Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Aug 1.
Published in final edited form as: Clin Neurol Neurosurg. 2022 Jun 7;219:107331. doi: 10.1016/j.clineuro.2022.107331

Treatment of tegmen dehiscence using a middle fossa approach and autologous temporalis fascia graft: Outcomes from a single center☆

James Swanson a, William Oetojo a, Zachary Uram a, Ignacio Jusue-Torres b, Jehad Zakaria b, Matthew L Kircher a,c, Anand V Germanwala a,b,c,*
PMCID: PMC10171465  NIHMSID: NIHMS1887588  PMID: 35724613

Abstract

Introduction:

Although bony defects of the tegmen surface are relatively common, the majority of dehiscences are asymptomatic. For those who experience symptoms, there is a wide spectrum of relatively benign manifestations such as hearing loss and otorrhea to potentially more serious but rare sequelae such as epilepsy and meningitis. Surgical management of tegmen dehiscences (TDs) can help prevent these symptoms. In this manuscript, we present one of the largest reported single team experiences of using a temporal craniotomy with middle cranial fossa approach and temporalis fascia graft in the treatment of tegmen defects.

Methods:

We retrospectively reviewed every case of a TD surgically repaired by the same neurosurgeon/otolaryngologist team at Loyola University Medical Center from May 2015 to January 2022. In our chart review, we identified 44 patients with 48 cases of tegmen defect repair. We analyzed patient characteristics, operative details, and postoperative outcomes.

Results:

44 patients met inclusion criteria for the presence of TD (mean age 55 years, 55% male, and average body mass index 35.6). 89% of these patients had no clear etiology for the dehiscence. Commonly reported symptoms were hearing loss (89%) and CSF otorrhea (82%). The least reported presenting signs and symptoms were seizures (5%) and meningitis (2%). Most defects were repaired with both temporalis fascial and calvarial bone grafts (63%), while a minority were treated with temporalis fascia only (33%), temporalis fascia with muscle (2%), or fascia lata (2%). Every patient in our sample experienced resolution of CSF otorrhea after tegmen repair and 81% of the sample reported subjective hearing improvements after surgery. 6% of our sample had post-operative infections and 8% of patients underwent repeat unilateral surgery for a surgical complication.

Conclusion:

Craniotomy for middle fossa approach using autologous temporalis fascial grafts is a safe and effective method for the treatment of TD. These procedures should be performed by experienced and multidisciplinary teams.

Keywords: Tegmen Dehiscence, Middle Fossa Approach

1. Introduction

Tegmen dehiscence (TD) describes a bony defect in the lateral skull base overlying the middle ear and mastoid air cells. While the prevalence of TD is not well documented, there may be as high as 20–35% of adults who develop temporal bone defects and 6% of adults who develop a tegmental dehiscence [1,2]. The etiology of TD ranges from traumatic, iatrogenic, congenital, secondary to cholesteatoma or chronic otitis media, and idiopathic [3]. Idiopathic TD is a common and increasingly prevalent form with various theorized pathophysiologic mechanisms [4]. One etiological theory, the arachnoid granulation theory, posits that arachnoid granulations without direct venous terminations recirculate cerebrospinal fluid, ultimately increasing the pulsatile pressure on the nearby thin temporal bone [5]. Another theory, the obesity theory, suggests that the rise in obesity explains the increase in spontaneous TD [4,6]. High body mass increases intra-abdominal and intrathoracic pressures. With greater thoracic pressure, there is a corresponding decrease in venous return due to higher resistance within the internal jugular vein. This impaired venous drainage increases intracranial and dural pulsatile pressures. Thus, obesity may predispose one to slow, chronic erosion of the middle fossa skull base.

TDs can lead to herniation of the meninges (meningocele), brain parenchyma (encephalocele), and creation of cerebrospinal fluid (CSF) fistula, causing otorrhea. Symptoms usually include hearing loss, aural fullness, otalgia, and ipsilateral fluid drainage [7,8]. If left untreated, patients with TD can rarely develop meningitis or epilepsy [9]. Thus, proper surgical management can avoid potentially serious sequelae.

Surgical management for TD is limited to three main surgical approaches: transmastoid approach (TMA), middle fossa approach (MFA), and combined transmastoid-middle fossa approach. The transmastoid approach is performed through a retroauricular incision and subsequent mastoidectomy. There are several advantages to using the TMA. Namely, the required incision is much smaller than any other approach, and the procedure is usually quicker and can be carried out without brain manipulation. The TMA, however, limits the operative view and restricts the surgeon from reaching more anteromedial defects along the middle fossa floor. The MFA is performed through a temporal craniotomy and provides a panoramic view of the middle fossa floor, allowing surgeons to repair defects outside the scope of a TMA. There is also evidence to support improved symptom resolution in MFA repaired defects compared to TMA [10]. Disadvantages of the MFA include complications associated with temporal lobe retraction, a larger incision, longer operative time, and a longer postoperative recovery. While less common, the combined transmastoid-middle fossa approach requires a mastoidectomy and a subsequent temporal craniotomy. Advocates of the combined approach note that mastoidectomy may provide better planning for the position of the subsequent craniotomy; this dual approach provides simultaneous superior and inferior views of the middle fossa skull base [11]. In reality, the combined approach is often used when otologic pathology, such as cholesteatoma, and TD need to be addressed simultaneously.

