Journal of Otology & RhinologyISSN: 2324-8785

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.
bahis siteleri bahis siteleri bahis siteleri casino siteleri

Case Report, J Otol Rhinol Vol: 2 Issue: 1

Trans-Mastoid Management of Temporal Bone Tegmen Defects, Encephaloceles and CSF Leaks

Neil S. Patel1, Erin Canopy2 and Kianoush Sheykholeslami1,3*
1University of Illinois College of Medicine at Rockford, Rockford, IL, USA
2Department of Surgery, University of Missouri – Columbia, Columbia, MO, USA
3OSF Saint Anthony Medical Center, Rockford, IL, USA
Corresponding author : Kianoush Sheikholeslami, M.D
Guilford Square Speciallity Clinic, 698 Featherstone Rd. Rockford, IL 61107, USA
Tel: +1-815-484-7021; Fax: +1-815-484-7021
E-mail: [email protected]
Received: October 29, 2012 Accepted: December 10, 2012 Published: December 17, 2012
Citation: Patel NS, Canopy E, Sheykholeslami K (2013) Trans-Mastoid Management of Temporal Bone Tegmen Defects, Encephaloceles and CSF Leaks. J Otol Rhinol 2:1. doi:10.4172/2324-8785.1000109


Trans-Mastoid Management of Temporal Bone Tegmen Defects, Encephaloceles and CSF Leaks

A review of records revealed fifteen post-surgical patients who underwent tegmen defect repair from 2009-2012. We reviewed patients’ surgical logs, pre- and postoperative charts, imaging, and billing records. At six to nine months postoperatively, all patients underwent MRI with appropriate sequencing to evaluate the repair site and evidence of residual tegmen defect, encephalocele, or CSF leaks.



Brain herniation and cerebrospinal fluid (CSF) leak has been described in the otology literature for over a century [1]. With the advent of the operating microscope and antibiotics, the incidence of encephalocele was greatly reduced. Unfortunately, this rare but serious complication still occurs. With a defective middle fossa floor, increased intracranial pressure can lead to the herniation of meningeal and/or cerebral tissue into the middle ear or mastoid cavity. There are several etiologies, including chronic otitis media (COM) with or without cholesteatoma, head trauma, or previous surgical procedures involving the temporal bone. Herniations or leaks can occur spontaneously due to a congenital tegmen defect. Most often, however, the cause of temporal lobe brain herniation is previous mastoid surgery [2-6]. The clinical presentation of middle fossa encephaloceles (MFEs) is generally mild, with symptoms of serous otitis media, aural fullness, hearing loss, and CSF otorrhea. Life-threatening complications such as meningitis, brain abscess, and temporal lobe seizures are occasionally sentinel events. Because of the potential morbidity associated with MFE, rapid diagnosis and surgical management are of paramount importance.
The diagnosis of brain herniation and/or CSF leak is facilitated by combined imaging with computed tomography (CT) and magnetic resonance imaging (MRI). CT provides details of bony anatomy that will alert the surgeon to tegmen defects and soft tissue through the defect. MRI provides improved soft tissue resolution over CT and allows the surgeon to differentiate between recurrent disease in the mastoid and brain herniation [7].
Several considerations, including the site and size of the herniated tissue, preoperative auditory function, and the presence of active infection or other coexisting pathology, must be taken into account before selecting the most appropriate surgical technique. In this study, the trans-mastoid repair of tegmen mastoideum defects, spontaneous and acquired encephaloceles and otogenic CSF leaks in a series of fifteen cases at a single hospital from 2009 to 2012 are discussed. Based on this experience, we remark on the rationale for attempting transmastoid surgical repair for tegmen defects.


