TITLE: GLOMUS TUMORS
SOURCE: Dept. of Otolaryngology, UTMB, Grand Rounds
DATE: January 11, 1995
RESIDENT PHYSICIAN: Michael Bryan, MD
FACULTY: Jeffrey T. Vrabec, MD
SERIES EDITOR: Francis B. Quinn, Jr., M.D.

|Return to Grand Rounds Index|

"This material was prepared by resident physicians in partial fulfillment of educational requirements established for the Postgraduate Training Program of the UTMB Department of Otolaryngology/Head and Neck Surgery and was not intended for clinical use in its present form. It was prepared for the purpose of stimulating group discussion in a conference setting. No warranties, either express or implied, are made with respect to its accuracy, completeness, or timeliness. The material does not necessarily reflect the current or past opinions of members of the UTMB faculty and should not be used for purposes of diagnosis or treatment without consulting appropriate literature sources and informed professional opinion."

Introduction

Glomus tumors are generally benign neoplasms of arising from neuroectodermal tissues, found in various parts of the body. They consist of clusters of catecholamine containing chief cells surrounded by sustentacular cells (modified Schwann cells) intimately surrounded and interlaced with a rich network of capillaries and venules. These clusters of cells are called "zellballen", and can generally be separated into "light" and "dark" cell subpopulations, referring to the density of intracellular neurosecretory granules. Chief cells are members of the amine precursor and uptake decarboxylase (APUD) family, recently referred to as the DNES (diffuse neuroendocrine system). They are related to the adrenal medulla, and their neoplastic counterpart in the adrenal gland is the pheochromocytoma. The extra-adrenal versions have various names including non-chromaffin paragangliomas, paragangliomas, chemodectomas, glomerocytomas, receptomas, tympanic body tumors, carotid body tumors, and a few others. The term glomus was applied originally because it was erroneously thought that the chief cells arose from specialized pericytes as seen in true arteriovenous complexes ("glomus complexes" ). Despite more appropriate descriptive terms, they will undoubtedly continue to be called glomus tumors in most instances.

Chief cells have the capability of synthesizing catecholamines. Infrequently paragangliomas secrete catecholamines in significant amounts and can produce symptoms (HTN, arrhythmias, excessive perspiration, headaches, nausea, pallor, etc.). The predominant catecholamine product is norepinephrine because of a lack of the enzyme (phenylethanolamine-N-methyltransferase) needed to convert norepinephrine to epinephrine. Elevated levels of epinephrine should alert to the possibility of a concurrent pheochromocytoma.

Synchronous glomus tumors are not the only associated neoplasms. Several authors have reported an association with other neoplasms (benign and malignant). Spector et. al. reported a 7% incidence in his series of 95 patients. Those most commonly associated include thyroid C-cell carcinoma, parathyroid adenomas, pheochromocytomas, the MEN syndromes, and visceral neoplasms of neural crest origin.

In the head and neck, two anatomic groups of paragangliomas can be differentiated: cervical paragangliomas and temporal bone (aka jugulotympanic) paragangliomas. The cervical group includes primarily carotid body tumors and glomus vagale or intravagale tumors, while the jugulotympanic refers to glomus jugulare and glomus tympanicum tumors. Jugulotympanic glomus tumors are the second most common temporal bone tumors (after acoustic neurinomas). The remainder of this discussion will focus on the temporal bone lesions.

Historical Perspective

In 1941, Guild first described "glomic tissue" in the temporal bone. He described a vascular tissue in the dome of the jugular bulb and on the promontory of the middle ear. The correlation between this tissue and what we now call glomus tumors was not made until 1945, by Rosenwasser, who reported a "carotid body tumor" of the middle ear and mastoid.

