Because of recent advances in microsurgical technique, the morbidity and mortality of acoustic neuroma surgery have dramatically fallen over the past several decades. 1–7 However, despite these advances, cerebrospinal fluid (CSF) leak continues to occur. A recent large series of acoustic neuroma cases documented an overall CSF leak rate of 7%. CSF leak is not typically a life-threatening complication, but it does predispose the patient to an increased risk of developing meningitis or brain abscess.
Various techniques have been described to prevent CSF leak after the translabyrinthine approach for acoustic neuroma removal. The most commonly reported technique involves packing of the eustachian tube after opening the facial recess. In this technique the incus is removed, the mucosa of the eustachian tube is scarified, and the eustachian tube is plugged with muscle, fat, Proplast, ossicles, or bone wax. The middle ear cavity is filled with muscle or fat tissue. 4,8 Fibrin glue or polytetrafluorethylene has also been use to prevent CSF leak. 8
Previously, the senior authors (r.j.w., r.a.b.) scarified the eustachian tube and middle ear mucosa and packed the eustachian tube with the incus, which was followed by packing the middle ear with muscle. The technique to prevent CSF leak has been modified and the results of the modified technique are described.
The records of all patients with the diagnosis of an acoustic neuroma who were referred to the Chicago Otology Group between October 1997 and December 1999 were reviewed. Ninety patients were evaluated and 44 patients underwent surgical resection of an acoustic neuroma via translabyrinthine approach. The modified technique to pack the middle ear cavity and mastoid defect was used in all 44 patients.
The surgical incision is a standard U-shaped incision approximately two finger-breadths behind the postauricular fold. A complete mastoidectomy and labyrinthectomy with removal of bone of the petrous apex to expose the posterior fossa are accomplished. Bone dust is collected from the beginning of the procedure for obliteration.
After the removal of the tumor, a 1-cm-diameter piece of muscle is excised from the temporalis or sternomastoid muscle at the upper or lower edges of the postauricular incision and divided into small (“grains of rice”) pieces. The incus is removed. The middle ear cavity is filled with small pieces of muscle through the aditus (Fig. 1). A piece of bone wax is covered by bone dust and is placed in the aditus to seal the middle ear cavity (Fig. 2). Strips of abdominal fat are placed to seal the dural defect and to fill the mastoid cavity as described by other authors. 8 Skin and subcutaneous tissues are closed in three layers to create a watertight closure, and a pressure dressing is applied.
Excision of acoustic neuroma through translabyrinthine approach was performed in 21 male and 23 female patients ranging in age from 25 to 81 years, with a mean age of 57.8 years. The tumors were classified into the following groups: 1) intracanalicular, 2) small tumors extending 1 to 15 mm into the cerebellopontine angle, 3) medium-sized tumors (16 to 35 mm), and 4) large tumors (<36 mm) into cerebellopontine angle (Fig. 3). The mean tumor size was 19.09 mm (SD, 9.4 mm).
In these 44 consecutive patients, there has been just one case of CSF leakage rhinorrhea. No CSF leakage occurred from the scalp wound site.
Consensus is to use abdominal fat to seal the dural opening created by the translabyrinthine approach. Montgomery 3 introduced the use of fat as a graft material to pack into the mastoid and petrous apex. There are some variations according to authors about how to place the fat in the mastoid. Pulec 4 believed that best results are obtained by attempting to fill the mastoid and apex with a single, solid piece of fat. The senior authors (r.j.w., r.a.b.) prefer to use the strips of abdominal fat as introduced by House et al., 9 because this helps form a watertight seal of the mastoid. Jackler 10 advises use of a large piece of muscle fascia obtained from groin to cover the dural opening and placement of strips of fat over that fascia.
The usage of bone wax and bone pate in acoustic neuroma surgery is not new. Many surgeons routinely seal all the exposed mastoid air cells with bone wax in the retrosigmoid approach. Falcioni 11 seals any suspicious mastoid air cells with bone wax. The senior author (r.j.w.) credits Dr. Sanna for part of his technique with bone wax. In the method presented in this report, bone wax and bone dust are used to separate the middle ear space from mastoid by blocking the aditus ad antrum.
The major changes in our technique are 1) forcibly packing grains of muscle into the middle ear cavity after removal of the incus, 2) sealing of the aditus area with the bone wax and bone pate, and 3) the avoidance of opening the facial recess. The facial recess is not opened because the recess would create another potential communication to the middle ear cavity. If the facial recess is opened, it should be sealed with the bone wax covered with bone dust. There is no reason to strip entire middle ear or eustachian tube mucosa, because this may increase a local inflammatory response and interfere with wound healing.
The one patient in this series who developed a CSF leak failed conservative treatment (bed rest and head elevation) and required surgical re-exploration on postoperative day 3. The aditus area was found to be inadequately sealed. Therefore the middle ear cavity and the aditus area were repacked and resealed with the bone wax and bone dust combination. Any suspect air cells were sealed with bone wax as well. After surgery a lumbar drain was placed for 2 days with no further evidence of a CSF leak. It is our impression that this complication could have been prevented if critical attention had been given to the wound closure during the obliteration of the middle ear cavity and the sealing of the aditus area before the mastoid defect was packed with fat.
Our experience suggests that this modified closure technique has resulted in a declined incidence of postoperative CSF leakage in our patients after acoustic neuroma removal via the translabyrinthine approach.