Surgical logs, pre- and post-operative charts, imaging, and billing records were reviewed from 2009 to 2012. Charts of patients with the diagnosis of cholesteatoma, cranioplasty, CSF otorrhea, meningocele, and/or encephalocele were retrospectively reviewed. Data collected included age, sex, laterality, size, site, hearing status, prior surgical intervention, pathologic process, etiology (spontaneous, COM with or without cholesteatoma, and trauma including iatrogenic), method of repair, operative and perioperative complications, and post surgical complications such as recurrence of CSF otorrhea, persistence of the tegmen defect, and cholesteatoma. The initial suspicion of CSF leak was based on a history consistent with CSF rhinorrhea or otorrhea, commonly after myringotomy tube placement for middle ear effusion. Encephaloceles and tegmen mastoideum defect were usually diagnosed on preoperative imaging (CT, MRI) or as an incidental finding during mastoidectomy. Patients were regularly followed postoperatively for signs of recurrence, with clinical examination and imaging studies (CT and MRI) in all cases. Patients were followed for late complications including meningitis, extradural abscess, recurrence of underlying disease, and recurrent otorrhea. Pure-tone average (PTA) threshold was calculated, including air- and boneconduction thresholds in dB HL, by averaging the frequencies 0.5, 1, 2, and 3 (or 4) kHz. Air–bone gap (ABG) was calculated from air and bone conduction thresholds obtained pre- and post-operatively.


Patient demographics
Fifteen patients, aged 16-68 (mean: 48.3) were identified. Two-thirds were males. Nine patients had defects in the right middle cranial fossa. In the majority of cases, presenting symptoms were related to the concomitant disease (for example, otorrhea in COM or hearing loss in cholesteatoma). The etiology in the majority of cases was cholesteatoma, COM, or previous surgery. The most frequent presenting symptom was conductive or mixed hearing loss, present in 13 patients (92%). Two cases were identified as having CSF otorrhea, while 13 patients had a meningocele or meningoencephalocele without obvious CSF leak at the time of surgery. Audiologic evaluation pre-operatively showed presence of ABG (mean 53 dB, range 20-60 dB). The hearing mechanism rehabilitation using autologous or prosthetic materials (mostly total ossicular replacement prostheses) resulted in improvement of the ABG within 20 dB in all of the cases except one.
All patients had fine-cut CT of the temporal bones (650 micrometer slice thickness). A tegmen dehiscence was identified in 11 CT scans preoperatively. One case presented with spontaneous CSF otorrhea without middle ear disease at the time of surgery. Previous surgery in patients with an iatrogenic defect was canal-wall down tympanomastoidectomy in two cases, and canal-wall up tympanomastoidectomy in eight cases. Six of these patients had previously undergone more than one surgery on the same ear at another surgical center. Five encephaloceles were identified as incidental findings during cholesteatoma surgery. None of these cases were associated with CSF leak and the encephalocele was not in the immediate vicinity of the cholesteatoma. Two patients had multiple-site defects in the tegmen mastoideum.
Surgical approach, intraoperative findings, and repair materials
Making a classical retroauricular incision and a mastoidectomy, the defect, encephalocele and herniated brain tissue is exposed, coagulated, and sectioned at the hernia waist. Simultaneously, middle ear pathology is treated, if present. Then, the repair of the bony defect is performed by dissecting the dura off the defect edges and opening a space intracranially between the dura and bony tegmen (Figures 1A and 1B). Next, a piece of auricular cartilage inserted through the defect into the space between dura and tegmen intracranially, covering the defect and anchoring the cartilage on the free edges of bony defect. The reconstruction may be reinforced with bone dust. The defect was covered with temporalis fascia and fibrin glue (Figures 1C and 1D). Before complete closure of the retroauricular incision, the adequacy of the repair and its tight closure was assessed by indirectly increasing intracranial and intrathoracic pressure with a Valsalva maneuver to a pressure of 40 mm Hg for a few seconds. Efforts to preserve and/or reconstruct the middle ear conductive mechanism were attempted in all of the patients as a single-stage resection and reconstruction.
Figure 1: Intra-operative image demonstrating the tegmen dehiscence and herniated brain tissue. Left half of photograph was taken prior to cartilage graft placement.
Postoperative course
The mean follow up time was 18 months (range: 4-26 months). Four patients developed late complications during the postoperative period. One patient developed otitis media with effusion 6 months after surgery. This was successfully treated medically. One patient developed conductive hearing loss but refused to undergo revision surgery. Another two patients developed a recurrence of cholesteatoma at a different location than the previous cranioplasty site. This was treated with revision surgery in which the repair site was evaluated. The previous cranioplasty site showed complete closure of the defect with mucosa covering the tegmen plate. Hearing restoration using autologous and artificial commercially-available prostheses was performed in all of the cases except one. Post-operative audiogram performed three months after the repair revealed closure of ABG within 20 dB, with the exception of one case, which showed persistence of the initial air-bone gap.


