Original Research—Otology and Neurotology
Transmastoid Approach to Spontaneous Temporal Bone Cerebrospinal Fluid Leaks: Hearing Improvement and Success of Repair
Otolaryngology– Head and Neck Surgery 2014, Vol. 150(3) 472–478 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599813518173 http://otojournal.org
Leslie Kim, MD, MPH1, Clayton Ellis Wisely1, and Edward E. Dodson, MD1
No sponsorships or competing interests have been disclosed for this article.
Abstract Objective. To determine whether the transmastoid approach to repair of spontaneous temporal bone cerebrospinal fluid (CSF) leak is safe and effective and if improvement in conductive hearing loss is an achievable goal with this approach. Study Design. Case series with chart review. Setting. Tertiary academic medical center. Subjects and Methods. Fifteen consecutive patients (16 ears) presented with spontaneous temporal bone CSF leaks over a 6-year period. Clinical data, imaging, audiometry, operative reports, and postoperative course were reviewed. Results. Median age was 59.5 years. Mean body mass index was 40.7 kg/m2. All presented with chronic otorrhea after tympanostomy tube placement and conductive/mixed hearing loss. The mean preoperative air-bone gap was 19 dB. A transmastoid approach alone was used in 15 cases; 1 underwent middle fossa craniotomy. Most defects were located in the tegmen mastoideum and tympani. All repairs were multilayered, typically using autologous mastoid bone, temporalis fascia, and tissue sealant. Primary repair was successful in 15 cases; 1 patient with persistent postoperative otorrhea subsequently underwent middle fossa craniotomy, but no frank leakage was found. No serious complications were encountered. Following transmastoid repair, postoperative audiograms were available for 14 patients. The mean improvement in air-bone gap was 12 dB. Closure of the air-bone gap to 12 dB occurred in 100% of cases. Conclusion. The transmastoid approach to repair of spontaneous temporal bone CSF leak is highly successful. Furthermore, patients in this series had excellent hearing results with closure of their air-bone gap to 12 dB, which has not been previously described.
Keywords spontaneous cerebrospinal fluid leak, cerebrospinal fluid otorrhea, temporal bone encephalocele, temporal bone
meningoencephalocele, transmastoid approach, idiopathic intracranial hypertension Received August 23, 2013; revised November 25, 2013; accepted December 5, 2013.
C
erebrospinal fluid (CSF) leakage from the temporal bone is a rare clinical entity. It most commonly occurs as a result of trauma and iatrogenic injury, although infectious and neoplastic etiologies are also possible.1 Less often, it may arise spontaneously, in the absence of any precipitating factors. Spontaneous CSF leaks typically occur in 2 distinct populations: young children and middleaged adults.2 Spontaneous CSF otorrhea in young children frequently presents with an episode of meningitis and is typically attributable to a specific congenital defect, mostly commonly Mondini dysplasia of the labyrinth3 or, rarely, secondary to enlargement of preexisting bony pathways such as a widened cochlear aqueduct, petromastoid canal fistula, patent Hyrtl’s fissure, or a widened facial canal.4 The pathophysiology of adult-onset spontaneous CSF leak, however, remains less certain. Some suggest that small bony defects of the middle fossa tegmen resulting from abnormal embryologic development enlarge over time with constant CSF pressure.5,6 A second theory by Gacek et al1,7 implicates aberrant arachnoid granulations in the middle and posterior fossa dura; persistent pressure at these arachnoid granulations results in thinning and subsequent erosion of the underlying bone. Meningoencephaloceles, or herniations
1 Department of Otolaryngology–Head and Neck Surgery, The Ohio State University Wexner Medical Center, Columbus, Ohio, USA
This article was presented at the 2013 AAO-HNSF Annual Meeting & OTO EXPO; September 29–October 3, 2013; Vancouver, British Columbia, Canada. Corresponding Author: Leslie Kim, MD, MPH, Department of Otolaryngology–Head and Neck Surgery, The Ohio State University Wexner Medical Center, 915 Olentangy River Rd, Suite 4000, Columbus, OH 43212, USA. Email: [email protected]
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
Kim et al
473
of dura and brain through the skull base, can also develop at these sites. Postmortem and anatomical temporal bone studies have shown an approximately 20% incidence of bony tegmen defects.8,9 It is thought that years of constant, pulsating intracranial pressure eventually causes dural thinning to the point of rupture through these areas of bony dehiscence.2,10-12 However, as the incidence of tegmen defects is much higher than that of spontaneous CSF leaks, several authors propose that idiopathic intracranial hypertension (IIH) and morbid obesity play an etiologic role.2,12-21 Patients with IIH tend to be middle-aged, obese women, much like patients with spontaneous CSF otorrhea, suggesting that the underlying pathophysiology may be similar in both patient populations.2,12,14,15,17-19,21 Although rare, spontaneous CSF leaks are often challenging to diagnose and manage. Patients typically present with complaints of aural fullness, hearing loss, and/or pulsatile tinnitus, with a unilateral persistent or recurrent middle ear effusion noted on examination.11,13,22,23 Clinical presentations may be subtle and may be present for years before the diagnosis is made. The diagnosis ultimately depends on a high degree of clinical suspicion and is frequently made after myringotomy or tympanostomy tube placement results in persistent clear otorrhea.6,11,13,23 Ancillary tests available to facilitate the diagnosis include ear fluid biochemistry, b2-transferrin assay, radionuclide studies, and imaging, including high-resolution computed tomography (HRCT) with or without magnetic resonance imaging (MRI). The most significant complication of CSF leak is meningitis. Definitive and expedited surgical repair is therefore critical to prevent this potentially devastating complication. Options for surgical repair include transmastoid, middle fossa craniotomy, and combined approaches. To our knowledge, no prospective randomized trials exist comparing the success of the various approaches; therefore, the choice remains surgeon dependent, with factors such as defect location and size in mind. Many neurotologists advocate for the middle cranial fossa approach.5,6,9,11,18,24-26 Its advantages include a direct route and improved exposure of the entire tegmen and petrous apex. Hearing is generally thought to be preservable in most patients with this approach.6,11,18,22,24,25,27 However, the middle fossa craniotomy approach is associated with greater morbidity and the risk of neurological complications, including stroke and seizures.24,27 The transmastoid approach avoids these complications that may occur secondary to temporal lobe retraction, but potential disadvantages described include decreased exposure of the defect, lower initial success rate, and increased risk of hearing loss.24,27 Recent studies suggest that the transmastoid approach alone is sufficient for the repair of most temporal bone CSF leaks.12,27-31 Some advocate for a combined transmastoidtranscranial approach, especially in cases of large, multiple, or medial defects, or revision surgery for recurrent CSF leaks.27,32-34 Traditionally, the middle fossa craniotomy approach is thought to result in better hearing outcomes since the ossicular chain does not need to be manipulated to
improve exposure. Hearing improvement with the transmastoid approach has been previously mentioned, but mostly in the setting of chronic ear disease in which disarticulation and subsequent reconstruction of the ossicular chain is required.25,29,31 Improvement in conductive hearing loss in patients with spontaneous CSF leaks repaired by the transmastoid approach has not been well described in the literature. The objective of this study is to describe the presentation, management, and outcomes of patients who present with spontaneous temporal bone CSF leaks and to determine whether the transmastoid approach to repair is safe and effective. The secondary objective is to determine if improvement in conductive hearing loss is an achievable goal with this approach.
Methods Subjects Approval was obtained from The Ohio State University Wexner Medical Center’s Institutional Review Board. Retrospective chart review was conducted on all consecutive cases of spontaneous CSF otorrhea, temporal bone encephalocele, or meningoencephalocele that presented to the senior author (E.E.D.) and underwent surgical repair between January 2007 and January 2013. Patients with CSF leakage secondary to iatrogenic injury, chronic infection, or trauma were excluded. Only adults 18 years or older were considered. Fifteen adult patients underwent surgical repair of spontaneous temporal bone CSF leak in 16 ears (1 patient was treated for bilateral disease) over the 6-year period. Data collected included demographic characteristics, clinical presentation, imaging, audiometry, operative reports, and postoperative course. The median follow-up period was 9 months. All but 1 patient underwent a transmastoid approach for their initial surgical repair. One patient underwent the middle fossa craniotomy approach. This patient was found to have multiple defects extending across the middle fossa plate on preoperative computed tomography (CT). After a discussion of the risks, benefits, and alternatives with the patient, the decision was made to proceed with middle fossa craniotomy.
