55.3.1 Anatomical Considerations

The facial nerve lies in direct anatomical apposition to the tumor capsule in all cases regardless of tumor size. Small tumors typically interface the facial nerve along its intracanalicular course, with the nerve located along the anterior and superior aspect in most cases. As tumors enlarge and extend out of the internal auditory canal (IAC), the facial nerve invariably has more extensive length of contact with the tumor capsule. In part, this relationship between tumor size and length of facial nerve contact explains the greater risk of facial neuropathy when resecting or radiating larger tumors.

In the context of VS microsurgery, there are only two locations where the facial nerve position is constant: the facial nerve departs the brainstem at the pontomedullary junction, just anteroinferior to the cochlear nerve, and the facial nerve enters the temporal bone at the meatal foramen at the fundus of the IAC. Apart from this, significant variability arises with larger tumors with respect to the location of the facial nerve along the tumor capsule. In most VS, the facial nerve follows a course along the ventral aspect of the tumor in the cerebellopontine angle (CPA); however, not uncommonly it may be encountered along the superior or inferior aspect of the tumor and rarely along the dorsal capsule.s. Literatur ^,​ s. Literatur Another curious observation is that with larger tumors, the facial nerve does not always take the shortest route from the brainstem to the fundus. For example, in some cases, the facial nerve may follow a meandering course over the superior pole of the tumor in the cistern before acquiring a more ventral position at the porus acusticus. As a general rule, the risk of facial nerve injury or need for less than gross total resection (GTR) is greater in cases where the facial nerve courses over the dorsal side of the tumor, over the superior pole, or has significant redundancy.s. Literatur ^,​ s. Literatur ^,​ s. Literatur

A final important consideration is the organization and microscopic anatomy of the facial nerve. True fascicular organization of the facial nerve with perineurium and epineurium begins at the geniculate ganglion.s. Literatur ^,​ s. Literatur The perineurium provides tensile strength to the nerve and resists mechanical stretch, while the epineurium contains the vasa nervorum of the nerve and is important for nutrient delivery. Together, the perineurium and epineurium support bundling of facial nerve axons. Medial to the geniculate ganglion, the facial nerve is more vulnerable to mechanical or ischemic injury during dissection and more susceptible to splay. In the CPA and IAC, the facial nerve is without epineurium and is covered by pia mater and bathed in CSF.s. Literatur The proximal cisternal facial nerve receives its blood supply from the anterior inferior cerebellar artery, while a branch of the middle meningeal artery feeds the perigeniculate nerve.s. Literatur ^,​ s. Literatur ^,​ s. Literatur During tumor dissection, parasitized tumor feeders are systematically coagulated and divided, while every attempt is made to preserve brainstem perforators. Few anastomotic connections exist along the course of the subarachnoid segment of the facial nerve and as a result, varying degrees of devascularization inevitably occur.

The facial nerve can be injured at any point along its course from the brainstem to the fundus of the IAC. Of particular importance is the area of the facial nerve just proximal to the porus acusticus. Given the abrupt change in direction as the nerve transitions from the CPA to the IAC in addition to the compressive force of the cisternal component of the tumor as it enters the IAC, the nerve fibers can be splayed making safe dissection of the tumor capsule away from the nerve difficult. This is the most common site where the facial nerve is injured and the most frequent location to leave residual tumor should STR be performed. Early identification of the facial nerve at the brainstem or fundus, as well as frequent low threshold stimulation during tumor dissection, is imperative for optimal postoperative facial function. Chapters 30 and 32 offer further discussion on facial nerve anatomy and course. The interested reader is encouraged to review Chapters 27, 32, and 41 for further discussion relevant to intraoperative facial nerve monitoring and intraoperative decision making as it pertains to optimizing facial nerve function.

