Original Research—Head and Neck Surgery
Natural History of Glomus Jugulare: A Review of 16 Tumors Managed with Primary Observation
Otolaryngology– Head and Neck Surgery 2015, Vol. 152(1) 98–105 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599814555839 http://otojournal.org
Matthew L. Carlson, MD1, Alex D. Sweeney, MD1, George B. Wanna, MD1, James L. Netterville, MD1, and David S. Haynes, MD1
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
Received May 8, 2014; revised August 5, 2014; accepted September 26, 2014.
Abstract Objective. To characterize clinical disease progression and radiologic growth in a series of observed, previously untreated, glomus jugulare tumors (GJT).
G
Study Design. Retrospective review. Setting. Tertiary neurotologic referral center. Subjects and Methods. All patients with primary GJTs that were observed without intervention for a minimum of 2 years. Primary outcome measures included progression of cranial neuropathy and/or radiologic growth. Results. A total of 15 patients (80% female; median age, 69.6 years) with 16 GJTs met inclusion criteria. The most common indications for observation included advanced age (11; 73%) and patient preference (11; 73%). Cranial nerve function remained stable in most subjects over a median clinical follow-up period of 86.4 months. Among the 12 with serial imaging, 5 (42%) GJTs demonstrated radiologic growth, while 7 (58%) remained stable. The median growth rate of the 5 enlarging tumors using the maximum linear dimension was 0.8 mm/y (range, 0.6-1.6 mm/y) or 0.4 cm3/y (0.1-0.9 cm3/y) using volumetric analysis. There were no deaths attributable to tumor progression or treatment. Conclusion. In an older subset of patients, we found that a significant number of GJTs do not grow after time of diagnosis and symptoms frequently remain stable for many years. Even with disease progression, most GJTs exhibit indolent growth with slowly progressive cranial neuropathy, affording satisfactory physiologic compensation in most patients. In the absence of brainstem compression or concern for malignancy, observation of GJTs is a viable initial management option for elderly patients.
Keywords jugular paraganglioma, glomus jugulare, natural history
lomus jugulare (GJT) are rare, slow-growing tumors that arise from the paraganglia cells within the adventia of the jugular bulb. GJTs most commonly present during the fourth to fifth decade of life following symptoms of pulsatile tinnitus and hearing loss, with women being affected more frequently than men. Owing to the low prevalence of disease, inaccessible location for examination, and insidious growth, many patients have advanced disease at diagnosis and endorse a long duration of preceding symptoms. Following the introduction of modern microsurgical techniques, most patients diagnosed with GJTs received upfront microsurgery with the goal of gross total resection; radiation was commonly reserved for recurrent disease, patients with advanced medical comorbidities, or palliation.1 Even with refined surgical techniques, most patients with moderate to large GJTs are left with a maximum conductive hearing loss, at least transient facial palsy, and worsening lower cranial neuropathy.1-4 Within the past 15 years, several reports emerged demonstrating the utility of stereotactic radiosurgery (SRS) for treatment of primary and residual progressive GJTs.5-7 Data from the radiosurgical literature have demonstrated a reduced risk of cranial neuropathy compared with gross total resection, with varying rates of tumor control.8,9 Paralleling the emergence of these reports, several centers began adopting a conservative surgical strategy, using planned subtotal resection.10,11 Most such reports
1 The Otology Group of Vanderbilt, Department of Otolaryngology–Head and Neck Surgery, Vanderbilt University Medical Center, Nashville, Tennessee, USA
This article was presented at the 2014 AAO-HNS/F Annual Meeting & OTO EXPO; September 21-24, 2014; Orlando, Florida. Corresponding Author: David S. Haynes, MD, The Otology Group of Vanderbilt, Department of Otolaryngology–Head and Neck Surgery, Vanderbilt University Medical Center, Nashville, TN 37232, USA. Email: [email protected]
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Carlson et al
99
describe removal of mastoid and middle ear disease, without engaging tumor in the jugular bulb, to improve audiologic complaints while minimizing risk to lower cranial nerve function. As with any medical condition, there must be clear advantages of therapy over the natural history of disease to justify the potential risks and expense of intervention. Despite the very large number of publications reporting outcomes following microsurgery and radiation therapy, data concerning the natural history of primary GJT are scarce and greatly needed.8,9 To the best of our knowledge, only 2 centers have specifically examined the biological behavior and clinical course of previously untreated GJTs.12,13 Such data will help best counsel patients of advanced age, poor surgical candidacy, those with bilateral head and neck paragangliomas, and subjects with minimal symptoms. With increasing patient access to computed tomography and magnetic resonance imaging, and improved surveillance protocols for patients with familial paragangliomas, we will undoubtedly witness a rise in the number of smaller tumors diagnosed in patients with minimal symptoms who may appropriately question the need for immediate intervention. Finally, quoted rates of tumor control and neurologic morbidity following SRS are meaningless without knowledge of the prevalence of nongrowth and cranial nerve preservation of untreated tumors.13,14 In the authors’ practice, most tumors are treated with upfront microsurgery; however, over the past 2 decades, we have cared for a handful of patients who were managed with primary observation. Herein, we describe the clinical and radiologic outcomes of this unique population.
Subjects and Methods Following institutional review board approval (IRB 140299), a prospectively maintained clinical database and the electronic medical record were queried for all patients diagnosed with GJT after January 1, 1995. Subjects who were managed with primary observation, having a minimum of 24 months of clinical follow-up, were identified. Patients with prior treatment, incomplete medical records, less than 2 years of clinical follow-up, or a questionable diagnosis were excluded. The primary outcomes of interest were disease progression, defined by worsening cranial neuropathy or radiologic growth, and need for intervention. The following data were collected via retrospective chart review: baseline patient demographics (age at diagnosis, sex); nature of disease (isolated or multicentric); duration and character of symptoms prior to diagnosis; symptoms, hearing capacity, and cranial nerve function at time of diagnosis; tumor volume and stage at time of diagnosis; progression of symptoms, hearing loss, and cranial neuropathy during observation; volumetric changes on serial imaging during observation; surgical rehabilitative procedures; need for tumor treatment; duration of radiologic and clinical follow-up; and disease-specific and overall survival. A telephone questionnaire was conducted to augment chart review regarding recent symptom progression
and course of disease (see Supplemental Appendix at www.otojournal.org). Finally, to better estimate overall survival, the Social Security Death Index was searched for patients who were lost to follow-up, without having documentation of death within the electronic medical record. Tumors were staged according to the Glasscock-Jackson and Fisch classification systems.15,16 Tumor volume was calculated using 3 perpendicular axes, acquired from axial and coronal series, applying the ellipsoid volume calculation (4/ 3pr1r2r3; cm3).12 Growth was defined as a minimum of a 20% volume increase on serial imaging.12,17 The volumetric and maximum linear dimension rate of growth was calculated by the net change in tumor size on serial imaging, divided by the time interval between studies. Audiometric outcomes were determined by comparing the first and most recent 4frequency pure tone average (0.5, 1, 2, 4 kHz) and word recognition score. Continuous features were summarized with medians and ranges; categorical features were summarized with frequency counts and percentages. Comparisons of continuous features between groups of interest were evaluated using Wilcoxon rank sum tests. Associations between 2 continuous features were evaluated using Spearman rank correlation coefficients. All tests were 2-sided, and P values less than .05 were considered statistically significant.
