AMER IC AN JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D NE CK M E D ICI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
Available online at www.sciencedirect.com
ScienceDirect www.elsevier.com/locate/amjoto
Paraganglioma of the facial canal: A report of 2 cases and literature review☆,☆☆ Christopher D. Frisch, MD⁎, Colin L.W. Driscoll, MD, Brian A. Neff, MD The Department of Otorhinolaryngology, Mayo Clinic Rochester, Rochester, MN
ARTI CLE I NFO
A BS TRACT
Article history:
Purpose: 1. Describe the presentation, imaging, and outcome of two cases of paraganglioma of the facial
Received 22 July 2014
canal at our institution. 2. Summarize existing literature to better understand this lesion. 3. Clarify terminology. Methods: Retrospective chart review at single tertiary academic referral center. Literature review using the PubMed electronic database. Results: There are 12 cases of histologically-proven paraganglioma of the facial canal published in the English literature. We present two additional cases that were encountered at our institution. We also include three additional cases from a separate institution that have been accepted for publication in a different journal. We found that patients most commonly present with slowly progressive facial paralysis; though paralysis can be acute. The second most common symptom was pulsatile tinnitus, which was the only symptom in one patient. Radiographically, tumor location was in the descending segment in all but one case. The mass was often centered directly over the stylomastoid foramen, mimicking a parotid mass. We found circumferential widening of the proximal fallopian canal, and a “moth-eaten” bony destruction distally on CT imaging in several patients. In patients with poor facial nerve function (HB IV-VI), complete tumor removal with nerve sacrifice followed by great auricular or sural nerve grafting was performed with acceptable facial function results. Conclusions: Paraganglioma of the facial canal remains a rare etiology of progressive facial paralysis. The lesions are found in the distal descending segment, or centered over the stylomastoid foramen 94% of the time. In patients with poor facial function, surgical removal with nerve sacrifice, followed by great auricular or sural nerve grafting, yields an acceptable functional result. Published by Elsevier Inc.
1.
Introduction
In the patient presenting with progressive facial paresis, the differential diagnosis remains fairly broad. However,
when that patient also has an imaging study that reveals a mass lesion along the course of the facial nerve, particularly the descending portion, that differential becomes more focused. Of the options, paraganglioma is the least common
☆
Conflicts of interest: The authors have no conflicts of interest to disclose. Funding: The authors required no external funds or grants for this work. Minor internal department funds were used for searching our electronic medical record. ⁎ Corresponding author at: Department of Otorhinolaryngology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905. Tel.: + 1 937 308 9705; fax: +1 507 284 8855. E-mail addresses: [email protected] (C.D. Frisch), [email protected] (C.L.W. Driscoll), [email protected] (B.A. Neff). ☆☆
http://dx.doi.org/10.1016/j.amjoto.2014.08.006 0196-0709/Published by Elsevier Inc.
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
2
AMER ICA N JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D N E CK M EDI CI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
and the most newly recognized. There appears to be a lack of consensus on what these lesions should be defined as. Indeed, to date there are only 12 reports for a total of 14 patients published in the English literature. Analysis of this body of work is made more difficult when three different names (paraganglioma of the facial nerve, paraganglioma of the facial canal, and glomus faciale) are used to describe the lesion. Herein, we add our experience with two patients, and summarize the presentation and imaging of existing reports as well as those of three patients in a separate pending publication from another institution. In doing so, it is our hope to better define this relatively unrecognized pathology, discuss its presentation, exam findings, and imaging characteristics. We also hope to move toward a unified diagnostic term for this lesion. Armed with increased knowledge of this tumor's presentation, perhaps clinicians may change their pre-operative discussions and operative plan in patients presenting with this lesion.
2.
Materials and methods
After IRB approval, the electronic medical record was searched to identify patients with paraganglioma involving the facial nerve or facial canal. Cases of extension from glomus jugulare or tympanicum tumors were excluded. Permission for the use of photography in our two cases was obtained prior to proceeding to the operating room. Existing reports were identified through a PubMed search. All references within existing published reports were crosschecked in order to identify any missing cases. Data accepted for publication but not yet in print on three patients from one other center were shared electronically after de-identification.
3.
Case 1
Our first case is a 72 year-old lady with a history of prior thyroidectomy for papillary carcinoma 5 years previous, and a remote history of right lip basal cell carcinoma and left leg melanoma removal. She began having progressive left facial weakness dating four-months previously. Approximately one month after onset of weakness, she then began having left hemifacial spasms. She was initially evaluated locally, and after multiple imaging and laboratory studies she was ultimately given a diagnosis of Bell's palsy. She first presented to our department of neurology for multiple other, unrelated neurologic complaints. Careful evaluation of her images revealed enhancement along the distal descending segment of the facial nerve. Our evaluation elicited no further symptoms or findings. Her progressive weakness had stabilized to HB III/VI over the several weeks prior to our evaluation. There was evidence of mid and lower-facial spasm on exam. There were no abnormalities on otoscopy, or palpation of the parotid and neck. Her MRI showed a moderate enhancement along the distal descending segment of the facial
nerve near the stylomastoid foramen, but no discrete mass was appreciated (Fig. 1). CT imaging revealed circumferential enlargement of the distal fallopian canal without further bony erosion. During mastoidectomy and facial nerve exploration, a vascular mass was found situated on the lower half of the descending facial nerve extending to the stylomastoid foramen. Initial frozen pathology demonstrated dense lymphocytic infiltrates from two different biopsy sites. The case was terminated in anticipation of final pathology, which revealed paraganglioma. Postoperatively, the patient has not demonstrated radiographic tumor growth and her facial function remains stable at HB grade III.
