Acoustic neuromas (ANs) are the most common tumors of the cerebellopontine angle. Although numerous advances have occurred in the operative management of AN and perioperative care leading to a significant decrease in associated morbidity and mortality, there are several characteristic complications that accompany microsurgical resection of AN. Understanding the types and rates of complications in association with the various approaches is essential in patient counseling, establishing patient expectations, and ensuring the best patient outcome. In this article, the justification for incomplete surgical resection is discussed. Also, the most common complications of AN microsurgery and the associated management are reviewed.
- •
Incomplete resection may be used in larger tumors to decrease the risk of certain complications.
- •
Expedient detection and management of cerebrospinal fluid leak is essential in decreasing the risk of further complications, including bacterial meningitis.
- •
The most common cranial nerve deficits encountered involve the cochlear and facial nerves.
- •
The highest rates of hearing preservation are reported with the middle fossa approach.
- •
Long-term facial nerve preservation is most affected by tumor size.
- •
Headache is most commonly associated with the retrosigmoid approach and can be diminished with meticulous closure technique.
- •
Seizure and intracranial hemorrhage are relatively uncommon complications associated with acoustic neuroma microsurgery.
| ABR | Auditory brainstem response |
| CPA | Cerebellopontine angle |
| TEOAE | Transient evoked otoacoustic emissions |
Acoustic neuromas (ANs) or vestibular schwannomas (VSs) are relatively common benign tumors of Schwann cell origin that originate within or lateral to the Obersteiner-Redlich zone on the vestibular portion of the vestibulocochlear nerve. ANs account for approximately 6% of all intracranial tumors and 80% of all tumors of the cerebellopontine angle (CPA). Microsurgical resection is frequently used in the treatment. The surgical approach is governed mainly by preoperative hearing level, tumor size, and tumor location. The middle fossa approach is used in patients with serviceable hearing (American Academy of Otolaryngology–Head and Neck Surgery hearing classification grade A or B and Gardner-Robertson scale grade 1 or 2) and intracanalicular lesions with less than 1 cm of CPA involvement. An extended middle fossa approach can be used for larger tumors. The retrosigmoid approach is used for patients with:
- •
Serviceable hearing
- •
Greater involvement of the CPA
- •
No lateral involvement within the internal auditory canal (IAC)
- •
Tumors medially positioned within the IAC.
These criteria vary from center to center.
The translabyrinthine approach is used in patients
- •
Without serviceable hearing
- •
With larger tumors for whom hearing preservation is not possible.
There is no size limitation associated with the translabyrinthine approach.
Over the past century the morbidity and mortality of AN resection has declined precipitously secondary to advances in microsurgical technique and perioperative care. However, as with any medical intervention, there exists the potential for complications to occur in association with AN microsurgery. Keen surgical technique is essential in preventing complications. In addition, early diagnosis and treatment are critical in the management of postsurgical complications of AN microsurgery to minimize sequelae.
Incomplete resection and recurrence
Because ANs are benign neoplasms, the risk of facial nerve injury, cochlear nerve injury, brainstem injury, and other complications must be weighed against the benefit of complete tumor resection. This risk/benefit conundrum presents itself most commonly when resecting larger ANs. Near-complete or subtotal tumor resection is used for several reasons. Near-complete AN resection most commonly is undertaken when there is a risk of neural injury in association with complete tumor resection. Conversely, subtotal tumor resection is undertaken in elderly or debilitated patients to minimize operative length and risk of morbidity. In either instance, microsurgical dissection is believed to compromise tumor blood supply, thereby inhibiting further tumor growth.
Within a review of 81 patients with large and giant ANs, subtotal tumor resection was performed in 5% of patients because of the patients’ general health or was performed secondary to cerebellar herniation. Similarly, in a review of 79 patients who underwent near-total and subtotal AN resection, a 10% rate of incomplete microsurgical AN resections was reported at the investigators’ institution during the study period. Within this cohort, a 3% regrowth rate was reported for patients who underwent near-total tumor resection and 32% regrowth rate for patients who underwent subtotal tumor resection with a mean time interval to detection of 3 years. In a review of patients with large ANs, complete tumor resection was performed in 26% of patients, whereas 58% and 16% of patients underwent near-total and subtotal resection, respectively. A 6% regrowth rate was reported within this cohort, which corroborates the results of other reports within the literature.