Results A total of 15 patients, with 16 GJTs, met inclusion criteria. All 15 patients had a minimum clinical follow-up of 2 years, and 12 (80%) had serial imaging studies available for review. The median age at diagnosis was 69.6 years (range, 37.6-79.5), and 12 (80%) were women (Table 1). Most patients presented with hearing loss (12; 75%) and pulsatile tinnitus (11; 69%), while vagal (6; 38%), accessory (3; 19%), and hypoglossal (2; 12%) paralyses were uncommon at the time of diagnosis (Table 2 and 3). The median duration between symptom onset and diagnosis was 30.5 months (range, 0-144 months). Three (20%) patients had multiple head and neck paragangliomas: one with an ipsilatateral glomus vagale (patient 10); a second with bilateral GJTs, one of which was resected years earlier (patient 4); and a third with bilateral GJTs that remain untreated (patients 8a and 8b). With only 3 patients we could not perform statistical comparisons between groups. However, based on reviewing these cases, there did not appear to be a clear difference in tumor behavior between patients with multicentric disease and those without. The most common indications for initial observation were advanced age (11; 73%) and patient preference (11; 73%); 5 (33%) patients refused treatment despite receiving recommendations for surgery or radiotherapy, and 2 (13%) had contralateral skull base or cervical lesions. The distribution of tumor stage is outlined in Table 4; 6 (38%) GJTs had intracranial extension at time of diagnosis. The median tumor volume was 2.7 cm3 (range, 0.9-19.9 cm3), and the median maximum linear dimension in the axial plane was 2.0 cm (range, 1.4-3.9 cm). There was not a
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
100
Otolaryngology–Head and Neck Surgery 152(1)
Table 1. Baseline Patient Features, Indications for Primary Observation, and Interventions. Baseline Features Patient ID
Age
Sex
1 2 3 4 5 6 7 8a 8b 9 10 11 12 13 14 15 Summary
46.6 42.9 74.6 56.5 74.6 69.7 69.6 37.6
F M F M F F F M
72.7 71.3 59.6 66.6 79.5 74.7 66.9 69.6 (38-80)
F F F F F F F F, 12 (80%) M, 3 (20%)
Intervention(s) Multiple Tumorsa N N N Y, bilat N N N Y; bilat N Y; ipsilat N N N N N Y, 3 (20%) N, 12 (80%)
Clinical Follow-up
Resection
Radiation
Vocal Cord Procedure
N N N N N N N N N N N N N N N STR Y, 1 (6%) N, 15 (94%)
N SRT N N N N N N N N N N N N N N Y, 1 (6%) N, 15 (94%)
T1T1AA N N N T1T1AA N T1T1AA N N Injection T1T1AA N N N N N Y, 5 (31%) N, 11 (69%)
Duration, y
Status and Age
8.1 2.9 2.4 6.2 5.7 2.0 10.0 9.7
AWD, 54.6 AWD, 45.8 AWD, 77.0 AWD, 62.7 AWD, 80.3 AWD, 71.7 AWD, 79.7 AWD 47.3
7.2 13.2 7.7 9.3 6.5 3.3 8.6 7.2 (2.0-13.2)
AWD, 79.9 DUC, 84.5 AWD, 67.3 AWD, 75.9 AWD, 86.0 AWD, 78.0 DUC, 75.6 AWD, 13 (87%), 69.7 DUC, 2 (13%), 80.1
Abbreviations: AWD, alive with disease; DUC, death from unrelated cause; F, female; H&N, head and neck; M, male; N, no; T1T 1 AA, type 1 thyroplasty with arytenoid adduction; Y, yes. a At least 1 additional skull base or cervical paraganglioma present.
Table 2. Baseline and Progression of Symptoms and Pattern of Cranial Neuropathy during the Course of Observation. Patient ID
Hearing Loss
Pulsatile Tinnitus
Bloody Otorrhea
1 2 3 4 5 6 7 8a 8b 9 10 11 12 13 14 15 Never present, n (%) Stable, n (%) Progressed, n (%) New deficit, n (%)
NP Stable NP NP Stable Stable Progressed Progressed NP Progressed Stable Progressed Progressed Stable Stable Progressed 4 (25) 6 (38) 6 (38) 0 (0)
Stable NP NP NP NP Stable Stable Stable NP Stable Stable Stable Stable Stable Stable Stable 5 (31) 11 (69) 0 (0) 0 (0)
NP NP NP NP NP NP NP NP NP NP NP NP NP New deficit NP New deficit 14 (88) 0 (0) 0 (0) 2 (13)
Vertigo
Dysphonia
Dysphagia
NP NP NP NP NP Stable NP New Deficit NP Stable NP Progressed NP NP NP NP 12 (75) 2 (13) 1 (6) 1 (6)
Stable Progressed NP NP Stable Stable New deficit NP NP NP Stable NP NP Progressed NP NP 9 (56) 4 (25) 2 (13) 1 (6)
Stable Progressed NP NP Stable Stable New deficit NP NP NP Stable NP NP NP NP NP 10 (63) 4 (25) 1 (6) 1 (6)
Abbreviation: NP, never present.
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Carlson et al
101
Table 3. Baseline and Progression of Cranial Neuropathy during the Course of Observation. Patient ID 1 2 3 4 5 6 7 8a 8b 9 10 11 12 13 14 15 Never present, n (%) Stable, n (%) Progressed, n (%) New deficit, n (%)
CN 7
CN 10
CN 11
CN 12
NP NP NP NP NP NP NP New deficit NP NP NP NP NP NP NP NP 15 (94) 0 (0) 0 (0) 1 (6)
Stable Progresseda NP NP Stable NP New deficit Progressedb NP New deficit Stable NP NP Stable NP NP 8 (50) 4 (25) 2 (13) 2 (13)
Stable NP NP NP Stable NP New deficit NP NP New deficit Progressed NP NP NP NP NP 11 (69) 2 (13) 1 (6) 2 (13)
NP New deficita NP NP Stable NP New deficit NP NP NP Progressed NP NP New Deficit NP NP 11 (69) 1 (6) 1 (6) 3 (19)
Abbreviations: CN, cranial nerve; NP, never present. a Progression of vagal paralysis and new onset of hypoglossal paralysis occurred immediately following radiotherapy. b Patient also with ipsilateral vagal paraganglioma threatening cranial nerve 10 function.
Table 4. Extent of Disease and Radiologic Growth over the Course of Follow-up. Stage at Diagnosis
Patient ID
GJ Grade
Fisch Type
Size at Diagnosis Volume, cm3
1 2 C2 5.5 2 2 C2 2.2 3 4 D1 5.8 4 1 C1 1.6 5 2 C2 2.6 6 2 C2 1.4 7 3 D1 7.7 8a 3 D1 19.9 8b 1 C1 1.9 9 2 D1 5.0 10 3 D1 7.5 11 2 C2 1.1 12 2 C2 2.4 13 3 D1 5.8 14 2 C2 2.8 15 2 C2 0.9 Summary 1, 2 (13%) C1, 2 (13%) 2.7 (0.9-19.9) 2, 9 (56%) C2, 8 (50%) 3, 4 (25%) D1, 6 (38%) 4, 1 (6%)
Max Linear Dimension, cm 2.3 1.9 2.5 1.8 1.7 2 2.9 3.9 1.8 2 2.5 1.9 2.3 2.6 1.7 1.4 2.0 (1.4-3.9)
Pattern of Growth
Growtha Growth Growth Stable Stable Stable Stable Stable NA NA Growth Growth Stable Stable NA NA Growth Growth, 5 (42%) Stable, 7 (58%)
Max Linear Dimension Growth Volumetric Rate, mm/y Growth Rate, cm3/y 0.6 1.6 NA NA NA NA NA NA NA 1.0 0.8 NA NA NA NA 0.6 0.8 (0.6-1.6)
Abbreviations: GJ, Glasscock-Jackson; NA, not applicable; Rad FU, radiologic follow-up. a Growth was defined as a greater than a 20% increase in tumor volume on serial imaging. Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
0.4 0.9 NA NA NA NA NA NA NA 0.4 0.6 NA NA NA NA 0.1 0.4 (0.1-0.9)
Rad FU, y 7.2 2.8 2.1 4.8 2.7 2.0 5.8 NA NA 7.6 12.0 3.7 5.6 NA NA 4.8 4.8 (2.0-12.0)
102
Otolaryngology–Head and Neck Surgery 152(1)
Figure 1. Serial axial and coronal gadolinium enhanced T1weighted magnetic resonance imaging sequences of a right-sided glomus jugulare demonstrating stable tumor size over the course of 5.8 years of observation (patient 7).