4.
F1
Case 2
Patient 2 is a 45 year-old female who presented for a second opinion regarding the onset of left facial paresis progressing to total paralysis over the course of four days. She was initially seen at an outside local institution, where a diagnosis of Bell's palsy was made. Following this, a course of oral prednisone and valacyclovir was instituted. With her condition unchanged, a gadolinium-enhanced head MRI was obtained. The outside physicians were not able to identify a cause of the paralysis at that time. She presented to our institution two months later for followup surveillance of her history of familial paraganglioma syndrome with an SDHB mutation. At that visit it was felt she would benefit from review in our department for ongoing complete facial paralysis. Her physical exam was significant for HB VI facial paralysis. There were no abnormalities on otoscopy, or palpation of the parotid and neck. Review of her outside MRI revealed a mass centered at the stylomastoid foramen. The mass was isointense to muscle on both T1and T2-weighted imaging and mildly enhanced with contrast (Fig. 2). On CT scan, the mass appeared to expand the stylomastoid foramen without further bone erosion. The patient underwent a combined trans-mastoid/transcervical approach for mass removal. A vascular, fleshy appearing mass was found centered over the stylomastoid foramen, and extending slightly proximally along the descending segment of the facial nerve (Fig. 3). A biopsy was sent, revealing paraganglioma. After unsuccessful attempts at removing the tumor without damaging the nerve, the tumor was removed en-bloc with a segment of the nerve. A great auricular nerve graft was utilized for facial nerve reconstruction (Fig. 3). At her latest follow-up (approximately 2 months from surgery), as expected, she has not yet recovered facial function.
5.
Results
5.1.
Demographics
A total of 19 cases, including 2 from this report, of paraganglioma within the facial canal were found within 14 existing reports. After careful review, 17 of these tumors were confirmed histologically, with 2 cases being presumptive based on symptomatology and imaging features. These two reports
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
F2
F3
AMER IC AN JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D NE CK M E D ICI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
3
Fig. 1 – (A) Axial T1 pre-contrast showing abnormality along distal descending segment (open arrow). (B) Axial T1 post contrast showing mild enhancement at the same level. (C) Coronal T1 post contrast corresponds to abnormality on axial view. (D) Coronal view on CT showing circumferential widening of the distal canal without erosion.
were excluded from analysis. Of the 17, 13 (76%) were female, 4 male, and 9 cases were right ear. Average age at diagnosis was 49.6 years (range 20–74).
5.2.
Symptoms and physical exam
Twelve patients (70%) had facial paresis or full paralysis, 3 of which were acute (1 acute of which occurred repeatedly at the onset of menses), and 9 which were progressive. House– Brackmann (HB) grading was documented in 14 cases, with an average grade of 3.9 (range 1–6). The second most-common symptom was pulsatile tinnitus, occurring in 6 patients (35%). Four patients (24%) experienced facial twitching, 4 with hearing loss, 3 with otalgia, and 2 with fullness. On exam, 8 of 17 patients (47%) had normal otoscopy. Of the 11 with otoscopic findings, 5 had a mass in the external auditory canal (EAC), 5 had a retrotympanic mass, and 1 had a mass visible in both areas. All but one mass was described as red or vascular. Three of the masses appeared visibly pulsatile. Two patients had a palpable parotid mass (12%).
5.3.
Imaging
Of 17 patients, 15 (88%) had a CT scan, 10 (59%) had an MRI, and 7 (41%) had angiography (3 with embolization concurrently). Of CT findings, a widened descending facial nerve canal was found in
66%, and 50% of these displayed a “moth-eaten” erosion pattern. One patient had extensive erosion of bone of the mastoid cortex, EAC, sigmoid sinus and tegmen. One patient had widening and erosion of bone at the geniculate. MRI findings were variable, with some reporting specifics of T1- and/or T2-weighted signal characteristics, with others just reporting enhancement. T1- and T2-signals varied from iso-intermediate intensities. All were described as enhancing with contrast, and two displayed a “salt and pepper” pattern of enhancement. Of the 7 patients with angiography, 3 described vascular supply, including ascending pharyngeal, occipital or post-auricular branches.
5.4. Treatment, tumor findings at time of surgery, and follow-up Of 17 patients, 16 (94%) underwent surgery. One patient received radiotherapy due to a description of unresectability. Of the 16 patients who had surgery, 13 (81%) had full tumor resection (3 of which had biopsy followed by surgical resection at a later date), 2 (13%) had tumor debulking and descending facial nerve decompression, and 1 (6%) had biopsy only. In those patients undergoing surgery, 14 (88%) had standard mastoidectomy approaches. One patient (case involving geniculate ganglion) required combined middle-cranial fossa/mastoidectomy approach, and 1 patient (case with mastoid/EAC/sigmoid/tegmen erosion) had a post-auricular, infratemporal fossa approach.
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
4
AMER ICA N JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D N E CK M EDI CI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
Fig. 2 – (A) Axial CT with contrast shows 1 cm mass situated at the stylomastoid foramen, separate from the jugular foramen (open arrow). (B) Coronal CT at the same level. There is no evidence for bony erosion on either view. (C) Axial T1 post-contrast at same level as CT in A. (D) Coronal T1 post-contrast at the same level.