There are also variations in graft material used to repair the dural and bony defects. Autologous graft options include temporalis fascia, septal cartilage, and calvarial bone extracted from the craniotomy. Allograft options include bone cement, polymers, and ceramics. Xenogeneic grafts such as fetal bovine collagen have also been used [12].

While there are a few case series currently published, there are no studies with large case volumes of a single team. Moreover, these studies included a patient population with mixed etiology, varied surgical approaches, varied use of graft material, and all have a relatively low case volume rate [3,13–17]. To our knowledge, no study with a high case volume has reported surgical outcomes following an exclusively MFA repair of TD. In the present study, we present the experience of a single neurosurgeon/otolaryngologist team utilizing a temporal craniotomy for middle cranial fossa approach with autologous temporalis fascia grafts used in the vast majority of largely idiopathic TDs.

2. Methods

2.1. Data collection

This project received Institutional Review Board approval. All patients with tegmen dehiscence treated by the senior neurosurgeon (AG) and otolaryngologist (MK) from May 2015 to January 2022 were retrospectively included in this study. Patient demographics were then reviewed for the following variables: age at surgery, gender, body mass index, diabetes, hypertension, smoking history, hearing loss, cerebrospinal fluid fistula, pre-operative seizures, symptom onset, pre-operative audiogram results, type of middle fossa defect, etiology of defect, laterality of defect, presence of encephalocele, history of prior surgery using a middle fossa approach, placement of lumbar drain, surgical approach, graft type, length of surgery, length of postoperative hospital stay, postoperative complications, patient-reported improvement of hearing, and post-operative audiogram findings.

2.2. Surgical technique

All patients had preoperative CT imaging identifying a TD; most also had MR imaging to evaluate for the presence of an encephalocele. All patients provided informed consent. Patients were positioned supine on a horseshoe with their heads turned to the contralateral side with the aid of an ipsilateral shoulder roll. Neurophysiological monitoring was used in all cases and consisted of somatosensory evoked potentials, electroencephalography, and ipsilateral facial nerve electromyography. Monitoring is utilized to provide real-time feedback given the duration of temporal lobe retraction needed to expose the middle fossa floor. Additionally, given the fact that the geniculate ganglion may be dehiscent and exposed during the extradural dissection, facial nerve monitoring is used due to the potential risk of injury to the facial nerve. An approximate 8 cm straight preauricular incision was made extending from the root of the zygoma to the superior temporal line and a roughly 3 cm x 3 cm free graft of deep temporalis fascial was harvested during the opening (Fig. 1 and Fig. 2). A 5 cm x 5 cm temporal craniotomy centered over the external auditory canal was then made. The inferior aspect of the temporal bone was then drilled and removed using Kerrison rongeurs such that the bony exposure was flush with the middle cranial fossa (Fig. 3). Under high magnification and illumination, the middle fossa floor was identified and the dura was further elevated off the floor in a posterior to anterior direction (Fig. 4). Dural defects and TDs were identified in all cases. The temporalis fascia was then advanced along the skull base to a position medial to all dural defects (Figs. 5 and 6). A small amount of fibrin glue was placed over the repair. The bone flap was then replaced and standard wound closure was performed (Fig. 7).

Fig. 1.

Fig. 1.

Open in a new tab

Intraoperative photo showing linear incision with identification of temporalis fascia during opening.

Fig. 2.

Fig. 2.

Open in a new tab

Intraoperative photo demonstrating resected temporalis fascia graft.

Fig. 3.

Fig. 3.

Open in a new tab

Intraoperative photo depicting temporal craniotomy with dural exposure.

Fig. 4.

Fig. 4.

Open in a new tab

A: Laboratory cadaveric photo showing extradural temporal lobe retraction following craniotomy and exposure of the middle fossa floor. B: The arrow demonstrates a tegmen dehiscence (created for demonstration purposes).

Fig. 5.

Fig. 5.

Open in a new tab

Laboratory cadaveric photo demonstrating temporalis fascial graft covering the middle fossa floor and tegmen dehiscence.

Fig. 6.

Fig. 6.

