The Laryngoscope C 2015 The American Laryngological, V
Rhinological and Otological Society, Inc.
Conservative Management of Vestibular Schwannoma: Predictors of Growth and Hearing Daniel Jethanamest, MD; Andrew M. Rivera, MD; Hongzhao Ji, MD; Venkatakarthikeyan Chokkalingam, MD; Fred F. Telischi, MD; Simon I. Angeli, MD Objective: To describe the clinical outcomes of patients undergoing serial observation for vestibular schwannoma (VS) and identify factors that may predict tumor growth or hearing loss. Study Design: Retrospective review. Methods: A retrospective review was conducted of patients seen at a tertiary care medical center between 2002 and 2013 with an International Classification of Diseases-9 diagnosis code of 225.1. Patients electing observation as initial management, with at least two documented imaging results, were included. Exclusion criteria comprised bilateral VS, diagnosis of neurofibromatosis type 2, and neoplasms other than VS. Decline in serviceable hearing, tumor growth, and changes in management strategy were recorded. Survival analysis to assess median time to outcomes and multiple logistic regression analyses were performed. Results: A total of 94 patients met inclusion criteria. While undergoing observation, 22.3% of patients underwent a change in management strategy to microsurgical excision or stereotactic radiotherapy. For patients with initial serviceable hearing, 24.3% observed a decline to a nonserviceable level. No significant clinical factors were identified to predict changes in hearing. Survival analysis revealed that an estimated 69.1% of patients electing observation as initial management continued to do so at 5 years. Imbalance or disequilibrium at presentation was found to be associated with an increased adjusted odds ratio (OR) (OR 2.96; 95% confidence interval, 1.03–8.50; P 5 0.04) for tumor growth. Conclusion: Serial observation of VS is a viable treatment strategy for selected patients, with two-thirds of patients electing to continue this management option after 5 years. Disequilibrium as a presenting symptom may be associated with subsequent tumor growth. Key Words: Vestibular schwannoma, acoustic neuroma, observation, conservative management, disequilibrium, imbalance, hearing loss, cranial nerve, tumor. Level of Evidence: 4. Laryngoscope, 00:000–000, 2015
INTRODUCTION Vestibular schwannomas (VS) are benign neoplasms of the eighth cranial nerve that derive from Schwann cells. They commonly present with asymmetric hearing loss that may be accompanied by tinnitus, dizziness, facial weakness, or a combination of these. The estimated annual incidence of these tumors ranges from approximately 0.6 to 1.9 per 100,000.1 This figure has been increasing, likely due to improved detection with evolving technology such as magnetic resonance imaging. In addition, due to an enhanced capability to detect
From the Department of Otolaryngology, New York University School of Medicine (D.J.), New York, New York; and the Department of Otolaryngology, University of Miami Miller School of Medicine (A.M.R., H.J., V.C., F.F.T., S.I.A.), Miami, Florida, U.S.A. Editor’s Note: This Manuscript was accepted for publication December 18, 2014. Presented as a poster at the Triological Society Combined Sections Meeting, Miami Beach, Florida, U.S.A., January 10–12, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Simon I. Angeli, MD, Department of Otolaryngology, University of Miami Miller School of Medicine, 1120 NW 14th St. 5th Floor, Miami, FL 33136. E-mail: [email protected] DOI: 10.1002/lary.25159
Laryngoscope 00: Month 2015
these lesions, tumor size at diagnosis has decreased significantly over time as well.2 Because our ability to identify and diagnose these lesions has improved, the treatment options have evolved and undergone refinement as well. Microsurgical resection and stereotactic radiotherapy have been the traditional mainstays of treatment; however, the detection of smaller tumors with minimal symptoms has led to an increasing trend toward conservative management with observation and serial imaging. Many factors influence the management of patients with VS, including the basic demographic factors of the patient, tumor size, hearing, clinical symptoms, comorbidities, and patient preferences. Because our understanding of the natural history of VS has improved, there has been a change in treatment trends toward conservative management without significant degradation of patients’ quality of life.3–5 Stangerup et al. reviewed their 35-year experience with the conservative management of these tumors in Denmark and found that 17% of intrameatal tumors grew within the first 5 years, whereas 28.9% of extrameatal tumors grew during the same time period. The authors found no correlation between tumor growth, gender, or age.6 Yoshimoto conducted a systematic review of the literature and reported Jethanamest et al.: Conservative Management of VS
1
tosis type 2, neoplasms other than VS, and insufficient followup or data. The outcomes measured were hearing changes to nonserviceable classification (50/50 pure tone average > 50 dB, word recognition score < 50%), tumor growth (1 mm/year), and change in management strategy. The independent variables at presentation that were assessed included: age, sex, laterality, symptoms (tinnitus, subjective hearing loss, vertigo, imbalance or disequilibrium, facial nerve dysfunction), tumor size, tumor location (internal auditory canal, cerebellopontine angle, or both), and fundal involvement. Tumor size was estimated by maximal linear measurement in any dimension; or if physical images were unavailable, it was based on radiologic reports if maximal dimension was clearly stated. Volumetric analysis could not be completed due to insufficient imaging data. Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 21.0 (IBM Corp., Armonk, NY). Kaplan-Meier survival analyses to assess median time to hearing loss, tumor growth, and change in management strategy were completed. Comparisons between groups for the various independent variables were performed with Student t test, Mann Whitney U test, Pearson’s chi-squared test, and Fisher’s exact test where appropriate, in addition to multiple logistic regression analyses. Fig. 1. Flow diagram of patients evaluated for inclusion in this study. ICD-9 5 International Classification of Diseases-9; VS 5 vestibular schwannoma. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
an annual tumor growth rate of approximately 1.2 millimeters per year.7 Changes in hearing have also been studied, and although tumor growth less than or equal to 2.5 millimeters per year has been shown to lead to higher rates of hearing preservation, no identifiable variable at the time of diagnosis has been found that can predict outcomes regarding change in tumor size or hearing.8,9 Although our understanding of the natural history of VS has improved, there is currently no consensus regarding their optimal management because the growth and hearing outcomes remain unpredictable at the time of diagnosis. Identification of clinical factors at presentation that can predict tumor growth or changes in hearing would be tremendously beneficial because it would allow for early recognition of patients who might be more likely to experience clinically significant tumor growth or deterioration in hearing. Such patients could potentially benefit from early intervention as opposed to conservative management. The objective of our study is to review our experience at a tertiary care referral center and describe the outcomes of the conservative management of VS with the aim of identifying clinical factors at presentation that can predict tumor growth or hearing loss.