Of those who underwent tumor resection, 11 (85%) had enbloc resection with nerve sacrifice, and 2 (15%) had resection with total nerve preservation. Of the 11 with nerve sacrifice, a great auricular nerve graft was used in 7 (63%) patients. Two cases (18%) used sural nerve grafting, 1 used a novel processed cadaveric allograft, and 1 had a static sling. In all surgical cases, tumor was seen to involve the descending segment to some degree, with 4 (25%) having a mass centered at the stylomastoid foramen. The case
involving a mass centered at the geniculate had tumor tracking inferior to the second genu. Some degree of nerve swelling was seen in all cases. Of 17 patients, 12 had some degree of follow-up documented. Ten patients have documented follow-up past 6 months. Of those 10 patients, 7 had some degree of facial weakness documented on initial presentation. Five improved post-operatively. Of those with both pre- and post-op HB grade documented (4 patients), average HB grade following
Fig. 3 – (A) Intra-operative photo showing a fleshy, hypervascular mass in the distal descending segment near the stylomastoid foramen. A small diamond burr is shown for size comparison. (B) Post-resection view with arrowheads corresponding to anastomotic sites. Pes = pes anserinus. FR = facial recess. A = antrum. Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
AMER IC AN JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D NE CK M E D ICI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
intervention was 3.25. Two patients had stable post-operative weakness. No patient had worse facial function. One patient with pulsatile tinnitus as their main complaint experienced resolution. There was only 1 patient with a complication, which included a hematoma and skin flap necrosis requiring drainage and debridement.
6.
Discussion
Paraganglioma is the second most common tumor of the temporal bone, behind only schwannoma [1]. However, paraganglioma within the facial canal is exceedingly rare. Larger glomus jugulare and tympanicum tumors can and have been documented to cause facial paralysis [2,3]. To date, there are 10 published reports for a total of 12 cases of histologicallyproven paraganglioma within the facial canal [4–13]. Two additional reports of suspected cases exist, but these are based on patient presentation and imaging features, without tissue confirmation. They were excluded from our analysis [14,15]. The two cases in this report and the three additional in press from a separate institution total 17 cases [16]. There is variability in the presenting symptoms, physical exam and imaging findings. Most commonly, these patients present with progressive facial weakness, but acute-onset facial paralysis has occurred as well. The most peculiar case involved repeated episodes of acute-onset paralysis at the start of menarche, in a lesion whose epicenter was the geniculate ganglion. It must be re-iterated that no degree of facial weakness was observed in 35% of patients. The differential diagnosis for a descending segment facial nerve tumor should include schwannoma, hemangioma, extension from adjacent glomus tumors of the middle ear cleft or jugular foramen, tumor invasion from a primary parotid tumor, and cholesteatoma. While there is no one pathognomonic radiographic finding, imaging of these lesions remains the most useful tool in narrowing the differential using a combination of CT and MRI findings. CT most commonly shows circumferential widening of the facial canal, and may or may not have an irregular, “moth-eaten” bony destruction 50% of the time. This tends to be seen if the widening is just proximal to the stylomastoid foramen. There is occasionally tumor extension through the posterior wall of the EAC or into the middle ear cleft. MRI appearance is variable, though these lesions tend to be iso-intense to muscle on both T1- and T2-weighted imaging. All tumors enhanced with gadolinium. The classic “salt-andpepper” pattern described in cervical or jugular foramen paragangliomas was seen in only two patients [17]. This appearance is due to internal vascular flow voids which are more commonly seen in tumors greater than 2 cm. Most paragangliomas of the facial canal are much smaller than 2 cm, helping to explain a lack of flow voids in most of this series' tumors. As only a handful of authors described use of angiography, and even fewer of those with embolization, its use is not advocated on a routine basis. The majority of resections were performed uneventfully without addition of this costly procedure that carries a small risk of increased morbidity. After review of published reports, we find that appropriate surgical treatment for this lesion depends on a combination
5
of the health of the patient, findings at the time of surgical exploration, results of frozen pathology, and pre-operative facial nerve function. Skill and experience of the pathologist will play a key role in completing treatment in one operation. As seen in this review, 3 of 17 cases required multiple operations as initial pathology was inconclusive. Prior to resection, if possible, a biopsy of an exophytic portion of the tumor away from nerve fibers or the main nerve trunk should be submitted for frozen section analysis. Following a diagnosis of paraganglioma, lesion removal should not occur in a patient with good pre-operative facial nerve function (HB III or better). Removal is not recommended unless an easy plane of dissection from the nerve is achieved as this will help prevent undue morbidity or potential complication. It was our experience that the plane of dissection between the tumor and nerve may often be indistinct. In these cases, biopsy for diagnosis, followed by very conservative tumor debulking and/or descending facial nerve decompression, should suffice. The patient may then be followed clinically and radiographically. In a patient with poor pre-operative facial nerve function (HB IV or worse), resection of tumor can initially be attempted without nerve sacrifice. This was achieved in 3 of 13 patients who underwent tumor removal. If a plane of dissection is not achieved, the tumor can be successfully resected en bloc with the involved segment of facial nerve, followed by either great auricular or sural nerve grafting, depending on length needed. In the reports of patients who had pre-op facial weakness, and who had documented follow-up, these approaches were successful in improving facial function 5 of 7 times (2 of 7 remained stable). In those with pre- and post-operative HB grading documented, an average final HB class of 3.25 was achieved. In one of these cases, novel use of a processed cadaveric allograft was used for reconstruction, but unfortunately, improvement was limited to HB IV [personal communication]. The terminology surrounding glomus tumors is somewhat confusing, and unfortunately, analysis on histological sourcing of these 17 cases provides little in terms of clarity. Guild was the first to describe glomus tumors and their histology in a case involving the jugular foramen in 1941 [18]. Guild then provided the landmark temporal bone study revealing the distribution of paraganglia within the temporal bone in 1953 [19]. After review of 88 temporal bones, 5 bones were found to have clusters of paraganglia (or clusters of glomus bodies, as it were) along a branch of Arnold's nerve which ascends within the fallopian canal [8]. Special attention directed at the course of Arnold's nerve reveals that it enters the mastoid through one of two canaliculi, then crosses the distal descending facial nerve just proximal to the stylomastoid foramen. At this point, it gives off a branch which ascends within the facial canal [20]. With this anatomical description in mind, examination of the published operative reports and imaging findings would suggest that these tumors do not arise from the facial nerve, itself, but from paraganglia surrounding this branch of Arnold's nerve. Poor function of the facial nerve, as seen in many of these cases, appears to be due to facial nerve compression from tumors emanating from this branch of Arnold's nerve within the facial nerve canal. Therefore, we wonder if it is accurate to label these tumors “(primary)
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
6
AMER ICA N JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D N E CK M EDI CI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
paraganglioma of the facial nerve?” Although a minor point, we feel that paraganglioma of the facial canal is a more accurate descriptive term. Additionally, this single term would provide a consensus on reporting in the literature which would be helpful for future research and evidencebased medicine.