Figure 2. Serial axial and coronal gadolinium enhanced T1weighted magnetic resonance imaging sequences demonstrating consistent radiologic growth of a right-sided glomus jugulare over the course of 7.6 years (patient 9).
statistically significant correlation between age at diagnosis and tumor volume (correlation coefficient = 0.17; P = .53). Over a median radiologic follow-up of 57.6 months (range, 24-144 months), 7 of 12 (58%) remained stable while 5 (42%) demonstrated serial growth (Figures 1 and 2). The median growth rate of the 5 enlarging tumors was 0.8 mm/y (range, 0.6-1.6 mm/y) using the maximum linear dimension or 0.44 cm3/y (range, 0.14-0.87 cm3/y) using volumetric analysis. The median age for patients with GJTs that grew was 66.9 years (range, 42.9-72.7 years) compared with 69.2 years (range, 56.5-74.6 years) for those with stable disease (P = .27). Overall, patients with growing tumors had a longer duration of radiologic follow-up (median 86.4 vs 44.4 years; P = .07). One elderly patient with a growing GJT and new-onset bloody otorrhea underwent conservative mastoid and middle ear tumor resection with ear canal closure at the age of 72 years, 6 years after initial diagnosis. A second patient received fractionated radiotherapy (3 sessions, 700 cGy/fraction) at the age of 45 years, 30 months following initial diagnosis. Following completion of treatment, the latter patient experienced progressive 10th and 12th nerve palsies. After a median clinical follow-up of 86.4 months (range, 24-158 months), 6 (38%) patients endured progressive hearing loss and 2 (12.5%) experienced bloody otorrhea from tumor eruption into the external auditory canal. Pulsatile tinnitus remained stable in all 11 patients who reported symptoms at initial diagnosis. Over the course of observation, 8 (50%) patients maintained normal vagal function and 11 (69%) normal accessory and hypoglossal function. Following initial presentation, new or worsening lower cranial nerve deficits occurred in less than a third of cases. One (6%) patient experienced partial facial paralysis during
the course of observation. Four (25%) patients underwent type 1 thyroplasty with arytenoid adduction, and 1 (6%) received injection laryngoplasty. No patients required feeding tube placement, tracheostomy, or ventriculoperitoneal shunt placement. At last follow-up, 13 (87%) patients are alive while 2 (13.3%) died of unrelated causes at the ages of 76 and 85 years.
Discussion In 1992, van der Mey and colleagues from Leiden University in the Netherlands reviewed the clinical outcomes of 13 observed jugulotympanic paragliomas, comparing them to 16 patients who underwent gross total removal and 23 who received subtotal resection.18 They concluded that surgical treatment of skull base paragangliomas did not appear to improve survival but resulted in a 2-fold increase in cranial neuropathy. A detailed report regarding tumor growth, cranial nerve outcome, and duration of clinical follow-up of the 13 observed patients was not given. However, this landmark article was the first to provide important evidence to support an initial wait-and-scan policy for select patients. In a later series from the same center, Jansen et al12 reviewed 11 jugulotympanic tumors that were conservatively observed. After a mean follow-up of 46 months, 55% of tumors demonstrated radiologic progression with a median growth rate 0.8 mm/y. The authors found that jugulotympanic paragangliomas tended to be smaller and exhibit a more indolent growth pattern compared with other head and neck paragangliomas such as carotid body and vagal paragangliomas. In both series, it is not clear how many of the observed tumors were glomus tympanicum (Fisch type A and B) rather than GJT (Fisch type C and D). The authors reported that the average tumor volume was 0.8 cm3
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Carlson et al
103
(equating to a ~1.1-cm diameter), and none of the patients exhibited cranial neuropathy, suggesting that at least a fraction of tumors were glomus tympanicum. Most recently, Prasad et al13 analyzed the outcomes of 23 GJTs (Fisch type C and D) that were observed for a minimum of 3 years. They found that 65% of tumors remained stable or regressed over a median follow-up of 61 months. Of the 8 tumors that demonstrated progression, 7 were slow growing (\3 mm/y) and were managed with continued observation, while 1 fast-growing tumor was treated with radiotherapy. Importantly, only 30% of patients developed new cranial nerve deficits. Two subjects had facial paresis at the time of diagnosis, and no patient experienced newonset or progression of facial weakness during the course of observation. Data concerning the natural history of GJTs are scarce, and reports such as these are greatly needed to provide accurate patient counseling and informed consent as well as to determine optimal tailored treatment. Paralleling other benign skull base tumors such as vestibular schwannoma, gross total resection has been the default treatment strategy for decades. With time, many centers began using radiotherapy with increasing frequency given an excellent safety profile and durable tumor control. However, it was from reports such as these that a ‘‘wait-and-scan’’ strategy was ultimately legitimized, eventually becoming the default strategy at many centers for small- to medium-sized tumors.19 Within the past 3 years, there have been several large reviews and 1 large multicenter study evaluating the outcomes of SRS for treatment of GJT. In 2011, Ivan and colleagues9 performed a meta-analysis and identified 869 patients from 109 studies that met inclusion criteria. Of these, 351 underwent gross total resection, 82 subtotal resection alone, 97 with subtotal resection plus SRS, and 339 received SRS alone. The mean length of follow-up was approximately 1.5 years longer in the gross total resection group (88 months) compared with the SRS arm (71 months). Overall, tumor control was greatest in patients who received SRS alone (95%) compared with gross total resection (86%) and subtotal resection (69%). When examining rates of new or worsening cranial neuropathy, IX, X, or XI nerve palsies occurred in 18% to 40% of patients following GTR and 9% to 12% following SRS. In a large Gamma Knife SRS multicenter study, Sheehan and colleagues5 found that overall tumor control was achieved in 93% of patients after a median follow-up of 50.5 months, with an actuarial tumor control rate of 88% at 5 years. Pulsatile tinnitus was improved in 49%, and new or progressive deficits occurred in 15% of patients. Based on the emerging but limited data concerning primary observation of GJT, it appears that cranial nerve outcomes following SRS are roughly comparable to observation, and both are superior to gross total tumor resection.12,13 However, long-term tumor control is greater for SRS and microsurgery than for observation. From the current data, we know that at least half of GJTs do not grow for extended periods of time following diagnosis, and the overwhelming
majority of tumors with radiologic progression exhibit a slow growth rate, with an average of less than 1 mm per year.13 Combining our data with that of Prasad et al,13 only 3 of 39 observed patients eventually required treatment. Collectively, these data provide evidence supporting the option of an initial wait-and-scan policy for patients of advanced age. However, because most patients who have been analyzed are of older age and since follow-up beyond 10 years is very limited, it would be premature to conclude that observation should be routinely used for younger patients who may harbor more aggressive tumors and who are expected to live decades beyond diagnosis. In the current study, we found that there was a trend toward higher rates of tumor progression in patients who were followed longer—a finding that was also reported by Prasad et al.13 Thus, a healthy 30-year-old patient diagnosed with a medium-sized GJT with intracranial extension is likely to live long enough to experience significant growth and complications from disease. Since the outcome of treatment is more favorable for smaller tumors, early surgical intervention or stereotactic radiosurgery should be strongly considered in such cases. On the other end of the spectrum, a 75-year-old patient with average age-related comorbidities is expected to live less than 10 more