MATERIALS AND METHODS A retrospective review was conducted of patients seen between January 2002 and May 2013 with an International Classification of Diseases (ICD)-9 diagnosis code of 225.1. Patients electing serial observation as initial management, with at least two documented imaging results, were included. Exclusion criteria comprised bilateral VS, diagnosis of neurofibroma-
Laryngoscope 00: Month 2015
2
TABLE I. Patient and Tumor Characteristics. Mean age
59.3 (SD 13.1)
Sex Female
62 (66%)
Male
32 (34%)
Symptoms Hearing loss at presentation
77 (81.9%)
Tinnitus at presentation
44 (46.8%)
Imbalance at presentation Vertigo at presentation
37 (39.4%) 22 (23.4%)
Facial nerve symptoms at presentation
3 (3.2%)
Median pure tone average1 Median word recognition score1 Change in management
42.5 dB (IQR 32.5 dB) 80.0% (IQR 60.0%)
Mean follow-up
34.8 months (SD 32.8 months)
Mean tumor size2 Mean tumor growth rate2
9.41 mm (SD 5.70 mm) 1.14 mm/year (SD 2.67 mm)
21 (22.3%)
Laterality Left Right
50 (53.2%) 44 (46.8%)
Location Internal auditory canal only Cerebellopontine angle only
58 (61.7%) 4 (4.3%)
Both
32 (34.1%)
Fundal involvement
15 (16%)
1
Seven cases with missing values. Three cases with missing values. dB 5 decibel; IQR 5 interquartile range; mm 5 millimeter; SD 5 standard deviation. 2
Jethanamest et al.: Conservative Management of VS
RESULTS A total of 976 patients were identified with the ICD-9 diagnosis code, and 94 met inclusion criteria. A flow diagram of the patients who were identified and the exclusions are shown in Figure 1. The baseline patient and tumor characteristics are shown in Table I. While undergoing serial observation, 22.3% of patients underwent a change in management strategy from observation to microsurgical excision or stereotactic radiotherapy, and 37.8% showed tumor growth. Among the patients undergoing observation, 75 had complete audiometric data at presentation and follow-up for review. Hearing status at baseline and after observation, as depicted by the American Academy of Otolaryngology–Head and Neck Surgery (AAO–HNS) Hearing Committee standard scattergram format, are presented in Figure 2.10 Among patients with initial serviceable hearing (n 5 37), nine patients (24.3%) observed a decline to nonserviceable levels at last follow-up. In multivariate regression analyses, no significant clinical factors were identified to predict changes in hearing. In univariate comparisons of patients with and without tumor growth 1 mm/year, two factors were found to be statistically significant: tinnitus at presentation (P 5 0.038, Pearson chi-squared) and disequilibrium at presentation (P 5 0.033, Pearson chi-squared). However, in multivariate analysis, only disequilibrium at presentation was statistically found to be significantly associated with tumor growth, adjusted odds ratio (OR) 2.962 (95% confidence interval [CI], 1.033–8.496; P 5 0.043). The presenting symptom of tinnitus was not statistically significant, OR 2.355 (95% CI, 0.933–5.948; P 5 0.07). Survival analysis revealed that, for patients with serviceable hearing at onset of observation, the median time to hearing loss worsening to a nonserviceable level (50/50 rule) was 76 months (Fig. 3), and the median time to detecting tumor growth ( 1 mm/year) was 67 months (Fig. 4). Using a change in management strategy away from initial observation as an endpoint, Kaplan-Meier survival analysis estimated a 5-year survival of continued serial observation of 69.1% (Fig. 5). Approximately two-thirds of patients electing observation as initial management continued to do so at 5 years.