Once these tumors were recognized and reports of their occurrence increased, attempts were made to classify them. In the 1960's, Guilford and Alford coined the term glomus tympanicum to describe those paragangliomas limited to the middle ear. Their classification system was limited by the diagnostic armamentarium available at the time of its derivation and was based on symptomology. As diagnostic imaging and surgical techniques advanced, more extensive classification systems were devised. In 1969, McCabe and Fletcher offered an updated classification methodology. This was followed in 1978, by Fisch's classification system, and subsequently, in 1981, by Glasscock and Jackson's proposed scheme. The latter two are used currently.

In the 1940's and 1950's, surgical approaches to the temporal bone were limited and recurrence after resection of glomus tumors was common, as were injuries to the facial nerve. Radiation therapy for glomus tumors became popular largely for these reasons. Since the 1970's, the advances in skull base neurotologic surgery have allowed a resurgence in the surgical management of temporal bone glomus tumors. Acoustic neuroma surgery approaches paved the way for these developments. In 1974, Fisch described the infratemporal fossa approach to the lateral skull base, which probably has had the biggest impact on surgical management of these tumors. Despite surgical advances, arguments over the relative merits of radiation therapy vs surgery as treatment modalities for glomus tumors continue today.

Gross Anatomy

There are typically three glomus bodies in each ear. The glomus bodies are usually found accompanying Jacobsen's (CN IX) or Arnold's (CN X) nerve or in the adventitia of the jugular bulb. However, the physical location is usually the mucosa of the promontory(glomus tympanicums), or the jugular bulb (glomus jugulare).

The main blood supply to the glomus tumors is the ascending pharyngeal artery, usually via inferior tympanic and neuromeningeal branches. Numerous other arteries can contribute, or occasionally be found to be the primary supply, particularly when the tumor is large or has intracranial or intradural extension. Angiography shows that 85% of cases demonstrate non-communicating multiple vascular compartments with small feeding vessels.

Behavior

Although usually histologically benign (see below), glomus tumors are locally destructive, spreading along paths of least resistance. Spread is multidirectional and simultaneous. The main routes of spread (because of lack of barriers) are the air cell tracts of the temporal bone, but spread through and beyond the temporal bone is not uncommon. Bone erosion is usually evident as crescentic lucencies on CT scan. These tumors spread through the bone's Haversian systems, and also tend to invade adjacent nerves.

A key characteristic of glomus tumors is their slow growth rate. The typical patient goes undiagnosed for 3 to 6 years after the initial onset of symptoms because of slow insidious progression. Some patients have remained asymptomatic or with stable deficits for up to 40 years after detection of glomus tumors.

Epidemiology

The incidence of glomus jugulare tumors is 1:1,300,000 population.

The most striking bit of epidemiology is the predominant incidence in females. The female:male incidence ratio is at least 4:1.

The incidence of malignancy in glomus tumors is believed to be low (<5%). There is no histologic distinction between the benign and malignant versions, and malignancy can only be proven by the presence of metastasis.

Patient age averages 50 to 60 years at presentation, but this is highly variable.

Catecholamine secreting (aka "functional") tumors occur in 1% - 3% of cases.

There is no racial or ethnic predilection.

There are familial patterns of occurrence, although the genetic model of inheritance is not agreed upon. The familial form is associated with a high incidence (25% - 50%) of multicentric paragangliomas. Multicentricity is found in about 5-15% of non-familial form patients. Interestingly, tumors are not typically found above and below the clavicles simultaneously.

Clinical Presentation

Symptoms are insidious in onset. Because of the location and the vascular nature of the tumors, it is not surprising that the most common complaint of symptomatic patients is pulsatile tinnitus. Its believed that the tinnitus is secondary to mechanical impingement on the umbo is most cases. Other common symptoms are aural fullness, and (conductive) hearing loss.

