47.1 Introduction

Postoperative cerebrospinal fluid (CSF) leak is the second most common complication of vestibular schwannoma (VS) surgery, besides facial paralysis. Because postoperative CSF leak leads to prolonged hospitalization, additional procedures, and increased risk for meningitis, reducing the incidence of CSF is of utmost importance. In the 1960s, the introduction of the abdominal fat graft significantly reduced postoperative CSF leak rates in the translabyrinthine approach. Moreover, the introduction of muscle or fat grafts into the internal auditory canal (IAC) and bone wax obliteration of craniotomy and meatal air cells has reduced the incidence of CSF leaks in the retrosigmoid and middle fossa approaches as well. With greater attention to CSF leak prevention, the prevalence of this complication continues to decline but, nevertheless, the reported rates in the literature continue to range from 2 to 30%s. Literatur ^,​ s. Literatur ^,​ s. Literatur with a realistic estimate of approximately 10%.s. Literatur The objectives of this chapter are to (1) highlight risk factors and approach-based differences that affect the incidence of CSF leak; (2) review various techniques and principles of optimal surgical closure; (3) outline successful treatment strategies for CSF leak, pseudomeningocele, and/or meningitis.

47.2 Risk Factors

The risk of CSF leak has been investigated against tumor, patient, and surgical parameters. Patient characteristics including age, sex, and body mass index (BMI), as well as surgical parameters such as approach, specific closure techniques, length of surgery, use of lumbar drains, and even who performed closure (resident vs. attending) have been investigated. Although, comorbidities such as anemia, diabetes mellitus, renal failure, or malnourishment might be expected to have an important bearing on healing and CSF leak rates, none of these has been specifically studied to the authors’ knowledge.

Of the aforementioned variables, none have been conclusively shown to affect risk of the postoperative CSF leak. However, for some factors, there is potential conflicting evidence. For example, several studies have shown an association between CSF leak risk and tumor size, with higher rates being reported with larger tumors,s. Literatur others showing no association with size at alls. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur and still others suggesting a higher rate of leak with smaller tumors.s. Literatur ^,​ s. Literatur ^,​ s. Literatur Reoperation has also been suggested as a risk factor for CSF leak, where Stieglitz et als. Literatur reported leak rates escalating to 11.1% in revision cases as compared to 4.5% at primary surgery. Copeland et al demonstrated, in a series of greater than 450 patients, that increasing BMI is associated with an increase in postoperative CSF leak.s. Literatur Other factors potentially associated with a postoperative CSF leak are degree of aeration of the petrous bones. Literatur ^,​ s. Literatur with a greater risk in well-pneumatized temporal bones, closure with fat versus muscle,s. Literatur and usage of perioperative lumbar drains.s. Literatur ^,​ s. Literatur

Regarding the influence of surgical approach, a systematic review of 35 studies by Ansari et als. Literatur found the incidence of CSF leak was significantly greater after the retrosigmoid approach than after either the middle cranial fossa or translabyrinthine approaches (10.3, 5.3, and 7.1%, respectively; p = 0.001). However, a large meta-analysis of 25 studies by Selesnick et als. Literatur showed no difference in leak rates among the different surgical approaches and another report by Mangus et al reached the same conclusion after evaluating more than 1,900 cases performed at a single center.s. Literatur In contrast, Sughrue et als. Literatur found that rates of CSF leak were similar between middle fossa and retrosigmoid approaches and were higher with the translabyrinthine approach. The remaining literature is diverse in terms of the incidence of CSF leak rates comparing translabyrinthine and retrosigmoid approaches, with some reporting higher rates for translabyrinthine,s. Literatur ^,​ s. Literatur whereas others reporting no difference based on approach.s. Literatur ^,​ s. Literatur ^,​ s. Literatur It is notable that the type of CSF leak may be different according to approach. For example, Mangus et al reported a higher rate of incision leak with translabyrinthine compared to retrosigmoid craniotomy and higher rates of CSF rhinorrhea with retrosigmoid and middle fossa approachess. Literatur (Table 47‑1 ).

47.3 General Principles of Closure

All surgeons agree on the need for a watertight multilayered closure. The five layers from deep to superficial include the dura, bone, muscle/fascia, subcutaneous tissue, and skin and attempts to reconstitute all these layers will reduce the risk of postoperative CSF leak. Particular to the retrosigmoid approach, watertight dural closure is key in preventing postoperative CSF leak and pseudomeningocele formation. In contrast, primary dural closure with the translabyrinthine approach is less feasible, and dural substitutes in addition to mastoid obliteration with fat may be useful.

Several allografts have been successfully developed and implemented over the years. For the retrosigmoid approach, if primary watertight dural closure cannot be achieved, the authors prefer a pericranial autograft. Alternatively, suturable collagen patches are used to achieve tight closure as tested by a Valsalva maneuver performed by anesthesia. Subsequently, we reinforce the dural closure with dural sealant and perform another Valsalva test. Rigid support of the dural closure using the craniotomy bone flap may act to decrease the differential pressure across the dura and resist breakdown against transmitted dural pulsations. Additionally, replacement of the bone flap may potentially reduce postoperative headache by preventing adhesions between the nuchal muscles and exposed dura.

For the retrosigmoid or translabyrinthine approaches, cranioplasty can be performed using titanium or resorbable plates/mesh and screws. For the retrosigmoid approach, the bone flap may be incorporated in this closure. Alternatively, hydroxyapatite (HA) bone cement can be used, which avoids hardware under the thin skin and provides a sealant effect of the mastoid air cells. Some surgeons use titanium mesh in combination with HA cement. For all approaches, it is paramount to thoroughly plug the perimeatal and mastoid air cells with bone wax in order to avoid CSF otorhinorrhea. It is a good habit to wax the craniotomy air cells “on the way in and the way out.”

When possible, reapproximation of the musculoperiosteal layer should be performed. While this layer rarely achieves a truly watertight closure, it serves to buttress the replaced craniotomy plate or obliterating fat, depending on the approach. Reapproximation of the musculoperiosteal layer may also reduce the risk of plate or HA cement extrusion or exposure. Finally, watertight closure of the subcutaneous tissue and skin serve as the last line of defense against percutaneous CSF leakage. There is no evidence to suggest that staples or sutures are superior. Regardless of the final closure technique, generally sutures or staples are not removed until at least 2 weeks after surgery to reduce the risk of wound dehiscence and CSF leak, particularly if pseudomeningocele is present.