years based on current life-expectancy estimates—hardly long enough to experience clinically significant tumor growth or symptom progression. In this case, initial observation can be considered. In the rare case that rapid growth does occur, radiosurgery could be pursued. This paradigm shift toward conservatism in the elderly is further supported by the observation that physiologic compensation of lower cranial neuropathy is slow and frequently incomplete in older patients.20 Following abrupt loss of cranial nerve IX, X, and XII function, many young adults can resume a normal diet within a year of surgery; however, adequate upper aerodigestive competency is rarely fully regained in older patients. Furthermore, adaptation to lower cranial nerve paralysis is more favorable with progressive loss of function from disease, rather than abrupt loss from surgery. We have noticed that even in cases of seemingly complete preoperative IX, X, and XII paralysis, a significant number of patients experience worsening dysphagia following gross total resection with nerve sacrifice, suggesting that in the absence of motor function, residual pharyngeal and glottic tone may be present in a number of patients. Reflecting on the available data, we are reluctant to construct a rigid treatment algorithm or make any universal management recommendations. Rather, we believe treatment should be highly individualized; priorities between patients are often different, and chronological age does not always match biological age. Until 20 years ago, upfront radical resection was considered the standard of care. More recently, subtotal resection and radiosurgery have been used with increasing frequency in an attempt to reduce neurologic morbidity. The current study, along with the 2 previous reports by Prasad et al13 and Jansen et al,12 support that initial observation is a viable option for select patients, particularly those with advanced age or limited life expectancy. If
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
104
Otolaryngology–Head and Neck Surgery 152(1)
growth should occur, radiosurgery should be considered in such patients given the favorable safety profile and high likelihood of tumor control. As a final note, the authors want to emphasize the distinction between management of glomus tympanicum and GJT since the clinical course, risks of surgery and treatment outcomes are different. By definition, glomus tympanicum tumors arise from within the middle ear cleft. Even with advanced disease, intracranial extension and/or cranial neuropathy are exceedingly rare. While observation could be entertained in very poor surgical candidates, we generally advocate early surgical resection in all patients given the low risk of surgery and the significant benefit most patients enjoy from relief of pulsatile tinnitus and potential improvement in hearing. In closing, several strengths and limitations of the current study deserve mention. To the authors’ knowledge, this is only the third study to report detailed longitudinal data on the natural history of previously untreated GJTs, providing a median clinical follow-up of approximately 7 years. To improve the accuracy of data and ease of future interstudy comparisons, volumetric data are reported in addition to maximum linear dimension measurements. Without management randomization, our conclusions may be confounded by treatment selection bias. It could be argued that since most of the study population is older, we are inherently selecting for patients with a less aggressive disease variant. Further studies are needed to determine if these conclusions also apply to a younger GJT population.
Conclusions Among an older cohort of patients with GJTs, the authors found that a significant percentage of tumors do not grow after time of diagnosis and symptoms frequently remain stable for many years. Even with disease progression, most GJTs exhibit indolent growth (~1 mm/y) with slowly progressive cranial neuropathy affording satisfactory physiologic compensation in most patients. These data affirm that initial observation of GJTs is a viable option for patients of advanced age. Future studies are needed to determine if this strategy can be applied to younger patients who may harbor more aggressive disease. Author Contributions Matthew L. Carlson, concept, review of data, manuscript writing, manuscript revision; Alex D. Sweeney, concept, review of data, manuscript revision; George B. Wanna, concept, review of data, manuscript revision; James L. Netterville, concept, review of data, manuscript revision; David S. Haynes, concept, review of data, manuscript revision.
Disclosures Competing interests: James L. Netterville, consultant, Medtronic. David S. Haynes, consultant Advanced Bionics Corp, Cochlear Corp, MED-EL GmbH, Stryker/Synthes; Anspach Corp; Grace medical. Sponsorships: None. Funding source: None.
Supplemental Material Additional supporting information may be found at http://otojournal .org/supplemental.
References 1. Jackson CG, McGrew BM, Forest JA, et al. Lateral skull base surgery for glomus tumors: long-term control. Otol Neurotol. 2001;22:377-382. 2. Bacciu A, Ait Mimoune H, D’Orazio F, et al. Management of facial nerve in surgical treatment of previously untreated fisch class C tympanojugular paragangliomas: long-term results. J Neurol Surg B Skull Base. 2014;75:1-7. 3. Bacciu A, Medina M, Ait Mimoune H, et al. Lower cranial nerves function after surgical treatment of Fisch Class C and D tympanojugular paragangliomas [published online December 12, 2013]. Eur Arch Otorhinolaryngol. 4. Jackson CG, Haynes DS, Walker PA, et al. Hearing conservation in surgery for glomus jugulare tumors. Am J Otol. 1996; 17:425-437. 5. Sheehan JP, Tanaka S, Link MJ, et al. Gamma Knife surgery for the management of glomus tumors: a multicenter study. J Neurosurg. 2012;117:246-254. 6. Foote RL, Pollock BE, Gorman DA, et al. Glomus jugulare tumor: tumor control and complications after stereotactic radiosurgery. Head Neck. 2002;24:332-338. 7. Eustacchio S, Trummer M, Unger F, et al. The role of Gamma Knife radiosurgery in the management of glomus jugular tumours. Acta Neurochir Suppl. 2002;84:91-97. 8. van Hulsteijn LT, Corssmit EP, Coremans IE, et al. Regression and local control rates after radiotherapy for jugulotympanic paragangliomas: systematic review and meta-analysis. Radiother Oncol. 2013;106:161-168. 9. Ivan ME, Sughrue ME, Clark AJ, et al. A meta-analysis of tumor control rates and treatment-related morbidity for patients with glomus jugulare tumors. J Neurosurg. 2011;114:1299-1305. 10. Cosetti M, Linstrom C, Alexiades G, et al. Glomus tumors in patients of advanced age: a conservative approach. Laryngoscope. 2008;118:270-274. 11. Willen SN, Einstein DB, Maciunas RJ, et al. Treatment of glomus jugulare tumors in patients with advanced age: planned limited surgical resection followed by staged gamma knife radiosurgery: a preliminary report. Otol Neurotol. 2005;26:1229-1234. 12. Jansen JC, van den Berg R, Kuiper A, et al. Estimation of growth rate in patients with head and neck paragangliomas influences the treatment proposal. Cancer. 2000;88:2811-2816. 13. Prasad SC, Mimoune HA, D’Orazio F, et al. The role of waitand-scan and the efficacy of radiotherapy in the treatment of temporal bone paragangliomas. Otol Neurotol 2014;35:922931. 14. Miller T, Lau T, Vasan R, et al. Reporting success rates in the treatment of vestibular schwannomas: are we accounting for the natural history? J Clin Neurosci. 2014;21:914-918. 15. Oldring D, Fisch U. Glomus tumors of the temporal region: surgical therapy. Am J Otol. 1979;1:7-18. 16. Jackson CG, Glasscock ME III, Harris PF. Glomus tumors: diagnosis, classification, and management of large lesions. JAMA Otolaryngol Head Neck Surg. 1982;108:401-410.
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Carlson et al
105
17. Lundin P, Pedersen F. Volume of pituitary macroadenomas: assessment by MRI. J Comput Assist Tomogr. 1992;16:519528. 18. van der Mey AG, Frijns JH, Cornelisse CJ, et al. 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;101:635-642.
19. Strasnick B, Glasscock ME III, Haynes D, et al. The natural history of untreated acoustic neuromas. Laryngoscope. 1994; 104:1115-1119. 20. Netterville JL, Civantos FJ. Rehabilitation of cranial nerve deficits after neurotologic skull base surgery. Laryngoscope. 1993;103:45-54.