Transmastoid Surgical Technique A C-shaped postauricular incision is made. Prior to beginning the mastoidectomy, a large portion of the lateral cortical mastoid plate is outlined and undermined using a 3-mm coarse diamond burr and freed from the underlying bone using a curved osteotome. This is then placed in bacitracin-impregnated solution until its later use. The cortical mastoidectomy is then performed, and the tegmen defect(s) is exposed. In cases with associated meningocele or meningoencephalocele, the stalk of the herniated tissue is cauterized with bipolar cautery and amputated at its base. It is subsequently sent for pathologic analysis to confirm diagnosis. All tegmen defects are repaired in a multilayered fashion, most often using a combination of autologous mastoid bone,
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
474
Otolaryngology–Head and Neck Surgery 150(3)
temporalis fascia, and tissue sealant. The dura is dissected from the defect edges, creating a small space intracranially between the dura and the bony tegmen. The previously harvested cortical mastoid bone graft is then shaped and inserted through the bony defect and into this space between the dura and tegmen. The bone graft is sized slightly larger than the defect in one dimension so that it can be locked into place with a gentle turn, to secure it above the bony edges of the defect. Temporalis fascia is then placed in an overlay fashion and covered with tissue sealant (ie, DuraSeal; Confluent Surgical Inc, Waltham, Massachusetts). The reconstruction may also include temporalis muscle graft, bone pate, and collagen matrix (ie, DuraGen; Integra LifeSciences, Plainsboro, New Jersey). The middle ear structures are preserved in all patients and, where possible, protected with Gelfoam (Pharmacia & Upjohn, Kalamazoo, Michigan) prior to placement of the tissue sealant. Tympanostomy tubes are removed at the time of surgery, but no concurrent repairs are performed.
Results Demographics and Clinical Presentation The patients ranged in age from 36 to 77 years, with a median age of 59.5 years (Table 1). There were 9 women and 6 men in the cohort. Mean body mass index was 40.7 kg/m2. All patients (100%) presented with persistent otorrhea following tympanostomy tube placement and subjective hearing loss. Laterality was split evenly with 8 of 16 (50%) rightsided and 8 of 16 (50%) left-sided cases. b-2-Transferrin was sent on otorrhea specimens in 12 of 16 cases and returned positive in 9 of 12 (75%). All 16 cases (100%) underwent HRCT preoperatively, all of which demonstrated a tegmen defect with middle ear and/or mastoid opacification, suggestive of the diagnosis. Magnetic resonance imaging was also obtained in 4 of 16 cases (25%) (see Table 1).
Operative Findings All but 1 patient underwent a transmastoid approach for their initial surgical repair; 1 patient underwent the middle fossa craniotomy approach. The surgical procedure is described in the Methods section. All defects were located in the floor of the middle cranial fossa, and 1 patient had an additional defect in the posterior fossa. Five of 16 (31%) had multiple defects, whereas the remainder had single defects. Defect sizes were documented in 12 of 16 cases, all of which were no larger than 2.1 cm. Defects were located in the tegmen tympani/antri in 6 of 16 (38%), tegmen mastoideum in 6 of 16 (38%), tegmen tympani and tegmen mastoideum in 3 of 16 (19%), and tegmen mastoideum and posterior fossa in 1 of 16 (6%). Encephaloceles were found in 13 of 16 cases (81%).
Audiometric Testing Preoperative audiograms were available in 16 of 16 cases (100%). Postoperative audiograms were available in 15 of 16 (94%). Audiometric data included air-conduction and bone-conduction thresholds reported in decibels hearing level (dB HL). Pure-tone average (PTA) thresholds were
calculated according to the American Academy of Otolaryngology—Head and Neck Surgery guidelines by averaging the frequencies at 0.5, 1, 2, and 3 kHz.35 Airbone gaps (ABGs) were calculated by subtracting the bone PTA from air PTA. The mean (SD) preoperative PTA was 41 (13) dB. The mean (SD) preoperative ABG was 19 (7) dB. Fourteen patients underwent the transmastoid approach to repair and had postoperative audiograms (Table 2, Figure 1). Of these, the mean (SD) postoperative PTA was 29 (11) dB. The mean (SD) postoperative ABG was 7 (3) dB. The change in hearing was statistically significant by the paired-sample t test, 2-tailed at 95% confidence, with P \ .001 for both PTA and ABG. Closure of the ABG to 10 dB occurred in 11 of 14 cases (79%). Closure of the ABG to 12 dB occurred in 14 of 14 cases (100%).
Outcomes The median follow-up was 9 months with a range of 2 to 54 months. Primary surgical repair by the transmastoid approach was successful in 14 of 15 cases (93%). Success was defined by the absence of CSF otorrhea at the 3-week postoperative visit. All patients with successful repair had spontaneous resolution of their tympanic membrane perforations and absence of middle ear effusion and/or leakage from their incision. One patient with persistent otorrhea 3 weeks after the initial operation subsequently underwent middle fossa craniotomy, but no frank leakage or encephalocele was found. This same patient developed a wound hematoma requiring surgical management on postoperative day 1 following transmastoid repair. No other recurrences or complications were encountered. Ten of 15 patients (67%) underwent preoperative or postoperative ophthalmology and/or neurosurgery consultation to evaluate for IIH. Only 1 of these 10 patients who were tested had an elevated opening pressure of 38 mm Hg on postoperative lumbar puncture, and she underwent ventriculoperitoneal shunt placement prior to discharge from the hospital. Postoperative lumbar drain or acetazolamide treatment was not routinely used.
Discussion Spontaneous temporal bone CSF leaks and encephaloceles are rare. Two primary theories on its pathogenesis exist; one theory implicates small congenital bony defects5,6 and the second, aberrant arachnoid granulations,1,7 as the etiology for bony skull base erosion. Although autopsy and anatomical studies show a relatively high 20% incidence of tegmen defects,8,9 the relative rarity of spontaneous temporal bone CSF leaks suggests that a physiologic predisposition may exist.15 Recent literature proposes morbid obesity and IIH as risk factors in the development of spontaneous CSF otorrhea.2,12-21 Consistent with many other reports in the literature, there was a preponderance of middle-aged, obese women in this study.2,12,14,15,17-19,21 Although 10 of 15 patients underwent ophthalmology or neurosurgery evaluation due to clinical suspicion of IIH, only 1 patient was
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
Kim et al
475
Table 1. Summary of results. Demographics
Presentation
Ear Age, No. Sex y BMI Side 1
F
36
2
M
65
3
M
50
4
F
69
5
F
77
6
F
59
7
F
65
8
M
60
9
F
45
10
F
56
11
M
62
12
F
64
13
F
45
14
M
52
15
M
46
16
M
72
Chief Complaint
42.8 AD Otorrhea s/p PET, HL 30.8 AD Otorrhea s/p PET, HL 37.0 AD Otorrhea s/p PET, HL 48.2 AS Otorrhea s/p PET, HL 39.4 AS Otorrhea s/p PET, HL 52.3 AD Otorrhea s/p PET, HL 30.6 AS Otorrhea s/p PET, HL 48.8 AD Otorrhea s/p PET, HL 55.4 AD Otorrhea s/p PET, HL 53.0 AS Otorrhea s/p PET, HL 31.3 AS Otorrhea s/p PET, HL 38.2 AS Otorrhea s/p PET, HL 32.6 AS Otorrhea s/p PET, HL 34.6 AD Otorrhea s/p PET, HL 36.0 AS Otorrhea s/p PET, HL NA AD Otorrhea s/p PET, HL
Workup
Operative Findings and Treatment
CT MRI b-2-T Approach Location Defects
Follow-up
Size, Time, cm Encephalocele Recurrence mo
MT 1 PF TT 1 MT MT
Multiple
NA
Ya
N
7
Single
1.75
Y
N
7
Multiple
2.1
Y
N
3
Y (1) TM
MT
Multiple
2.1
Y
N
7
N
Y (1) TM
TT
Single
1.8
Y
N
2
Y
N
Y (–)
TM
MT
Single
0.9
Ya
N
11
Y
N
Y (1) TM
TT
Single
1.2
N
N
16
Y
N
Y (1) TM
MT
Single
0.3
Y
N
2
Y
N
Y (1) TM
TT
Single
1.8
Ya
N
5
Y
Y
Y (1) TM
Single
2
Y
N
26
Y
N
Y (1) TM
TT 1 MT MT
Single
NA
Y
N
3
Y
Y
Y (1) MCF
Multiple 1.75
N
N
13
Y
N
Y (1) TM
TT 1 MT MT
Single
NA
Y
N
34
Y
N
Y (–)
TM
TT
Multiple
NA
N
41
Y
N
N
TM
TT
Single
0.6
Y
Y; at 3 wk N
Y
Y
N
TM
TT
Single
0.4
Y
N
16
Y
N
N
TM
Y
N
N
TM
Y
Y
Y (–)
TM
Y
N
Y
54
Abbreviations: AD, right ear; AS, left ear; b-2-T, b-2-transferrin; BMI, body mass index; CT, computed tomography; F, female; HL, hearing loss; M, male; MCF, middle cranial fossa approach; MRI, magnetic resonance imaging; MT, mastoid tegmen; N, no; NA, not available; PF, posterior fossas/p PET, status post pressure equalization tube; TM, transmastoid approach; TT, tegmen tympani; Y, yes; (–), negative result; (1) positive result. a Encephalocele was abutting the ossicles.