On physical examination, the hallmark of a jugulotympanic glomus tumor is a red or reddish-blue mass seen behind the tympanic membrane. The diagnosis of glomus tympanicum can only be entertained if the examiner can see a full 360 degrees around the perimeter of the lesion, otherwise the presumptive diagnosis must be a glomus jugulare. The finding of a middle ear mass is fairly reliable, with 94 - 100% of untreated cases demonstrating this in reviews of large series of patients. Brown's sign (blanching of the mass with positive pressure pneumotoscopy) is often mentioned, but the frequency of this finding is not clear. Rarely, a friable or bleeding mass in the EAC may be the presenting sign with larger tumors. HTN, tachycardia, tremor, or complaints of vascular headaches should alert for the possibility of a functional tumor.

Cranial nerve deficits are seen primarily with larger tumors. Reports cite compression or invasion of CN's IX, and X most commonly, with CN's VII, VIII, XI, and XII affected less often. The presence of lower cranial nerve deficits rules out an isolated middle ear glomus tympanicum. Isolated deficits of CN's VII and VIII are more likely to be secondary to a tympanicum.

Patients who present with "idiopathic" lower cranial nerve deficits should have their lateral skull bases evaluated with CT or MRI if no cause can be found for their neuropathy.

Sensorineural hearing loss on the side of a glomus tumor is the hallmark of labyrinthine invasion.

Diagnosis and Preoperative Evaluation

The presumptive diagnosis may be made on clinical grounds but the diagnosis must be confirmed with medical imaging (CT, MRI, and sometimes angiography). A myringotomy for purposes of obtaining tissue should not be performed on a middle ear mass. If a tissue biopsy is to be obtained (not usually necessary) a transmastoid approach should be used.

Glomus tumors are much more common than almost any other middle ear neoplasm, and second only to acoustic tumors in the temporal bone overall, but other lesions should be considered. Although the list of differential diagnoses includes neural lesions (neurolemmoma, neurofibroma, chordoma), osteoblastoma, adenomas, adenocarcinomas, inflammatory polyps, cholesterol granulomas, and a host of rare lesions such as histiocytosis, fibromyxoma, melanoma, rhabdomyosarcoma, lipoma, plasmacytoma, or metastatic lesions (lung, breast, prostate), almost all of these can be excluded with the appropriate physical examination and radiological studies.

After diagnosis, the workup must include a search for other tumors (multicentricity). Screens for urine catecholamine metabolites (VMA, normetanephrine and metanephrine), or serum levels of catecholamines need only be done if there is reason to suspect a secreting tumor. It has been recommended that this be done before any angiography because of possible need for selective renal vein sampling if elevated levels of epinephrine are detected, but MRI screening of the abdomen in these cases may negate the need for this. Although there is a reported increased incidence of associated non-paraganglioma tumors, a search for them is not performed unless clinically indicated.

If surgery is planned, and its suspected that the carotid artery may be at high risk, angiography with balloon test occlusion is strongly advised, although 15-25 % of people who tolerate temporary balloon occlusion will have neurologic sequelae after ligation of the tested carotid artery. Those who have neurologic changes with test occlusion should not be operated on unless there are plans for interposition graft or other recollateralization. The predictability of ICA occlusion effects is the most controversial topic in skull base surgery. Our faculty feels that if carotid resection is likely to be required to remove the tumor, strong consideration should be given to radiation treatment.

Preoperative embolization of feeding vessels is a little controversial. Proponents argue that if embolization can be performed within 48 hours of planned surgery, the blood loss is significantly reduced. Some experts, most notably The Otology Group (Glasscock, Jackson, et. al.) disagree, claiming benefits of embolization are outweighed by the risks. In the absence of planned embolization there may be no need for angiography at all.

Imaging

CT scans and MRI are the primary modalities used in evaluating the size and extent of glomus tumors. Angiography is used for identifying and often embolizing the primary blood supply to tumor.

Glomus tympanicums appear on CT as a soft tissue mass abutting the promontory of the middle ear. There may be displacement of ossicles or bony erosion of the tympanic cavity. The finding of air or bone between the tumor and the jugular bulb virtually assures the diagnosis of a tympanicum and no further imaging is required.