55.3.2 Preoperative Predictors of Facial Nerve Outcome

One of the most intensely studied topics in the VS literature is pre- and intraoperative prognosticators of facial nerve outcome. Many candidate predictors have been extensively evaluated including age, sex, tumor size, degree of ventral tumor extension, degree of fundal obliteration, surgical approach, tumor cystic quality, tumor vascularity, facial nerve course, prior treatment, preoperative electromyography, intraoperative facial nerve stimulation testing, and extent of resection.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur

Of all baseline predictive factors, tumor size remains the most consistent predictor of early and long-term facial nerve function following microsurgical resection.s. Literatur ^,​ s. Literatur This relationship is most obvious when comparing the extremes: the risk of facial paralysis after resection of an intracanalicular tumor or one with minimal cisternal growth is approximately 5 to 10%, while the risk of facial paralysis with GTR of large tumors approaches 65%, according to a recent systematic review.s. Literatur Larger tumors have a lengthier nerve–capsule interface for dissection, in addition to a higher likelihood of acquiring an unfavorable nerve course, nerve adherence, nerve splay, and peritumoral edema.s. Literatur ^,​ s. Literatur ^,​ s. Literatur Related to tumor extent is the degree of fundus obliteration. The literature suggests that hearing preservation is more feasible in cases of a large lateral CSF cap, that is, greater distance between the lateral tumor margin and the fundus.s. Literatur Paralleling this, Rompaey et al found that fundus obliteration is associated with poorer early postoperative facial nerve outcome.s. Literatur Specifically, among patients with fundal obliteration, 30% had HB III or worse, compared to only 13% without fundal obliteration at 1 month following surgery. However, this significant difference was lost at 1 year.

A more recent factor that parallels tumor size is volume of tumor located anterior to the axis of the IAC.s. Literatur ^,​ s. Literatur ^,​ s. Literatur While most tumors are more or less centered on the porus acusticus, some medium- and large-size tumors acquire eccentric growth. Tumors that grow disproportionately more ventral have a greater risk of facial nerve paralysis for the same reasons a large tumor might. Wong et al reported that tumors extending greater than 1.5 cm anterior to the IAC were three times more likely to have postoperative HB grade III function or higher.s. Literatur Similarly, Grahnke et al found that after controlling for tumor size, there was a 3.8 times greater likelihood of higher HB score for every 1 standard deviation increase in the anterior-to-posterior tumor extension ratio.s. Literatur

Tumor consistency and tumor vascularity are two associated factors that have been studied in several recent publications. In 2013, Copeland et al examined MRI signal intensities to predict intraoperative tumor consistency and facial nerve outcome.s. Literatur Firm and soft tumors were equally hypointense on T1 sequences; however, soft VSs were more likely to be hyperintense on T2 sequences (88 vs. 14%, p < 0.005). This finding parallels the meningioma literature.s. Literatur In contrast, Rizk et al found that T2 signal was not predictive of tumor consistency, but rather widening of the IAC was.s. Literatur Specifically, the mean widening of the IAC in relation to the nontumor side was 1.9 mm for soft tumors and 3.6 mm for firm tumors. Both authors found better facial nerve outcomes in softer tumors. Recently, Patel et al found that tumors exhibiting prominent dural enhancement, reminiscent of dural tails in meningioma, at the porus acusticus had greater facial nerve tumor adhesion and were more likely to undergo STR compared to those without (31 vs. 3%, p = 0.01).s. Literatur While still under investigation, magnetic resonance elastography, slip interface imaging, and diffusion tensor imaging tractography may be used in the future to predict intraoperative VS tumor consistency, adhesion, and facial nerve course, respectively.s. Literatur ^,​ s. Literatur ^,​ s. Literatur This aspect is covered in greater detail in Chapter 9.

The association between facial nerve outcome and presence of prominent cystic growth has also been extensively examined. As reviewed in Chapter 76, one of the primary obstacles in characterizing this association is a lack of consensus of what constitutes a “cystic VS,” and the varying classification of intratumoral thick-walled cysts versus peripheral thin-walled cysts. The latter type, in particular, has been linked to poorer facial nerve outcome or increased need for STR.s. Literatur The mechanisms to explain the greater risk of facial nerve injury with macrocystic VS are twofold: first, many cystic tumors undergo periods of rapid expansion, which may result in peritumoral edema, facial nerve splay, or tumor adhesion. Second, dissection of a flimsy, thin-walled, adherent tumor capsule from the nerve can be challenging with less substance to work with when applying countertraction.s. Literatur ^,​ s. Literatur

