The transotic approach for vestibular schwannoma: indications and results

Vestibular schwannoma is the most common tumor of the cerebellopontine angle. Its treatment options include observation (wait-and-scan policy), radiation treatment and surgery [1]. The most commonly used surgical approaches are the translabyrinthine, retrosigmoid and middle cranial fossa approaches. The appropriate approach should be selected for the individual patient according to tumor size and location with the intent of radical tumor removal, preservation of the facial nerve and, if possible, preservation of the patient’s cochlear function [2]. In the 1960s, William F. House developed the translabyrinthine approach [3], which became very popular because it reduced the mortality of vestibular schwannoma extirpation from 20% (neurosurgical approaches) to 2%. The total loss of hearing was accepted in view of the radical removal of the tumor combined with better preservation of facial function with dramatically reduced postoperative morbidity and mortality [4].

In the 1970s, Ugo Fisch developed the transotic approach [5‐7] to avoid three main limitations inherent in the translabyrinthine approach [8]: (1) the reduction of surgical exposure given by the preservation of the middle ear spaces in cases of reduced pneumatization, anterior location of the sigmoid sinus, high jugular bulb and low middle cranial fossa dura; (2) the difficulty in exposing the anterior cerebellopontine angle and, therefore, in separating and preserving under direct vision the intracranial segment of the facial nerve from the anterior pole of the tumor; and (3) the danger of postoperative CSF leakage and meningitis due to the direct contact of the intact middle ear mucosa with the subarachnoid space in the operated mastoid cavity [8]. We introduced the transotic approach in our clinic in 2007, and would like to present a retrospective review of our experience with it.

Institutional Review Board (IRB) approval was obtained from our institution. Each patient, or his/her guardian(s), provided written informed consent for the surgical procedures. And all the cases operated by one surgeon.

Fisch et al. [5, 12] described the transotic approach, which Wang [13] called the “entire labyrinthine approach,” to provide better tumor exposure and facial nerve preservation than was possible with the translabyrinthine approach of W. House. Although the transotic approach may be used for all sizes of tumors, Fisch [5, 6] recommended its use for tumors up to 2.5 cm. This limitation was proposed to optimize the results by otologists performing skull base surgery, since neurosurgeons in Zürich preferred the retrosigmoid approach for the removal of larger tumors. Wang [13] proposed to use the transotic approach for tumors up to 3.5 cm. And Falcioni stated that the transotic approach was especially suitable for resection of vestibular schwannomas that involved the otic capsule [14], and Gantz et al. [15] for those that involved the vestibule and otic capsule.

The goal of the development of the transotic approach [5, 6, 14] was to provide the largest anterior transtemporal exposure of the cerebellopontine angle from the sigmoid sinus to the vertical ICA and from the superior petrosal sinus to the jugular bulb. Initially, the facial nerve was permanently transposed anteriorly [14], but very soon it was observed that the same surgical results could be obtained by leaving the fallopian canal intact and skeletonized [15].

The translabyrinthine approach provides no direct exposure of the anterior cerebello-pontine angle. With the transotic approach we were routinely able to obtain about 7 mm of exposure anteriorly to the skeletonized fallopian canal, and 4 mm posteriorly (Fig. 4). The separation of the intracranial facial nerve from the anterior pole of the tumor is done under direct vision working anteriorly to the skeletonized tympanic and mastoid segments of the fallopian canal without the need to displace the cerebellum (Figs. 3, 4).

The disadvantages of the transotic over the translabyrinthine approach are the resulting total loss of hearing and the longer operative time; on the other hand, the advantages of the transotic approach include better exposure and preservation of the facial nerve and significantly reduced incidence of immediate and delayed cerebrospinal fluid leaks with possible meningitis (Table 2) [6, 16]. Spinal fluid leaks are reduced by closure of the tympanic ostium of the Eustachian tube, obliteration of the operative cavity (subtotal petrosectomy) with abdominal fat and temporalis muscle flap, and by blind sac closure of the external auditory canal (Table 2; Fig. 2).

Our experience confirms these results (Table 2). With the transotic approach we were able to preserve the anatomic integrity of the facial nerve in all patients (Table 1). 94% of the patients had a postoperative HB grade of I or II and a Fisch DEFS of 100 or 77–83% at 6 weeks following surgery (Table 2). We believe that preservation of facial function is due to the excellent exposure given by the transotic approach for the separation of the intrameatal and intracranial segments of the nerve from the tumor (Figs. 3, 4).

Brackmann et al. [17] and Browne and Fisch [6] reported significantly fewer complications with the transotic than with the translabyrinthine approach. House [3] reported for the translabyrinthine approach the following results for postoperative facial function: 58.2% (HB I), 12.6% (HB II), 13.2% (HB III), 7.8% (HB IV), 3.3% (HB V) and 5.1% (HB VI) [3]. Fisch et al. [5] showed that the anatomic integrity of the facial nerve was preserved in 94% of the patients undergoing a transotic approach and that the functional recovery of facial function (Fisch DEFS) for tumors up to 2.5 cm was 70% (Fisch grade 100), 15% (Fisch grade 99–80), 9% (Fisch grade 79–60), 3% (Fisch grade 59–40) and 3% (Fisch grade 39–0). Our results (Table 2) for tumors up to 2.0 cm are in agreement with these results. House reported that at the time when the dura was closed with temporalis muscle (before 1974) the incidence of postoperative cerebrospinal fluid leaks for the translabyrinthine approach was as high as 20% [3], but that this rate dropped to 7% when abdominal fat was used instead of the temporalis muscle. House reported that most of the cerebrospinal fluid leaks resolved with a pressure dressing and by raising the head of the bed [7]. This indicates that the CSF leaks occurred through the surgical wound, a situation that is avoided if the abdominal fat is placed in a surgical cavity performed as a subtotal petrosectomy with occlusion of the ET and blind sack closure of the EAC. Fisch compared the incidence of postoperative CSF leaks for 147 transotic versus 114 translabyrinthine approaches [6]. Significantly fewer immediate leaks were found in the transotic (4%) than in the translabyrinthine approach (22%) in that study. The rate of delayed leaks, which can appear years after surgery and induce meningitis, was also significantly higher in cases operated with the translabyrinthine approach [6]. Our results confirm these findings (Table 2). The size of the tumor has implications for possible complications. Intracranial hemorrhage occurred in a tumor with a size of 3.8 cm and the CSF Leak in a tumor of 4.2 cm (Table 2).

Our data (Table 1) confirm that more time is needed to perform the transotic than for the translabyrinthine approach, but also that the total time of surgery is extended only a little. The longer time (2–3 h) spent developing the exposure is compensated by the more rapid removal of the tumor (1–1.5 h) (Table 1).

The transotic approach has proven of value for the removal of vestibular schwannomas up to 5.0 cm in the presence of temporal bone contraction (reduced pneumatization, anteriorly located sigmoid sinus, high jugular bulb, low middle cranial fossa) as determined by preoperative imaging (CT and MRI). Hearing was not preserved, but other outcomes were favorable, including the rates of total tumor removal, preservation of facial nerve function, avoidance of CSF leaks, and severe postoperative complications.