Violation of the leptomeninges of the temporal bone may lead to leakage of CSF from the subarachnoid space and present as CSF otorrhea or otorhinorrhea. If the leak is low in volume and is not associated with complications, it is little more than a nuisance. However, otorrhea and otorhinorrhea can be life threatening because these symptoms suggest that an aqueduct may be present for bacteria to enter the subarachnoid space, leading to meningitis and abscess. The risk of meningitis in patients with a CSF leak ranges from 4% to 50% depending on the cause and circumstances of the leak [8]. Earlier studies showed that prophylactic treatment with antibiotics is not effective in preventing the onset of meningitis in patients with a CSF leak at the base of the skull [9,10]. When a patient has a leak of fluid from the ear—especially a thin, watery fluid rather than a tenacious mucoid fluid that is suspected to be CSF—then the highly sensitive and specific β2 transferrin assay can be helpful in confirming the clinical suspicion of CSF leak [11]. Measuring β-trace protein [12] and the ratio of β-trace protein in secretions to serum [13] may be even more reliable, though interpretation may be difficult in certain cases (requiring confirmation using β2 transferrin or fluorescein). Other available tests include glucose testing using a multireagent strip. Quantitative laboratory glucose determination is effective for confirming the presence of CSF. However, contamination from blood or wound secretions and active meningitis can lower CSF glucose concentration and therefore confound the results of quantitative glucose analysis.
CSF otorrhea and otorhinorrhea demand the attention of an otolaryngologist for consideration of surgical repair because long-term morbidity with a persistent CSF leak is high, especially with concomitant middle ear and mastoid disease [14,15]. Primary dural closure alone has a much higher rate of failure than layered closures [16]. Many techniques to repair an encephalocele have been described in the literature. These include a middle cranial fossa craniotomy alone [5,17], a mini-middle cranial fossa approach [2,18], a transmastoid approach [6,19-22], and a combined middle cranial fossa-transmastoid approach [3,5,19,20,23-25]. The repair has been described with fascia alone, intra- and extradural [3,19-21,24], or in combination with allogenic materials, autologous tissue grafts such as auricular cartilage [2,6,18], split-thickness calvarial bone grafts [5,21,23], pedicled pericranial flap [25], and temporalis muscle flaps [17-20]. All of these repair options depend on surgeon experience, size and location of defect, and volume of herniated brain. No single supporting mechanism was shown to have a 100% successful closure rate. A multilayered closure technique is considered to result in the highest rate of definitive closure and the lowest rate of recurrence [22,26].
In 13 of our patients, severe disease was encountered in the OR, including brain herniation not diagnosed on preoperative imaging, erosion of bony structures of the labyrinth, mastoid tegmen and/or posterior fossa bony defect(s), erosion of the posterior wall of the external auditory canal (EAC), fallopian canal of facial nerve dehiscence, and ossicular chain disruption. In our experience, the TM approach is an adequate repair method for single or multiple defects with close proximity to each other. It can be used to reliably close a dehiscence that is smaller than 4 cm and localized to the level of the posterolateral tegmen mastoideum. The approach may also be used for more anteriorly positioned defects when the ossicular chain is removed for middle ear pathology. Preoperatively, it is necessary to evaluate and confirm, by means of CT scan and MRI, the size and location of the defect and the existence of other pathology. As well, it is important to ensure that there is adequate space between the EAC and the tegmen mastoideum to allow access to the anterior and medial epitympanum. In cases where the herniated tissue is large, it can be simply coagulated and removed with caution, with the bony defect covered with cartilage, bone dust and fascia.