Glomus jugulare tumors can expand in any direction, but they tend to erode the bony plate between the jugular bulb and the internal carotid artery. An intact petrous carotid canal is helpful in ruling out an aberrant carotid artery, along with intravenous contrast imaging. Contrast CT scans and/or angiography are usually adequate to detect this.

Radiographically, glomus tumors appear isodense to cerebellum on unenhanced CT scans, and enhancing masses with contrast. On post gadolinium T1 weighted MRI scans, there are hypodense areas as well as isodense areas, referred to as a"salt and pepper" appearance. The hypodense areas correspond to flow voids, or blood flow in vascular spaces. The isodense areas represent the solid portions of the tumor. On T2 scans, the flow voids remain, and the solid areas are slightly hyperintense.

CT scans are best for evaluating bony destruction and erosion, which is a hallmark of jugulotympanic glomus tumors. The CT may lead to the incorrect diagnosis of a malignancy because of the irregular edges of the bone destruction.

MRI in usually better than CT for delineating tumor edges and extent, especially intracranially. It is also better for evaluating the relationship of the tumor to adjacent jugular vein, carotid artery, membranous labyrinth, and cranial nerves.

Treatment

Glomus Tympanicum The appropriate treatment of glomus tympanicum tumors isolated to the middle ear, middle ear and mastoid, or even the inner ear (Fisch classifications A & B) is surgical excision. Glomus tympanicums can be completely excised in over 90% of cases with standard otologic approaches. Typically extended facial recess or canal wall down approaches are used. The facial nerve necessary nerve can be mobilized if necessary. Tympanicum tumors that have extended beyond the middle ear to involve the skull base can be treated in a manner similar to glomus jugulares as discussed below.

Glomus Jugulare Glomus jugulare tumors can be treated with surgical excision, radiation therapy, surgery and radiation, or observation. The argument over the relative merits of surgery vs radiation center on their complications and efficacy. With advances in skull base techniques, surgery is generally efficacious, so the decision on treatment mode often centers on the potential morbidity of the surgery, and other patient factors.

Tumor size, extent, and location dictate the surgical approach to be used as well as the potential for post operative problems. The larger the tumor the higher the morbidity, with rates of cranial nerve sacrifice directly related to tumor stage. However the larger the tumor, the more likely the lower cranial nerves will be affected even before surgery. Obviously there is little consequence to surgical sacrifice of a nonfunctioning nerves, so preoperative deficits must be weighed against the likelihood of postoperative deficit. Additionally, the patient's age and overall health influence the treatment plan. Elderly (>65 years old) patients are not ideal candidates for skull base surgery. On the other hand, the unproven efficacy of XRT over the long term and otherwise long life expectancies bias the decision toward an attempt at complete excision in younger patients. Young patients also adapt to loss of cranial nerve functions better than older patients. Most often primary radiation therapy is used in cases where the patient is a poor surgical candidate, the tumor is "unresectable" without carotid sacrifice and the patient fails balloon occlusion testing, or the patient refuses surgery.

In the event of multicentricity, careful consideration of post-operative morbidity must be given. If the skull base is involved bilaterally, there is significant risk of laryngeal and pharyngeal denervation with bilateral surgical intervention. In this circumstance the most life threatening lesion is addressed first, and the subordinate lesion(s) palliated until outcome is clear.

The presence of intracranial extension or symptomatic catecholamine excretion negate the drawbacks of surgery. These tumors must be operated on to prevent dire consequences.

Glasscock & Jackson Class I & II and Fisch C1 & C2 glomus jugulare tumors can be resected with an extended facial recess approach. The canal wall can be preserved or removed as needed for exposure. Some exposure of the area anterior to the internal carotid can be exposed with this approach.