7.
Conclusions
Paraganglioma of the facial canal is a rare etiology of progressive facial paralysis. Patients typically present with progressive facial paralysis, though it can be acute. The lesions are found in the distal descending segment, or centered over the stylomastoid foramen 94% of the time. CT imaging shows widening of the facial canal, often with bony destruction. In patients with poor facial function, surgical removal with nerve sacrifice, followed by great auricular or sural nerve grafting, yields an acceptable functional result.
8.
Comment
We recognize that this study is limited by its retrospective examination of existing reports. Reporting of symptom nature and duration, as well as physical exam (especially documentation of House–Brackmann grading class) and radiographic findings is not standardized, but every attempt was made to align their descriptions.
Acknowledgment The authors thank Loyola University, Department of Otolaryngology for providing data ahead of publication (in press). REFERENCES
[1] Spector GJ, Sobol S. Surgery for glomus tumors at the skull base. Otolaryngol Head Neck Surg 1980;88:524–30. [2] Makek M, Franklin DJ, Zhao JC, et al. Neural infiltration of glomus temporale tumors. Am J Otol 1990;11:1–5.
[3] Leonetti JP, Anderson DE, Marzo SJ, et al. Facial paralysis associated with glomus jugulare tumors. Otol Neurotol 2007; 28:104–6. [4] Dutcher PO, Brackmann DE. Glomus tumor of the facial canal: a case report. Arch Otolaryngol Head Neck Surg 1986;112: 986–7. [5] Bartels LJ, Pennington J, Kamerer DB, et al. Primary fallopian canal glomus tumors. Otolaryngol Head Neck Surg 1990;102: 101–5. [6] Petrus LV, Lo WWM. Primary paraganglioma of the facial nerve canal. AJNR Am J Neuroradiol 1996;17:171–4. [7] Kania RE, Bouccara D, Colombani J-M, et al. Primary facial canal paraganglioma. Am J Otolaryngol 1999;20: 318–22. [8] Mafee MF, Raofi B, Kumar A, et al. Glomus faciale, glomus jugulare, glomus tympanicum, glomus vagale, carotid body tumors, and simulating lesions: role of MR imaging. Radiol Clin N Am 2000;38:1059–76. [9] Magliulo G, Parnasi E, Savastano V, et al. Multiple familial facial glomus: case report and review of the literature. Ann Otol Rhinol Laryngol 2003;112:287–92. [10] Connor SEJ, Gleeson MJ, Odell E. Extracranial glomus faciale tumour. J Laryngol Otol 2008;122:986–9. [11] Gunn S, Cosetti M, Roland Jr JT. Processed allograft: novel use in facial nerve repair after resection of a rare facial nerve paraganglioma. Laryngoscope 2010;120:206. [12] Parker NP, Huang TC. Progressive facial paralysis secondary to a rare temporal bone tumor: glomus faciale. Arch Otolaryngol Head Neck Surg 2011;137:712–5. [13] Takahashi K, Yamamoto Y, Ohshima S, et al. Primary paraganglioma in the facial nerve canal. Auris Nasus Larynx 2014;41:93–6. [14] Waldron MNH, Flood LM, Clifford K. A primary glomus tumour of 243 the facial nerve canal. J Laryngol Otol 2002;116: 156–8. [15] Wippold FJ, Neely JG, Haughey BH. Primary paraganglioma of the facial nerve canal. Otol Neurotol 2004;25:79–80. [16] Nadimi S, Leonetti JP, Marzo SJ, Anderson DA. Glomus faciale tumors: a report of three cases and literature review. ENT J 2014 [in press]. [17] Olsen WL, Dillon WP, Kelly WM, et al. MR imaging of paragangliomas. Am J Roentgenol 1987;148:201–4. [18] Guild SR. A hitherto unrecognized structure, the glomus jugularis, in man. Anat Rec 1941;79:28. [19] Guild SR. The glomus jugulare, a nonchromaffin paraganglion, in man. Ann Otol Rhinol Laryngol 1953: 1045–71. [20] Gray AA. The course and relations of Arnold's nerve (auricular branch of the vagus) in the temporal bone. Proc R Soc Med 1922;15:15–8 [Otology section].
