Transcanal endoscopic carbon dioxide laser resection of early-stage (A1–B1) glomus tympanicum tumours: single-centre case series

Abstract

Objective

To report a single-centre experience in the endoscopic carbon dioxide laser-assisted approach to glomus tympanicum tumours.

Methods

A retrospective case review was conducted of patients diagnosed with class A1 to B1 glomus tympanicum tumours who underwent exclusive transcanal endoscopic carbon dioxide laser surgery.

Results

Seven patients fulfilled the inclusion criteria. All patients (100 per cent) were women, with a mean age of 65.4 years (standard deviation, 13.6). There were five A2 tumours, one A1 tumour and one B1 tumour. One patient presented with a delayed tympanic membrane perforation needing myringoplasty on follow up. There were no substantial post-operative complications. The mean hospitalisation time was 9.5 hours (standard deviation, 9.8). The mean follow-up period was 32.7 months (standard deviation, 13.1), with all cases having resolution of pulsatile tinnitus and no tumour recurrence.

Conclusion

The study provides further evidence on the safety and efficacy of endoscopic carbon dioxide laser surgery as a minimally invasive technique for treating early-stage glomus tympanicum tumours.

Introduction

The transcanal endoscopic approach to middle-ear paragangliomas is gaining popularity. Improved visualisation enables minimally invasive surgery. Post-auricular incisions and mastoidectomies can be avoided and, thus, post-operative morbidity reduced. However, one-handed management of these bleeding lesions through a narrow access can be challenging. In addition, experience in this field is limited as these tumours are rare, and various surgical techniques have been used.^1–5 Using specifically designed materials and techniques may facilitate surgical treatment of these tumours.^1,6

The present study aimed to illustrate current management of early-stage glomus tympanicum tumours by using a carbon dioxide (CO₂) laser fibre through an exclusive transcanal endoscopic approach.

Materials and methods

A retrospective case series with a chart review of patients who underwent endoscopic CO₂ laser-assisted surgery for glomus tympanicum tumours between June 2018 and October 2021 was performed in a third-level, adult, university hospital. This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the ethics committee of the participating institution (HCB/2022/0660). Informed consent was obtained from all individual participants included in the study. The Preferred Reporting of Case Series in Surgery (‘PROCESS’) 2020 guidelines were followed.

Tumour staging was described using the modified Fisch–Mattox classification.⁷ Inclusion criteria included early-stage tumours (classes A1 to B1). Higher-stage or syndromic disease were excluded. The pre-operative staging study consisted of a computed tomography (CT) scan and magnetic resonance imaging (MRI) (Figure 1a–f). All the cases were histopathologically confirmed paragangliomas.

Figure 1. Computed tomography scans: (a) patient one, axial; (b) patient two, coronal; and (c) patient four, coronal. Superior arrow points to previous embolisation material beneath the malleus head. Magnetic resonance imaging scans: (d) patient one, axial T2-weighted; (e) patient five, coronal T2-weighted; and (f) patient two, coronal dynamic vascular imaging.

All patients were operated on by the same experienced surgeon (FL). Surgery was performed under general anaesthesia. A 0°, 3 mm diameter, 14 cm length endoscope (Karl Storz, Tuttlingen, Germany) was used in all cases. The procedures began with local infiltration of the ear canal (with Ultracain® local anaesthetic). Neurosurgical patties soaked in lidocaine (B Braun, Melsungen, Germany; 20 mg/ml, 10 ml) with adrenaline (B Braun; 1 mg/ml, 0.5 ml) were helpful in controlling intra-operative bleeding, as well as lavages to clear the surgical field.