Glasscock & Jackson Class III lesions extend medially and anteriorly to involve the petrous apex, and usually the horizontal carotid canal. Class IV lesions extend beyond the petrous apex, into the clivus or into the infratemporal fossa. These are similar to the Fisch C3 & C4 classes, Both of these are most commonly treated by the infratemporal fossa approach or a variation of it. The infratemporal fossa approach has become the predominant one used for glomus jugulare resection. Extension of the dissection into the neck allows proximal and distal control of the jugular and carotid systems, which is considered essential for success in resection of larger anteromedial tumors. The exposure is adequate for removal of tumor up to and including cavernous sinus involvement, but it does require relocation of the facial nerve. It has the reputation of a significant percentage of residual lower cranial nerve deficits. However, it allows the resection of tumors that are not accessible with more conservative approaches.

Intracranial and/or intradural extension is a subclass (CxD1-3) of the Fisch classification system. The same approaches are used, but the extracranial portion of the tumor is resected first. This may help reduce bleeding during the intracranial dissection by removing the primary blood supply. Single stage complete resection is fairly common now.

The incorporation of subtemporal exposure can be combined with the infratemporal approach. An attraction of this technique is that it allows adequate exposure of medial structures without transposition of the facial nerve. Patel et. al. recently reported on a series of patients treated with this and other combined approaches and their incidence of facial and lower cranial nerve injury compared favorably to those reported in larger series of traditional infratemporal fossa resections.

Retrosigmoid, or suboccipital approaches can be incorporated as needed for posterior fossa extension.

Incomplete resection is not desirable but may be necessary in some cases. If this occurs, postoperative radiation therapy is usually indicated.

Controversy and Complications

The treatment of glomus tumors involving the unilateral skull base is an area of ongoing controversy. Both radiation therapy and surgical excision have been used, and proponents of each can site numerous reports that seem to support their point of view. Radiotherapy advocates claim that surgical resection carries too high a price in terms of morbidity due mainly to iatrogenic cranial nerve deficits and CSF leaks. But radiotherapy is not without serious complications, such as osteoradionecrosis (ORN) of the temporal bone, brain necrosis, pituitary-hypothalamic insufficiency, and secondary malignancy. Additionally, Wilson et. al, state the obvious when they report that surgery is complicated by previous radiation therapy. Objective evidence of this is found in the review by Green, et. al., wherein they note poorer cranial nerve preservation in previously irradiated patients undergoing salvage surgery. Additionally, histologic analysis of some irradiated glomus tumors reveal primarily fibro-vascular degenerative changes, but with formation of intact islands of chief cells.

Evaluating the reports of success or failure of various treatment options has been plagued by low numbers of patients, poor long term follow up, and the inability to generate data that represent results of "current state of the art" treatment for either modality. Additionally, nonuniformity in the presentation of data, frequent lack of distinction between different glomus tumor locations, combining of previously treated (ie - treatment failures) with previously untreated cases, indistinct therapeutic endpoints, and potential patient selection bias have led to difficulty making meaningful comparisons between reported surgical results and results of primary radiation therapy. Carrasco and Rosenman, in an effort to resolve some of the controversy, reviewed and analyzed 24 commonly cited treatment series published between 1964 and 1987. 582 patients were included in these studies. They point out shortcomings in the published studies, but make several pertinent observations.

Osteoradionecrosis

With respect to XRT side effects, one of the biggest concerns is the delayed occurrence of ORN of the temporal bone. Even at low radiation doses, some reports of ORN have been made. After review of several large series, they conclude that the risk is low if multiport megavoltage treatments are used and the total dose is less than 60 Gy/6 weeks. They review the reports by Kim et. al, and Cummings et al., involving 200 cases of primary XRT, and conclude that the appropriate megavoltage dose of 35 Gy/3 weeks to 45 Gy/4 weeks is optimal and avoids ORN (this was an AOE question last year).

