Autologous Mastoid Cortex Bone Graft for Multilayer Repair of Lateral Skull Base Defects Authors: Christopher Z Wen, Daniel C Fong, Vera Bzhilyanskaya, Adam C Kaufman, Ronna Hertzano, David J Eisenman Published: 2026 ABSTRACT AbstractMultilayer repair of lateral skull base cerebrospinal fluid leaks and encephaloceles can be reinforced with rigid autologous grafts. Harvesting a monocortical mastoid cortex bone graft is effective and easy to implement without conferring additional morbidity.Keywords: bone graft, CSF leak repair, CSF otorrhea Introduction Cerebrospinal fluid leaks are due to defects between the intracranial space, bony skull base, and overlying periosteum and soft tissue. They can be iatrogenic, post‐traumatic, or spontaneous in nature. Different surgical approaches (transmastoid, middle cranial fossa, combined) with a variety of reconstructive options (free flaps, pedicled flaps, free grafts, non‐autologous materials) have been described [1, 2]. Due to the need to individualize treatment for defect size and location, there is no widely agreed upon consensus for ideal technique and materials. For small to medium lateral skull base defects, multilayer closure with a variety of free grafts is a common option. While the ideal reconstructive option is to replace like with like, previously described bony grafts such as split calvarial bone grafts and nasal septal bone grafts carry additional surgical site morbidity [3]. For harvest of bone for use in CSF leak repair, we describe our experience using monocortical mastoid cortex bone as a free graft material within multilayer closure of lateral skull base defects. Methods A retrospective chart review of patients who underwent repair of lateral skull base CSF leaks or encephaloceles using autologous bone grafts was performed. Leaks that were reconstructed with other techniques and approaches, such as mastoid cavity obliteration, were excluded. CSF diversion with a lumbar drain was not routinely performed. Acetazolamide was prescribed for patients with other symptoms of idiopathic intracranial hypertension or who have had a history of prior CSF leak at the same or a different site. Patients were evaluated preoperatively with computed tomography for bony anatomy and magnetic resonance imaging if encephalocele was suspected. This study received approval from the Institutional Review Board at the University of Maryland (HP‐00115739). The representative procedure (Video 1) begins with a standard mastoidectomy approach via postauricular incision. Periosteal flaps are raised to expose the cortex of the mastoid temporal bone. Prior to commencing a standard cortical mastoidectomy, the mastoid cortex bone graft is harvested (Figure 1A–F). Design of the bone graft is based on preoperative imaging measurements to avoid violating critical structures such as the tegmen and sigmoid sinus. In this case, the posterior edge of the graft is measured 2 cm posterior to the external auditory canal, which is carried anteriorly to immediately posterior to the external auditory canal and inferiorly to the mastoid tip. Using a 2 mm cutting bur, the anterosuperior edge of the graft is defined until air cells are encountered over the mastoid antrum. This depth is used as a reference for the remainder of the graft harvest. A 5 mm cutting bur is used to thin the posterior and superior bony ledges opposite the bone graft to allow for better visualization and access deep to the bony graft for instruments. A 2 mm cutting bur is then used to undercut the bone graft posteriorly and superiorly to the extent visible or safe. This creates a ledge that allows an angled oscillating saw to further undermine the bone graft. Once the majority of the bone graft has been freed from the underlying mastoid air cells, the final harvest of the bone graft is done with a curved osteotome but can also be fractured off with a Freer elevator or similar sturdy instrument. Demonstration of lateral skull base defect repair with harvest of autologous mastoid cortex bone graft. This video demonstrates the technique of repair of a lateral skull base defect using an autologous harvest of monocortical mastoid cortex bone graft. This technique is suitable for bony dehiscences as well as encephaloceles. The patient in this representative case is a 70‐year‐old male with history of right tegmen mastoideum dehiscence who underwent transtemporal repair 4 years prior, which has healed well without complication. He has concern for focal dehiscence of the tegmen mastoideum on the left with new left‐sided hearing loss and type B tympanogram. Due to this prior history, he was preoperatively started on Diamox with some improvement in his hearing, suggesting resolution of active CSF leak. During evaluation of his imaging, we measure the distance from the external auditory canal to the tegmen and sigmoid sinus, as well as the depth of the sigmoid sinus from the surface of the mastoid cortex to inform the design of the bone graft. A postauricular incision is marked with an S‐shaped superior extension in the event a middle fossa approach is needed. A standard post‐auricular approach is performed, dissecting