actually confirmed to have IIH, with a postoperative lumbar puncture showing elevated opening pressure 20 mm Hg. Eight of the 10 patients were evaluated for IIH postoperatively. Of note, an IIH patient with active CSF otorrhea may not have the classic signs and symptoms of IIH secondary to a ‘‘pressure relief valve’’ effect from the leak.2,14,21 Therefore, we tend to defer ophthalmology and/or neurosurgery referral until after the CSF leak has been repaired. All patients in this series presented with persistent clear otorrhea following tympanostomy tube placement as well as conductive or mixed hearing loss, similar to other reports in the literature.6,11,13,23 Conductive hearing loss is common secondary to the effusion or due to encephalocele abutting
the ossicular chain.25,26,31,33 The presence of persistent otorrhea after tympanostomy tube insertion, particularly in a patient without nasopharyngeal obstruction or a history of chronic ear disease or trauma, should raise the suspicion of a spontaneous CSF leak. Fortunately, no patients in this group presented with meningitis, seizures, or other more serious complications. High-resolution CT of the temporal bones was obtained in 100% of the cases in this series. High-resolution CT with thin coronal and axial cuts is usually sufficient to identify the tegmen defect(s) but may be supplemented with T2 coronal MRI images, which can more clearly delineate the potential concomitant dural/arachnoid herniation. Magnetic
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
476
Otolaryngology–Head and Neck Surgery 150(3)
Table 2. Hearing outcomes following transmastoid approach. Pure-Tone Average, dB
Air-Bone Gap, dB
Ear No.
Preoperative
Postoperative
Change
Preoperative
Postoperative
Change
1 2 3 4 5 6 7 8 10 11 13 14 15 16 Mean SD P value
37.50 32.50 42.50 47.50 65.00 42.50 43.75 55.00 13.75 45.00 38.50 26.25 62.50 45.00 42.66 13.40
27.50 20.00 29.50 30.50 52.50 25.00 17.50 43.75 6.25 38.75 31.25 27.50 32.50 28.75 29.38 11.21
–10.00 –12.50 –13.00 –17.00 –12.50 –17.50 –26.25 –11.25 –7.50 –6.25 –7.25 1.25 –30.00 –16.25 –13.29 8.01 \.001
16.25 13.75 15.00 17.50 18.75 26.25 26.25 17.50 8.75 21.25 22.25 7.50 28.75 21.25 18.64 6.28
6.25 6.25 12.00 6.75 5.00 7.50 2.50 7.50 2.50 10.00 11.25 11.25 2.50 3.75 6.79 3.36
–10.00 –7.50 –3.00 –10.75 –13.75 –18.75 –23.75 –10.00 –6.25 –11.25 –11.00 3.75 –26.25 –17.50 –11.86 7.91 \.001
Abbreviations: dB, decibel; SD, standard deviation.
Figure 1. Hearing outcomes following the transmastoid approach.
resonance imaging may also diagnose an empty sella turcica, which is seen commonly in patients with IIH as well as spontaneous CSF otorrhea.2,14,15 Although analysis of middle ear otorrhea for b-2-transferrin is generally thought to be highly sensitive and specific for CSF, b-2-transferrin testing was found to be positive only in 75% of cases; this may be secondary to the slow and intermittent nature of some CSF leaks. In this series, all but 1 patient underwent the transmastoid approach to repair the spontaneous temporal bone CSF
leak; 1 patient underwent the middle fossa craniotomy approach. All patients had defects located in the tegmen tympani and/or tegmen mastoideum. This is consistent with previous reports that meningoencephalic herniation is most common through the middle cranial fossa, given the presence of thinner bone and the direct impact of the weight of the temporal lobe at this location.1,23,25 However, 1 patient did have a concomitant defect located in the posterior fossa. The transmastoid approach offers the advantage of visualizing both middle fossa and posterior fossa defects, no matter how uncommon the latter may be. Encephaloceles were found in 13 of 16 cases (81%) and were cauterized with bipolar cautery prior to being truncated. The herniated tissue is thought to be devitalized and essentially functionless and can be safely amputated.36 No intracranial sequelae was encountered as a result. The median follow-up of this series was 9 months, with a range between 2 and 54 months. During this follow-up period, 14 of the 15 patients who underwent the transmastoid approach (93%) did not develop any recurrences. One patient who underwent transmastoid repair had persistent otorrhea 3 weeks after the initial operation; he subsequently underwent middle fossa craniotomy, but no frank leakage or encephalocele was found. This same patient experienced a postoperative wound hematoma, which required reoperation on postoperative day 1. No other complications were encountered. Our review corroborates other studies that have shown that the transmastoid approach to spontaneous CSF leak repair is safe and effective.12,13,27-31 All defects were repaired in a multilayered fashion using autologous tissue (cortical mastoid bone graft, temporalis
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
Kim et al
477
fascia, temporalis muscle, etc) as well as allogeneic materials (Gelfoam, tissue sealant, etc). Better results and lower rates of recurrence are described when a multilayered approach is used.6,12,30 Of those who underwent the transmastoid approach, all but 2 patients had a repair that used an autologous cortical mastoid bone graft placed between the dura and bony tegmen defect edges, in an extraduralintracranial fashion, similar to those described by other studies.20,30,31 We agree that this maneuver is critical in stabilizing and securing the repair. As previously described, tissue sealant may be critical in the initial postoperative period in preventing leakage of CSF; by the time the tissue glue is resorbed, fibrous tissue has likely developed to seal the area of leakage.28 The use of tissue sealant does not appear to negatively affect the ossicular chain and hearing outcomes. In this review, multiple defects as well as defects involving the tegmen tympani were successfully repaired by the transmastoid approach. However, most of the defects were singular and all known defects were less than approximately 2 cm. Success with the transmastoid approach has been described in cases of multiple defects in close proximity to each other.28,29 For medial defects extending toward the petrous apex or for large, multiple, and separated defects, the middle fossa craniotomy or combined transmastoid–middle cranial fossa approach may offer better exposure of the defect for repair.* Success with the transmastoid approach with more anterior defects has also been described, but often in the setting of chronic ear disease where the ossicular chain is first removed to address middle ear pathology.25,29,31 In our experience, epitympanic defects were sufficiently accessible and reparable through the transmastoid approach without disruption of the ossicles. Traditionally, the middle fossa craniotomy approach is thought to result in better hearing outcomes given the extent of medial exposure ‘‘from above,’’ and a potentially less traumatic dissection of encephalocele off the ossicles is described.6,11,18,22,24,25,27 Hearing improvement with the transmastoid approach has been mentioned previously, but mostly in the setting of chronic ear disease in which disarticulation and subsequent reconstruction of the ossicular chain is required.27,29,31 Improvement in conductive hearing loss in the setting of spontaneous CSF leaks with the transmastoid approach has not been previously well described. In this series, all but 1 patient had improvement in their conductive hearing loss following transmastoid repair, with a mean improvement of 12 dB in the air-bone gap. All of these patients with spontaneous CSF leaks appeared to have intact ossicles, and no patient required ossicular chain disarticulation and/or reconstruction at the time of surgery, even though many defects occurred in the tegmen antri or more anteriorly above the ossicles. Limitations of this study include its retrospective nature, small sample size, and relatively short length of patient follow-up. Despite this, the results corroborate other reports in the literature that the transmastoid approach to repair *References 1, 3, 6, 11, 13, 18, 26-28, 32, 37.
spontaneous temporal bone CSF leaks and meningoencephaloceles is safe and effective. Furthermore, improvement in conductive hearing loss is shown to be a feasible goal with this approach, which has not been previously described. Author Contributions Leslie Kim, collected data, analyzed data, wrote article; Clayton Ellis Wisely, collected data, analyzed data, wrote article; Edward E. Dodson, collected data, analyzed data, revised article.