Radioresistance

Carrasco and Rosenman also examined the histologic studies of irradiated glomus tumors mentioned above and point out that these studies involved patients who were being treated with planned preoperative radiation, or were radiation failures. They note that in the group of patients undergoing preoperative radiation, not enough time had elapsed between the radiation and surgery to allow for substantial tumor death. Months are required after radiation treatments in order to obtain meaningful biopsy results if one is looking for eradication of disease. The time required is even longer for slower growing tumors as they express cellular damage at slower rate. In the treatment failure group, obviously tumor destruction would not be expected, as these were non-responders. Thus the most frequently cited literature used in espousing the radioresistance of glomus tumors may not really support this conclusion. What is needed is tissue from irradiated patients without recurrence, but this is not feasible.

Cranial Nerve Preservation

Carrasco & Rosenman do not address cranial nerve preservation data. Since this is probably the biggest objection to surgical intervention because of the associated morbidity of lower cranial nerve deficits, the incidence of iatrogenic cranial nerve injury is important. Two well known otologic surgery centers recently reported on probably the most current and comprehensive series of glomus tumor patients treated with surgery alone.

Green et. al., published their results of surgical treatment of 52 glomus jugulare patients between 1981, and 1991. The follow up period was relatively brief, averaging only 3.4 years. They reported that they were able to perform complete primary excision in 85% their cases, and there were no surgical mortalities. In all 8 cases of incomplete resection, the tumor had intradural spread, and 5 of these 8 cases had to be stopped secondary to excessive blood loss. They reported 83% of their cases performed via the infratemporal fossa approach. Compared to pre-operative status, post-surgical cranial nerve deficits occurred in 26%, 13%, 25%, and 6% of patients, for CN's IX, X, XI, and XII respectively.

Wilson et. al., reviewed the results of their primary surgical treatment of 71 patients with glomus jugulare tumors between 1971 and 1991. Sixty-seven of these patients underwent total resection, with 4 subtotal resections. These four underwent post operative XRT. There were two perioperative mortalities, one due to a pulmonary embolus, another due to stroke after a carotid injury intraoperatively. Forty-one percent of their patients had pre-operative lower cranial nerve deficits. Of the patients without pre-operative deficits, post- operative CN deficits were noted in 13%, 61%, 36%, 40%, and 33% with respect to CN's VII, IX, X, XI, and XII, respectively. Overall, the rates of sacrifice of CN's VII, IX, and X were 23, 63%, and 59%.

The fact that cranial nerve preservation decreases commensurate with increasing tumor size was dramatically illustrated in Wilson et. al.'s series. No patients with Type IV tumors had preservation of CN's VIII through XII. Contrarily, preservation rates in Glasscock & Jackson Type I tumors was greater than 90% for all cranial nerves except VII (83% reserved) and IX (74% preserved).

Other Complications

Other postoperative complications include CSF leaks, aspiration syndromes, meningitis, pneumonia, wound infections, and a number of other less frequent routine post-operative problems. In general the incidence of CSF leaks appears to be low. Although Glasscock et. al. reported a 14.5% incidence since 1971, improved reconstruction techniques have allowed them to reduce this to only one case in 34 operated on since 1987. Green et. al., reported a 4% incidence, with spontaneous cessation in all cases. Between a third and one half of all of the patients in the Green's and Glasscock's series had chronic dysphagia and/or hoarseness post-operatively. Many underwent vocal cord augmenting procedures, a few (4% in Green et.al.) needed cricopharyngeal myotomy, and a few required long term nasogastric or gastrostomy tubes for nutrition. Glasscock stated that none of the patients in their report was debilitated by their deficits, and Green reported that 85% of those returning questionnaires (34 of 52 patients) were able to return to preoperative activity levels.

Discussion

It is obvious that treatment plans must take into account the patient's age, the site, size, and extent of the tumor, the rate of symptom progression, pre-operative cranial nerve status, possibility of multicentricity, neurosecretory status, the experience of the surgeon, and perhaps most importantly patient preference. In most cases, unilateral lower cranial nerve loss is well compensated for and long term severe morbidity is not the rule. Tumor size and extent are probably the next most important factors, as surgical intervention with low stage tumors yields good results with relatively low morbidity. Improved diagnostic imaging allowing earlier detection should create a trend toward diagnosis of smaller lesions and better surgical results.