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Natural History of Glomus Jugulare: A Review of 16 Tumors Managed with Primary Observation
Otolaryngology– Head and Neck Surgery 2015, Vol. 152(1) 98–105 Ó American Academy of Otolaryngology—Head and Neck Surgery Foundation 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0194599814555839 http://otojournal.org
Matthew L. Carlson, MD1, Alex D. Sweeney, MD1, George B. Wanna, MD1, James L. Netterville, MD1, and David S. Haynes, MD1
Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.
Received May 8, 2014; revised August 5, 2014; accepted September 26, 2014.
Abstract Objective. To characterize clinical disease progression and radiologic growth in a series of observed, previously untreated, glomus jugulare tumors (GJT).
G
Study Design. Retrospective review. Setting. Tertiary neurotologic referral center. Subjects and Methods. All patients with primary GJTs that were observed without intervention for a minimum of 2 years. Primary outcome measures included progression of cranial neuropathy and/or radiologic growth. Results. A total of 15 patients (80% female; median age, 69.6 years) with 16 GJTs met inclusion criteria. The most common indications for observation included advanced age (11; 73%) and patient preference (11; 73%). Cranial nerve function remained stable in most subjects over a median clinical follow-up period of 86.4 months. Among the 12 with serial imaging, 5 (42%) GJTs demonstrated radiologic growth, while 7 (58%) remained stable. The median growth rate of the 5 enlarging tumors using the maximum linear dimension was 0.8 mm/y (range, 0.6-1.6 mm/y) or 0.4 cm3/y (0.1-0.9 cm3/y) using volumetric analysis. There were no deaths attributable to tumor progression or treatment. Conclusion. In an older subset of patients, we found that a significant number of GJTs do not grow after time of diagnosis and symptoms frequently remain stable for many years. Even with disease progression, most GJTs exhibit indolent growth with slowly progressive cranial neuropathy, affording satisfactory physiologic compensation in most patients. In the absence of brainstem compression or concern for malignancy, observation of GJTs is a viable initial management option for elderly patients.
Keywords jugular paraganglioma, glomus jugulare, natural history
lomus jugulare (GJT) are rare, slow-growing tumors that arise from the paraganglia cells within the adventia of the jugular bulb. GJTs most commonly present during the fourth to fifth decade of life following symptoms of pulsatile tinnitus and hearing loss, with women being affected more frequently than men. Owing to the low prevalence of disease, inaccessible location for examination, and insidious growth, many patients have advanced disease at diagnosis and endorse a long duration of preceding symptoms. Following the introduction of modern microsurgical techniques, most patients diagnosed with GJTs received upfront microsurgery with the goal of gross total resection; radiation was commonly reserved for recurrent disease, patients with advanced medical comorbidities, or palliation.1 Even with refined surgical techniques, most patients with moderate to large GJTs are left with a maximum conductive hearing loss, at least transient facial palsy, and worsening lower cranial neuropathy.1-4 Within the past 15 years, several reports emerged demonstrating the utility of stereotactic radiosurgery (SRS) for treatment of primary and residual progressive GJTs.5-7 Data from the radiosurgical literature have demonstrated a reduced risk of cranial neuropathy compared with gross total resection, with varying rates of tumor control.8,9 Paralleling the emergence of these reports, several centers began adopting a conservative surgical strategy, using planned subtotal resection.10,11 Most such reports
1 The Otology Group of Vanderbilt, Department of Otolaryngology–Head and Neck Surgery, Vanderbilt University Medical Center, Nashville, Tennessee, USA
This article was presented at the 2014 AAO-HNS/F Annual Meeting & OTO EXPO; September 21-24, 2014; Orlando, Florida. Corresponding Author: David S. Haynes, MD, The Otology Group of Vanderbilt, Department of Otolaryngology–Head and Neck Surgery, Vanderbilt University Medical Center, Nashville, TN 37232, USA. Email: [email protected]
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Carlson et al
99
describe removal of mastoid and middle ear disease, without engaging tumor in the jugular bulb, to improve audiologic complaints while minimizing risk to lower cranial nerve function. As with any medical condition, there must be clear advantages of therapy over the natural history of disease to justify the potential risks and expense of intervention. Despite the very large number of publications reporting outcomes following microsurgery and radiation therapy, data concerning the natural history of primary GJT are scarce and greatly needed.8,9 To the best of our knowledge, only 2 centers have specifically examined the biological behavior and clinical course of previously untreated GJTs.12,13 Such data will help best counsel patients of advanced age, poor surgical candidacy, those with bilateral head and neck paragangliomas, and subjects with minimal symptoms. With increasing patient access to computed tomography and magnetic resonance imaging, and improved surveillance protocols for patients with familial paragangliomas, we will undoubtedly witness a rise in the number of smaller tumors diagnosed in patients with minimal symptoms who may appropriately question the need for immediate intervention. Finally, quoted rates of tumor control and neurologic morbidity following SRS are meaningless without knowledge of the prevalence of nongrowth and cranial nerve preservation of untreated tumors.13,14 In the authors’ practice, most tumors are treated with upfront microsurgery; however, over the past 2 decades, we have cared for a handful of patients who were managed with primary observation. Herein, we describe the clinical and radiologic outcomes of this unique population.
Subjects and Methods Following institutional review board approval (IRB 140299), a prospectively maintained clinical database and the electronic medical record were queried for all patients diagnosed with GJT after January 1, 1995. Subjects who were managed with primary observation, having a minimum of 24 months of clinical follow-up, were identified. Patients with prior treatment, incomplete medical records, less than 2 years of clinical follow-up, or a questionable diagnosis were excluded. The primary outcomes of interest were disease progression, defined by worsening cranial neuropathy or radiologic growth, and need for intervention. The following data were collected via retrospective chart review: baseline patient demographics (age at diagnosis, sex); nature of disease (isolated or multicentric); duration and character of symptoms prior to diagnosis; symptoms, hearing capacity, and cranial nerve function at time of diagnosis; tumor volume and stage at time of diagnosis; progression of symptoms, hearing loss, and cranial neuropathy during observation; volumetric changes on serial imaging during observation; surgical rehabilitative procedures; need for tumor treatment; duration of radiologic and clinical follow-up; and disease-specific and overall survival. A telephone questionnaire was conducted to augment chart review regarding recent symptom progression
and course of disease (see Supplemental Appendix at www.otojournal.org). Finally, to better estimate overall survival, the Social Security Death Index was searched for patients who were lost to follow-up, without having documentation of death within the electronic medical record. Tumors were staged according to the Glasscock-Jackson and Fisch classification systems.15,16 Tumor volume was calculated using 3 perpendicular axes, acquired from axial and coronal series, applying the ellipsoid volume calculation (4/ 3pr1r2r3; cm3).12 Growth was defined as a minimum of a 20% volume increase on serial imaging.12,17 The volumetric and maximum linear dimension rate of growth was calculated by the net change in tumor size on serial imaging, divided by the time interval between studies. Audiometric outcomes were determined by comparing the first and most recent 4frequency pure tone average (0.5, 1, 2, 4 kHz) and word recognition score. Continuous features were summarized with medians and ranges; categorical features were summarized with frequency counts and percentages. Comparisons of continuous features between groups of interest were evaluated using Wilcoxon rank sum tests. Associations between 2 continuous features were evaluated using Spearman rank correlation coefficients. All tests were 2-sided, and P values less than .05 were considered statistically significant.