down to the level of the temporalis fascia, where the flap is then lifted anteriorly. A T‐shaped periosteal incision is marked and periosteal flaps are elevated until the bony ear canal is reached. The planned monocortical bone graft is marked based on preoperative imaging measurements. In the current case, the posterior edge of the graft is measured 2 cm posterior to the external auditory canal. This is carried anteriorly to immediately posterior to the external auditory canal and inferiorly to the mastoid tip. A 2 mm cutting bur is used to outline the graft until mastoid air cells are encountered over the mastoid antrum. The remainder of the graft is then outlined using this depth as a reference. A 5 mm cutting bur is then used to thin the posterior and superior bony ledges opposite the bone graft to allow for better visualization and access deep to the bony graft for instruments. The 2 mm cutting bur is used once more to undercut the bone graft posteriorly and superiorly to the extent visible or safe. While the bur is not always fully visualized at this step, measurements of preoperative imaging ensure we avoid any potentially critical structures, such as the tegmen and the sigmoid sinus. The ledge that this creates then allows us to use an angled oscillating saw to further undermine the bone graft. A Freer elevator may be used for stabilization of the graft. After the majority of the bone graft has been freed from the underlying mastoid air cells, the final harvest of the bone graft is done with a curved osteotome. It can also sometimes simply be fractured off with a Freer elevator or similar sturdy instrument. The graft is then kept in saline until it is ready to be used in reconstruction. A standard cortical mastoidectomy is then performed using progressively smaller cutting burs. The posterior canal wall is thinned, the aditus ad antrum is entered, and the entire tegmen is skeletonized to evaluate for any defects. It is not uncommon to find multiple defects, and each must be identified and addressed. In this case, a single tear‐drop shaped defect was identified. This was measured to be 17 by 5 mm. The dimensions of the bone graft needed are measured, and the bone graft is then thinned and sculpted to the appropriate shape and size using diamond burs. The fit is tested, and further refinement of the shape and size is performed as needed. A multilayer repair of the CSF leak is then performed. A variety of materials can be used depending on surgeon preference. If there is a distinct, well‐defined identifiable dural defect, we will usually place a dumbbell‐shaped free muscle graft initially. In the current case, DuraMatrix was placed first, followed by DuraFlex. The bone graft is then carefully placed through the tegmen defect into the epidural space to support the dural grafts. A TachoSil fibrin‐coated collagen patch was laid over the intracranial grafts and remainder of the tegmen. Gelfoam is then placed into the antrum to isolate the middle ear and further support the TachoSil. Tisseel is then used to cover the TachoSil. The mastoid periosteum and skin were then closed in layers. The bone graft is noted to be well seated on post‐operative CT scan. Video content can be viewed at https://onlinelibrary.wiley.com/doi/10.1002/lary.70662. Representative images of bone graft harvest are shown; (A) the graft is designed based on preoperative imaging measurements to ensure safe margins from critical structures, with the most posterior point in this case 2 cm posterior to the external auditory canal (EAC); (B) view after bone graft is initially outlined with 2 mm bur; (C) view after posterior and superior bony ledges are thinned with 5 mm bur; (D) initial undercutting of the bone graft with a 2 mm bur; (E) the angled oscillating saw is used to further undermine the bone graft; (F) view of mastoid air cells after bone graft harvest. A standard mastoidectomy is then performed to skeletonize the tegmen. It is not uncommon for multiple defects to be present, and each must be identified and addressed. In this case, a single 17 mm by 5 mm tear‐drop shaped defect was identified. The bone graft is thinned and sculpted to the appropriate shape and size using small diamond burs. A multilayer sandwich technique is employed for repair (Figure 2A–E). Surgeon preference may dictate the exact materials used. If a well‐defined identifiable dural defect is encountered, a dumbbell‐shaped free muscle graft is inspissated through the defect straddling the dura. In this case, the dura was reinforced with Duramatrix (Stryker Corporation, Kalamazoo, MI) and DuraFlex (Integra Life Sciences, Princeton, NJ) dural onlay grafts, composed of bovine collagen and bovine pericardium, respectively. These are supported by the bone graft, which is placed through the bony defect to support the intracranial grafts. A TachoSil (Corza Medical, Parsippany, NJ) fibrin‐coated collagen patch was laid over the bony defect and remainder of the tegmen. The middle ear was isolated from the mastoid cavity with Gelfoam (Pfizer, New York, NY) gelatin sponge, and further Gelfoam was placed over the TachoSil. Placement of