Open in a new tab

Laboratory cadaveric photo depicting position of the temporalis fascial graft along the middle fossa floor after removal of temporal lobe retraction.

Fig. 7.

Fig. 7.

Open in a new tab

Intraoperative photo demonstrating closure of wound with absorbable suture. Dashed lines represent potential angling of linear incision to provide individualized skull base exposure.

2.3. Statistical analysis

Statistical analyses including the frequency, mean, range and median were computed using R studio V 4.1.0 (The R Foundation for Statistical Computing) and Microsoft Excel.

3. Results

3.1. Patient characteristics

There were 44 patients that met our inclusion criteria (Table 1). Of this sample, four patients received two separate operations due to bilateral involvement, (Case 19, Case 27, Case 34, and Case 38). While 39 of these patients had no clear etiology for the TD, five patients had a traumatic etiology (Table 2). The average age for all patients was 55 years. 55% of the patients were male and the average body mass index was 35. 63% of the patients had hypertension, 34% had diabetes, and 45% were smokers.

Table 1.

Patient Demographics and Health History.

Case # Age Sex Hypertension Diabetes BMI Smoking Hx
1 67 Female Yes No 43.32 Yes
2 69 Male Yes No 29.4 Yes
3 33 Male No No 24.94 No
4 56 Male No No 32.1 No
5 41 Female Yes Yes 49.62 Yes
6 71 Male Yes Yes 28.59 Yes
7 72 Male Yes Yes 39.68 Yes
8 26 Female No No 21.73 No
9 66 Female Yes No 40.74 Yes
10 22 Male No No 38.27 No
11 66 Male Yes Yes 30.56 Yes
12 59 Male No No 28.2 No
13 49 Female No No 32.69 No
14 61 Male Yes No 29.16 Yes
15 60 Male Yes No 38.84 Yes
16 66 Female Yes No 53.26 Yes
17 69 Male Yes No 30.85 Yes
18 35 Female No No 25.58 No
19 53 Female Yes No 52.73 Yes
20 50 Male No No 33.47 No
21 49 Female No No 24.03 No
22 52 Male No No 32.28 No
23 72 Male Yes Yes 37.31 Yes
24 64 Male Yes Yes 40.45 Yes
25 60 Female Yes No 28.25 Yes
26 61 Female No No 28.02 No
27 59 Female Yes Yes 42.13 Yes
28 61 Male Yes Yes 30.7 Yes
29 50 Male Yes No 30.74 Yes
30 51 Female No No 25.6 No
31 52 Female Yes No 26.9 Yes
32 63 Male Yes Yes 40.37 Yes
33 58 Female Yes Yes 27.5 Yes
34 59 Male Yes No 42.45 Yes
35 35 Female No No 40.24 No
36 62 Male Yes No 37.32 Yes
37 59 Female Yes Yes 33.8 Yes
38 49 Female Yes Yes 50.27 Yes
39 65 Male Yes No 38.11 Yes
40 62 Male No No 29.29 No
41 56 Male No Yes 34.75 No
42 61 Male Yes Yes 43.63 Yes
43 47 Female No No 22.61 No
44 54 Female Yes Yes 40.08 Yes
Open in a new tab

Table 2.

Patient Presenting Characteristics.