The transmastoid approach allows otologists to access to the middle ear and the mastoid cavity in a patient with active disease and conductive hearing loss in the presence of a middle cranial fossa defect. In the presence of cholesteatoma and conductive hearing loss, this approach enables the surgeon to remove the diseased tissue and reconstruct the ossicular chain in a single surgical attempt. Most patients in our series had long-standing disease, such as COM and cholesteatoma, and had undergone multiple previous operations. In the office, patients present with signs and symptoms of recalcitrant COM. In this setting, dural/brain herniation and/or CSF leak arise quietly and are often discovered on imaging. The chronic ear disease is managed at the same time as ossicular chain reconstruction (if needed) and tegmen defect repair, with or without brain herniation and/or CSF leak. Most of the cases were managed with the canal wall up mastoidectomy with an emphasis on safety and the prevention of intracranial catastrophe versus preservation of the posterior EAC.
Subjective hearing loss was a common preoperative complaint in all of our patients. Most of our patients had chronic ear discharge interfering with their hearing aid use. We attempted to reconstruct the hearing mechanism during the tegmen repair in each case. In their series of 31 cases of tegmen defect and meningoencephalocele, Semaan et al. [22] reported some success with two patients that required hearing rehabilitation. In our series, all of the cases required hearing rehabilitation due to advanced middle ear disease and previous surgical resections. Two patients without cholesteatoma and ossicular chain damage required mobilization of herniated material after ossicular disarticulation. In those cases, the ossiculoplasty was performed with autologous incus strut; in other cases with middle ear pathology and a disrupted ossicular chain, a TORP was used to re-establish ossicular chain continuity. Our results reveal that a carefully planned transmastoid approach can safely repair the defect, restore hearing, and improve quality of life.
The transmastoid approach allows extracranial visualization and has the advantage of allowing assessment of the middle and posterior fossa plates while avoiding craniotomy and temporal lobe retraction. However, it is not ideal for managing defects larger than 4 cm, multiple defects with significant separation, or defects in the tegmen tympani that extend toward the petrous apex. In these situations, a supplemental middle fossa exposure may be used. Herniated brain tissue is usually nonviable and can be cauterized with bipolar cautery and excised under direct vision. The size, number, and location of bony defects can be assessed with greater visualization. Dural defects can be packed with free soft tissue plugs and covered with a layer of fascia between the dura mater and tegmen plate. Bone or conchal cartilage also can be inserted to bridge bony defects in the tegmen and to support the primary graft. Bone wax is sometimes used to occlude small bony defects or to support tissue grafts. The mastoid is then covered with a layer of fascia, or filled with abdominal fat or pedicled temporalis muscle.
Encephaloceles have been described, occurring as a result of infectious, neoplastic, congenital, and traumatic injury to the tegmen [5,6,20,23]. A histopathologic examination of the herniated tissue usually reveals a layer-dependent organization: modified glial cells or middle ear mucosa comprise the superficial layer and organized neural networks constitute the deep tissue [24]. Jackson et al. found 86% of his series of 35 patients to have a history of COM with or without cholesteatoma, and 77% to have a previous history of mastoid surgery [3]. Iurato et al. in a review of the literature, found 82 of 139 patients with brain herniation occurring as a complication of previous mastoid surgery. However, a bony defect alone probably does not lead to herniation [24]. Neely and Kuhn showed that dural injury was necessary for encephalocele formation [6]. There are many theories for the pathogenesis of the dural injuries. These include local ischemia caused by the compression of an enlarging cholesteatoma, enzymatic degradation of the dura from proteins secreted by cholesteatoma epithelial cells, and direct iatrogenic injury from drilling [3,5,20]. In our series, it is likely that encephaloceles encountered incidentally during cholesteatoma surgery were secondary to a chronic inflammatory process and/or cholesteatoma weakening the mastoid tegmen (in the setting of COM) [3] or iatrogenic defect (in the setting of prior mastoid surgery). There is specific evidence supporting an imbalance of oxidants and antioxidants in the cholesteatoma, suggesting oxidant-induced bony erosion [27].