Disclosures Competing interests: None. Sponsorships: None. Funding source: None.
References 1. Gacek RR, Gacek MR, Tart R. Adult spontaneous cerebrospinal fluid otorrhea: diagnosis and management. Am J Otol. 1999;20:770-776. 2. Goddard JC, Meyer T, Nguyen S, Lambert PR. New considerations in the cause of spontaneous cerebrospinal fluid otorrhea. Otol Neurotol. 2010;31:940-945. 3. Wetmore SJ, Herrmann P, Fisch U. Spontaneous cerebrospinal fluid otorrhea. Am J Otol. 1987;8:96-102. 4. Gacek RR, Leipzig B. Congenital cerebrospinal otorrhea. Ann Otol. 1979;88:358-365. 5. Merchant SN, McKenna MJ. Neurotologic manifestations and treatment of multiple spontaneous tegmental defects. Am J Otol. 2000;21:234-239. 6. Brown NE, Grundfast KM, Jabre A, Megerian CA, O’Malley BW, Rosenberg SI. Diagnosis and management of spontaneous cerebrospinal fluid-middle ear effusion and otorrhea. Laryngoscope. 2004;114:800-805. 7. Gacek RR, Gacek MR. Arachnoid granulations of the temporal bone. Am J Otol. 1999;20:405-406. 8. Ahren C, Thulin CA. Lethal intracranial complications following inflation in treatment of serous otitis media and due to defects in the petrous bone. Acta Otolaryngol. 1965;60:407421. 9. Ferguson BJ, Wilkins RH, Hudson W, Farmer J. Spontaneous CSF otorrhea from tegmen and posterior fossa defects. Laryngoscope. 1986;96:635-644. 10. Ommaya AK. Cerebrospinal fluid rhinorrhea. Neurology. 1964;15:106-113. 11. Leonetti JP, Marzo S, Anderson D, Origitano T, Vukas DD. Spontaneous transtemporal CSF leakage: a study of 51 cases. Ear Nose Throat J. 2005;84:702-704, 706. 12. Rao AK, Merenda DM, Wetmore SJ. Diagnosis and management of spontaneous cerebrospinal fluid otorrhea. Otol Neurotol. 2005;26:1171-1175. 13. Pappas DG Jr, Hoffman RA, Cohen NL, Pappas DG Sr. Spontaneous temporal bone cerebrospinal fluid leak. Am J Otol. 1992;13:534-539. 14. Schlosser RJ, Wilensky EM, Grady MS, Bolger WE. Elevated intracranial pressures in spontaneous cerebrospinal fluid leaks. Am J Rhinol. 2003;17:191-195.
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
478
Otolaryngology–Head and Neck Surgery 150(3)
15. Prichard CN, Isaacson B, Oghalai JS, Coker NJ, Vrabec JT. Adult spontaneous CSF otorrhea: correlation with radiographic empty sella. Otolaryngol Head Neck Surg. 2006;134:767-771. 16. Scurry WC, Ort SA, Peterson WM, Sheehan JM, Isaacson JE. Idiopathic temporal bone encephaloceles in the obese patient. Otolaryngol Head Neck Surg. 2007;136:961-965. 17. LeVay AJ, Kveton JF. Relationship between obesity, obstructive sleep apnea, and spontaneous cerebrospinal fluid otorrhea. Laryngoscope. 2008;118:275-278. 18. Kutz JW, Husain IA, Isaacson B, Roland PS. Management of spontaneous cerebrospinal fluid otorrhea. Laryngoscope. 2008; 118:2195-2199. 19. Stucken EZ, Selesnick SH, Brown KD. The role of obesity in spontaneous temporal bone encephaloceles and CSF leak. Otol Neurotol. 2012;33:1412-1417. 20. Kari E, Mattox DE. Transtemporal management of temporal bone encephaloceles and CSF leaks: review of 56 consecutive patients. Acta Otolaryngol. 2011;131:391-394. 21. Brainard L, Chen DA, Aziz KM, Hillman TA. Association of benign intracranial hypertension and spontaneous encephalocele with cerebrospinal fluid leak. Otol Neurotol. 2012;33: 1621-1624. 22. Pappas DG, Hoffman RA, Holliday RA, Hammerschlag PE, Pappas DG, Swaid SN. Evaluation and management of spontaneous temporal bone cerebrospinal fluid leaks. Skull Base. 1995;5:1-7. 23. Nahas Z, Tatlipinar A, Limb CJ, Francis HW. Spontaneous meningoencephalocele of the temporal bone: clinical spectrum and presentation. Arch Otolaryngol Head Neck Surg. 2008; 134:509-518. 24. Dutt SN, Mirza S, Irving RM. Middle cranial fossa approach for the repair of spontaneous cerebrospinal fluid otorrhoea using autologous bone pate. Clin Otolaryngol Allied Sci. 2001; 26:117-123. 25. Sanna M, Fois P, Russo A, Falcioni M. Management of meningoencephalic herniation of the temporal bone: personal experience and literature review. Laryngoscope. 2009;119: 1579-1585.
26. Gubbels SP, Selden NR, Delashaw JB, McMenomey SO. Spontaneous middle fossa encephalocele and cerebrospinal fluid leakage: diagnosis and management. Otol Neurotol. 2007;28:1131-1139. 27. Pelosi S, Bederson JB, Smouha EE. Cerebrospinal fluid leaks of temporal bone origin: selection of surgical approach. Skull Base. 2010;20:253-259. 28. Oliaei S, Mahboubi H, Djalilian HR. Transmastoid approach to temporal bone cerebrospinal fluid leaks. Am J Otolaryngol. 2012;33:556-561. 29. Patel NS, Canopy E, Sheykholeslami K. Trans-mastoid management of temporal bone tegmen defects, encephaloceles and CSF leaks. J Otol Rhinol. 2012;2:1-5. 30. Savva A, Taylor MJ, Beatty CW. Management of cerebrospinal fluid leaks involving the temporal bone: report on 92 patients. Laryngoscope. 2003;113:50-56. 31. Semaan M, Gilpin D, Hsu D, Wasman J, Megerian C. Transmastoid extradural-intracranial approach for repair of transtemporal meningoencephalocele: a review of 31 consecutive cases. Laryngoscope. 2011;121:1765-1772. 32. Patel RB, Kwartler JA, Hodosh RM, Baredes S. Spontaneous cerebrospinal fluid leakage and middle ear encephalocele in seven patients. Ear Nose Throat J. 2000;79:372-373, 376-378. 33. Lundy LB, Graham MD, Kartush JM, LaRouere MJ. Temporal bone encephalocele and cerebrospinal fluid leaks. Am J Otol. 1996;17:461-469. 34. Kemink JL, Graham MD, Kartush JM. Spontaneous encephalocele of the temporal bone. Arch Otolaryngol Head Neck Surg. 1986;112:558-561. 35. Gurgel RK, Jackler RK, Dobie RA, Popelka GR. A new standardized format for reporting hearing outcome in clinical trials. Otolaryngol Head Neck Surg. 2012;147:803-807. 36. Jackson C, Pappas DG, Manolidis S, et al. Brain herniation into the middle ear and mastoid: concepts in diagnosis and surgical management. Am J Otol. 1997;18:198-205. 37. Stenzel M, Preuss S, Orloff L, Jecker P, Mann W. Cerebrospinal fluid leaks of temporal bone origin: etiology and management. ORL J Otorhinolaryngol Relat Spec. 2005;67:51-55.