Case # Type of Hearing loss CSF Leak Seizures Symptom Onset (days) Etiology Type of Defect
1 Mixed Otorrhea N 1034 Spontaneous L. Tegmen
2 Mixed Otorrhea N 720 Spontaneous L. Tegmen
3 Sensorineural Otorrhea N 5 Trauma R. Otic capsule and MF dehiscence
4 NA Otorrhea N 460 Spontaneous R. Tegmen mastoidium
5 Conductive Otorrhea and Rhinorrea Y 66 Spontaneous L. Tegmen
6 Mixed Otorrhea N 1460 Spontaneous R. Tegmen
7 Mixed Otorrhea N 380 Spontaneous R. Tegmen
8 NA NA N 433 Spontaneous R. SSCD
9 Mixed Otorrhea and Throat N 128 Spontaneous R. Tegmen
10 Conductive Otorrhea N 106 Spontaneous L. Tegmen mastoidium
11 Mixed Otorrhea N 64 Spontaneous R. Tegmen
12 Conductive Otorrhea N 116 Spontaneous R. Tegmen
13 Conductive NA N 241 Trauma L. SSCD and Tegmen
14 Mixed Otorrhea N 142 Spontaneous L. Tegmen
15 Mixed Otorrhea N 323 Spontaneous Tegmen
16 Mixed Otorrhea N 400 Spontaneous L. Tegmen
17 Mixed Otorrhea N 547 Spontaneous L. Tegmen
18 Mixed Otorrhea N 369 Spontaneous R. Tegmen
19 Conductive Otorrhea and Rhinorrea N 1095 Spontaneous B. Tegmen
20 Conductive Otorrhea N 1800 Spontaneous R. Tegmen
21 Conductive Otorrhea N 316 Spontaneous R. Tegmen
22 Mixed NA N 912 Spontaneous R. SSCD
23 Mixed Otorrhea N 476 Spontaneous R. Tegmen
24 Mixed Otorrhea N 360 Spontaneous L. Tegmen
25 Conductive NA N 380 Spontaneous L. SSCD
26 Mixed NA N 590 Trauma L. SSCD
27 Conductive Otorrhea N 107 Spontaneous B. Tegmen
28 Mixed Otorrhea N 913 Spontaneous R. Tegmen
29 Mixed NA N 2555 Spontaneous L. SSCD
30 NA NA N 1095 Spontaneous L. SSCD
31 Conductive Otorrhea N 106 Trauma R. SSCD and Tegmen
32 Conductive Otorrhea N 400 Spontaneous R. Tegmen
33 Mixed Otorrhea N 425 Spontaneous B. Tegmen
34 Conductive Otorrhea N 1634 Spontaneous L. Tegmen
35 Conductive Otorrhea N 61 Spontaneous R. Tegmen
36 NA Otorrhea N 730 Spontaneous R. Tegmen
37 NA Otorrhea N 152 Spontaneous R. Tegmen
38 Conductive Otorrhea N 183 Spontaneous B. Tegmen
39 Mixed Otorrhea and Rhinorrea Y 183 Spontaneous B. Tegmen
40 Sensorineural NA N 305 Trauma B. SSCD and Tegmen mastoidium
41 Conductive Otorrhea N 6000 Spontaneous R. Tegmen
42 Mixed Otorrhea N 114 Spontaneous L. Tegmen mastoidium
43 Conductive Otorrhea N 414 Spontaneous R. Tegmen
44 Mixed Otorrhea and Throat N 207 Spontaneous R. Tegmen
Open in a new tab

3.2. Operative Details

There were 48 initial operations carried out on the 44 patients (Table 3). Repair for right-sided defects comprised 29 (60%) of the total operations (Table 4). From incision to the end of the surgery, the average operation lasted 185 min (median, 161; range [101–582 min]). There were six intraoperative lumbar drains placed to relax the brain due to preoperative concerns of elevated intracranial pressure. The most common grafts were temporalis fascia and cortical bone (63%), temporalis fascia only (33%), temporalis fascia and muscle (2%), and fascia lata (2%). The patient receiving fascia lata graft had previously received a temporalis fascia graft for tegmen dehiscence repair at a separate institution. Patients on average stayed at the hospital for four days.

Table 3.

Operative Details from 48 Initial Cases of Tegmen Repair Using A Middle Fossa Approach.

Operation # Graft Type Length of Surgery Hospital Stay Follow up surgery Complications Leak Resolution Hearing
1 Temporalis Fascia and Cortical Bone 197 2 0 None Yes Improvement
2 Temporalis Fascia and Cortical Bone 109 3 0 None Yes NA
3 Temporalis Fascia 114 5 0 None Yes None
4 Temporalis Fascia and Muscle 149 5 0 None Yes NA
5 Temporalis Fascia and Cortical Bone 171 73 0 None Yes Improvement
6 Temporalis Fascia 140 3 0 None Yes Improvement
7 Temporalis Fascia 122 2 0 None Yes Improvement
8 Temporalis Fascia 138 2 0 None NA NA
9 Temporalis Fascia 162 2 0 None Yes Improvement
10 Temporalis Fascia and Cortical Bone 343 2 0 None Yes None
11 Temporalis Fascia and Cortical Bone 168 3 0 None Yes None
12 Temporalis Fascia and Cortical Bone 183 4 1 Infection Yes Improvement
13 Temporalis Fascia and Cortical Bone 125 2 0 None NA Improvement
14 Temporalis Fascia 147 3 0 None Yes Improvement
15 Temporalis Fascia 235 2 0 None Yes Improvement
16 Temporalis Fascia 156 2 0 None Yes Improvement
17 Temporalis Fascia 176 2 0 None Yes Improvement
18 Temporalis Fascia 114 2 0 None Yes Improvement
19 Temporalis Fascia and Cortical Bone 126 4 0 None Yes Improvement
20 Temporalis Fascia and Cortical Bone 122 3 2 Infection Yes Improvement
21 Temporalis Fascia and Cortical Bone 128 2 0 None Yes None
22 Temporalis Fascia and Cortical Bone 101 2 0 None Yes Improvement
23 Temporalis Fascia 132 3 0 None NA Improvement
24 Temporalis Fascia 151 6 0 None Yes Improvement
25 Temporalis Fascia and Cortical Bone 155 2 0 None Yes None
26 Temporalis Fascia 169 2 0 None NA Improvement
27 Temporalis Fascia 278 2 0 None NA None
28 Temporalis Fascia and Cortical Bone 199 2 0 None Yes Improvement
29 Temporalis Fascia and Cortical Bone 212 2 0 None Yes Improvement
30 Temporalis Fascia and Cortical Bone 221 3 0 None Yes Improvement
31 Temporalis Fascia 217 2 0 None NA Improvement
32 Temporalis Fascia 273 3 0 None NA NA
33 Temporalis Fascia and Cortical Bone 261 1 0 None Yes Improvement
34 Temporalis Fascia and Cortical Bone 326 3 0 None Yes Improvement
35 Temporalis Fascia and Cortical Bone 147 3 0 None Yes Improvement
36 Temporalis Fascia and Cortical Bone 149 2 0 None Yes Improvement
37 Temporalis Fascia and Cortical Bone 144 2 0 None Yes Improvement
38 Temporalis Fascia and Cortical Bone 146 2 0 None Yes Improvement
39 Temporalis Fascia and Cortical Bone 171 3 0 None Yes NA
40 Fascia Lata 286 6 0 None Yes NA
41 Temporalis Fascia and Cortical Bone 167 5 0 None Yes Improvement
42 Temporalis Fascia and Cortical Bone 582 6 0 Seizure Yes Improvement
43 Temporalis Fascia and Cortical Bone 292 3 0 None Yes None
44 Temporalis Fascia and Cortical Bone 173 2 0 None NA None
45 Temporalis Fascia and Cortical Bone 160 2 2 Infection Yes Improvement
46 Temporalis Fascia and Cortical Bone 153 8 0 None Yes Improvement
47 Temporalis Fascia and Cortical Bone 136 3 0 None Yes Improvement
48 Temporalis Fascia and Cortical Bone 188 1 1 Seizure Yes NA
Open in a new tab