DISCUSSION Conservative management of VS is an option for selected patients and is increasingly proposed as an initial strategy. Although benign, some tumors will grow or cause hearing loss and other morbidities, but both microsurgical excision and stereotactic radiotherapy carry potential risks and adverse effects as well. Many clinical scenarios, including asymptomatic patients with small tumors contemplating treatment, would benefit from indicators for future progression of symptoms. Currently, no prognostic factors or markers have been widely accepted to help select those patients at risk. Some authors have suggested that presenting symptoms may be associated with tumor growth, whereas many could not identify predictive factors for Laryngoscope 00: Month 2015
growth.11,12 Previously, a clinical growth index utilizing duration of clinical symptoms and maximum tumor diameter was proposed but later found to not correlate with tumor growth.13 Others have found no correlation between patient age, sex, or tumor laterality at presentation with subsequent tumor growth.14,15 Larger initial tumor size or early growth have been proposed as factors for future growth.11,16 Rather than demographic or radiologic features, symptoms at initial presentation have been suggested as possible predictors of future tumor behavior. Agrawal et al. studied 180 patients initially managed with serial observation and found the presence of tinnitus to be associated with a nearly three-fold adjusted OR of tumor growth.17 In that series, 37% of tumors showed growth defined as 1 mm/year or greater increase in tumor size, and 73% of patients had tinnitus at presentation. In this study, univariate analysis of patients with and without growing tumors found tinnitus to be a significant factor; however, this was not statistically significant in multiple logistic regression (P 5 0.07). A very similar growth rate was found in our series (37.8%). However, a much lower proportion (only 46.8%) of patients in our study were found to have tinnitus at presentation, which is below many other reports and the 73% of patients in the most recent Acoustic Neuroma Association survey.18 The generally high incidence of tinnitus may make it a relatively nonspecific predictive factor. Although not investigated in this retrospective review, the degree or other characteristics of tinnitus in VS potentially may be worthy of study as factors for future growth or intervention. One large review of 212 patients undergoing observation found that a presenting symptom of disequilibrium carried a substantial increased risk of failing observation, and the authors noted that the symptom of disequilibrium itself did not trigger any of the patients to leave the observation group.19 Unsteadiness was also one of a number of predictors for growth that was found in a model of high-risk profiles for tumor growth in 234 patients, in addition to the presence of an extrameatal component, short duration of hearing loss, and the lack of a sudden sensorineural loss.20 This study supports the importance of disequilibrium or imbalance at presentation as a potential significant predictive factor, with a nearly three-fold adjusted OR for future growth. Although fewer studies have searched for predictors of hearing loss, none have clearly emerged. Additionally, none of the independent variables available at presentation in this series was found to be associated with subsequent hearing loss. Changes in hearing during the conservative management of VS have been described in the literature, although primarily reported using broad classification schemes. This study contributes a large cohort, with detailed hearing data following recently adopted reporting standards of hearing outcomes presented in scattergram format for potential future data aggregation, as well as survival estimates graphically depicting the maintenance of serviceable hearing over time (Figs. 2 and 3). In a review of data from a national Danish database of over 2,500 VS patients, 17% of intrameatal tumors Jethanamest et al.: Conservative Management of VS
3
Fig. 2. (A) Scattergram of baseline PTA and WRS of study patients. (B) Scattergram showing changes in PTA and WRS during the period of observation. PTA 5 pure tone average; WRS 5 word recognition score. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
were found to increase in size to extrameatal extension, and 28.9% of intraextrameatal tumors were noted to increase in size during observation.6 The Danish review Laryngoscope 00: Month 2015
4
also noted that of patients initially with AAO class A hearing at presentation in the tumor ear, 45% lost class A hearing after 5 years. Using speech discrimination of Jethanamest et al.: Conservative Management of VS
Fig. 3. Kaplan-Meier survival plot of time to loss of serviceable hearing for patients starting with serviceable hearing who chose observation. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
70% or greater alone as a measure of good hearing, Stangerup et al. noted 59% of patients preserved good hearing after a mean of 4.7 years.9 A meta-analysis of studies from 1989 to 2003 on conservative management of VS comprising 1,345 patients found 43% of cases demonstrating positive growth during observation.11 Only a small subset of the patients in the analysis had hearing status available for analysis, but in that group hearing was considered preserved in 49%. Likely, some degree of patient selection bias and publication bias may contribute to the differences in rates of tumor growth between the national Danish data and the meta-analysis. In this series, a 37.8% rate of tumor growth was found, and for