Results A total of 15 patients, with 16 GJTs, met inclusion criteria. All 15 patients had a minimum clinical follow-up of 2 years, and 12 (80%) had serial imaging studies available for review. The median age at diagnosis was 69.6 years (range, 37.6-79.5), and 12 (80%) were women (Table 1). Most patients presented with hearing loss (12; 75%) and pulsatile tinnitus (11; 69%), while vagal (6; 38%), accessory (3; 19%), and hypoglossal (2; 12%) paralyses were uncommon at the time of diagnosis (Table 2 and 3). The median duration between symptom onset and diagnosis was 30.5 months (range, 0-144 months). Three (20%) patients had multiple head and neck paragangliomas: one with an ipsilatateral glomus vagale (patient 10); a second with bilateral GJTs, one of which was resected years earlier (patient 4); and a third with bilateral GJTs that remain untreated (patients 8a and 8b). With only 3 patients we could not perform statistical comparisons between groups. However, based on reviewing these cases, there did not appear to be a clear difference in tumor behavior between patients with multicentric disease and those without. The most common indications for initial observation were advanced age (11; 73%) and patient preference (11; 73%); 5 (33%) patients refused treatment despite receiving recommendations for surgery or radiotherapy, and 2 (13%) had contralateral skull base or cervical lesions. The distribution of tumor stage is outlined in Table 4; 6 (38%) GJTs had intracranial extension at time of diagnosis. The median tumor volume was 2.7 cm3 (range, 0.9-19.9 cm3), and the median maximum linear dimension in the axial plane was 2.0 cm (range, 1.4-3.9 cm). There was not a
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
100
Otolaryngology–Head and Neck Surgery 152(1)
Table 1. Baseline Patient Features, Indications for Primary Observation, and Interventions. Baseline Features Patient ID
Age
Sex
1 2 3 4 5 6 7 8a 8b 9 10 11 12 13 14 15 Summary
46.6 42.9 74.6 56.5 74.6 69.7 69.6 37.6
F M F M F F F M
72.7 71.3 59.6 66.6 79.5 74.7 66.9 69.6 (38-80)
F F F F F F F F, 12 (80%) M, 3 (20%)
Intervention(s) Multiple Tumorsa N N N Y, bilat N N N Y; bilat N Y; ipsilat N N N N N Y, 3 (20%) N, 12 (80%)
Clinical Follow-up
Resection
Radiation
Vocal Cord Procedure
N N N N N N N N N N N N N N N STR Y, 1 (6%) N, 15 (94%)
N SRT N N N N N N N N N N N N N N Y, 1 (6%) N, 15 (94%)
T1T1AA N N N T1T1AA N T1T1AA N N Injection T1T1AA N N N N N Y, 5 (31%) N, 11 (69%)
Duration, y
Status and Age
8.1 2.9 2.4 6.2 5.7 2.0 10.0 9.7
AWD, 54.6 AWD, 45.8 AWD, 77.0 AWD, 62.7 AWD, 80.3 AWD, 71.7 AWD, 79.7 AWD 47.3
7.2 13.2 7.7 9.3 6.5 3.3 8.6 7.2 (2.0-13.2)
AWD, 79.9 DUC, 84.5 AWD, 67.3 AWD, 75.9 AWD, 86.0 AWD, 78.0 DUC, 75.6 AWD, 13 (87%), 69.7 DUC, 2 (13%), 80.1
Abbreviations: AWD, alive with disease; DUC, death from unrelated cause; F, female; H&N, head and neck; M, male; N, no; T1T 1 AA, type 1 thyroplasty with arytenoid adduction; Y, yes. a At least 1 additional skull base or cervical paraganglioma present.
Table 2. Baseline and Progression of Symptoms and Pattern of Cranial Neuropathy during the Course of Observation. Patient ID
Hearing Loss
Pulsatile Tinnitus
Bloody Otorrhea
1 2 3 4 5 6 7 8a 8b 9 10 11 12 13 14 15 Never present, n (%) Stable, n (%) Progressed, n (%) New deficit, n (%)
NP Stable NP NP Stable Stable Progressed Progressed NP Progressed Stable Progressed Progressed Stable Stable Progressed 4 (25) 6 (38) 6 (38) 0 (0)
Stable NP NP NP NP Stable Stable Stable NP Stable Stable Stable Stable Stable Stable Stable 5 (31) 11 (69) 0 (0) 0 (0)
NP NP NP NP NP NP NP NP NP NP NP NP NP New deficit NP New deficit 14 (88) 0 (0) 0 (0) 2 (13)
Vertigo
Dysphonia
Dysphagia
NP NP NP NP NP Stable NP New Deficit NP Stable NP Progressed NP NP NP NP 12 (75) 2 (13) 1 (6) 1 (6)
Stable Progressed NP NP Stable Stable New deficit NP NP NP Stable NP NP Progressed NP NP 9 (56) 4 (25) 2 (13) 1 (6)
Stable Progressed NP NP Stable Stable New deficit NP NP NP Stable NP NP NP NP NP 10 (63) 4 (25) 1 (6) 1 (6)
Abbreviation: NP, never present.
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Carlson et al
101
Table 3. Baseline and Progression of Cranial Neuropathy during the Course of Observation. Patient ID 1 2 3 4 5 6 7 8a 8b 9 10 11 12 13 14 15 Never present, n (%) Stable, n (%) Progressed, n (%) New deficit, n (%)
CN 7
CN 10
CN 11
CN 12
NP NP NP NP NP NP NP New deficit NP NP NP NP NP NP NP NP 15 (94) 0 (0) 0 (0) 1 (6)
Stable Progresseda NP NP Stable NP New deficit Progressedb NP New deficit Stable NP NP Stable NP NP 8 (50) 4 (25) 2 (13) 2 (13)
Stable NP NP NP Stable NP New deficit NP NP New deficit Progressed NP NP NP NP NP 11 (69) 2 (13) 1 (6) 2 (13)
NP New deficita NP NP Stable NP New deficit NP NP NP Progressed NP NP New Deficit NP NP 11 (69) 1 (6) 1 (6) 3 (19)
Abbreviations: CN, cranial nerve; NP, never present. a Progression of vagal paralysis and new onset of hypoglossal paralysis occurred immediately following radiotherapy. b Patient also with ipsilateral vagal paraganglioma threatening cranial nerve 10 function.
Table 4. Extent of Disease and Radiologic Growth over the Course of Follow-up. Stage at Diagnosis
Patient ID
GJ Grade
Fisch Type
Size at Diagnosis Volume, cm3
1 2 C2 5.5 2 2 C2 2.2 3 4 D1 5.8 4 1 C1 1.6 5 2 C2 2.6 6 2 C2 1.4 7 3 D1 7.7 8a 3 D1 19.9 8b 1 C1 1.9 9 2 D1 5.0 10 3 D1 7.5 11 2 C2 1.1 12 2 C2 2.4 13 3 D1 5.8 14 2 C2 2.8 15 2 C2 0.9 Summary 1, 2 (13%) C1, 2 (13%) 2.7 (0.9-19.9) 2, 9 (56%) C2, 8 (50%) 3, 4 (25%) D1, 6 (38%) 4, 1 (6%)
Max Linear Dimension, cm 2.3 1.9 2.5 1.8 1.7 2 2.9 3.9 1.8 2 2.5 1.9 2.3 2.6 1.7 1.4 2.0 (1.4-3.9)
Pattern of Growth
Growtha Growth Growth Stable Stable Stable Stable Stable NA NA Growth Growth Stable Stable NA NA Growth Growth, 5 (42%) Stable, 7 (58%)
Max Linear Dimension Growth Volumetric Rate, mm/y Growth Rate, cm3/y 0.6 1.6 NA NA NA NA NA NA NA 1.0 0.8 NA NA NA NA 0.6 0.8 (0.6-1.6)
Abbreviations: GJ, Glasscock-Jackson; NA, not applicable; Rad FU, radiologic follow-up. a Growth was defined as a greater than a 20% increase in tumor volume on serial imaging. Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
0.4 0.9 NA NA NA NA NA NA NA 0.4 0.6 NA NA NA NA 0.1 0.4 (0.1-0.9)
Rad FU, y 7.2 2.8 2.1 4.8 2.7 2.0 5.8 NA NA 7.6 12.0 3.7 5.6 NA NA 4.8 4.8 (2.0-12.0)
102
Otolaryngology–Head and Neck Surgery 152(1)
Figure 1. Serial axial and coronal gadolinium enhanced T1weighted magnetic resonance imaging sequences of a right-sided glomus jugulare demonstrating stable tumor size over the course of 5.8 years of observation (patient 7).