Even with complete tumor resection, residual cell rests may remain adherent to intact or remnant nerves. Recurrence typically represents regrowth of these residual tumor rests. Mamikoglu and colleagues reported a tumor recurrence rate of 1% after complete resection at 5-year follow-up. In a review of more than 1500 translabyrinthine approach microsurgical resections of AN, a 0.3% rate of recurrence was described. Similarly, a review of 735 cases of complete tumor resection via the middle fossa approach revealed a 0.3% rate of tumor recurrence. Residual and recurrent tumors typically remain asymptomatic until tumor regrowth is sizable ( Table 1 ).
| Extent of Resection | Number of Patients | Percent with Resection | Recurrence and Time | Study |
|---|---|---|---|---|
| Subtotal | 81 | 5% | — | Mamikoglu et al |
| Complete | — | — | 1% at 5 y | Mamikoglu et al |
| Near total and subtotal | 79 | 10% | 3% regrowth for near total at 3 y 32% regrowth for subtotal at 3 y | Bloch et al |
| Near total Subtotal Complete | — | 58% 16% 26% | 6% regrowth | Godefroy et al |
| Translabyrinthine complete | 1500 | All | 0.3% | Shelton |
| Middle fossa complete | 735 | All | 0.3% | Gjuric et al |
For this reason, patients are best followed up with serial magnetic resonance imaging (MRI). Fat suppression protocol MRI facilitates postoperative surveillance in patients who underwent reconstruction of operative defects with adipose tissue grafting. Patients should be regularly monitored for evidence of tumor growth within at least the first 5 years postoperatively.
Subsequent Treatment of Tumor Regrowth
Within most patients, tumor regrowth does not occur. However, in the selected cohort of patients in whom regrowth is encountered, subsequent treatment is almost always necessary to control brainstem compression by tumor volume expansion. In the series by Bloch and colleagues, all patients required subsequent treatment secondary to continued tumor growth. Fifty-nine percent of patients within the series by El-Keshlan and colleagues required subsequent treatment.
Treatment options include repeating microsurgical resection and stereotactic radiosurgery. Repeating microsurgical resection has been associated with elevated rates of facial nerve injury. Stereotactic radiosurgery for recurrence and regrowth has been associated with facial nerve deficit rate of approximately 10%, trigeminal deficits of 4%, and a control rate of 94%. The patient’s age, functionality, medical comorbidities, and rate of tumor growth must be factored into the clinical decision making of the management of recurrence and regrowth.
Cerebrospinal fluid leak
Cerebrospinal fluid (CSF) leak is the most commonly reported complication in association with AN microsurgery. CSF leak rates of 4% to 20% have been reported inassociation with middle fossa approach AN resection. With the retrosigmoid approach, CSF leak rates of 3% to 12% have been reported. CSF leak rates of 0% to 31% have been reported in association with the translabyrinthine approach. However, a meta-analysis of more than 3351 procedures revealed no difference in CSF leak rates across approaches. The only variable found to correlate with an increased rate of CSF leak was age, with a 5% increase rate of leaks in patients older than 50 years. Similarly, a meta-analysis of 5964 patients who underwent AN resection revealed no difference in CSF leak rate across approaches.
Controversial Relationship between Tumor Size and Rate of CSF Leak
The relationship between tumor size and the rate of CSF leak remains controversial. Large series by Slattery and colleagues, Bryce and colleagues, and Brennan and colleagues revealed a correlation between larger tumor size and CSF leak. However, other reports challenge this notion.
CSF Leak Routes
CSF leaks may occur via direct extension through the cutaneous incision or via air cell tracts to the middle ear and Eustachian tube. Multiple reviews have revealed a slight increased preponderance for CSF leak via air cell tracts as compared with direct incisional extension, with an average approximate 60% of CSF leaks having occurred via the mastoid air cells irrespective of surgical approach.
CSF Closure Techniques
Various closure techniques have been used in an attempt to decrease the rate of CSF leak encountered after AN surgery, including:
- •
Autologous fascial grafts
- •
Artificial dural grafts
- •
Adipose tissue grafts
- •
Muscle grafts
- •
Fibrin glue.
In addition, various materials have been used to obliterate the air cell tracts, including:
- •
Bone wax
- •
Bone pate
- •
Hydroxyapatite granules
- •
Alloplastic glue
- •
Biomaterials.
Wu and colleagues reported a decrease in CSF leak rate from 28.2% to 7.4% with a musculoperiosteal flap closure of translabyrinthine approach defects. They also noted a decrease in reoperation rate from 7.7% to 3.7% with this reconstructive technique. However, the use of musculoperiosteal flaps limit the opportunity for fat saturation imaging in the postoperative follow-up of patients with AN, complicating the postoperative follow-up. Fat grafting assists in the imaging follow-up of patients with AN. In addition, fat grafting has been found to decrease the rate of CSF leak from 5.7% to 2.2% when compared with muscle flap reconstruction. With limited facial recess dissection, minimal dissection of the incus buttress, and packing of temporalis fascia around the incus to obstruct the mastoid antrum in addition to fat obliteration of the mastoid cavity, no CSF leaks were encountered in 61 patients after translabyrinthine approach AN resection.