Table 4.

Operative Details From 48 Initial Cases of Tegmen Defect Repair Using a Middle Fossa Approach.

Variable N or Mean %
Surgery length (min.) 185 –
Lumbar Drain 6 of 48 12.5%
Laterality
Left 19 of 48 40.0%
Right 29 of 48 60.4%
Graft type
Temporalis fascia 16 of 48 33.3%
Temporalis fascia + bone 30 of 48 62.5%
Temporalis fascia + muscle 1 of 48 2.1%
Fascia lata 1 of 48 2.1%
Open in a new tab

3.3. Post-operative outcomes

Post-operative outcomes were recorded in subsequent patient encounters. In our sample, the average follow-up was 252 days (Table 5). The average hospital stay was four days (median, 2; range [1,73]). All patients were placed on levetiracetam only while inpatient for seizure prophylaxis. The patient with a hospital stay of 73 days had no complications from MFA tegmen repair, but her stay was elongated by pre-existing cardiovascular issues (Operation #5). Only 23 cases of a patient receiving preoperative and postoperative audiograms were recorded. Out of those 23 cases, comparison between preoperative and postoperative audiograms showed that 18 patients (78%) had objective improvements in hearing. There were 33 cases out of 41 (81%) of patients reporting subjective hearing improvement. Of these cases, 16 of 18 (89%) cases had improvement in conductive hearing loss, 17 of 20 (85%) cases had improvements in mixed hearing loss, and zero of three (0%) cases had improvements in sensorineural hearing loss. Additionally, there were 37 of 45 patients (82%) reporting subjective resolution of otorrhea. Four patients required repeat surgical intervention due to a complication. The most common complications were infections (6%) and wound dehiscence (2%). Two patients had postoperative seizures, one of which presented with preoperative seizures while the other presented with preoperative meningitis. There were zero cases of facial nerve injury.

Table 5.

Postoperative Outcomes Following 48 Initial Cases of Tegmen Repair using a Middle Fossa Approach.

Outcome N or Mean Range or %
Follow-up (days) 252 Range 29–1478
Hospital stay (day) 4.4 Range 1–73
Complications
Facial nerve injury 0 of 48 0%
Infection 3 of 48 6.30%
Seizure 2 of 48 4.17%
Follow-up Surgery 4 of 48 8.30%
Wound dehiscence 1 of 48 2.10%
Symptom Resolution
Overall hearing loss 33 of 41 80.50%
Audiogram improvement 18 of 23 78.30%
Conductive loss 16 of 18 88.90%
Mixed loss 17 of 20 85.00%
Sensorineural loss 0 of 3 0.00%
Otorrhea 40 of 40 100.00%
Open in a new tab

Surgical outcomes were further stratified-based on the type of graft used in the repair. Repairs with temporalis fascia and cortical bone were compared to those using fascia alone. Patients receiving both fascial and cortical bone grafts had slightly lower hearing improvement rates (78%) compared to the fascia only group (86%). The cohort of patients receiving temporalis fascia and bone grafts had higher complication rates on multiple measures: seizures (10% vs. 0%), infection (10% vs. 0%), wound dehiscence (3% vs. 0%), and hospital stay (median of 2.5 days vs. 2 days).