The surgical phases were as follows. The first phase was tumour exposure. The technique involved elevation of an extense, anterosuperior-based, tympanomeatal flap. In addition, the tympanic membrane was separated from the malleus handle and umbo (180° tympanomeatal flap technique) (Figure 2a). The second phase concerned tumour reduction. This involved coagulation of the tumour surface using a CO₂ laser fibre (continuous wave 5W, 0.30 seconds/pulse) and otology handpiece (AcuPulse^™ system with a MicroLase^™ CO₂ fibre, Micro-H^™ handpiece; Lumenis, Yokneam, Israel). The third phase involved tumour dissection and excision from a posterior to anterior direction, searching when possible the neurovascular pedicle, with coagulation using the laser (Figure 2b and c). A suction dissector (Karl Storz, Tuttlingen, Germany) was very useful at this stage of the surgery. Traction dissection and tumour fragmentation helped in the final stage of tumour extraction (Figure 2d). In the fourth phase of surgery, haemostasis and vaporisation of the tumour bed residues was performed using a laser (Figure 2e). In the last two phases of the procedure, two-handed surgery (with a second surgeon providing a third hand) was performed, as shown in photographs and video (Figure 2c and d, and supplementary video, available on The Journal of Laryngology & Otology website (Appendix 1)). The procedure ended with replacement of the tympanomeatal flap. The packing consisted of an absorbable gelatine sponge. The cases were treated in accordance with a major out-patient surgery regimen.

Figure 2. Intra-operative images (patient one): (a) anterosuperior tympanomeatal flap elevated and glomus exposed; (b) arrows pointing to main vascular supply; (c) two-handed carbon dioxide (CO₂) laser coagulation with suction (with a third hand from a second surgeon); (d) two-handed excision using forceps and suction; and (e) CO₂ laser vaporisation of tumour residues. ET = Eustachian tube; MH = malleus handle; ISJ = incudo-stapedial joint; P = promontorium; T = tumour; LF = CO₂ laser fibre; S = suction; CF = ear cup forceps

The main outcome measure was disease clearance. The monitoring for recurrence included clinical examination and imaging. Secondary measures were: resolution of pulsatile tinnitus, hearing results (six months after surgery), complications, duration of surgery, and length of stay. Hearing results were evaluated in accordance with the American Academy of Otolaryngology – Head and Neck Surgery Committee on Hearing and Equilibrium guidelines.⁸

Statistical methods

All continuous data were expressed as mean (± standard deviation (SD)) values, and all non-continuous variables were expressed as percentages. Analysis was performed using SPSS statistical software, version 22.0 for Windows (SPSS, Chicago, Illinois, USA).

Results

During the study period, seven patients diagnosed with class A1 to B1 middle-ear paragangliomas were admitted for surgery. Seven patients underwent exclusive transcanal endoscopic CO₂ laser surgery.

Demographics, initial symptoms, tumour location, disease extent, surgical notes, complications and follow-up time are summarised in Table 1. All seven patients (100 per cent) were women with a mean age of 65.4 years (SD, 13.6). Four cases (57.1 per cent) were right-sided. There were five class A2 tumours, one class A1 tumour and one class B1 tumour. Interestingly, the patient with the class B1 tumour had undergone endovascular embolisation two years before surgery (Figure 1c). It was initially a class B2 tumour; it protruded through the tympanic membrane into the external auditory canal and involved the mastoid. In accordance with our series, the protympanic space was involved in five cases and the Eustachian tube was involved in two cases. The hypotympanum was affected in one case. Ossicular chain removal was not necessary in any case. All tumours were coagulated using a CO₂ laser fibre. Tumour bed residues were treated by laser vaporisation in all cases. The mean surgical duration was 186 minutes (range, 150–230 minutes). The mean tumour size was 6.1 mm (SD, 1.8).

Table 1. Clinical summary of patients

Y = year; F = female; L = left; CO₂ = carbon dioxide; TM = tympanic membrane; R = right; CHL = conductive hearing loss

There were no significant post-operative complications. Two patients had tympanic membrane perforations repaired intra-operatively (via a cartilage palisade technique) without residual perforation. One case presented with delayed tympanic membrane perforation needing myringoplasty on follow up. Cartilage reinforcement under the tympanic membrane was performed at the end of surgery in four cases (57.1 per cent). All patients had normal post-operative facial nerve function.

Patients were discharged, without incident, a few hours after the intervention. Two patients (the patient with a class B1 tumour and one patient from a rural area) were discharged after overnight observation. Mean hospitalisation time was 9.5 hours (SD, 9.8).

The mean follow-up period was 32.7 months (SD, 13.1), with all cases having resolution of pulsatile tinnitus and with no tumour recurrence. Complete audiometric data were obtained in all cases. The hearing results are summarised in Table 2. The mean pure tone average worsened by 6.2 dB (SD, 7.4) after surgery.