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
Transmastoid Approach to Spontaneous Temporal Bone Cerebrospinal Fluid Leaks: Hearing Improvement and Success of Repair
Otolaryngology– Head and Neck Surgery 2014, Vol. 150(3) 472–478 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599813518173 http://otojournal.org
Leslie Kim, MD, MPH1, Clayton Ellis Wisely1, and Edward E. Dodson, MD1
No sponsorships or competing interests have been disclosed for this article.
Abstract Objective. To determine whether the transmastoid approach to repair of spontaneous temporal bone cerebrospinal fluid (CSF) leak is safe and effective and if improvement in conductive hearing loss is an achievable goal with this approach. Study Design. Case series with chart review. Setting. Tertiary academic medical center. Subjects and Methods. Fifteen consecutive patients (16 ears) presented with spontaneous temporal bone CSF leaks over a 6-year period. Clinical data, imaging, audiometry, operative reports, and postoperative course were reviewed. Results. Median age was 59.5 years. Mean body mass index was 40.7 kg/m2. All presented with chronic otorrhea after tympanostomy tube placement and conductive/mixed hearing loss. The mean preoperative air-bone gap was 19 dB. A transmastoid approach alone was used in 15 cases; 1 underwent middle fossa craniotomy. Most defects were located in the tegmen mastoideum and tympani. All repairs were multilayered, typically using autologous mastoid bone, temporalis fascia, and tissue sealant. Primary repair was successful in 15 cases; 1 patient with persistent postoperative otorrhea subsequently underwent middle fossa craniotomy, but no frank leakage was found. No serious complications were encountered. Following transmastoid repair, postoperative audiograms were available for 14 patients. The mean improvement in air-bone gap was 12 dB. Closure of the air-bone gap to 12 dB occurred in 100% of cases. Conclusion. The transmastoid approach to repair of spontaneous temporal bone CSF leak is highly successful. Furthermore, patients in this series had excellent hearing results with closure of their air-bone gap to 12 dB, which has not been previously described.
Keywords spontaneous cerebrospinal fluid leak, cerebrospinal fluid otorrhea, temporal bone encephalocele, temporal bone
meningoencephalocele, transmastoid approach, idiopathic intracranial hypertension Received August 23, 2013; revised November 25, 2013; accepted December 5, 2013.
C
erebrospinal fluid (CSF) leakage from the temporal bone is a rare clinical entity. It most commonly occurs as a result of trauma and iatrogenic injury, although infectious and neoplastic etiologies are also possible.1 Less often, it may arise spontaneously, in the absence of any precipitating factors. Spontaneous CSF leaks typically occur in 2 distinct populations: young children and middleaged adults.2 Spontaneous CSF otorrhea in young children frequently presents with an episode of meningitis and is typically attributable to a specific congenital defect, mostly commonly Mondini dysplasia of the labyrinth3 or, rarely, secondary to enlargement of preexisting bony pathways such as a widened cochlear aqueduct, petromastoid canal fistula, patent Hyrtl’s fissure, or a widened facial canal.4 The pathophysiology of adult-onset spontaneous CSF leak, however, remains less certain. Some suggest that small bony defects of the middle fossa tegmen resulting from abnormal embryologic development enlarge over time with constant CSF pressure.5,6 A second theory by Gacek et al1,7 implicates aberrant arachnoid granulations in the middle and posterior fossa dura; persistent pressure at these arachnoid granulations results in thinning and subsequent erosion of the underlying bone. Meningoencephaloceles, or herniations
1 Department of Otolaryngology–Head and Neck Surgery, The Ohio State University Wexner Medical Center, Columbus, Ohio, USA
This article was presented at the 2013 AAO-HNSF Annual Meeting & OTO EXPO; September 29–October 3, 2013; Vancouver, British Columbia, Canada. Corresponding Author: Leslie Kim, MD, MPH, Department of Otolaryngology–Head and Neck Surgery, The Ohio State University Wexner Medical Center, 915 Olentangy River Rd, Suite 4000, Columbus, OH 43212, USA. Email: [email protected]
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
Kim et al
473
of dura and brain through the skull base, can also develop at these sites. Postmortem and anatomical temporal bone studies have shown an approximately 20% incidence of bony tegmen defects.8,9 It is thought that years of constant, pulsating intracranial pressure eventually causes dural thinning to the point of rupture through these areas of bony dehiscence.2,10-12 However, as the incidence of tegmen defects is much higher than that of spontaneous CSF leaks, several authors propose that idiopathic intracranial hypertension (IIH) and morbid obesity play an etiologic role.2,12-21 Patients with IIH tend to be middle-aged, obese women, much like patients with spontaneous CSF otorrhea, suggesting that the underlying pathophysiology may be similar in both patient populations.2,12,14,15,17-19,21 Although rare, spontaneous CSF leaks are often challenging to diagnose and manage. Patients typically present with complaints of aural fullness, hearing loss, and/or pulsatile tinnitus, with a unilateral persistent or recurrent middle ear effusion noted on examination.11,13,22,23 Clinical presentations may be subtle and may be present for years before the diagnosis is made. The diagnosis ultimately depends on a high degree of clinical suspicion and is frequently made after myringotomy or tympanostomy tube placement results in persistent clear otorrhea.6,11,13,23 Ancillary tests available to facilitate the diagnosis include ear fluid biochemistry, b2-transferrin assay, radionuclide studies, and imaging, including high-resolution computed tomography (HRCT) with or without magnetic resonance imaging (MRI). The most significant complication of CSF leak is meningitis. Definitive and expedited surgical repair is therefore critical to prevent this potentially devastating complication. Options for surgical repair include transmastoid, middle fossa craniotomy, and combined approaches. To our knowledge, no prospective randomized trials exist comparing the success of the various approaches; therefore, the choice remains surgeon dependent, with factors such as defect location and size in mind. Many neurotologists advocate for the middle cranial fossa approach.5,6,9,11,18,24-26 Its advantages include a direct route and improved exposure of the entire tegmen and petrous apex. Hearing is generally thought to be preservable in most patients with this approach.6,11,18,22,24,25,27 However, the middle fossa craniotomy approach is associated with greater morbidity and the risk of neurological complications, including stroke and seizures.24,27 The transmastoid approach avoids these complications that may occur secondary to temporal lobe retraction, but potential disadvantages described include decreased exposure of the defect, lower initial success rate, and increased risk of hearing loss.24,27 Recent studies suggest that the transmastoid approach alone is sufficient for the repair of most temporal bone CSF leaks.12,27-31 Some advocate for a combined transmastoidtranscranial approach, especially in cases of large, multiple, or medial defects, or revision surgery for recurrent CSF leaks.27,32-34 Traditionally, the middle fossa craniotomy approach is thought to result in better hearing outcomes since the ossicular chain does not need to be manipulated to
improve exposure. Hearing improvement with the transmastoid approach has been previously mentioned, but mostly in the setting of chronic ear disease in which disarticulation and subsequent reconstruction of the ossicular chain is required.25,29,31 Improvement in conductive hearing loss in patients with spontaneous CSF leaks repaired by the transmastoid approach has not been well described in the literature. The objective of this study is to describe the presentation, management, and outcomes of patients who present with spontaneous temporal bone CSF leaks and to determine whether the transmastoid approach to repair is safe and effective. The secondary objective is to determine if improvement in conductive hearing loss is an achievable goal with this approach.
Methods Subjects Approval was obtained from The Ohio State University Wexner Medical Center’s Institutional Review Board. Retrospective chart review was conducted on all consecutive cases of spontaneous CSF otorrhea, temporal bone encephalocele, or meningoencephalocele that presented to the senior author (E.E.D.) and underwent surgical repair between January 2007 and January 2013. Patients with CSF leakage secondary to iatrogenic injury, chronic infection, or trauma were excluded. Only adults 18 years or older were considered. Fifteen adult patients underwent surgical repair of spontaneous temporal bone CSF leak in 16 ears (1 patient was treated for bilateral disease) over the 6-year period. Data collected included demographic characteristics, clinical presentation, imaging, audiometry, operative reports, and postoperative course. The median follow-up period was 9 months. All but 1 patient underwent a transmastoid approach for their initial surgical repair. One patient underwent the middle fossa craniotomy approach. This patient was found to have multiple defects extending across the middle fossa plate on preoperative computed tomography (CT). After a discussion of the risks, benefits, and alternatives with the patient, the decision was made to proceed with middle fossa craniotomy.
Transmastoid Surgical Technique A C-shaped postauricular incision is made. Prior to beginning the mastoidectomy, a large portion of the lateral cortical mastoid plate is outlined and undermined using a 3-mm coarse diamond burr and freed from the underlying bone using a curved osteotome. This is then placed in bacitracin-impregnated solution until its later use. The cortical mastoidectomy is then performed, and the tegmen defect(s) is exposed. In cases with associated meningocele or meningoencephalocele, the stalk of the herniated tissue is cauterized with bipolar cautery and amputated at its base. It is subsequently sent for pathologic analysis to confirm diagnosis. All tegmen defects are repaired in a multilayered fashion, most often using a combination of autologous mastoid bone,
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
474
Otolaryngology–Head and Neck Surgery 150(3)
temporalis fascia, and tissue sealant. The dura is dissected from the defect edges, creating a small space intracranially between the dura and the bony tegmen. The previously harvested cortical mastoid bone graft is then shaped and inserted through the bony defect and into this space between the dura and tegmen. The bone graft is sized slightly larger than the defect in one dimension so that it can be locked into place with a gentle turn, to secure it above the bony edges of the defect. Temporalis fascia is then placed in an overlay fashion and covered with tissue sealant (ie, DuraSeal; Confluent Surgical Inc, Waltham, Massachusetts). The reconstruction may also include temporalis muscle graft, bone pate, and collagen matrix (ie, DuraGen; Integra LifeSciences, Plainsboro, New Jersey). The middle ear structures are preserved in all patients and, where possible, protected with Gelfoam (Pharmacia & Upjohn, Kalamazoo, Michigan) prior to placement of the tissue sealant. Tympanostomy tubes are removed at the time of surgery, but no concurrent repairs are performed.
Results Demographics and Clinical Presentation The patients ranged in age from 36 to 77 years, with a median age of 59.5 years (Table 1). There were 9 women and 6 men in the cohort. Mean body mass index was 40.7 kg/m2. All patients (100%) presented with persistent otorrhea following tympanostomy tube placement and subjective hearing loss. Laterality was split evenly with 8 of 16 (50%) rightsided and 8 of 16 (50%) left-sided cases. b-2-Transferrin was sent on otorrhea specimens in 12 of 16 cases and returned positive in 9 of 12 (75%). All 16 cases (100%) underwent HRCT preoperatively, all of which demonstrated a tegmen defect with middle ear and/or mastoid opacification, suggestive of the diagnosis. Magnetic resonance imaging was also obtained in 4 of 16 cases (25%) (see Table 1).
Operative Findings All but 1 patient underwent a transmastoid approach for their initial surgical repair; 1 patient underwent the middle fossa craniotomy approach. The surgical procedure is described in the Methods section. All defects were located in the floor of the middle cranial fossa, and 1 patient had an additional defect in the posterior fossa. Five of 16 (31%) had multiple defects, whereas the remainder had single defects. Defect sizes were documented in 12 of 16 cases, all of which were no larger than 2.1 cm. Defects were located in the tegmen tympani/antri in 6 of 16 (38%), tegmen mastoideum in 6 of 16 (38%), tegmen tympani and tegmen mastoideum in 3 of 16 (19%), and tegmen mastoideum and posterior fossa in 1 of 16 (6%). Encephaloceles were found in 13 of 16 cases (81%).
Audiometric Testing Preoperative audiograms were available in 16 of 16 cases (100%). Postoperative audiograms were available in 15 of 16 (94%). Audiometric data included air-conduction and bone-conduction thresholds reported in decibels hearing level (dB HL). Pure-tone average (PTA) thresholds were
calculated according to the American Academy of Otolaryngology—Head and Neck Surgery guidelines by averaging the frequencies at 0.5, 1, 2, and 3 kHz.35 Airbone gaps (ABGs) were calculated by subtracting the bone PTA from air PTA. The mean (SD) preoperative PTA was 41 (13) dB. The mean (SD) preoperative ABG was 19 (7) dB. Fourteen patients underwent the transmastoid approach to repair and had postoperative audiograms (Table 2, Figure 1). Of these, the mean (SD) postoperative PTA was 29 (11) dB. The mean (SD) postoperative ABG was 7 (3) dB. The change in hearing was statistically significant by the paired-sample t test, 2-tailed at 95% confidence, with P \ .001 for both PTA and ABG. Closure of the ABG to 10 dB occurred in 11 of 14 cases (79%). Closure of the ABG to 12 dB occurred in 14 of 14 cases (100%).
Outcomes The median follow-up was 9 months with a range of 2 to 54 months. Primary surgical repair by the transmastoid approach was successful in 14 of 15 cases (93%). Success was defined by the absence of CSF otorrhea at the 3-week postoperative visit. All patients with successful repair had spontaneous resolution of their tympanic membrane perforations and absence of middle ear effusion and/or leakage from their incision. One patient with persistent otorrhea 3 weeks after the initial operation subsequently underwent middle fossa craniotomy, but no frank leakage or encephalocele was found. This same patient developed a wound hematoma requiring surgical management on postoperative day 1 following transmastoid repair. No other recurrences or complications were encountered. Ten of 15 patients (67%) underwent preoperative or postoperative ophthalmology and/or neurosurgery consultation to evaluate for IIH. Only 1 of these 10 patients who were tested had an elevated opening pressure of 38 mm Hg on postoperative lumbar puncture, and she underwent ventriculoperitoneal shunt placement prior to discharge from the hospital. Postoperative lumbar drain or acetazolamide treatment was not routinely used.
Discussion Spontaneous temporal bone CSF leaks and encephaloceles are rare. Two primary theories on its pathogenesis exist; one theory implicates small congenital bony defects5,6 and the second, aberrant arachnoid granulations,1,7 as the etiology for bony skull base erosion. Although autopsy and anatomical studies show a relatively high 20% incidence of tegmen defects,8,9 the relative rarity of spontaneous temporal bone CSF leaks suggests that a physiologic predisposition may exist.15 Recent literature proposes morbid obesity and IIH as risk factors in the development of spontaneous CSF otorrhea.2,12-21 Consistent with many other reports in the literature, there was a preponderance of middle-aged, obese women in this study.2,12,14,15,17-19,21 Although 10 of 15 patients underwent ophthalmology or neurosurgery evaluation due to clinical suspicion of IIH, only 1 patient was
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
Kim et al
475
Table 1. Summary of results. Demographics
Presentation
Ear Age, No. Sex y BMI Side 1
F
36
2
M
65
3
M
50
4
F
69
5
F
77
6
F
59
7
F
65
8
M
60
9
F
45
10
F
56
11
M
62
12
F
64
13
F
45
14
M
52
15
M
46
16
M
72
Chief Complaint
42.8 AD Otorrhea s/p PET, HL 30.8 AD Otorrhea s/p PET, HL 37.0 AD Otorrhea s/p PET, HL 48.2 AS Otorrhea s/p PET, HL 39.4 AS Otorrhea s/p PET, HL 52.3 AD Otorrhea s/p PET, HL 30.6 AS Otorrhea s/p PET, HL 48.8 AD Otorrhea s/p PET, HL 55.4 AD Otorrhea s/p PET, HL 53.0 AS Otorrhea s/p PET, HL 31.3 AS Otorrhea s/p PET, HL 38.2 AS Otorrhea s/p PET, HL 32.6 AS Otorrhea s/p PET, HL 34.6 AD Otorrhea s/p PET, HL 36.0 AS Otorrhea s/p PET, HL NA AD Otorrhea s/p PET, HL
Workup
Operative Findings and Treatment
CT MRI b-2-T Approach Location Defects
Follow-up
Size, Time, cm Encephalocele Recurrence mo
MT 1 PF TT 1 MT MT
Multiple
NA
Ya
N
7
Single
1.75
Y
N
7
Multiple
2.1
Y
N
3
Y (1) TM
MT
Multiple
2.1
Y
N
7
N
Y (1) TM
TT
Single
1.8
Y
N
2
Y
N
Y (–)
TM
MT
Single
0.9
Ya
N
11
Y
N
Y (1) TM
TT
Single
1.2
N
N
16
Y
N
Y (1) TM
MT
Single
0.3
Y
N
2
Y
N
Y (1) TM
TT
Single
1.8
Ya
N
5
Y
Y
Y (1) TM
Single
2
Y
N
26
Y
N
Y (1) TM
TT 1 MT MT
Single
NA
Y
N
3
Y
Y
Y (1) MCF
Multiple 1.75
N
N
13
Y
N
Y (1) TM
TT 1 MT MT
Single
NA
Y
N
34
Y
N
Y (–)
TM
TT
Multiple
NA
N
41
Y
N
N
TM
TT
Single
0.6
Y
Y; at 3 wk N
Y
Y
N
TM
TT
Single
0.4
Y
N
16
Y
N
N
TM
Y
N
N
TM
Y
Y
Y (–)
TM
Y
N
Y
54
Abbreviations: AD, right ear; AS, left ear; b-2-T, b-2-transferrin; BMI, body mass index; CT, computed tomography; F, female; HL, hearing loss; M, male; MCF, middle cranial fossa approach; MRI, magnetic resonance imaging; MT, mastoid tegmen; N, no; NA, not available; PF, posterior fossas/p PET, status post pressure equalization tube; TM, transmastoid approach; TT, tegmen tympani; Y, yes; (–), negative result; (1) positive result. a Encephalocele was abutting the ossicles.
actually confirmed to have IIH, with a postoperative lumbar puncture showing elevated opening pressure 20 mm Hg. Eight of the 10 patients were evaluated for IIH postoperatively. Of note, an IIH patient with active CSF otorrhea may not have the classic signs and symptoms of IIH secondary to a ‘‘pressure relief valve’’ effect from the leak.2,14,21 Therefore, we tend to defer ophthalmology and/or neurosurgery referral until after the CSF leak has been repaired. All patients in this series presented with persistent clear otorrhea following tympanostomy tube placement as well as conductive or mixed hearing loss, similar to other reports in the literature.6,11,13,23 Conductive hearing loss is common secondary to the effusion or due to encephalocele abutting
the ossicular chain.25,26,31,33 The presence of persistent otorrhea after tympanostomy tube insertion, particularly in a patient without nasopharyngeal obstruction or a history of chronic ear disease or trauma, should raise the suspicion of a spontaneous CSF leak. Fortunately, no patients in this group presented with meningitis, seizures, or other more serious complications. High-resolution CT of the temporal bones was obtained in 100% of the cases in this series. High-resolution CT with thin coronal and axial cuts is usually sufficient to identify the tegmen defect(s) but may be supplemented with T2 coronal MRI images, which can more clearly delineate the potential concomitant dural/arachnoid herniation. Magnetic
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
476
Otolaryngology–Head and Neck Surgery 150(3)
Table 2. Hearing outcomes following transmastoid approach. Pure-Tone Average, dB
Air-Bone Gap, dB
Ear No.
Preoperative
Postoperative
Change
Preoperative
Postoperative
Change
1 2 3 4 5 6 7 8 10 11 13 14 15 16 Mean SD P value
37.50 32.50 42.50 47.50 65.00 42.50 43.75 55.00 13.75 45.00 38.50 26.25 62.50 45.00 42.66 13.40
27.50 20.00 29.50 30.50 52.50 25.00 17.50 43.75 6.25 38.75 31.25 27.50 32.50 28.75 29.38 11.21
–10.00 –12.50 –13.00 –17.00 –12.50 –17.50 –26.25 –11.25 –7.50 –6.25 –7.25 1.25 –30.00 –16.25 –13.29 8.01 \.001
16.25 13.75 15.00 17.50 18.75 26.25 26.25 17.50 8.75 21.25 22.25 7.50 28.75 21.25 18.64 6.28
6.25 6.25 12.00 6.75 5.00 7.50 2.50 7.50 2.50 10.00 11.25 11.25 2.50 3.75 6.79 3.36
–10.00 –7.50 –3.00 –10.75 –13.75 –18.75 –23.75 –10.00 –6.25 –11.25 –11.00 3.75 –26.25 –17.50 –11.86 7.91 \.001
Abbreviations: dB, decibel; SD, standard deviation.
Figure 1. Hearing outcomes following the transmastoid approach.
resonance imaging may also diagnose an empty sella turcica, which is seen commonly in patients with IIH as well as spontaneous CSF otorrhea.2,14,15 Although analysis of middle ear otorrhea for b-2-transferrin is generally thought to be highly sensitive and specific for CSF, b-2-transferrin testing was found to be positive only in 75% of cases; this may be secondary to the slow and intermittent nature of some CSF leaks. In this series, all but 1 patient underwent the transmastoid approach to repair the spontaneous temporal bone CSF
leak; 1 patient underwent the middle fossa craniotomy approach. All patients had defects located in the tegmen tympani and/or tegmen mastoideum. This is consistent with previous reports that meningoencephalic herniation is most common through the middle cranial fossa, given the presence of thinner bone and the direct impact of the weight of the temporal lobe at this location.1,23,25 However, 1 patient did have a concomitant defect located in the posterior fossa. The transmastoid approach offers the advantage of visualizing both middle fossa and posterior fossa defects, no matter how uncommon the latter may be. Encephaloceles were found in 13 of 16 cases (81%) and were cauterized with bipolar cautery prior to being truncated. The herniated tissue is thought to be devitalized and essentially functionless and can be safely amputated.36 No intracranial sequelae was encountered as a result. The median follow-up of this series was 9 months, with a range between 2 and 54 months. During this follow-up period, 14 of the 15 patients who underwent the transmastoid approach (93%) did not develop any recurrences. One patient who underwent transmastoid repair had persistent otorrhea 3 weeks after the initial operation; he subsequently underwent middle fossa craniotomy, but no frank leakage or encephalocele was found. This same patient experienced a postoperative wound hematoma, which required reoperation on postoperative day 1. No other complications were encountered. Our review corroborates other studies that have shown that the transmastoid approach to spontaneous CSF leak repair is safe and effective.12,13,27-31 All defects were repaired in a multilayered fashion using autologous tissue (cortical mastoid bone graft, temporalis
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
Kim et al
477
fascia, temporalis muscle, etc) as well as allogeneic materials (Gelfoam, tissue sealant, etc). Better results and lower rates of recurrence are described when a multilayered approach is used.6,12,30 Of those who underwent the transmastoid approach, all but 2 patients had a repair that used an autologous cortical mastoid bone graft placed between the dura and bony tegmen defect edges, in an extraduralintracranial fashion, similar to those described by other studies.20,30,31 We agree that this maneuver is critical in stabilizing and securing the repair. As previously described, tissue sealant may be critical in the initial postoperative period in preventing leakage of CSF; by the time the tissue glue is resorbed, fibrous tissue has likely developed to seal the area of leakage.28 The use of tissue sealant does not appear to negatively affect the ossicular chain and hearing outcomes. In this review, multiple defects as well as defects involving the tegmen tympani were successfully repaired by the transmastoid approach. However, most of the defects were singular and all known defects were less than approximately 2 cm. Success with the transmastoid approach has been described in cases of multiple defects in close proximity to each other.28,29 For medial defects extending toward the petrous apex or for large, multiple, and separated defects, the middle fossa craniotomy or combined transmastoid–middle cranial fossa approach may offer better exposure of the defect for repair.* Success with the transmastoid approach with more anterior defects has also been described, but often in the setting of chronic ear disease where the ossicular chain is first removed to address middle ear pathology.25,29,31 In our experience, epitympanic defects were sufficiently accessible and reparable through the transmastoid approach without disruption of the ossicles. Traditionally, the middle fossa craniotomy approach is thought to result in better hearing outcomes given the extent of medial exposure ‘‘from above,’’ and a potentially less traumatic dissection of encephalocele off the ossicles is described.6,11,18,22,24,25,27 Hearing improvement with the transmastoid approach has been mentioned previously, but mostly in the setting of chronic ear disease in which disarticulation and subsequent reconstruction of the ossicular chain is required.27,29,31 Improvement in conductive hearing loss in the setting of spontaneous CSF leaks with the transmastoid approach has not been previously well described. In this series, all but 1 patient had improvement in their conductive hearing loss following transmastoid repair, with a mean improvement of 12 dB in the air-bone gap. All of these patients with spontaneous CSF leaks appeared to have intact ossicles, and no patient required ossicular chain disarticulation and/or reconstruction at the time of surgery, even though many defects occurred in the tegmen antri or more anteriorly above the ossicles. Limitations of this study include its retrospective nature, small sample size, and relatively short length of patient follow-up. Despite this, the results corroborate other reports in the literature that the transmastoid approach to repair *References 1, 3, 6, 11, 13, 18, 26-28, 32, 37.
spontaneous temporal bone CSF leaks and meningoencephaloceles is safe and effective. Furthermore, improvement in conductive hearing loss is shown to be a feasible goal with this approach, which has not been previously described. Author Contributions Leslie Kim, collected data, analyzed data, wrote article; Clayton Ellis Wisely, collected data, analyzed data, wrote article; Edward E. Dodson, collected data, analyzed data, revised article.