For the six patients receiving lumbar drains, the drain was removed an average of 2 days after the operation. Overall hearing improvement in this subset was found to be 80% with no incidences of infections or wound dehiscence.

4. Discussion

Bony defects of the tegmen surface can lead to hearing loss and predispose patients to potentially serious consequences. In one of the largest reported case series, we reviewed the outcomes of a single multidisciplinary team at a single institution predominantly using a free autograft for the TD repair through a linear incision.

A review of the patient characteristics reveals that patients presenting with symptomatic TDs were more likely to be obese (BMI > 30), advanced in age (> 50 years), and hypertensive. These findings support prior studies suggesting the incidence of spontaneous tegmen defect is on the rise, and that obesity is the main driver of this increase [6]. Though it is still unclear whether smoking or diabetes has a significant impact on the development of TD, almost half of our patients had a history of smoking and about a third had diabetes. Of the three post-operative wound infections and one poorly-healing wound, three of these patients had a long history of smoking tobacco products, one patient had a body mass index greater than 50, and one patient continuously picked at his wound closure.

The middle fossa approach using a temporalis fascia graft led to acceptable outcomes with regards to presenting symptoms. The vast majority (81%) of patients in our sample reported subjective improvements in hearing; this is within the 64–96% reported range of other series [12,16,17]. Furthermore, 78% of tested patients had objective improvements on postoperative audiograms. CSF otorrhea improved in all patients and is similar to the 95%– 100% reported range [12,16,17]. While studies reporting TM approach are small, we found CSF otorrhea resolution ranged from 80% to 94% following this approach [18–20].

Following our repair of TD, there were complications from infection (6%), poor wound healing (2%), and seizure (4%). Our rate of repeat surgery for complications parallels the rate of revision surgeries reported across all three surgical approaches [3]. Our post-operative infection rate was 6%. All three of these patients were successfully treated with a course of antibiotics following surgical exploration, irrigation, wound debridement, and bone flap removal. Other studies investigating similarly invasive approaches have reported a range in infection rates from 4.5% to 8% [16,21]. Despite a subset of patients developing postoperative infections, no patient ever developed meningitis following our MFA surgery.

While 95% of the patients who underwent MFA surgery received temporalis fascial grafts, there were variations in treatment application. We utilized four graft variations, including temporalis fascia with cortical bone graft (63%), temporalis fascia graft only (33%), temporalis fascia with muscle graft (2%), and fascia lata graft (2%). Temporalis fascia and cortical bone graft were generally used to repair larger bony defects compared to patients receiving temporalis fascia only. Patients receiving temporalis fascia and cortical bone were noted to have more post-operative seizures, infections, and longer hospital stay. There were marginal differences in hearing improvement between those receiving cortical bone and fascial grafts compared to those receiving fascia only (78% vs 86%). Lumbar drain utilization comprised another treatment variation. The lumbar drain was used to decrease intracranial pressures and facilitate temporal lobe retraction in those patients suspected to have elevated intracranial pressures (based on imaging findings) and facilitate wound healing. The subset of six patients who received intraoperative lumbar drains experienced CSF leak resolution and had similar hearing improvements (80%) to the overall sample with no post-operative infections or wound dehiscences.

There were multiple variables in our dataset coded as NA. These missing values did not significantly impact data interpretation. Eight cases were coded as NA for CSF otorrhea resolution due to these patients having superior semicircular canal dehiscences and not presenting symptoms or signs of CSF otorrhea. Five cases were coded as NA for hearing loss improvement due to these patients not presenting with hearing loss. For two other cases of tegmen defects, the post-operative hearing improvements could not be confirmed and were thus coded as NA for the hearing loss improvement variable.

Investigating the results of a single treating team at a single institution provides the consistency to evaluate outcomes in these rare conditions that have a variety of treatment approaches and modifications. A review of the literature shows that the volume in our series exceeded that of other published tegmen defect repair articles to date. Our surgical team repaired 48 individual TDs over an interval of approximately six years. In a geographically comparable region, Eddelman et al. reviewed 37 operative cases within seven years across four different neurosurgical teams [17]. One of the largest studies in the United States reported treating 86 patients across 12 years and two different institutions [15].