Figure 2. Serial axial and coronal gadolinium enhanced T1weighted magnetic resonance imaging sequences demonstrating consistent radiologic growth of a right-sided glomus jugulare over the course of 7.6 years (patient 9).
statistically significant correlation between age at diagnosis and tumor volume (correlation coefficient = 0.17; P = .53). Over a median radiologic follow-up of 57.6 months (range, 24-144 months), 7 of 12 (58%) remained stable while 5 (42%) demonstrated serial growth (Figures 1 and 2). The median growth rate of the 5 enlarging tumors was 0.8 mm/y (range, 0.6-1.6 mm/y) using the maximum linear dimension or 0.44 cm3/y (range, 0.14-0.87 cm3/y) using volumetric analysis. The median age for patients with GJTs that grew was 66.9 years (range, 42.9-72.7 years) compared with 69.2 years (range, 56.5-74.6 years) for those with stable disease (P = .27). Overall, patients with growing tumors had a longer duration of radiologic follow-up (median 86.4 vs 44.4 years; P = .07). One elderly patient with a growing GJT and new-onset bloody otorrhea underwent conservative mastoid and middle ear tumor resection with ear canal closure at the age of 72 years, 6 years after initial diagnosis. A second patient received fractionated radiotherapy (3 sessions, 700 cGy/fraction) at the age of 45 years, 30 months following initial diagnosis. Following completion of treatment, the latter patient experienced progressive 10th and 12th nerve palsies. After a median clinical follow-up of 86.4 months (range, 24-158 months), 6 (38%) patients endured progressive hearing loss and 2 (12.5%) experienced bloody otorrhea from tumor eruption into the external auditory canal. Pulsatile tinnitus remained stable in all 11 patients who reported symptoms at initial diagnosis. Over the course of observation, 8 (50%) patients maintained normal vagal function and 11 (69%) normal accessory and hypoglossal function. Following initial presentation, new or worsening lower cranial nerve deficits occurred in less than a third of cases. One (6%) patient experienced partial facial paralysis during
the course of observation. Four (25%) patients underwent type 1 thyroplasty with arytenoid adduction, and 1 (6%) received injection laryngoplasty. No patients required feeding tube placement, tracheostomy, or ventriculoperitoneal shunt placement. At last follow-up, 13 (87%) patients are alive while 2 (13.3%) died of unrelated causes at the ages of 76 and 85 years.
Discussion In 1992, van der Mey and colleagues from Leiden University in the Netherlands reviewed the clinical outcomes of 13 observed jugulotympanic paragliomas, comparing them to 16 patients who underwent gross total removal and 23 who received subtotal resection.18 They concluded that surgical treatment of skull base paragangliomas did not appear to improve survival but resulted in a 2-fold increase in cranial neuropathy. A detailed report regarding tumor growth, cranial nerve outcome, and duration of clinical follow-up of the 13 observed patients was not given. However, this landmark article was the first to provide important evidence to support an initial wait-and-scan policy for select patients. In a later series from the same center, Jansen et al12 reviewed 11 jugulotympanic tumors that were conservatively observed. After a mean follow-up of 46 months, 55% of tumors demonstrated radiologic progression with a median growth rate 0.8 mm/y. The authors found that jugulotympanic paragangliomas tended to be smaller and exhibit a more indolent growth pattern compared with other head and neck paragangliomas such as carotid body and vagal paragangliomas. In both series, it is not clear how many of the observed tumors were glomus tympanicum (Fisch type A and B) rather than GJT (Fisch type C and D). The authors reported that the average tumor volume was 0.8 cm3
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Carlson et al
103
(equating to a ~1.1-cm diameter), and none of the patients exhibited cranial neuropathy, suggesting that at least a fraction of tumors were glomus tympanicum. Most recently, Prasad et al13 analyzed the outcomes of 23 GJTs (Fisch type C and D) that were observed for a minimum of 3 years. They found that 65% of tumors remained stable or regressed over a median follow-up of 61 months. Of the 8 tumors that demonstrated progression, 7 were slow growing (\3 mm/y) and were managed with continued observation, while 1 fast-growing tumor was treated with radiotherapy. Importantly, only 30% of patients developed new cranial nerve deficits. Two subjects had facial paresis at the time of diagnosis, and no patient experienced newonset or progression of facial weakness during the course of observation. Data concerning the natural history of GJTs are scarce, and reports such as these are greatly needed to provide accurate patient counseling and informed consent as well as to determine optimal tailored treatment. Paralleling other benign skull base tumors such as vestibular schwannoma, gross total resection has been the default treatment strategy for decades. With time, many centers began using radiotherapy with increasing frequency given an excellent safety profile and durable tumor control. However, it was from reports such as these that a ‘‘wait-and-scan’’ strategy was ultimately legitimized, eventually becoming the default strategy at many centers for small- to medium-sized tumors.19 Within the past 3 years, there have been several large reviews and 1 large multicenter study evaluating the outcomes of SRS for treatment of GJT. In 2011, Ivan and colleagues9 performed a meta-analysis and identified 869 patients from 109 studies that met inclusion criteria. Of these, 351 underwent gross total resection, 82 subtotal resection alone, 97 with subtotal resection plus SRS, and 339 received SRS alone. The mean length of follow-up was approximately 1.5 years longer in the gross total resection group (88 months) compared with the SRS arm (71 months). Overall, tumor control was greatest in patients who received SRS alone (95%) compared with gross total resection (86%) and subtotal resection (69%). When examining rates of new or worsening cranial neuropathy, IX, X, or XI nerve palsies occurred in 18% to 40% of patients following GTR and 9% to 12% following SRS. In a large Gamma Knife SRS multicenter study, Sheehan and colleagues5 found that overall tumor control was achieved in 93% of patients after a median follow-up of 50.5 months, with an actuarial tumor control rate of 88% at 5 years. Pulsatile tinnitus was improved in 49%, and new or progressive deficits occurred in 15% of patients. Based on the emerging but limited data concerning primary observation of GJT, it appears that cranial nerve outcomes following SRS are roughly comparable to observation, and both are superior to gross total tumor resection.12,13 However, long-term tumor control is greater for SRS and microsurgery than for observation. From the current data, we know that at least half of GJTs do not grow for extended periods of time following diagnosis, and the overwhelming
majority of tumors with radiologic progression exhibit a slow growth rate, with an average of less than 1 mm per year.13 Combining our data with that of Prasad et al,13 only 3 of 39 observed patients eventually required treatment. Collectively, these data provide evidence supporting the option of an initial wait-and-scan policy for patients of advanced age. However, because most patients who have been analyzed are of older age and since follow-up beyond 10 years is very limited, it would be premature to conclude that observation should be routinely used for younger patients who may harbor more aggressive tumors and who are expected to live decades beyond diagnosis. In the current study, we found that there was a trend toward higher rates of tumor progression in patients who were followed longer—a finding that was also reported by Prasad et al.13 Thus, a healthy 30-year-old patient diagnosed with a medium-sized GJT with intracranial extension is likely to live long enough to experience significant growth and complications from disease. Since the outcome of treatment is more favorable for smaller tumors, early surgical intervention or stereotactic radiosurgery should be strongly considered in such cases. On the other end of the spectrum, a 75-year-old patient with average age-related comorbidities is expected to live less than 10 more years based on current life-expectancy