It seems that the overall number of iatrogenic dural injuries in the United States is increasing because of an increase in revision mastoidectomy surgery for chronic ear disease [3,6]. This may be due, in part, to poor surgical technique with failure to recognize dural injury. Obviously, while it is important not to cause a dural injury in the first place, recognizing an inadvertent dural injury and repairing it primarily are even more important. In our experience, more tegmen violation was found on the right mastoid, which may due to the angle at which the drill approaches the tegmen by a right-handed surgeon.
The role of lumbar drainage intra- and post-operatively remains to be elucidated. In their series of six patients who underwent combined transmastoid and middle fossa approach tegmen defect repair, Patel et al. routinely used lumbar drainage to prevent venous infarction of the temporal lobe, divert CSF flow, and promote defect closure. In our practice, we reserve the use of lumbar drainage for patients with 1) high flow CSF leaks from trauma or prior surgery, 2) spontaneous CSF leaks with signs or symptoms of hydrocephalus, or 3) defects requiring a transcranial repair (i.e. middle cranial fossa).
The CT scans of the temporal bone ordered for all of our patients for evaluation of otorrhea in the setting of COM with a history of cholesteatoma suggested tegmen defects (Figure 2). However, in cases in which the tegmen is quite thin, CT alone can be misleading because tissue averaging may hide an intact bony plate. Therefore, if there is suspicion of an encephalocele with CSF leakage, an MRI is invaluable for differentiating between granulation tissue, cholesteatoma, and herniated brain. In our opinion, the best view of the region is provided by T2-weighted coronal images, because brain and CSF entering the mastoid or middle ear are easily distinguished from the temporal bone or air in the middle ear/mastoid system (Figure 3). Both imaging modalities also prove instrumental in surgical planning because they help to define the size and location of the tegmen defect. Three-dimensional reconstructions of CT images were not used for preoperative planning in this series, but may be utilized to estimate the size of defects and to visualize the topography of the lateral skull base (Figures 4 and 5). Postoperatively, imaging may be used to visualize the repair site and to document adequate repair, as well as the presence or absence of CSF signal in the mastoid and middle ear. Imaging was performed in all of our patients 6-9 months after surgical repair, and complete closure was observed.
Figure 2: Pre-operative CT image demonstrating a representative right-sided tegmen mastoideum defect. Vestibulocochlear nerve identified as a landmark.
Figure 3: Post-operative MRI shows complete closure of the defect with no residual brain herniation. Again, the eighth cranial nerve is shown as a landmark.
Figure 4: Three-dimensional reconstruction of fine-cut CT showing left-sided tegmen defect and associated landmarks in the right middle cranial fossa.
Figure 5: Three-dimensional reconstruction of fine-cut CT showing right-sided tegmen defect and associated landmarks in the right middle cranial fossa. Note the middle ear ossicles visible through the defect.
In conclusion, the transmastoid approach is a feasible option for the management of primary tegmen defects, defects secondary to chronic middle ear disease, and defects complicated by CSF leak or brain herniation. We were able to manage multiple defects as well as those located more antero-medially in the tegmen. Hearing restoration was possible using prostheses in all but one patient in this series. In select situations where the ossicular chain is unaffected (and the transmastoid approach would result in ossicular chain disruption), or if the transmastoid approach is simply not feasible, one may choose to perform a middle cranial fossa craniotomy approach as opposed to tampering with ossicular chain continuity. Both may result in complications that should be discussed with patients prior to surgery. The importance of multi-modality imaging prior to and following surgery cannot be overemphasized. Given that iatrogenic tegmen defects are increasing in incidence, we recommend that otologic surgeons pay close attention to the temporal tegmen during middle ear procedures to avoid injury to the tegmen and overlying dura. If such an injury is caused or found incidentally, primary repair is important to avoid complications including herniation, CSF leak, and potential infection of the CNS.


Track Your Manuscript

Media Partners