Graded Management of CSF Leak
When a CSF leak occurs, it may successfully be managed in a graded manner. The initial conservative management of CSF leaks may include head of bed elevation, bed rest, acetazolamide, and compressive dressing reapplication. If leak persists despite these measures, lumbar subarachnoid spinal fluid diversion drain may be placed to decrease intracranial pressure and allow for fluid siphoning from the CSF space. In most instances, lumbar drains are maintained for 3 to 5 days. If the lumbar drainage of CSF is unsuccessful in relieving leakage, reoperation is used. These procedures may include repeated obliteration of temporal bone cavity, obliteration of the Eustachian tube, repeated dural closure with or without muscle reinforcement, and oversewing the cutaneous incision.
In a review of 412 patients who underwent retrosigmoid approach posterior cranial fossa surgery, of whom 137 patients had AN, conservative management was successful in the treatment of those with CSF leak in 31% of patients, whereas 38% were successfully managed with lumbar subarachnoid spinal fluid diversion drain, and 31% required reoperation. Surgical repair included Eustachian tube obliteration for CSF rhinorrhea and dural repair with muscle graft for incisional leaks. Within a review of complications in 1687 patients who underwent AN surgery,
- •
53% of CSF leaks resolved with conservative management
- •
16% resolved with lumbar drain
- •
23% resolved after reoperation
- •
The additional patients underwent unspecified combined management.
The reported rates of success in the management of CSF leak with lumbar drain placement range from 31% to 83% in the recent literature, and reoperation rates have been reported to range from 21% to 61%. The highest rates of reoperation have been reported with the translabyrinthine approach. The successful management of CSF leaks is critical to prevent infection-related morbidity and mortality as seen with meningitis and brain abscess.
Cerebrospinal fluid leak
Cerebrospinal fluid (CSF) leak is the most commonly reported complication in association with AN microsurgery. CSF leak rates of 4% to 20% have been reported inassociation with middle fossa approach AN resection. With the retrosigmoid approach, CSF leak rates of 3% to 12% have been reported. CSF leak rates of 0% to 31% have been reported in association with the translabyrinthine approach. However, a meta-analysis of more than 3351 procedures revealed no difference in CSF leak rates across approaches. The only variable found to correlate with an increased rate of CSF leak was age, with a 5% increase rate of leaks in patients older than 50 years. Similarly, a meta-analysis of 5964 patients who underwent AN resection revealed no difference in CSF leak rate across approaches.
Controversial Relationship between Tumor Size and Rate of CSF Leak
The relationship between tumor size and the rate of CSF leak remains controversial. Large series by Slattery and colleagues, Bryce and colleagues, and Brennan and colleagues revealed a correlation between larger tumor size and CSF leak. However, other reports challenge this notion.
CSF Leak Routes
CSF leaks may occur via direct extension through the cutaneous incision or via air cell tracts to the middle ear and Eustachian tube. Multiple reviews have revealed a slight increased preponderance for CSF leak via air cell tracts as compared with direct incisional extension, with an average approximate 60% of CSF leaks having occurred via the mastoid air cells irrespective of surgical approach.
CSF Closure Techniques
Various closure techniques have been used in an attempt to decrease the rate of CSF leak encountered after AN surgery, including:
- •
Autologous fascial grafts
- •
Artificial dural grafts
- •
Adipose tissue grafts
- •
Muscle grafts
- •
Fibrin glue.
In addition, various materials have been used to obliterate the air cell tracts, including:
- •
Bone wax
- •
Bone pate
- •
Hydroxyapatite granules
- •
Alloplastic glue
- •
Biomaterials.
Wu and colleagues reported a decrease in CSF leak rate from 28.2% to 7.4% with a musculoperiosteal flap closure of translabyrinthine approach defects. They also noted a decrease in reoperation rate from 7.7% to 3.7% with this reconstructive technique. However, the use of musculoperiosteal flaps limit the opportunity for fat saturation imaging in the postoperative follow-up of patients with AN, complicating the postoperative follow-up. Fat grafting assists in the imaging follow-up of patients with AN. In addition, fat grafting has been found to decrease the rate of CSF leak from 5.7% to 2.2% when compared with muscle flap reconstruction. With limited facial recess dissection, minimal dissection of the incus buttress, and packing of temporalis fascia around the incus to obstruct the mastoid antrum in addition to fat obliteration of the mastoid cavity, no CSF leaks were encountered in 61 patients after translabyrinthine approach AN resection.
Graded Management of CSF Leak
When a CSF leak occurs, it may successfully be managed in a graded manner. The initial conservative management of CSF leaks may include head of bed elevation, bed rest, acetazolamide, and compressive dressing reapplication. If leak persists despite these measures, lumbar subarachnoid spinal fluid diversion drain may be placed to decrease intracranial pressure and allow for fluid siphoning from the CSF space. In most instances, lumbar drains are maintained for 3 to 5 days. If the lumbar drainage of CSF is unsuccessful in relieving leakage, reoperation is used. These procedures may include repeated obliteration of temporal bone cavity, obliteration of the Eustachian tube, repeated dural closure with or without muscle reinforcement, and oversewing the cutaneous incision.
In a review of 412 patients who underwent retrosigmoid approach posterior cranial fossa surgery, of whom 137 patients had AN, conservative management was successful in the treatment of those with CSF leak in 31% of patients, whereas 38% were successfully managed with lumbar subarachnoid spinal fluid diversion drain, and 31% required reoperation. Surgical repair included Eustachian tube obliteration for CSF rhinorrhea and dural repair with muscle graft for incisional leaks. Within a review of complications in 1687 patients who underwent AN surgery,
- •
53% of CSF leaks resolved with conservative management
- •
16% resolved with lumbar drain
- •
23% resolved after reoperation
- •
The additional patients underwent unspecified combined management.
The reported rates of success in the management of CSF leak with lumbar drain placement range from 31% to 83% in the recent literature, and reoperation rates have been reported to range from 21% to 61%. The highest rates of reoperation have been reported with the translabyrinthine approach. The successful management of CSF leaks is critical to prevent infection-related morbidity and mortality as seen with meningitis and brain abscess.
Meningitis
Meningitis is reportedly the second most common complication of AN microsurgery. The reported rates of meningitis after AN microsurgical resection range from 0.14% to 9.90%. A meta-analysis of 13 studies including 2316 AN procedures revealed a meningitis rate of 3.7%.
The presence of CSF leak increases the risk of meningitis in patients with AN after microsurgical resection. A meta-analysis of AN microsurgery revealed a statistically significant ( P <.0001) increase in the incidence of meningitis in patients with CSF leak (14%) when compared with those without CSF leaks (2.5%). Similarly, a review of 3000 patients who underwent AN resection revealed a meningitis rate of 21% in patients with CSF leak compared with no encountered episodes of meningitis in patients without CSF leaks. In an analysis of patients with meningitis after AN resection, 82% of patients had a history of postoperative CSF leak. For this reason, judicious management of CSF leaks as described earlier is of utmost importance.
Two forms of meningitis are encountered in postoperative patients after AN microsurgery: bacterial meningitis and chemical (aseptic) meningitis.
Bacterial meningitis typically presents with:
- •
Meningismus
- •
Severe headache
- •
Mental status changes
- •
Malaise
- •
High fever.
Analysis of CSF may reveal
- •
Leukocytosis
- •
Decreased glucose level
- •
Elevated protein level
- •
Positive gram stain
- •
Positive culture result.
A review of the literature regarding postoperative meningitis in patients who underwent craniotomy revealed Staphylococcus species, Enterobacter , and Propionibacterium acnes to be the most commonly encountered pathogens. Bacterial meningitis, however, represented the minority of postoperative cases of meningitis.
Most patients with postoperative meningitis after AN microsurgery present with chemical meningitis. Chemical meningitis, also known as aseptic meningitis, is believed to be the result of an inflammatory response to foreign substances within the subarachnoid space, namely, blood, hemoglobin, bilirubin, cotton wool lint, and bone dust. Animal studies support this assertion.
In a review of 1146 patients who underwent CPA surgery, patients with bacterial meningitis confirmed by positive culture results accounted for only 22% of patients. Presumed bacterial meningitis not confirmed by gram stain or culture results accounted for 22% of patients, whereas chemical meningitis accounted for 56% of patients. In a review of patients with postoperative meningitis after posterior fossa craniotomy, 72% of patients presented with aseptic meningitis. Chemical meningitis also typically presents with meningismus, headache, fever, and malaise.
A retrospective review of postoperative meningitis in patients after CPA surgery identified key clinical factors that are helpful in the differentiation of bacterial meningitis from chemical meningitis ( Table 2 ).
