s. Literatur However, standard MAC values are defined for and apply only to surgical incision. The intraoperative use of EMG stimulation of the larynx and the potential for laryngeal tracheal stimulation by inadvertent movement of the endotracheal tube, directly or via attached ventilator tubing, argue for using the MAC for endotracheal intubation (MAC-EI) which is approximately 1.5 times standard MAC.

While the nonrelaxant approach is possible with volatile anesthetic as the sole anesthetic agent, a greatly prolonged time to emergence from anesthesia will result, as well as more hemodynamic instability. A more balanced approach has been made possible by the introduction of potent, ultrashort-acting anesthetics, particularly remifentanil, with a context-sensitive half-life of <5 minutes regardless of infusion duration.s. Literatur The nonrelaxant approach combines a dose of general anesthetic, which allows reasonably timely emergence after prolonged anesthesia, with a high-dose remifentanil infusion. Much of the data available on adequate levels of anesthetic are anecdotal. One well-designed study found that 21% of patients without relaxant moved during craniotomy when given isoflurane 0.6% (0.5 MAC) and remifentanil 0.21 µg/kg/min.s. Literatur This has been adapted in the author’s practice to successfully prevent movement with isoflurane 1.0% (measured as end-tidal concentration) and remifentanil 0.25 to 1.0 µg/kg/min, although this should also be regarded as anecdotal until further study is done.

The use of desflurane or sevoflurane rather than isoflurane during EMG monitoring would allow much faster emergence from anesthesia after a long surgery, and all of the volatile anesthetics preserve neuromuscular function and should be compatible with EMG monitoring. However, there is concern that desflurane and sevoflurane may cause a higher baseline level of spontaneous neuronal discharge than isoflurane, so that subtle neurotonic discharges may be missed. This has not been formally studied or reported. The author’s current practice is to use isoflurane during EMG monitoring, and then switch to desflurane when monitoring is finished.