estimates—hardly long enough to experience clinically significant tumor growth or symptom progression. In this case, initial observation can be considered. In the rare case that rapid growth does occur, radiosurgery could be pursued. This paradigm shift toward conservatism in the elderly is further supported by the observation that physiologic compensation of lower cranial neuropathy is slow and frequently incomplete in older patients.20 Following abrupt loss of cranial nerve IX, X, and XII function, many young adults can resume a normal diet within a year of surgery; however, adequate upper aerodigestive competency is rarely fully regained in older patients. Furthermore, adaptation to lower cranial nerve paralysis is more favorable with progressive loss of function from disease, rather than abrupt loss from surgery. We have noticed that even in cases of seemingly complete preoperative IX, X, and XII paralysis, a significant number of patients experience worsening dysphagia following gross total resection with nerve sacrifice, suggesting that in the absence of motor function, residual pharyngeal and glottic tone may be present in a number of patients. Reflecting on the available data, we are reluctant to construct a rigid treatment algorithm or make any universal management recommendations. Rather, we believe treatment should be highly individualized; priorities between patients are often different, and chronological age does not always match biological age. Until 20 years ago, upfront radical resection was considered the standard of care. More recently, subtotal resection and radiosurgery have been used with increasing frequency in an attempt to reduce neurologic morbidity. The current study, along with the 2 previous reports by Prasad et al13 and Jansen et al,12 support that initial observation is a viable option for select patients, particularly those with advanced age or limited life expectancy. If
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
104
Otolaryngology–Head and Neck Surgery 152(1)
growth should occur, radiosurgery should be considered in such patients given the favorable safety profile and high likelihood of tumor control. As a final note, the authors want to emphasize the distinction between management of glomus tympanicum and GJT since the clinical course, risks of surgery and treatment outcomes are different. By definition, glomus tympanicum tumors arise from within the middle ear cleft. Even with advanced disease, intracranial extension and/or cranial neuropathy are exceedingly rare. While observation could be entertained in very poor surgical candidates, we generally advocate early surgical resection in all patients given the low risk of surgery and the significant benefit most patients enjoy from relief of pulsatile tinnitus and potential improvement in hearing. In closing, several strengths and limitations of the current study deserve mention. To the authors’ knowledge, this is only the third study to report detailed longitudinal data on the natural history of previously untreated GJTs, providing a median clinical follow-up of approximately 7 years. To improve the accuracy of data and ease of future interstudy comparisons, volumetric data are reported in addition to maximum linear dimension measurements. Without management randomization, our conclusions may be confounded by treatment selection bias. It could be argued that since most of the study population is older, we are inherently selecting for patients with a less aggressive disease variant. Further studies are needed to determine if these conclusions also apply to a younger GJT population.
Conclusions Among an older cohort of patients with GJTs, the authors found that a significant percentage of tumors do not grow after time of diagnosis and symptoms frequently remain stable for many years. Even with disease progression, most GJTs exhibit indolent growth (~1 mm/y) with slowly progressive cranial neuropathy affording satisfactory physiologic compensation in most patients. These data affirm that initial observation of GJTs is a viable option for patients of advanced age. Future studies are needed to determine if this strategy can be applied to younger patients who may harbor more aggressive disease. Author Contributions Matthew L. Carlson, concept, review of data, manuscript writing, manuscript revision; Alex D. Sweeney, concept, review of data, manuscript revision; George B. Wanna, concept, review of data, manuscript revision; James L. Netterville, concept, review of data, manuscript revision; David S. Haynes, concept, review of data, manuscript revision.
Disclosures Competing interests: James L. Netterville, consultant, Medtronic. David S. Haynes, consultant Advanced Bionics Corp, Cochlear Corp, MED-EL GmbH, Stryker/Synthes; Anspach Corp; Grace medical. Sponsorships: None. Funding source: None.
Supplemental Material Additional supporting information may be found at http://otojournal .org/supplemental.
References 1. Jackson CG, McGrew BM, Forest JA, et al. Lateral skull base surgery for glomus tumors: long-term control. Otol Neurotol. 2001;22:377-382. 2. Bacciu A, Ait Mimoune H, D’Orazio F, et al. Management of facial nerve in surgical treatment of previously untreated fisch class C tympanojugular paragangliomas: long-term results. J Neurol Surg B Skull Base. 2014;75:1-7. 3. Bacciu A, Medina M, Ait Mimoune H, et al. Lower cranial nerves function after surgical treatment of Fisch Class C and D tympanojugular paragangliomas [published online December 12, 2013]. Eur Arch Otorhinolaryngol. 4. Jackson CG, Haynes DS, Walker PA, et al. Hearing conservation in surgery for glomus jugulare tumors. Am J Otol. 1996; 17:425-437. 5. Sheehan JP, Tanaka S, Link MJ, et al. Gamma Knife surgery for the management of glomus tumors: a multicenter study. J Neurosurg. 2012;117:246-254. 6. Foote RL, Pollock BE, Gorman DA, et al. Glomus jugulare tumor: tumor control and complications after stereotactic radiosurgery. Head Neck. 2002;24:332-338. 7. Eustacchio S, Trummer M, Unger F, et al. The role of Gamma Knife radiosurgery in the management of glomus jugular tumours. Acta Neurochir Suppl. 2002;84:91-97. 8. van Hulsteijn LT, Corssmit EP, Coremans IE, et al. Regression and local control rates after radiotherapy for jugulotympanic paragangliomas: systematic review and meta-analysis. Radiother Oncol. 2013;106:161-168. 9. Ivan ME, Sughrue ME, Clark AJ, et al. A meta-analysis of tumor control rates and treatment-related morbidity for patients with glomus jugulare tumors. J Neurosurg. 2011;114:1299-1305. 10. Cosetti M, Linstrom C, Alexiades G, et al. Glomus tumors in patients of advanced age: a conservative approach. Laryngoscope. 2008;118:270-274. 11. Willen SN, Einstein DB, Maciunas RJ, et al. Treatment of glomus jugulare tumors in patients with advanced age: planned limited surgical resection followed by staged gamma knife radiosurgery: a preliminary report. Otol Neurotol. 2005;26:1229-1234. 12. Jansen JC, van den Berg R, Kuiper A, et al. Estimation of growth rate in patients with head and neck paragangliomas influences the treatment proposal. Cancer. 2000;88:2811-2816. 13. Prasad SC, Mimoune HA, D’Orazio F, et al. The role of waitand-scan and the efficacy of radiotherapy in the treatment of temporal bone paragangliomas. Otol Neurotol 2014;35:922931. 14. Miller T, Lau T, Vasan R, et al. Reporting success rates in the treatment of vestibular schwannomas: are we accounting for the natural history? J Clin Neurosci. 2014;21:914-918. 15. Oldring D, Fisch U. Glomus tumors of the temporal region: surgical therapy. Am J Otol. 1979;1:7-18. 16. Jackson CG, Glasscock ME III, Harris PF. Glomus tumors: diagnosis, classification, and management of large lesions. JAMA Otolaryngol Head Neck Surg. 1982;108:401-410.
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
Carlson et al
105
17. Lundin P, Pedersen F. Volume of pituitary macroadenomas: assessment by MRI. J Comput Assist Tomogr. 1992;16:519528. 18. van der Mey AG, Frijns JH, Cornelisse CJ, et al. 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;101:635-642.
19. Strasnick B, Glasscock ME III, Haynes D, et al. The natural history of untreated acoustic neuromas. Laryngoscope. 1994; 104:1115-1119. 20. Netterville JL, Civantos FJ. Rehabilitation of cranial nerve deficits after neurotologic skull base surgery. Laryngoscope. 1993;103:45-54.
Downloaded from oto.sagepub.com at Bobst Library, New York University on January 7, 2015