Other issues

Some have questioned the necessity of any treatment in patients who are not experiencing significant symptom progression. This is sensible if you consider survival as an endpoint. Van der May et. al., reviewed results after treatment of 52 glomus jugulare tumors. He concluded that treatment, or rather the lack of it, did not seem to affect survival. Carrasco and Rosenman found that in their review of 582 patients, only 6.2% died of any cause during the follow up periods of the reports studied. They also concluded that treatment method did not statistically affect survival. Other important conclusions derived from the review by Carrasco and Rosenman are that statistical differences in outcome, with respect to recurrence is not significant, with failure rates of less than 10% for each. Additionally, salvage rates for each modality are about the same, with successful salvage of 88% overall.

Finally, what constitutes successful treatment? Surgery supporters insist eradication of disease is the true measure of success, whereas radiotherapy advocates generally define successful treatment as an arrest in the progression of symptoms and tumor growth. It is difficult to prove a cure because of the impossibility of proving the absence of microscopic disease in post surgical patients, and the observation that recurrences can be extremely delayed (in some cases 20 or more years post treatment). The validity of claims of curative therapy, with what we traditionally consider to be adequate follow up, may not be certain.

More data and long term follow up from this generation of skull base surgery patients will be needed before the controversy is settled (if it ever is). Included will no doubt be additional data about radiotherapeutic measures, such as the Gamma Knife, for which there is a dearth of data in regard to glomus tumor treatment.

BIBLIOGRAPHY

1. Green JD Jr; Brackmann DE; Nguyen CD; Arriaga MA; and others: Surgical management of previously untreated glomus jugulare tumors. Laryngoscope 1994 Aug;104(8 Pt 1):917-21

2. Pulec JL; Deguine C: Glomus jugulare tumor. Ear Nose Throat J 1994 Apr;73(4)

3. Pluta RM; Ram Z; Patronas NJ; Keiser H: Long-term effects of radiation therapy for a catecholamine-producing glomus jugulare tumor. Case report. J Neurosurg 1994 Jun;80(6):1091-4

4. Patel SJ; Sekhar LN; Cass SP; Hirsch BE: Combined approaches for resection of extensive glomus jugulare tumors. A review of 12 cases. J Neurosurg 1994 Jun;80(6):1026-38

5. Karas DE; Kwartler JA: Glomus tumors: a fifty-year historical perspective. Am J Otol 1993 Sep;14(5):495-500

6. Gulya AJ: The glomus tumor and its biology. Laryngoscope 1993 Nov;103(11 Pt 2 Suppl 60):7-15

7. Woods CI; Strasnick B; Jackson CG: Surgery for glomus tumors: The Otology Group experience. Laryngoscope 1993 Nov;103(11 Pt 2 Suppl 60):65-70

8. Netterville JL; Civantos FJ: Rehabilitation of cranial nerve deficits after neurotologic skull base surgery. Laryngoscope 1993 Nov;103(11 Pt 2 Suppl 60):45-54

9. Glasscock ME 3d: The history of glomus tumors: a personal perspective. Laryngoscope 1993 Nov;103(11 Pt 2 Suppl 60):3-6

10. Jackson CG: Basic surgical principles of neurotologic skull base surgery. Laryngoscope 1993 Nov;103(11 Pt 2 Suppl 60):29-44

11. Carrasco V; Rosenman J: Radiation therapy of glomus jugulare tumors. Laryngoscope 1993 Nov;103(11 Pt 2 Suppl 60):23-7

12. Jackson CG: Neurotologic skull base surgery for glomus tumors. Diagnosis for treatment planning and treatment options. Laryngoscope 1993 Nov;103(11 Pt 2 Suppl 60):17-22

13. Anand VK; Leonetti JP; al-Mefty O: Neurovascular considerations in surgery of glomus tumors with intracranial extensions. Laryngoscope 1993 Jul;103(7):722-8