Fig. 4. Kaplan-Meier survival plot of time to tumor growth defined as > 5 1 mm/year. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
Laryngoscope 00: Month 2015
Fig. 5. Kaplan-Meier survival plot of time to change in management for patients electing initial serial observation. At 5 years (dashed line), 69.1% of patients choose to continue this management strategy. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
patients who initially presented with serviceable hearing, as defined by the 50/50 rule, a total of 22.3% declined to nonserviceable levels at last follow-up. In survival analysis, the estimated median time to loss of serviceable hearing was 76 months (6.3 years). For patients with serviceable hearing who are contemplating either microsurgical intervention or stereotactic radiotherapy, these results on observation are useful for counseling. In addition to changes in hearing and tumor growth, perhaps one of the most important estimates for a patient choosing initial conservative management is the risk of requiring future interventions. In published reports, as in this study, potential factors including selection criteria for observation, physician preferences in decision making, and medical comorbidities may influence the ultimate individual patient choices of management. Nevertheless, the need to change from an initially agreed upon conservative strategy is a good reflection of the management’s effectiveness. In this series, KaplanMeier survival analysis estimated a 5-year survival of continued observation of 69.1%, which correlates with contemporary published literature. Although one large review of 202 patients reported a very low 9.4% failure rate of conservative management, the mean follow-up period was relatively brief at 2.48 years, and most other studies show higher rates.19 Fayad et al. reported on the outcomes of 114 patients undergoing conservative management of VS and found that 31% of patients required further treatment with surgery or radiation therapy, at a mean interval of 3.8 years.21 Conservative management in a separate study of 72 patients with long-term, median 121 months of follow-up, found a 35% rate of failure in conservative management.22 Similarly, a recent study of 180 patients reported that 35% of observed patients pursued a change in treatment Jethanamest et al.: Conservative Management of VS
5
strategy and that tumor growth was the primary factor prompting this change.10 This study is affected by selection bias and referral bias, with only a proportion of VS patients pursuing conservative management and with no formal protocol or criteria during the period of this retrospective review. Nevertheless, the majority of these selected patients were successfully observed, and available hearing outcome data is reported in scattergram format for future comparisons. Available imaging data limited tumor-size estimates to maximal linear dimensions rather than volumetric assessment in this study, although this has been noted to be potentially a better parameter for comparisons of tumor size than volume calculations in practice.23 Further study of predictive factors, and disequilibrium in particular at presentation, is warranted and may aid future risk stratification for patients. Median times to clinically significant outcomes such as nonserviceable hearing loss and the need to change management strategy provided by the survival analyses are valuable in counseling patients and managing expectations.
CONCLUSION Serial observation of VS is a viable treatment strategy for selected patients, with approximately two-thirds of patients electing to continue observation at 5 years. No factors or symptoms at presentation were identified to be predictive of future hearing loss; and for patients choosing observation with good initial hearing, the estimated median time to loss of serviceable hearing is 6.3 years. Disequilibrium or imbalance as a presenting symptom may be associated with an increased risk of subsequent tumor growth.
Acknowledgment The authors wish to acknowledge Ka-Ming Lo, MPH (University of Miami Department of Epidemiology and Public Health), for his assistance with statistical analysis.
BIBLIOGRAPHY 1. Babu R, Sharma R, Bagley JH, Hatef J, Friedman AH, Adamson C. Vestibular schwannomas in the modern era: epidemiology, treatment trends, and disparities in management. J. Neurosurg 2013;119:121–130.
Laryngoscope 00: Month 2015
6