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
Available online at www.sciencedirect.com
ScienceDirect www.elsevier.com/locate/amjoto
Paraganglioma of the facial canal: A report of 2 cases and literature review☆,☆☆ Christopher D. Frisch, MD⁎, Colin L.W. Driscoll, MD, Brian A. Neff, MD The Department of Otorhinolaryngology, Mayo Clinic Rochester, Rochester, MN
ARTI CLE I NFO
A BS TRACT
Article history:
Purpose: 1. Describe the presentation, imaging, and outcome of two cases of paraganglioma of the facial
Received 22 July 2014
canal at our institution. 2. Summarize existing literature to better understand this lesion. 3. Clarify terminology. Methods: Retrospective chart review at single tertiary academic referral center. Literature review using the PubMed electronic database. Results: There are 12 cases of histologically-proven paraganglioma of the facial canal published in the English literature. We present two additional cases that were encountered at our institution. We also include three additional cases from a separate institution that have been accepted for publication in a different journal. We found that patients most commonly present with slowly progressive facial paralysis; though paralysis can be acute. The second most common symptom was pulsatile tinnitus, which was the only symptom in one patient. Radiographically, tumor location was in the descending segment in all but one case. The mass was often centered directly over the stylomastoid foramen, mimicking a parotid mass. We found circumferential widening of the proximal fallopian canal, and a “moth-eaten” bony destruction distally on CT imaging in several patients. In patients with poor facial nerve function (HB IV-VI), complete tumor removal with nerve sacrifice followed by great auricular or sural nerve grafting was performed with acceptable facial function results. Conclusions: Paraganglioma of the facial canal remains a rare etiology of progressive facial paralysis. The lesions are found in the distal descending segment, or centered over the stylomastoid foramen 94% of the time. In patients with poor facial function, surgical removal with nerve sacrifice, followed by great auricular or sural nerve grafting, yields an acceptable functional result. Published by Elsevier Inc.
1.
Introduction
In the patient presenting with progressive facial paresis, the differential diagnosis remains fairly broad. However,
when that patient also has an imaging study that reveals a mass lesion along the course of the facial nerve, particularly the descending portion, that differential becomes more focused. Of the options, paraganglioma is the least common
☆
Conflicts of interest: The authors have no conflicts of interest to disclose. Funding: The authors required no external funds or grants for this work. Minor internal department funds were used for searching our electronic medical record. ⁎ Corresponding author at: Department of Otorhinolaryngology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905. Tel.: + 1 937 308 9705; fax: +1 507 284 8855. E-mail addresses: [email protected] (C.D. Frisch), [email protected] (C.L.W. Driscoll), [email protected] (B.A. Neff). ☆☆
http://dx.doi.org/10.1016/j.amjoto.2014.08.006 0196-0709/Published by Elsevier Inc.
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
2
AMER ICA N JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D N E CK M EDI CI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
and the most newly recognized. There appears to be a lack of consensus on what these lesions should be defined as. Indeed, to date there are only 12 reports for a total of 14 patients published in the English literature. Analysis of this body of work is made more difficult when three different names (paraganglioma of the facial nerve, paraganglioma of the facial canal, and glomus faciale) are used to describe the lesion. Herein, we add our experience with two patients, and summarize the presentation and imaging of existing reports as well as those of three patients in a separate pending publication from another institution. In doing so, it is our hope to better define this relatively unrecognized pathology, discuss its presentation, exam findings, and imaging characteristics. We also hope to move toward a unified diagnostic term for this lesion. Armed with increased knowledge of this tumor's presentation, perhaps clinicians may change their pre-operative discussions and operative plan in patients presenting with this lesion.
2.
Materials and methods
After IRB approval, the electronic medical record was searched to identify patients with paraganglioma involving the facial nerve or facial canal. Cases of extension from glomus jugulare or tympanicum tumors were excluded. Permission for the use of photography in our two cases was obtained prior to proceeding to the operating room. Existing reports were identified through a PubMed search. All references within existing published reports were crosschecked in order to identify any missing cases. Data accepted for publication but not yet in print on three patients from one other center were shared electronically after de-identification.
3.
Case 1
Our first case is a 72 year-old lady with a history of prior thyroidectomy for papillary carcinoma 5 years previous, and a remote history of right lip basal cell carcinoma and left leg melanoma removal. She began having progressive left facial weakness dating four-months previously. Approximately one month after onset of weakness, she then began having left hemifacial spasms. She was initially evaluated locally, and after multiple imaging and laboratory studies she was ultimately given a diagnosis of Bell's palsy. She first presented to our department of neurology for multiple other, unrelated neurologic complaints. Careful evaluation of her images revealed enhancement along the distal descending segment of the facial nerve. Our evaluation elicited no further symptoms or findings. Her progressive weakness had stabilized to HB III/VI over the several weeks prior to our evaluation. There was evidence of mid and lower-facial spasm on exam. There were no abnormalities on otoscopy, or palpation of the parotid and neck. Her MRI showed a moderate enhancement along the distal descending segment of the facial
nerve near the stylomastoid foramen, but no discrete mass was appreciated (Fig. 1). CT imaging revealed circumferential enlargement of the distal fallopian canal without further bony erosion. During mastoidectomy and facial nerve exploration, a vascular mass was found situated on the lower half of the descending facial nerve extending to the stylomastoid foramen. Initial frozen pathology demonstrated dense lymphocytic infiltrates from two different biopsy sites. The case was terminated in anticipation of final pathology, which revealed paraganglioma. Postoperatively, the patient has not demonstrated radiographic tumor growth and her facial function remains stable at HB grade III.
4.