14. Lloyd GA; Barker PG; Phelps PD: Subtraction gadolinium enhanced magnetic resonance for head and neck imaging. Br J Radiol 1993 Jan;66(781):12-6

15. Lloyd GA; Cheesman AD; Phelps PD; King CM: The demonstration of glomus tumours by subtraction MRI. Neuroradiology 1992;34(6):470-4

16. Vogl TJ; Mack MG; Juergens M; Bergman C; and others: Skull base tumors: gadodiamide injection-enhanced MR imaging drop-out effect in the early enhancement pattern of paragangliomas versus different tumors. Radiology 1993 Aug;188(2):339-46

17. van der Mey AG; Frijns JH; Cornelisse CJ; Brons EN; and others: Does intervention improve the natural course of glomus tumors? A series of 108 patients seen in a 32-year period. Ann Otol Rhinol Laryngol 1992 Aug;101(8):635-42

18. Maniglia AJ; Sprecher RC; Megerian CA; Lanzieri C: Inferior mastoidectomy-hypotympanicapproach for surgical removal of glomus jugulare tumors: an anatomical and radiologic study emphasizing distances between critical structures.Laryngoscope 1992 Apr;102(4):407-14

19. Larner JM; Hahn SS; Spaulding CA; Constable WC: Glomus jugulare tumors. Long-term control by radiation therapy. Cancer 1992 Apr 1;69(7):1813-7

20. DeAndrea RA; Josephson JS: Pathologic quiz case 4. Glomus jugulare tumor. Arch Otolaryngol Head Neck Surg 1992 Feb;118(2):216-7, 219

21. Jackson CG; Cueva RA; Thedinger BA; Glasscock ME 3d: Cranial nerve preservation inlesions of the jugular fossa. Otolaryngol Head Neck Surg 1991 Nov;105(5):687-93

22. Souliere CR Jr; Telian SA; Kemink JL: The infratemporal fossa approach to skull base surgery. Ear Nose Throat J 1991 Sep;70(9):620-36

23. Poe DS; Jackson G; Glasscock ME; Johnson GD: Long-term results after lateral cranial base surgery. Laryngoscope 1991 Apr;101(4 Pt 1):372-8

24. Springate SC; Weichselbaum RR: Radiation or surgery for chemodectoma of the temporal bone: a review of local control and complications. Head Neck 1990 Jul-Aug;12(4):303-7

25. Murphy TP; Brackmann DE: Effects of preoperative embolization on glomus jugulare tumors. Laryngoscope 1989 Dec;99(12):1244-7

26. Brackmann DE; House WF; Terry R; Scanlan RL: Glomus jugulare tumors: effect of irradiation. Trans Am Acad Ophth & Otol 1972 Nov-Dec; 76:1423-31

27. Spector GJ; Maisel RH; Ogura JH: Glomus Jugulare Tumors II. A clinicopathologic analysis of the effect of radiotherapy. Ann Otol Rhinol Laryngol 1974; 83: 26-32

28. Gardner G; Cocke EW; Robertson JT; and others: Combined approach surgery for removal of glomus jugulare tumors. Laryngoscope 1977; 87: 665-688

29. Kim JA; Elkon D; Lim ML; amd others: Optimum dose of radiotherapy for chemodectomas of the middle ear. Int J Radiat Oncol Biol Phys 1980; 6: 815-19

30. Cummings BJ; Beale FA; Garrett PG; and others: The treatment of glomus tumors in the temporal bone by megavoltage radiation. Cancer 1984; 53: 2635-40

31. Konefal JB; Pilepich MV; Spector GJ; and others: Radiation therapy in the treatment of chemodectomas. Laryngoscope 1987; 97: 1331-35

32. Fisch U: Infratemporal fossa approach for glomus tumors of the temporal bone. Ann Otol Rhinol Laryngol 1982; 91: 474-9