In our experience, the middle fossa approach using autologous transplant grafts leads to reasonably good outcomes with limited post-operative complications. The MFA allowed for an optimal panoramic view of the middle fossa floor and graft placement. Additionally, our team utilized a vertical linear incision, opposed to a horseshoe incision, to expose the lateral aspect of the cranium. This approach contrasts with the horseshoe incision approach, where one begins the incision preauricularly and ends post-auricular. While the vertical linear incision may limit the anterior to posterior range of exposure along the skull base, our team successfully visualized and repaired all cases of tegmental defects. This requires proper review of pre-operative imaging and strategic angling of the vertical incision; the incision is angled slightly anterior for anterior middle fossa defects and angled slightly posterior for posterior skull middle fossa defects (Fig. 7).

There are some limitations in interpreting our study results, the most important of which being that it is a retrospective study that reviewed only one type of surgical approach by a single team at a single institution. Additionally, due to the nature of our surgical approach, all patients were healthy enough to withstand a roughly 3-hour neurosurgical operation. Patients who are surgically unfit for this length of operation may need a less invasive approach or require alternative therapy to the one described in this paper. This study is also one of only a handful of studies to examine the long-term outcomes of tegmen dehiscence repairs using the MFA, but comparing our outcomes to other institutions and approaches is currently difficult due to the variations in case numbers, surgical nuances, and different operative techniques. Regardless, our data suggests that the middle fossa approach with autologous temporalis fascia using a linear incision is safe and has a comparable low risk of complications to that described in other series. The middle fossa approach can be used to effectively repair tegmen dehiscence, improve symptoms, and prevent potential sequelae.

5. Conclusion

The MFA can lead to safe outcomes with a high rate of symptomatic improvement and the mitigation of potentially serious consequences. Repair of tegmen defects using autologous temporalis fascia through a linear incision is effective. Such procedures should be performed by experienced, multidisciplinary teams in otolaryngology and neurosurgery.

Acknowledgments

☆ This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Footnotes

CRediT authorship contribution statement

James Swanson: Investigation, Writing – original draft, Formal analysis, Writing – review & editing. William Oetojo: Investigation, Writing – original draft, Formal analysis. Zachary Uram: Investigation, Writing – original draft. Ignacio Jusue-Torres: Resources, Writing – review & editing. Jehad Zakaria: Resources, Writing – review & editing. Matthew Kircher: Writing – review & editing. Anand Germanwala: Conceptualization, Supervision, Project administration, Writing – review & editing.

Declaration of interest

None.