Table 2. Comparison of mean pre-operative and six-month post-operative hearing values

Data represent mean ± standard deviation values (in dB). AC = air conduction; BC = bone conduction; ABG = air–bone gap

Discussion

The results of the present study indicate that the CO₂ laser fibre seems to be a safe and effective coagulation tool for the transcanal approach to the studied bleeding lesions, thus enabling a brief hospitalisation time.

To our knowledge, this is the largest reported single-centre case series on middle-ear paragangliomas treated using a CO₂ laser through a transcanal endoscopic approach. In terms of completeness of tumour resection, this study's results are comparable to those previously reported by other authors, and 100 per cent of the cases could be treated through an exclusive transcanal approach.^1,2,4,5 Pre-operatively, imaging studies (CT and MRI) may be used to detect mastoid extension, dehiscences in the bony wall covering the internal carotid artery, or a high, dehiscent jugular bulb; these prevent transcanal attempts on B2–B3 tumours, which we excluded from this treatment option. Exceptionally, tumour extension could be reduced by embolisation, as seen in the class B1 tumour in the present series.⁹

Most authors use bipolar micro-forceps to reduce the dimensions of the mass and to avoid bleeding. However, in larger paragangliomas, Marchioni et al.¹ used a CO₂ laser. The CO₂ laser fibre handpieces are of a convenient size for transcanal work and enable two-handed surgery. Even though drilling of the tumour bed is recommended to eliminate any residual disease and prevent recurrence, a CO₂ laser fibre may be an alternative both for haemostasis and vaporisation of tumour bed residues toward the end of the procedure. The duration of surgery seemed to be longer in our series. Killeen et al.² and Quick et al.⁵ referred to surgery times of approximately 100 minutes. However, mean hospitalisation time was lower in the present study than Fermi and colleagues’ 1.6 days.⁴

The results of the present work reveal similarities with previous studies. We agree with other authors on complete resolution of pulsatile tinnitus after surgery.^2,5,10 As expected by the indications for this approach, the average tumour size in this study was comparable with that reported by Killeen et al.² and Quick et al.,⁵ of 6.1 mm and 6.2 mm, respectively. We concur with other authors that complications were limited to tympanic membrane perforations.^3,4 In our first case, we observed a delayed tympanic membrane perforation, which we speculate was related to the poor vascular supply in the extense tympanomeatal flap created. At the end of the procedure, the authors recommend placing cartilage pieces in an underlay fashion, to prevent this complication. In the present study, the authors observed worsened hearing, although it was clinically insubstantial (less than 10 dB). Previous studies referred to stability or even decreases in the air–bone gap.^2,3,5 The cartilage reinforcement performed in our cases may account for this difference. The tumour recurrence rate did not differ from that of a larger series.⁴

The main limitation of our study is the small number of cases. However, this is a homogeneous series, as all patients were treated using the same protocol. In addition, the lack of a control group (e.g. a microscopic transcanal approach) could be another source of bias.

• • Exclusive transcanal endoscopic management of glomus tympanicum tumours can be challenging; a carbon dioxide (CO₂) laser fibre may facilitate surgical treatment of early-stage tumours

• • Pre-operative imaging can detect mastoid extension, dehiscences in the bony wall covering the internal carotid artery, or a high, dehiscent jugular bulb, which prevent transcanal attempts

• • Regarding completeness of tumour resection, all cases could be treated through an exclusive transcanal endoscopic approach

• • Vaporisation of tumour bed residues avoided drilling of the tumour bed and improved haemostasis at the end of surgery

• • Complications were limited to tympanic membrane perforations; patients were discharged, without incident, a few hours after intervention

• • The CO₂ laser fibre seems to be a safe and effective coagulation tool for the transcanal approach to these lesions

Using a CO₂ laser fibre for transcanal excision of middle-ear paragangliomas may avoid drilling of the tumour bed and improve haemostasis at the end of surgery, thus reducing hospitalisation time and cost. This is a minimally invasive surgery that may improve patient's quality of life compared with other surgical interventions.

Conclusion

The present study provides further evidence on the efficacy and safety of a minimally invasive transcanal endoscopic CO₂ laser fibre approach for treating A1 to B1 middle-ear paragangliomas. Although this technique may increase the duration of surgery compared with other surgical interventions, there is reduced hospitalisation time.