Tisseel (Baxter International, Deerfield, IL) fibrin sealant over the grafts concluded the reconstruction, and the surgical wound was closed in a standard multilayer fashion. (A) The bony defect to undergo multilayer repair is demonstrated (B) DuraMatrix (Stryker, Kalamazoo, MI) was placed first; (C) DuraFlex (Integra LifeSciences, Princeton, NJ) was placed next; (D) The bone graft is sculpted and thinned, then placed through the tegmen defect into the epidural space to support the dural grafts; (E) A TachoSil patch (Corza Medical, Parsippany, NJ) was laid over the defect and remainder of the tegmen; (F) Gelfoam (Pfizer, New York, NY) was then placed into the aditus and to further support the Tachosil. Tisseel (Baxeter, Deerfield, IL) is then placed to cover the Tachosil. Results A total of 34 patients were included, with 25 (74%) female. Average age at surgery was 60.3 years (SD 9.2). The majority of patients had encephaloceles (n = 28, 82%) and were spontaneous (n = 32, 94.1%). Defects included the tegmen mastoideum in 16 cases, tegmen tympani in 3 cases, and both in 15 cases. The largest defects were measured at 2.5 cm in longest dimension. No patients were found to have persisting post‐operative CSF otorrhea or recurrence. One patient had symptoms concerning for recurrent leak but no leak or dehiscence was identified on exploration. Average follow‐up time was 13.3 months (0.4–72.6 months, SD 18.3 months). There were no major complications. Two patients (5.9%) had minor wound healing complications. Discussion The use of an autologous mastoid cortex bone graft has been underreported as a source of bone for CSF leak repairs. It provides several distinct advantages over other bony sources. As it is routinely within the planned operative field of a transmastoid approach, it minimizes donor site morbidity and the need for additional incisions or special equipment. Even harvest of a cartilage graft such as the tragal cartilage adds an additional surgical site and operative time. In contrast, the monocortical mastoid cortex bone can be harvested quickly, carries little additional risk, and can be easily incorporated into existing surgical workflows. If a 90° oscillating saw is unavailable, it is possible to also harvest the bone graft using a curved osteotome. At the same time, it maintains the structural integrity of cortical bone, as opposed to converting it into bone pate for use as an autologous graft. One consideration in using this site for bony graft material is in patients with prior history of mastoidectomy, as typically a majority of the mastoid cortex is drilled away. In one patient with such history, the cortex overlying the mastoid tip was harvested. While the harvested bone initially appeared sufficient, it was ultimately not large enough to repair the bony defect and was subsequently not used. Additionally, because the monocortical bone graft is outlined without the traditional wide exposure from saucerization, awareness of individual patient anatomy is critical. Though this did not happen in this series, a low‐lying tegmen or dehiscent sigmoid sinus may be at risk if not carefully avoided. There are many surgical options for safe and effective repair of lateral skull base CSF leaks. Middle cranial fossa approaches allow for a wide view of the entire tegmen and allow for adequate surgical exposure of the tegmen tympani, though they require a larger craniotomy with increased dural manipulation, as well as creation of a larger epidural potential space which can accumulate CSF or blood [4]. Exclusive repair with non‐autologous materials can result in equivalent rates of post‐operative recurrence and infection, and may reduce operative duration [2, 4, 5]. However, for patients with defects accessible through a transmastoid approach and with adequate pneumatization, use of the mastoid cortex as an autologous graft offers an option for anatomic repair without additional morbidity. Especially as the rates of idiopathic intracranial hypertension and obesity are increasing, both of which have a meaningful impact on the success rate of CSF leak repair, this technique can be an important surgical option in repairing lateral skull base defects and can complement other approaches [4]. Conclusion The described surgical technique is an effective, accessible, and easy to implement alternative to other rigid autologous grafts when repairing lateral skull base defects. This bolsters a multilayer reconstruction and is successful without a routine need for CSF diversion or diuretic therapy. Funding The authors have nothing to report. Conflicts of Interest The authors declare no conflicts of interest. Wen C. Z., Fong D. C., Bzhilyanskaya V., Kaufman A. C., Hertzano R., and Eisenman D. J., “Autologous Mastoid Cortex Bone Graft for Multilayer Repair of Lateral Skull Base Defects ,” The Laryngoscope 136, no. 11 (2026): 4991–4995, 10.1002/lary.70662. This work was presented at the Triological Society Combined Sections Meeting in Orlando, FL on January 22–24, 2026. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. References Associated Data Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.