F1
Case 2
Patient 2 is a 45 year-old female who presented for a second opinion regarding the onset of left facial paresis progressing to total paralysis over the course of four days. She was initially seen at an outside local institution, where a diagnosis of Bell's palsy was made. Following this, a course of oral prednisone and valacyclovir was instituted. With her condition unchanged, a gadolinium-enhanced head MRI was obtained. The outside physicians were not able to identify a cause of the paralysis at that time. She presented to our institution two months later for followup surveillance of her history of familial paraganglioma syndrome with an SDHB mutation. At that visit it was felt she would benefit from review in our department for ongoing complete facial paralysis. Her physical exam was significant for HB VI facial paralysis. There were no abnormalities on otoscopy, or palpation of the parotid and neck. Review of her outside MRI revealed a mass centered at the stylomastoid foramen. The mass was isointense to muscle on both T1and T2-weighted imaging and mildly enhanced with contrast (Fig. 2). On CT scan, the mass appeared to expand the stylomastoid foramen without further bone erosion. The patient underwent a combined trans-mastoid/transcervical approach for mass removal. A vascular, fleshy appearing mass was found centered over the stylomastoid foramen, and extending slightly proximally along the descending segment of the facial nerve (Fig. 3). A biopsy was sent, revealing paraganglioma. After unsuccessful attempts at removing the tumor without damaging the nerve, the tumor was removed en-bloc with a segment of the nerve. A great auricular nerve graft was utilized for facial nerve reconstruction (Fig. 3). At her latest follow-up (approximately 2 months from surgery), as expected, she has not yet recovered facial function.
5.
Results
5.1.
Demographics
A total of 19 cases, including 2 from this report, of paraganglioma within the facial canal were found within 14 existing reports. After careful review, 17 of these tumors were confirmed histologically, with 2 cases being presumptive based on symptomatology and imaging features. These two reports
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
F2
F3
AMER IC AN JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D NE CK M E D ICI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
3
Fig. 1 – (A) Axial T1 pre-contrast showing abnormality along distal descending segment (open arrow). (B) Axial T1 post contrast showing mild enhancement at the same level. (C) Coronal T1 post contrast corresponds to abnormality on axial view. (D) Coronal view on CT showing circumferential widening of the distal canal without erosion.
were excluded from analysis. Of the 17, 13 (76%) were female, 4 male, and 9 cases were right ear. Average age at diagnosis was 49.6 years (range 20–74).
5.2.
Symptoms and physical exam
Twelve patients (70%) had facial paresis or full paralysis, 3 of which were acute (1 acute of which occurred repeatedly at the onset of menses), and 9 which were progressive. House– Brackmann (HB) grading was documented in 14 cases, with an average grade of 3.9 (range 1–6). The second most-common symptom was pulsatile tinnitus, occurring in 6 patients (35%). Four patients (24%) experienced facial twitching, 4 with hearing loss, 3 with otalgia, and 2 with fullness. On exam, 8 of 17 patients (47%) had normal otoscopy. Of the 11 with otoscopic findings, 5 had a mass in the external auditory canal (EAC), 5 had a retrotympanic mass, and 1 had a mass visible in both areas. All but one mass was described as red or vascular. Three of the masses appeared visibly pulsatile. Two patients had a palpable parotid mass (12%).
5.3.
Imaging
Of 17 patients, 15 (88%) had a CT scan, 10 (59%) had an MRI, and 7 (41%) had angiography (3 with embolization concurrently). Of CT findings, a widened descending facial nerve canal was found in
66%, and 50% of these displayed a “moth-eaten” erosion pattern. One patient had extensive erosion of bone of the mastoid cortex, EAC, sigmoid sinus and tegmen. One patient had widening and erosion of bone at the geniculate. MRI findings were variable, with some reporting specifics of T1- and/or T2-weighted signal characteristics, with others just reporting enhancement. T1- and T2-signals varied from iso-intermediate intensities. All were described as enhancing with contrast, and two displayed a “salt and pepper” pattern of enhancement. Of the 7 patients with angiography, 3 described vascular supply, including ascending pharyngeal, occipital or post-auricular branches.
5.4. Treatment, tumor findings at time of surgery, and follow-up Of 17 patients, 16 (94%) underwent surgery. One patient received radiotherapy due to a description of unresectability. Of the 16 patients who had surgery, 13 (81%) had full tumor resection (3 of which had biopsy followed by surgical resection at a later date), 2 (13%) had tumor debulking and descending facial nerve decompression, and 1 (6%) had biopsy only. In those patients undergoing surgery, 14 (88%) had standard mastoidectomy approaches. One patient (case involving geniculate ganglion) required combined middle-cranial fossa/mastoidectomy approach, and 1 patient (case with mastoid/EAC/sigmoid/tegmen erosion) had a post-auricular, infratemporal fossa approach.
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
4
AMER ICA N JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D N E CK M EDI CI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
Fig. 2 – (A) Axial CT with contrast shows 1 cm mass situated at the stylomastoid foramen, separate from the jugular foramen (open arrow). (B) Coronal CT at the same level. There is no evidence for bony erosion on either view. (C) Axial T1 post-contrast at same level as CT in A. (D) Coronal T1 post-contrast at the same level.
Of those who underwent tumor resection, 11 (85%) had enbloc resection with nerve sacrifice, and 2 (15%) had resection with total nerve preservation. Of the 11 with nerve sacrifice, a great auricular nerve graft was used in 7 (63%) patients. Two cases (18%) used sural nerve grafting, 1 used a novel processed cadaveric allograft, and 1 had a static sling. In all surgical cases, tumor was seen to involve the descending segment to some degree, with 4 (25%) having a mass centered at the stylomastoid foramen. The case
involving a mass centered at the geniculate had tumor tracking inferior to the second genu. Some degree of nerve swelling was seen in all cases. Of 17 patients, 12 had some degree of follow-up documented. Ten patients have documented follow-up past 6 months. Of those 10 patients, 7 had some degree of facial weakness documented on initial presentation. Five improved post-operatively. Of those with both pre- and post-op HB grade documented (4 patients), average HB grade following
Fig. 3 – (A) Intra-operative photo showing a fleshy, hypervascular mass in the distal descending segment near the stylomastoid foramen. A small diamond burr is shown for size comparison. (B) Post-resection view with arrowheads corresponding to anastomotic sites. Pes = pes anserinus. FR = facial recess. A = antrum. Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
AMER IC AN JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D NE CK M E D ICI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
intervention was 3.25. Two patients had stable post-operative weakness. No patient had worse facial function. One patient with pulsatile tinnitus as their main complaint experienced resolution. There was only 1 patient with a complication, which included a hematoma and skin flap necrosis requiring drainage and debridement.
6.
Discussion
Paraganglioma is the second most common tumor of the temporal bone, behind only schwannoma [1]. However, paraganglioma within the facial canal is exceedingly rare. Larger glomus jugulare and tympanicum tumors can and have been documented to cause facial paralysis [2,3]. To date, there are 10 published reports for a total of 12 cases of histologicallyproven paraganglioma within the facial canal [4–13]. Two additional reports of suspected cases exist, but these are based on patient presentation and imaging features, without tissue confirmation. They were excluded from our analysis [14,15]. The two cases in this report and the three additional in press from a separate institution total 17 cases [16]. There is variability in the presenting symptoms, physical exam and imaging findings. Most commonly, these patients present with progressive facial weakness, but acute-onset facial paralysis has occurred as well. The most peculiar case involved repeated episodes of acute-onset paralysis at the start of menarche, in a lesion whose epicenter was the geniculate ganglion. It must be re-iterated that no degree of facial weakness was observed in 35% of patients. The differential diagnosis for a descending segment facial nerve tumor should include schwannoma, hemangioma, extension from adjacent glomus tumors of the middle ear cleft or jugular foramen, tumor invasion from a primary parotid tumor, and cholesteatoma. While there is no one pathognomonic radiographic finding, imaging of these lesions remains the most useful tool in narrowing the differential using a combination of CT and MRI findings. CT most commonly shows circumferential widening of the facial canal, and may or may not have an irregular, “moth-eaten” bony destruction 50% of the time. This tends to be seen if the widening is just proximal to the stylomastoid foramen. There is occasionally tumor extension through the posterior wall of the EAC or into the middle ear cleft. MRI appearance is variable, though these lesions tend to be iso-intense to muscle on both T1- and T2-weighted imaging. All tumors enhanced with gadolinium. The classic “salt-andpepper” pattern described in cervical or jugular foramen paragangliomas was seen in only two patients [17]. This appearance is due to internal vascular flow voids which are more commonly seen in tumors greater than 2 cm. Most paragangliomas of the facial canal are much smaller than 2 cm, helping to explain a lack of flow voids in most of this series' tumors. As only a handful of authors described use of angiography, and even fewer of those with embolization, its use is not advocated on a routine basis. The majority of resections were performed uneventfully without addition of this costly procedure that carries a small risk of increased morbidity. After review of published reports, we find that appropriate surgical treatment for this lesion depends on a combination
5
of the health of the patient, findings at the time of surgical exploration, results of frozen pathology, and pre-operative facial nerve function. Skill and experience of the pathologist will play a key role in completing treatment in one operation. As seen in this review, 3 of 17 cases required multiple operations as initial pathology was inconclusive. Prior to resection, if possible, a biopsy of an exophytic portion of the tumor away from nerve fibers or the main nerve trunk should be submitted for frozen section analysis. Following a diagnosis of paraganglioma, lesion removal should not occur in a patient with good pre-operative facial nerve function (HB III or better). Removal is not recommended unless an easy plane of dissection from the nerve is achieved as this will help prevent undue morbidity or potential complication. It was our experience that the plane of dissection between the tumor and nerve may often be indistinct. In these cases, biopsy for diagnosis, followed by very conservative tumor debulking and/or descending facial nerve decompression, should suffice. The patient may then be followed clinically and radiographically. In a patient with poor pre-operative facial nerve function (HB IV or worse), resection of tumor can initially be attempted without nerve sacrifice. This was achieved in 3 of 13 patients who underwent tumor removal. If a plane of dissection is not achieved, the tumor can be successfully resected en bloc with the involved segment of facial nerve, followed by either great auricular or sural nerve grafting, depending on length needed. In the reports of patients who had pre-op facial weakness, and who had documented follow-up, these approaches were successful in improving facial function 5 of 7 times (2 of 7 remained stable). In those with pre- and post-operative HB grading documented, an average final HB class of 3.25 was achieved. In one of these cases, novel use of a processed cadaveric allograft was used for reconstruction, but unfortunately, improvement was limited to HB IV [personal communication]. The terminology surrounding glomus tumors is somewhat confusing, and unfortunately, analysis on histological sourcing of these 17 cases provides little in terms of clarity. Guild was the first to describe glomus tumors and their histology in a case involving the jugular foramen in 1941 [18]. Guild then provided the landmark temporal bone study revealing the distribution of paraganglia within the temporal bone in 1953 [19]. After review of 88 temporal bones, 5 bones were found to have clusters of paraganglia (or clusters of glomus bodies, as it were) along a branch of Arnold's nerve which ascends within the fallopian canal [8]. Special attention directed at the course of Arnold's nerve reveals that it enters the mastoid through one of two canaliculi, then crosses the distal descending facial nerve just proximal to the stylomastoid foramen. At this point, it gives off a branch which ascends within the facial canal [20]. With this anatomical description in mind, examination of the published operative reports and imaging findings would suggest that these tumors do not arise from the facial nerve, itself, but from paraganglia surrounding this branch of Arnold's nerve. Poor function of the facial nerve, as seen in many of these cases, appears to be due to facial nerve compression from tumors emanating from this branch of Arnold's nerve within the facial nerve canal. Therefore, we wonder if it is accurate to label these tumors “(primary)
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
6
AMER ICA N JOURNAL OF OT OLA RYNGOLOGY– H E A D A N D N E CK M EDI CI N E AN D S U RGE RY X X (2 0 1 4) XXX – XXX
paraganglioma of the facial nerve?” Although a minor point, we feel that paraganglioma of the facial canal is a more accurate descriptive term. Additionally, this single term would provide a consensus on reporting in the literature which would be helpful for future research and evidencebased medicine.
7.
Conclusions
Paraganglioma of the facial canal is a rare etiology of progressive facial paralysis. Patients typically present with progressive facial paralysis, though it can be acute. The lesions are found in the distal descending segment, or centered over the stylomastoid foramen 94% of the time. CT imaging shows widening of the facial canal, often with bony destruction. In patients with poor facial function, surgical removal with nerve sacrifice, followed by great auricular or sural nerve grafting, yields an acceptable functional result.
8.
Comment
We recognize that this study is limited by its retrospective examination of existing reports. Reporting of symptom nature and duration, as well as physical exam (especially documentation of House–Brackmann grading class) and radiographic findings is not standardized, but every attempt was made to align their descriptions.
Acknowledgment The authors thank Loyola University, Department of Otolaryngology for providing data ahead of publication (in press). REFERENCES
[1] Spector GJ, Sobol S. Surgery for glomus tumors at the skull base. Otolaryngol Head Neck Surg 1980;88:524–30. [2] Makek M, Franklin DJ, Zhao JC, et al. Neural infiltration of glomus temporale tumors. Am J Otol 1990;11:1–5.
[3] Leonetti JP, Anderson DE, Marzo SJ, et al. Facial paralysis associated with glomus jugulare tumors. Otol Neurotol 2007; 28:104–6. [4] Dutcher PO, Brackmann DE. Glomus tumor of the facial canal: a case report. Arch Otolaryngol Head Neck Surg 1986;112: 986–7. [5] Bartels LJ, Pennington J, Kamerer DB, et al. Primary fallopian canal glomus tumors. Otolaryngol Head Neck Surg 1990;102: 101–5. [6] Petrus LV, Lo WWM. Primary paraganglioma of the facial nerve canal. AJNR Am J Neuroradiol 1996;17:171–4. [7] Kania RE, Bouccara D, Colombani J-M, et al. Primary facial canal paraganglioma. Am J Otolaryngol 1999;20: 318–22. [8] Mafee MF, Raofi B, Kumar A, et al. Glomus faciale, glomus jugulare, glomus tympanicum, glomus vagale, carotid body tumors, and simulating lesions: role of MR imaging. Radiol Clin N Am 2000;38:1059–76. [9] Magliulo G, Parnasi E, Savastano V, et al. Multiple familial facial glomus: case report and review of the literature. Ann Otol Rhinol Laryngol 2003;112:287–92. [10] Connor SEJ, Gleeson MJ, Odell E. Extracranial glomus faciale tumour. J Laryngol Otol 2008;122:986–9. [11] Gunn S, Cosetti M, Roland Jr JT. Processed allograft: novel use in facial nerve repair after resection of a rare facial nerve paraganglioma. Laryngoscope 2010;120:206. [12] Parker NP, Huang TC. Progressive facial paralysis secondary to a rare temporal bone tumor: glomus faciale. Arch Otolaryngol Head Neck Surg 2011;137:712–5. [13] Takahashi K, Yamamoto Y, Ohshima S, et al. Primary paraganglioma in the facial nerve canal. Auris Nasus Larynx 2014;41:93–6. [14] Waldron MNH, Flood LM, Clifford K. A primary glomus tumour of 243 the facial nerve canal. J Laryngol Otol 2002;116: 156–8. [15] Wippold FJ, Neely JG, Haughey BH. Primary paraganglioma of the facial nerve canal. Otol Neurotol 2004;25:79–80. [16] Nadimi S, Leonetti JP, Marzo SJ, Anderson DA. Glomus faciale tumors: a report of three cases and literature review. ENT J 2014 [in press]. [17] Olsen WL, Dillon WP, Kelly WM, et al. MR imaging of paragangliomas. Am J Roentgenol 1987;148:201–4. [18] Guild SR. A hitherto unrecognized structure, the glomus jugularis, in man. Anat Rec 1941;79:28. [19] Guild SR. The glomus jugulare, a nonchromaffin paraganglion, in man. Ann Otol Rhinol Laryngol 1953: 1045–71. [20] Gray AA. The course and relations of Arnold's nerve (auricular branch of the vagus) in the temporal bone. Proc R Soc Med 1922;15:15–8 [Otology section].
Please cite this article as: Frisch CD, et al, Paraganglioma of the facial canal: A report of 2 cases and literature review, Am J Otolaryngol–Head and Neck Med and Surg (2014), http://dx.doi.org/10.1016/j.amjoto.2014.08.006
![[Retroperitoneal paraganglioma: report of a case and literature review].](/assets/img/hold.png)