A final area of discussion relates to preoperative subclinical facial nerve involvement. As discussed previously, HB grade I function does not inherently indicate a completely healthy nerve. Indeed, a substantial portion of facial nerve fibers may be compromised before gross motor weakness is seen.s. Literatur ^,​ s. Literatur ^,​ s. Literatur In line with this, several authors have correlated preoperative electroneurography (ENoG) amplitude testing with postoperative facial nerve outcome. They hypothesized that a tumor with a presurgical compromised nerve is more likely to develop a significant injury during microsurgical resection compared to a nerve that is not compromised at baseline. Kartush et al found that 8 of 10 patients with VS exhibited reduced ENoG amplitude on preoperative testing.s. Literatur The authors subsequently found that preoperative ENoG amplitude data inversely correlated with tumor size.s. Literatur While intriguing, most subsequent studies did not find a consistent association between preoperative reduction in ENoG amplitude and postoperative facial function.s. Literatur ^,​ s. Literatur As a result, most centers do not employ preoperative EMG testing given the added expense and uncertain overall predictive value.

55.3.3 Facial Nerve Outcome by Surgical Approach

Despite the increasing use of radiosurgery and observation within the last two decades for small- and medium-sized VS, microsurgery remains the most prevalent management strategy for VS worldwide.s. Literatur Various surgical approaches have been described; however, the retrosigmoid, translabyrinthine, and middle fossa approaches are used almost exclusively today. Controversy over which surgical approach is best for VS resection dates back over 100 years and is beautifully described in Chapter 1. Suffice it to say, no consensus has been reached.

The primary obstacles that impede objective comparison of surgical approaches include inherent selection biases, inconsistent reporting, confounding variables such as completeness of resection, and other less tangible aspects such as skill and experience of the surgical team. These challenges are well illustrated in the age-long debate comparing the middle fossa and retrosigmoid approaches for hearing preservation. Inherently, the average tumor size among middle cranial fossa cases is much smaller than retrosigmoid cases, since most surgeons will not use the former for tumors greater than 0.5 cm in the CPA. Additionally, the average pure-tone threshold and word recognition score is likely better for most middle fossa cases. Without accounting for tumor size and baseline hearing levels, one might prematurely conclude that facial nerve and hearing preservation outcomes are superior with the middle cranial fossa approach. However, when comparing comparable tumors, several studies have demonstrated superior facial nerve outcomes with the retrosigmoid approach and equivalent hearing preservation rates.s. Literatur ^,​ s. Literatur

In 2012, Gurgel et al published a systematic review of the literature examining facial nerve outcome according to surgical approach for large VS, defined as ≥ 2.5 cm in the CPA.s. Literatur Pooled data from 30 eligible studies and 1,688 tumor resections demonstrated that good facial nerve outcomes (HB I–II) were achieved in 62.5% of translabyrinthine cases and 65.2% of retrosigmoid approaches, a difference that was not statistically significant. The same year, Ansari et al published a systematic review analyzing facial nerve outcomes according to surgical approach, after stratifying according to tumor size. In total, 35 studies encompassing 5,064 patients were included.s. Literatur For intracanalicular tumors, the middle cranial fossa approach was associated with a higher risk of facial nerve injury compared to the retrosigmoid approach; however, both were not statistically different from the translabyrinthine approach (16.7, 4, 0%, respectively, p < 0.001). The middle cranial fossa approach exhibited poorer facial nerve outcomes compared to the translabyrinthine approach for tumors smaller than 1.5 cm, whereas neither differed from the retrosigmoid approach (11.5, 3.3, and 7.2%, respectively, p = 0.001). The retrosigmoid approach exhibited less facial nerve paresis than the middle cranial fossa or translabyrinthine approaches for tumors 1.5 to 3.0 cm (6.1, 17.3, and 15.8, respectively; p < 0.001). In contrast to the findings of Gurgel et al, Ansari and colleagues reported that facial nerve outcome with the retrosigmoid approach was superior to the translabyrinthine approach for tumors greater than 3.0 cm (30.2 vs. 42.5%, respectively, p < 0.001). In evaluating the current literature, and based on personal experience, it is fair to conclude that in many cases, more than one surgical approach is reasonable, and the best surgical approach in one surgeon’s hands may be different for another. It is also true that tumor-related factors, such as size, degree of ventral extension, course of the facial nerve, and tumor consistency, and other factors invariably impact tumor control and cranial nerve outcomes more than approach selection in many cases. Chapter 31 provides a detailed summary of select large clinical series comparing facial nerve and hearing preservation outcomes according to surgical approach.