2. Patel J, Vasan R, van Loveren H, Downes K, Agazzi S. The changing face of acoustic neuroma management in the USA: analysis of the 1998 and 2008 patient surveys from the acoustic neuroma association. Br J Neurosurg 2014;28:20–24. 3. Chen DA. Acoustic neuroma in a private neurotology practice: trends in demographics and practice patterns. Laryngoscope 2007;117:2003–2012. 4. Tan M, Myrie OA, Lin FR, et al. Trends in the management of vestibular schwannomas at Johns Hopkins 1997–2007. Laryngoscope 2010;120: 144–149. 5. Godefroy WP, Kaptein AA, Vogel JJ, van der Mey AGL. Conservative treatment of vestibular schwannoma: a follow-up study on clinical and quality-of-life outcome. Otol. Neurotol 2009;30:968–974. 6. Stangerup S-E, Caye-Thomasen P. Epidemiology and natural history of vestibular schwannomas. Otolaryngol Clin North Am 2012;45:257–268, vii. 7. Yoshimoto Y. Systematic review of the natural history of vestibular schwannoma. J Neurosurg 2005;103:59–63. 8. Stangerup S-E, Caye-Thomasen P, Tos M, Thomsen J. Change in hearing during “wait and scan” management of patients with vestibular schwannoma. J Laryngol Otol 2008;122:673–681. 9. Stangerup S-E, Thomsen J, Tos M, Caye-Thomasen P. Long-term hearing preservation in vestibular schwannoma. Otol Neurotol 2010;31:271–275. 10. Gurgel RK, Jackler RK, Dobie RA, Popelka GR. A new standardized format for reporting hearing outcome in clinical trials. Otolaryngol Head Neck Surg 2012;147:803–807. 11. Smouha EE, Yoo M, Mohr K, Davis RP. Conservative management of acoustic neuroma: a meta-analysis and proposed treatment algorithm. Laryngoscope 2005;115:450–454. 12. Timmer FCA, Artz JCJM, Beynon AJ, et al. Prediction of vestibular schwannoma growth: a novel rule based on clinical symptomatology. Ann Otol Rhinol Laryngol 2011;120:807–813. 13. Mohyuddin A, Vokurka EA, Evans DGR, Ramsden RT, Jackson A. Is clinical growth index a reliable predictor of tumour growth in vestibular schwannomas? Clin Otolaryngol Allied Sci 2003;28:85–90. 14. Herwadker A, Vokurka EA, Evans DGR, Ramsden RT, Jackson A. Size and growth rate of sporadic vestibular schwannoma: predictive value of information available at presentation. Otol Neurotol 2005;26:86–92. 15. Flint D, Fagan P, Panarese A. Conservative management of sporadic unilateral acoustic neuromas. J Laryngol Otol 2005;119:424–428. 16. Nutik SL, Babb MJ. Determinants of tumor size and growth in vestibular schwannomas. J Neurosurg 2001;94:922–926. 17. Agrawal Y, Clark JH, Limb CJ, Niparko JK, Francis HW. Predictors of vestibular schwannoma growth and clinical implications. Otol Neurotol 2010;31:807–812. 18. Van Gompel JJ, Patel J, Danner C, et al. Acoustic neuroma observation associated with an increase in symptomatic tinnitus: results of the 2007–2008 Acoustic Neuroma Association survey. J Neurosurg 2013;119: 864–868. 19. Malhotra PS, Sharma P, Fishman MA, et al. Clinical, radiographic, and audiometric predictors in conservative management of vestibular schwannoma. Otol Neurotol 2009;30:507–514. 20. Artz JCJM, Timmer FCA, Mulder JJS, Cremers CWRJ, Graamans K. Predictors of future growth of sporadic vestibular schwannomas obtained by history and radiologic assessment of the tumor. Eur Arch Otorhinolaryngol 2009;266:641–646. 21. Fayad JN, Semaan MT, Lin J, Berliner KI, Brackmann DE. Conservative Management of vestibular schwannoma: expectations based on the length of the observation period. Otol Neurotol 2014;35:1258–1265. doi: 10.1097/MAO.0000000000000285. 22. Hajioff D, Raut VV, Walsh RM, et al. Conservative management of vestibular schwannomas: third review of a 10-year prospective study. Clin Otolaryngol 2008;33:255–259. 23. Fiirgaard B, Pedersen CB, Lundorf E. The size of acoustic neuromas: CT and MRI. Neuroradiology 1997;39:599–601.
Jethanamest et al.: Conservative Management of VS
Rhinological and Otological Society, Inc.
Conservative Management of Vestibular Schwannoma: Predictors of Growth and Hearing Daniel Jethanamest, MD; Andrew M. Rivera, MD; Hongzhao Ji, MD; Venkatakarthikeyan Chokkalingam, MD; Fred F. Telischi, MD; Simon I. Angeli, MD Objective: To describe the clinical outcomes of patients undergoing serial observation for vestibular schwannoma (VS) and identify factors that may predict tumor growth or hearing loss. Study Design: Retrospective review. Methods: A retrospective review was conducted of patients seen at a tertiary care medical center between 2002 and 2013 with an International Classification of Diseases-9 diagnosis code of 225.1. Patients electing observation as initial management, with at least two documented imaging results, were included. Exclusion criteria comprised bilateral VS, diagnosis of neurofibromatosis type 2, and neoplasms other than VS. Decline in serviceable hearing, tumor growth, and changes in management strategy were recorded. Survival analysis to assess median time to outcomes and multiple logistic regression analyses were performed. Results: A total of 94 patients met inclusion criteria. While undergoing observation, 22.3% of patients underwent a change in management strategy to microsurgical excision or stereotactic radiotherapy. For patients with initial serviceable hearing, 24.3% observed a decline to a nonserviceable level. No significant clinical factors were identified to predict changes in hearing. Survival analysis revealed that an estimated 69.1% of patients electing observation as initial management continued to do so at 5 years. Imbalance or disequilibrium at presentation was found to be associated with an increased adjusted odds ratio (OR) (OR 2.96; 95% confidence interval, 1.03–8.50; P 5 0.04) for tumor growth. Conclusion: Serial observation of VS is a viable treatment strategy for selected patients, with two-thirds of patients electing to continue this management option after 5 years. Disequilibrium as a presenting symptom may be associated with subsequent tumor growth. Key Words: Vestibular schwannoma, acoustic neuroma, observation, conservative management, disequilibrium, imbalance, hearing loss, cranial nerve, tumor. Level of Evidence: 4. Laryngoscope, 00:000–000, 2015
INTRODUCTION Vestibular schwannomas (VS) are benign neoplasms of the eighth cranial nerve that derive from Schwann cells. They commonly present with asymmetric hearing loss that may be accompanied by tinnitus, dizziness, facial weakness, or a combination of these. The estimated annual incidence of these tumors ranges from approximately 0.6 to 1.9 per 100,000.1 This figure has been increasing, likely due to improved detection with evolving technology such as magnetic resonance imaging. In addition, due to an enhanced capability to detect
From the Department of Otolaryngology, New York University School of Medicine (D.J.), New York, New York; and the Department of Otolaryngology, University of Miami Miller School of Medicine (A.M.R., H.J., V.C., F.F.T., S.I.A.), Miami, Florida, U.S.A. Editor’s Note: This Manuscript was accepted for publication December 18, 2014. Presented as a poster at the Triological Society Combined Sections Meeting, Miami Beach, Florida, U.S.A., January 10–12, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Simon I. Angeli, MD, Department of Otolaryngology, University of Miami Miller School of Medicine, 1120 NW 14th St. 5th Floor, Miami, FL 33136. E-mail: [email protected] DOI: 10.1002/lary.25159
Laryngoscope 00: Month 2015
these lesions, tumor size at diagnosis has decreased significantly over time as well.2 Because our ability to identify and diagnose these lesions has improved, the treatment options have evolved and undergone refinement as well. Microsurgical resection and stereotactic radiotherapy have been the traditional mainstays of treatment; however, the detection of smaller tumors with minimal symptoms has led to an increasing trend toward conservative management with observation and serial imaging. Many factors influence the management of patients with VS, including the basic demographic factors of the patient, tumor size, hearing, clinical symptoms, comorbidities, and patient preferences. Because our understanding of the natural history of VS has improved, there has been a change in treatment trends toward conservative management without significant degradation of patients’ quality of life.3–5 Stangerup et al. reviewed their 35-year experience with the conservative management of these tumors in Denmark and found that 17% of intrameatal tumors grew within the first 5 years, whereas 28.9% of extrameatal tumors grew during the same time period. The authors found no correlation between tumor growth, gender, or age.6 Yoshimoto conducted a systematic review of the literature and reported Jethanamest et al.: Conservative Management of VS
1
tosis type 2, neoplasms other than VS, and insufficient followup or data. The outcomes measured were hearing changes to nonserviceable classification (50/50 pure tone average > 50 dB, word recognition score < 50%), tumor growth (1 mm/year), and change in management strategy. The independent variables at presentation that were assessed included: age, sex, laterality, symptoms (tinnitus, subjective hearing loss, vertigo, imbalance or disequilibrium, facial nerve dysfunction), tumor size, tumor location (internal auditory canal, cerebellopontine angle, or both), and fundal involvement. Tumor size was estimated by maximal linear measurement in any dimension; or if physical images were unavailable, it was based on radiologic reports if maximal dimension was clearly stated. Volumetric analysis could not be completed due to insufficient imaging data. Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 21.0 (IBM Corp., Armonk, NY). Kaplan-Meier survival analyses to assess median time to hearing loss, tumor growth, and change in management strategy were completed. Comparisons between groups for the various independent variables were performed with Student t test, Mann Whitney U test, Pearson’s chi-squared test, and Fisher’s exact test where appropriate, in addition to multiple logistic regression analyses. Fig. 1. Flow diagram of patients evaluated for inclusion in this study. ICD-9 5 International Classification of Diseases-9; VS 5 vestibular schwannoma. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
an annual tumor growth rate of approximately 1.2 millimeters per year.7 Changes in hearing have also been studied, and although tumor growth less than or equal to 2.5 millimeters per year has been shown to lead to higher rates of hearing preservation, no identifiable variable at the time of diagnosis has been found that can predict outcomes regarding change in tumor size or hearing.8,9 Although our understanding of the natural history of VS has improved, there is currently no consensus regarding their optimal management because the growth and hearing outcomes remain unpredictable at the time of diagnosis. Identification of clinical factors at presentation that can predict tumor growth or changes in hearing would be tremendously beneficial because it would allow for early recognition of patients who might be more likely to experience clinically significant tumor growth or deterioration in hearing. Such patients could potentially benefit from early intervention as opposed to conservative management. The objective of our study is to review our experience at a tertiary care referral center and describe the outcomes of the conservative management of VS with the aim of identifying clinical factors at presentation that can predict tumor growth or hearing loss.