References

  • [1].Kapur TR, Bangash W, Tegmental and petromastoid defects in the temporal bone, J. Laryngol. Otol 100 (10) (1986) 1129–1132, 10.1017/S0022215100100702. [DOI] [PubMed] [Google Scholar]
  • [2].Åhrén C, Thulin C-A, Lethal intracranial complications following inflation in the external auditory canal in treatment of serous otitis media and due to defects in the petrous bone, Acta Oto-Laryngol 60 (1–6) (1965) 407–421, 10.3109/00016486509127025. [DOI] [Google Scholar]
  • [3].McNulty B, Schutt CA, Bojrab D, Babu S, Middle cranial fossa encephalocele and cerebrospinal fluid leakage: etiology, approach, outcomes, J. Neurol. Surg. Part B, Skull Base 81 (3) (2020) 268–274, 10.1055/s-0039-1688793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Nelson RF, Gantz BJ, Hansen MR, The rising incidence of spontaneous cerebrospinal fluid leaks in the united states and the association with obesity and obstructive sleep apnea, Otol. Neurotol 36 (3) (2015) 476–480, 10.1097/MAO.0000000000000535. [DOI] [PubMed] [Google Scholar]
  • [5].Raine C, Diagnosis and management of otologic cerebrospinal fluid leak, Otolaryngol. Clin. North Am 38 (4) (2005) 583–595, 10.1016/j.otc.2005.03.009. [DOI] [PubMed] [Google Scholar]
  • [6].Agha M, Agha R, The rising prevalence of obesity: part A: impact on public health, Int. J. Surg. Oncol 2 (7) (2017), e17, 10.1097/IJ9.0000000000000017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Jeevan DS, Ormond DR, Kim AH, Meiteles LZ, Stidham KR, Linstrom C, Murali R, Cerebrospinal fluid leaks and encephaloceles of temporal bone origin: Nuances to diagnosis and management, World Neurosurg 83 (4) (2015) 560–566, 10.1016/j.wneu.2014.12.011. [DOI] [PubMed] [Google Scholar]
  • [8].Handzel O, Brenner-Ullman A, Cavel O, Oron Y, Wasserzug O, Fliss DM, Ungar OJ, Clinical Implications of the Association Between Temporal Bone Tegmen Defects and Superior Semicircular Canal Dehiscence, Otol. Neurotol 39 (6) (2018) 797–802, 10.1097/MAO.0000000000001814. [DOI] [PubMed] [Google Scholar]
  • [9].Kenning TJ, Willcox TO, Artz GJ, Schiffmacher P, Farrell CJ, Evans JJ, Surgical management of temporal meningoencephaloceles, cerebrospinal fluid leaks, and intracranial hypertension: treatment paradigm and outcomes, Neurosurg. Focus 32 (6) (2012), E6, 10.3171/2012.4.FOCUS1265. [DOI] [PubMed] [Google Scholar]
  • [10].Rodgers B, Lin J, Staecker H, Transmastoid resurfacing versus middle fossa plugging for repair of superior canal dehiscence: comparison of techniques from a retrospective cohort, World J. Otorhinolaryngol. - Head. Neck Surg 2 (3) (2016) 161–167, 10.1016/j.wjorl.2016.11.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Nahas Z, Tatlipinar A, Limb CJ, Francis HW, Spontaneous meningoencephalocele of the temporal bone: clinical spectrum and presentation, Arch. Otolaryngol. Neck Surg 134 (5) (2008) 509–518, 10.1001/archotol.134.5.509. [DOI] [PubMed] [Google Scholar]
  • [12].Patel A, Zakaria J, Bartindale MR, Germanwala AV, Anderson DE, Marzo SJ, Kircher ML, Leonetti JP, Prabhu VC, Fetal bovine collagen grafts for repair of tegmen defects and encephaloceles via middle cranial fossa approach, Ear, Nose,Throat J 100 (3_suppl) (2021) 347S–351S, 10.1177/0145561320906906. [DOI] [PubMed] [Google Scholar]
  • [13].Remangeon F, Lahlou G, Alciato L, Tankere F, Mosnier I, Sterkers O, Pyatigorskaya N, Bernardeschi D, Management of tegmen defects with mastoid and epitympanic obliteration using S53P4 bioactive glass, Laryngoscope Invest. Otolaryngol 5 (2) (2020) 297–304, 10.1002/lio2.374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Gupta A, Sikka K, Irugu DVK, Verma H, Bhalla AS, Thakar A, Temporal bone meningoencephaloceles and cerebrospinal fluid leaks: Experience in a tertiary care hospital, J. Laryngol. Otol 133 (3) (2019) 192–200, 10.1017/S0022215119000203. [DOI] [PubMed] [Google Scholar]
  • [15].Carlson ML, Copeland WR, Driscoll CL, Link MJ, Haynes DS, Thompson RC, Weaver KD, Wanna GB, Temporal bone encephalocele and cerebrospinal fluid fistula repair utilizing the middle cranial fossa or combined mastoid-middle cranial fossa approach, J. Neurosurg 119 (5) (2013) 1314–1322, 10.3171/2013.6.JNS13322. [DOI] [PubMed] [Google Scholar]
  • [16].Hoang S, Ortiz Torres MJ, Rivera AL, Litofsky NS, Middle cranial fossa approach to repair tegmen defects with autologous or alloplastic graft, World Neurosurg 118 (2018) e10–e17, 10.1016/j.wneu.2018.05.196. [DOI] [PubMed] [Google Scholar]
  • [17].Eddelman DB, Munich S, Kochanski RB, Eggerstedt M, Kazan RP, Moftakhar R, Munoz L, Byrne RW, Wiet RM, Repair of temporal bone defects via the middle cranial fossa approach: treatment of 2 pathologies with 1 operation,Neurosurgery 84 (6) (2019) 1290–1295, 10.1093/neuros/nyy198. [DOI] [PubMed] [Google Scholar]
  • [18].Oliaei S, Mahboubi H, Djalilian HR, Transmastoid approach to temporal bone cerebrospinal fluid leaks, Am. J. Otolaryngol 33 (5) (2012) 556–561, 10.1016/j.amjoto.2012.01.011. [DOI] [PubMed] [Google Scholar]
  • [19].Perez E, Carlton D, Alfarano M, Smouha E, Transmastoid repair of spontaneous cerebrospinal fluid leaks, J. Neurol. Surg. Part B, Skull Base 79 (5) (2018) 451–457, 10.1055/s-0037-1617439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Rao AK, Merenda DM, Wetmore SJ, Diagnosis and management of spontaneous cerebrospinal fluid otorrhea, Otol. Neurotol 26 (6) (2005) 1171–1175, 10.1097/01.mao.0000179526.17285.cc. [DOI] [PubMed] [Google Scholar]
  • [21].Marchioni D, Bonali M, Alicandri-Ciufelli M, Rubini A, Pavesi G, Presutti L, Combined approach for tegmen defects repair in patients with cerebrospinal fluid otorrhea or herniations: our experience, J. Neurol. Surg. Part B, Skull Base 75 (4) (2014) 279–287, 10.1055/s-0034-1371524. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES