Abstract
Objective
To report rates of cerebrospinal fluid leak, wound infection, and other complications after repair of retrosigmoid craniotomy with hydroxyapatite bone cement.
Methods
Retrospective case review at tertiary referral center of patients who underwent retrosigmoid craniotomy from 2013 to 2016 with hydroxyapatite cement cranioplasty.
Outcome measures
Presence of absence of cerebrospinal fluid leak, wound infection, and other complications.
Results
Twenty cases of retrosigmoid craniotomy repaired with hydroxyapatite cement were identified. Median length of follow up was 9.8 months. No cases of cerebrospinal fluid leak were identified. One patient developed a wound infection which was thought to be related to a chronic inflammatory response to the implanted dural substitute. No other major complications were noted.
Conclusions
A method and case series of suboccipital retrosigmoid cranioplasty using hydroxyapatite cement and a are reported. Hydroxyapatite cement cranioplasty is a safe and effective technique for repair of retrosigmoid craniotomy defects.
1
Introduction
The suboccipital retrosigmoid craniotomy is a valuable approach to tumors or vascular anomalies in the posterior fossa or cerebellopontine angle . Cerebrospinal fluid (CSF) leak has been reported with an incidence of approximately 10% after retrosigmoid craniotomy and can lead to intracranial hypotension, meningitis, neurologic deficits and death . Wound infection has been reported in retrosigmoid craniotomy with rates of 1–3% and can also cause meningitis as well as wound dehiscence and the need for further surgery. The rates of these complications are metrics of success in lateral skull base surgery and are routinely reported in retrosigmoid craniotomy case series .
Retrosigmoid craniotomy defects have historically been repaired by numerous methods including fat, muscle, or split calvarial grafts and titanium mesh or polymer implants, as well as combinations therein . Hydroxyapatite is a rapidly setting cement formed by the mixture of calcium phosphate mineral powder with a pH-neutral liquid. Hydroxyapatite cement (HAC) has several desirable characteristics as a cranial reconstruction material and has been used successfully in cranioplasty for frontal, parietal and temporal translabyrinthine and retromastoid (intact mastoid) defects . Reports of HAC in retrosigmoid craniotomy are scarce and inconsistent, with some groups reporting success and others reporting unacceptably high complication rates .
We presently report our method of HAC closure for retrosigmoid craniotomy. We further report a series of 20 consecutive cases of retrosigmoid craniotomy closed with HAC with an analysis of complications including CSF leak and wound infection. This report adds to a small but growing body of literature suggesting that HAC is a safe and effective method for repair of retrosigmoid craniotomy defects.
2
Materials and methods
2.1
Case selection
Patients were identified by searching the records of the Yale-New Haven Health System for Current Procedural Terminology code 61595 (transtemporal approach to posterior cranial fossa, jugular foramen or midline skull base) from March 2013 to July 2016. Cases were individually examined and excluded if retrosigmoid approach was not performed (incorrect coding) or if HAC was not used in repair. Cases in which bone grafts were used in conjunction with cement were also excluded.
2.2
Intraoperative details
After primary dural closure and/or dural closure with a processed collagen implant (DuraGen, Integra Lifesciences or Durepair, Medtronic) with no evidence of CSF leak, calcium phosphate bone cement (HydroSet, Stryker) was applied. The cement was mixed according to the manufacturer’s instructions to a homogenous consistency. Fifteen to twenty cubic centimeters of this cement were injected into the craniectomy site over 1–2 min, covering all bony areas and filling the entire wound bed. Any exposed posterior mastoid air cells were also filled. A neurosurgical patty was used to tamp the cement to eliminate air bubbles and to smooth the calvarial contour after which the cement was allowed to harden for 2 to 3 min ( Fig. 1 ). After thorough irrigation, the area was again inspected for CSF leak. When none was noted, the periosteum was closed primarily followed by muscle, deep dermal and skin layers. All cement cranioplasties were performed by one of two surgeons; one surgeon was responsible for 90% of cranioplasties in this study.
2.3
Postoperative evaluation
Patients were evaluated in an office setting. Presence or absence of complications was evaluated based on documentation from these encounters or documentation from intervening hospital visits and/or imaging studies. Clinical evidence of post-operative clear otorrhea, incisional fluid, rhinorrhea and radiographic evidence of fluid at or near the site of skull base defect were reported as CSF leak unless proven otherwise. Infection was reported in cases of purulent incisional drainage, positive cultures, or persistent incisional erythema or swelling. Other complications were reported as encountered. Duration of follow up was noted as the time between the day of surgery and the last noted clinical encounter with either the otolaryngological or neurosurgical attending surgeon of record.
2
Materials and methods
2.1
Case selection
Patients were identified by searching the records of the Yale-New Haven Health System for Current Procedural Terminology code 61595 (transtemporal approach to posterior cranial fossa, jugular foramen or midline skull base) from March 2013 to July 2016. Cases were individually examined and excluded if retrosigmoid approach was not performed (incorrect coding) or if HAC was not used in repair. Cases in which bone grafts were used in conjunction with cement were also excluded.
2.2
Intraoperative details
After primary dural closure and/or dural closure with a processed collagen implant (DuraGen, Integra Lifesciences or Durepair, Medtronic) with no evidence of CSF leak, calcium phosphate bone cement (HydroSet, Stryker) was applied. The cement was mixed according to the manufacturer’s instructions to a homogenous consistency. Fifteen to twenty cubic centimeters of this cement were injected into the craniectomy site over 1–2 min, covering all bony areas and filling the entire wound bed. Any exposed posterior mastoid air cells were also filled. A neurosurgical patty was used to tamp the cement to eliminate air bubbles and to smooth the calvarial contour after which the cement was allowed to harden for 2 to 3 min ( Fig. 1 ). After thorough irrigation, the area was again inspected for CSF leak. When none was noted, the periosteum was closed primarily followed by muscle, deep dermal and skin layers. All cement cranioplasties were performed by one of two surgeons; one surgeon was responsible for 90% of cranioplasties in this study.
2.3
Postoperative evaluation
Patients were evaluated in an office setting. Presence or absence of complications was evaluated based on documentation from these encounters or documentation from intervening hospital visits and/or imaging studies. Clinical evidence of post-operative clear otorrhea, incisional fluid, rhinorrhea and radiographic evidence of fluid at or near the site of skull base defect were reported as CSF leak unless proven otherwise. Infection was reported in cases of purulent incisional drainage, positive cultures, or persistent incisional erythema or swelling. Other complications were reported as encountered. Duration of follow up was noted as the time between the day of surgery and the last noted clinical encounter with either the otolaryngological or neurosurgical attending surgeon of record.
3
Results
Retrosigmoid craniotomies were performed on twenty temporal bones in nineteen patients from 2013 to 2016 ( Table 1 ). Average age at the time of surgery was 48.9 years. Fourteen surgeries (70%) were performed on women and six (30%) on men. Length of follow-up ranged from 1.2 to 34.8 months, with a median length of 9.8 months. The most common indication for surgery was microvascular decompression for hemifacial spasm; other indications included microvascular decompression for trigeminal neuralgia, vestibular schwannoma, meningioma, epidermoid cyst, and internal auditory canal osteoma. A summary of the twenty cases and their demographic and clinical characteristics is shown in Table 1 .