MATERIALS AND METHODS A retrospective review was conducted of patients seen between January 2002 and May 2013 with an International Classification of Diseases (ICD)-9 diagnosis code of 225.1. Patients electing serial observation as initial management, with at least two documented imaging results, were included. Exclusion criteria comprised bilateral VS, diagnosis of neurofibroma-
Laryngoscope 00: Month 2015
2
TABLE I. Patient and Tumor Characteristics. Mean age
59.3 (SD 13.1)
Sex Female
62 (66%)
Male
32 (34%)
Symptoms Hearing loss at presentation
77 (81.9%)
Tinnitus at presentation
44 (46.8%)
Imbalance at presentation Vertigo at presentation
37 (39.4%) 22 (23.4%)
Facial nerve symptoms at presentation
3 (3.2%)
Median pure tone average1 Median word recognition score1 Change in management
42.5 dB (IQR 32.5 dB) 80.0% (IQR 60.0%)
Mean follow-up
34.8 months (SD 32.8 months)
Mean tumor size2 Mean tumor growth rate2
9.41 mm (SD 5.70 mm) 1.14 mm/year (SD 2.67 mm)
21 (22.3%)
Laterality Left Right
50 (53.2%) 44 (46.8%)
Location Internal auditory canal only Cerebellopontine angle only
58 (61.7%) 4 (4.3%)
Both
32 (34.1%)
Fundal involvement
15 (16%)
1
Seven cases with missing values. Three cases with missing values. dB 5 decibel; IQR 5 interquartile range; mm 5 millimeter; SD 5 standard deviation. 2
Jethanamest et al.: Conservative Management of VS
RESULTS A total of 976 patients were identified with the ICD-9 diagnosis code, and 94 met inclusion criteria. A flow diagram of the patients who were identified and the exclusions are shown in Figure 1. The baseline patient and tumor characteristics are shown in Table I. While undergoing serial observation, 22.3% of patients underwent a change in management strategy from observation to microsurgical excision or stereotactic radiotherapy, and 37.8% showed tumor growth. Among the patients undergoing observation, 75 had complete audiometric data at presentation and follow-up for review. Hearing status at baseline and after observation, as depicted by the American Academy of Otolaryngology–Head and Neck Surgery (AAO–HNS) Hearing Committee standard scattergram format, are presented in Figure 2.10 Among patients with initial serviceable hearing (n 5 37), nine patients (24.3%) observed a decline to nonserviceable levels at last follow-up. In multivariate regression analyses, no significant clinical factors were identified to predict changes in hearing. In univariate comparisons of patients with and without tumor growth 1 mm/year, two factors were found to be statistically significant: tinnitus at presentation (P 5 0.038, Pearson chi-squared) and disequilibrium at presentation (P 5 0.033, Pearson chi-squared). However, in multivariate analysis, only disequilibrium at presentation was statistically found to be significantly associated with tumor growth, adjusted odds ratio (OR) 2.962 (95% confidence interval [CI], 1.033–8.496; P 5 0.043). The presenting symptom of tinnitus was not statistically significant, OR 2.355 (95% CI, 0.933–5.948; P 5 0.07). Survival analysis revealed that, for patients with serviceable hearing at onset of observation, the median time to hearing loss worsening to a nonserviceable level (50/50 rule) was 76 months (Fig. 3), and the median time to detecting tumor growth ( 1 mm/year) was 67 months (Fig. 4). Using a change in management strategy away from initial observation as an endpoint, Kaplan-Meier survival analysis estimated a 5-year survival of continued serial observation of 69.1% (Fig. 5). Approximately two-thirds of patients electing observation as initial management continued to do so at 5 years.
DISCUSSION Conservative management of VS is an option for selected patients and is increasingly proposed as an initial strategy. Although benign, some tumors will grow or cause hearing loss and other morbidities, but both microsurgical excision and stereotactic radiotherapy carry potential risks and adverse effects as well. Many clinical scenarios, including asymptomatic patients with small tumors contemplating treatment, would benefit from indicators for future progression of symptoms. Currently, no prognostic factors or markers have been widely accepted to help select those patients at risk. Some authors have suggested that presenting symptoms may be associated with tumor growth, whereas many could not identify predictive factors for Laryngoscope 00: Month 2015
growth.11,12 Previously, a clinical growth index utilizing duration of clinical symptoms and maximum tumor diameter was proposed but later found to not correlate with tumor growth.13 Others have found no correlation between patient age, sex, or tumor laterality at presentation with subsequent tumor growth.14,15 Larger initial tumor size or early growth have been proposed as factors for future growth.11,16 Rather than demographic or radiologic features, symptoms at initial presentation have been suggested as possible predictors of future tumor behavior. Agrawal et al. studied 180 patients initially managed with serial observation and found the presence of tinnitus to be associated with a nearly three-fold adjusted OR of tumor growth.17 In that series, 37% of tumors showed growth defined as 1 mm/year or greater increase in tumor size, and 73% of patients had tinnitus at presentation. In this study, univariate analysis of patients with and without growing tumors found tinnitus to be a significant factor; however, this was not statistically significant in multiple logistic regression (P 5 0.07). A very similar growth rate was found in our series (37.8%). However, a much lower proportion (only 46.8%) of patients in our study were found to have tinnitus at presentation, which is below many other reports and the 73% of patients in the most recent Acoustic Neuroma Association survey.18 The generally high incidence of tinnitus may make it a relatively nonspecific predictive factor. Although not investigated in this retrospective review, the degree or other characteristics of tinnitus in VS potentially may be worthy of study as factors for future growth or intervention. One large review of 212 patients undergoing observation found that a presenting symptom of disequilibrium carried a substantial increased risk of failing observation, and the authors noted that the symptom of disequilibrium itself did not trigger any of the patients to leave the observation group.19 Unsteadiness was also one of a number of predictors for growth that was found in a model of high-risk profiles for tumor growth in 234 patients, in addition to the presence of an extrameatal component, short duration of hearing loss, and the lack of a sudden sensorineural loss.20 This study supports the importance of disequilibrium or imbalance at presentation as a potential significant predictive factor, with a nearly three-fold adjusted OR for future growth. Although fewer studies have searched for predictors of hearing loss, none have clearly emerged. Additionally, none of the independent variables available at presentation in this series was found to be associated with subsequent hearing loss. Changes in hearing during the conservative management of VS have been described in the literature, although primarily reported using broad classification schemes. This study contributes a large cohort, with detailed hearing data following recently adopted reporting standards of hearing outcomes presented in scattergram format for potential future data aggregation, as well as survival estimates graphically depicting the maintenance of serviceable hearing over time (Figs. 2 and 3). In a review of data from a national Danish database of over 2,500 VS patients, 17% of intrameatal tumors Jethanamest et al.: Conservative Management of VS