Disclosures Competing interests: None. Sponsorships: None. Funding source: None.
References 1. Gacek RR, Gacek MR, Tart R. Adult spontaneous cerebrospinal fluid otorrhea: diagnosis and management. Am J Otol. 1999;20:770-776. 2. Goddard JC, Meyer T, Nguyen S, Lambert PR. New considerations in the cause of spontaneous cerebrospinal fluid otorrhea. Otol Neurotol. 2010;31:940-945. 3. Wetmore SJ, Herrmann P, Fisch U. Spontaneous cerebrospinal fluid otorrhea. Am J Otol. 1987;8:96-102. 4. Gacek RR, Leipzig B. Congenital cerebrospinal otorrhea. Ann Otol. 1979;88:358-365. 5. Merchant SN, McKenna MJ. Neurotologic manifestations and treatment of multiple spontaneous tegmental defects. Am J Otol. 2000;21:234-239. 6. Brown NE, Grundfast KM, Jabre A, Megerian CA, O’Malley BW, Rosenberg SI. Diagnosis and management of spontaneous cerebrospinal fluid-middle ear effusion and otorrhea. Laryngoscope. 2004;114:800-805. 7. Gacek RR, Gacek MR. Arachnoid granulations of the temporal bone. Am J Otol. 1999;20:405-406. 8. Ahren C, Thulin CA. Lethal intracranial complications following inflation in treatment of serous otitis media and due to defects in the petrous bone. Acta Otolaryngol. 1965;60:407421. 9. Ferguson BJ, Wilkins RH, Hudson W, Farmer J. Spontaneous CSF otorrhea from tegmen and posterior fossa defects. Laryngoscope. 1986;96:635-644. 10. Ommaya AK. Cerebrospinal fluid rhinorrhea. Neurology. 1964;15:106-113. 11. Leonetti JP, Marzo S, Anderson D, Origitano T, Vukas DD. Spontaneous transtemporal CSF leakage: a study of 51 cases. Ear Nose Throat J. 2005;84:702-704, 706. 12. Rao AK, Merenda DM, Wetmore SJ. Diagnosis and management of spontaneous cerebrospinal fluid otorrhea. Otol Neurotol. 2005;26:1171-1175. 13. Pappas DG Jr, Hoffman RA, Cohen NL, Pappas DG Sr. Spontaneous temporal bone cerebrospinal fluid leak. Am J Otol. 1992;13:534-539. 14. Schlosser RJ, Wilensky EM, Grady MS, Bolger WE. Elevated intracranial pressures in spontaneous cerebrospinal fluid leaks. Am J Rhinol. 2003;17:191-195.
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
478
Otolaryngology–Head and Neck Surgery 150(3)
15. Prichard CN, Isaacson B, Oghalai JS, Coker NJ, Vrabec JT. Adult spontaneous CSF otorrhea: correlation with radiographic empty sella. Otolaryngol Head Neck Surg. 2006;134:767-771. 16. Scurry WC, Ort SA, Peterson WM, Sheehan JM, Isaacson JE. Idiopathic temporal bone encephaloceles in the obese patient. Otolaryngol Head Neck Surg. 2007;136:961-965. 17. LeVay AJ, Kveton JF. Relationship between obesity, obstructive sleep apnea, and spontaneous cerebrospinal fluid otorrhea. Laryngoscope. 2008;118:275-278. 18. Kutz JW, Husain IA, Isaacson B, Roland PS. Management of spontaneous cerebrospinal fluid otorrhea. Laryngoscope. 2008; 118:2195-2199. 19. Stucken EZ, Selesnick SH, Brown KD. The role of obesity in spontaneous temporal bone encephaloceles and CSF leak. Otol Neurotol. 2012;33:1412-1417. 20. Kari E, Mattox DE. Transtemporal management of temporal bone encephaloceles and CSF leaks: review of 56 consecutive patients. Acta Otolaryngol. 2011;131:391-394. 21. Brainard L, Chen DA, Aziz KM, Hillman TA. Association of benign intracranial hypertension and spontaneous encephalocele with cerebrospinal fluid leak. Otol Neurotol. 2012;33: 1621-1624. 22. Pappas DG, Hoffman RA, Holliday RA, Hammerschlag PE, Pappas DG, Swaid SN. Evaluation and management of spontaneous temporal bone cerebrospinal fluid leaks. Skull Base. 1995;5:1-7. 23. Nahas Z, Tatlipinar A, Limb CJ, Francis HW. Spontaneous meningoencephalocele of the temporal bone: clinical spectrum and presentation. Arch Otolaryngol Head Neck Surg. 2008; 134:509-518. 24. Dutt SN, Mirza S, Irving RM. Middle cranial fossa approach for the repair of spontaneous cerebrospinal fluid otorrhoea using autologous bone pate. Clin Otolaryngol Allied Sci. 2001; 26:117-123. 25. Sanna M, Fois P, Russo A, Falcioni M. Management of meningoencephalic herniation of the temporal bone: personal experience and literature review. Laryngoscope. 2009;119: 1579-1585.
26. Gubbels SP, Selden NR, Delashaw JB, McMenomey SO. Spontaneous middle fossa encephalocele and cerebrospinal fluid leakage: diagnosis and management. Otol Neurotol. 2007;28:1131-1139. 27. Pelosi S, Bederson JB, Smouha EE. Cerebrospinal fluid leaks of temporal bone origin: selection of surgical approach. Skull Base. 2010;20:253-259. 28. Oliaei S, Mahboubi H, Djalilian HR. Transmastoid approach to temporal bone cerebrospinal fluid leaks. Am J Otolaryngol. 2012;33:556-561. 29. Patel NS, Canopy E, Sheykholeslami K. Trans-mastoid management of temporal bone tegmen defects, encephaloceles and CSF leaks. J Otol Rhinol. 2012;2:1-5. 30. Savva A, Taylor MJ, Beatty CW. Management of cerebrospinal fluid leaks involving the temporal bone: report on 92 patients. Laryngoscope. 2003;113:50-56. 31. Semaan M, Gilpin D, Hsu D, Wasman J, Megerian C. Transmastoid extradural-intracranial approach for repair of transtemporal meningoencephalocele: a review of 31 consecutive cases. Laryngoscope. 2011;121:1765-1772. 32. Patel RB, Kwartler JA, Hodosh RM, Baredes S. Spontaneous cerebrospinal fluid leakage and middle ear encephalocele in seven patients. Ear Nose Throat J. 2000;79:372-373, 376-378. 33. Lundy LB, Graham MD, Kartush JM, LaRouere MJ. Temporal bone encephalocele and cerebrospinal fluid leaks. Am J Otol. 1996;17:461-469. 34. Kemink JL, Graham MD, Kartush JM. Spontaneous encephalocele of the temporal bone. Arch Otolaryngol Head Neck Surg. 1986;112:558-561. 35. Gurgel RK, Jackler RK, Dobie RA, Popelka GR. A new standardized format for reporting hearing outcome in clinical trials. Otolaryngol Head Neck Surg. 2012;147:803-807. 36. Jackson C, Pappas DG, Manolidis S, et al. Brain herniation into the middle ear and mastoid: concepts in diagnosis and surgical management. Am J Otol. 1997;18:198-205. 37. Stenzel M, Preuss S, Orloff L, Jecker P, Mann W. Cerebrospinal fluid leaks of temporal bone origin: etiology and management. ORL J Otorhinolaryngol Relat Spec. 2005;67:51-55.
Downloaded from oto.sagepub.com at University of Leeds on May 18, 2015