3
Fig. 2. (A) Scattergram of baseline PTA and WRS of study patients. (B) Scattergram showing changes in PTA and WRS during the period of observation. PTA 5 pure tone average; WRS 5 word recognition score. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
were found to increase in size to extrameatal extension, and 28.9% of intraextrameatal tumors were noted to increase in size during observation.6 The Danish review Laryngoscope 00: Month 2015
4
also noted that of patients initially with AAO class A hearing at presentation in the tumor ear, 45% lost class A hearing after 5 years. Using speech discrimination of Jethanamest et al.: Conservative Management of VS
Fig. 3. Kaplan-Meier survival plot of time to loss of serviceable hearing for patients starting with serviceable hearing who chose observation. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
70% or greater alone as a measure of good hearing, Stangerup et al. noted 59% of patients preserved good hearing after a mean of 4.7 years.9 A meta-analysis of studies from 1989 to 2003 on conservative management of VS comprising 1,345 patients found 43% of cases demonstrating positive growth during observation.11 Only a small subset of the patients in the analysis had hearing status available for analysis, but in that group hearing was considered preserved in 49%. Likely, some degree of patient selection bias and publication bias may contribute to the differences in rates of tumor growth between the national Danish data and the meta-analysis. In this series, a 37.8% rate of tumor growth was found, and for
Fig. 4. Kaplan-Meier survival plot of time to tumor growth defined as > 5 1 mm/year. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
Laryngoscope 00: Month 2015
Fig. 5. Kaplan-Meier survival plot of time to change in management for patients electing initial serial observation. At 5 years (dashed line), 69.1% of patients choose to continue this management strategy. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
patients who initially presented with serviceable hearing, as defined by the 50/50 rule, a total of 22.3% declined to nonserviceable levels at last follow-up. In survival analysis, the estimated median time to loss of serviceable hearing was 76 months (6.3 years). For patients with serviceable hearing who are contemplating either microsurgical intervention or stereotactic radiotherapy, these results on observation are useful for counseling. In addition to changes in hearing and tumor growth, perhaps one of the most important estimates for a patient choosing initial conservative management is the risk of requiring future interventions. In published reports, as in this study, potential factors including selection criteria for observation, physician preferences in decision making, and medical comorbidities may influence the ultimate individual patient choices of management. Nevertheless, the need to change from an initially agreed upon conservative strategy is a good reflection of the management’s effectiveness. In this series, KaplanMeier survival analysis estimated a 5-year survival of continued observation of 69.1%, which correlates with contemporary published literature. Although one large review of 202 patients reported a very low 9.4% failure rate of conservative management, the mean follow-up period was relatively brief at 2.48 years, and most other studies show higher rates.19 Fayad et al. reported on the outcomes of 114 patients undergoing conservative management of VS and found that 31% of patients required further treatment with surgery or radiation therapy, at a mean interval of 3.8 years.21 Conservative management in a separate study of 72 patients with long-term, median 121 months of follow-up, found a 35% rate of failure in conservative management.22 Similarly, a recent study of 180 patients reported that 35% of observed patients pursued a change in treatment Jethanamest et al.: Conservative Management of VS
5
strategy and that tumor growth was the primary factor prompting this change.10 This study is affected by selection bias and referral bias, with only a proportion of VS patients pursuing conservative management and with no formal protocol or criteria during the period of this retrospective review. Nevertheless, the majority of these selected patients were successfully observed, and available hearing outcome data is reported in scattergram format for future comparisons. Available imaging data limited tumor-size estimates to maximal linear dimensions rather than volumetric assessment in this study, although this has been noted to be potentially a better parameter for comparisons of tumor size than volume calculations in practice.23 Further study of predictive factors, and disequilibrium in particular at presentation, is warranted and may aid future risk stratification for patients. Median times to clinically significant outcomes such as nonserviceable hearing loss and the need to change management strategy provided by the survival analyses are valuable in counseling patients and managing expectations.
CONCLUSION Serial observation of VS is a viable treatment strategy for selected patients, with approximately two-thirds of patients electing to continue observation at 5 years. No factors or symptoms at presentation were identified to be predictive of future hearing loss; and for patients choosing observation with good initial hearing, the estimated median time to loss of serviceable hearing is 6.3 years. Disequilibrium or imbalance as a presenting symptom may be associated with an increased risk of subsequent tumor growth.
Acknowledgment The authors wish to acknowledge Ka-Ming Lo, MPH (University of Miami Department of Epidemiology and Public Health), for his assistance with statistical analysis.
BIBLIOGRAPHY 1. Babu R, Sharma R, Bagley JH, Hatef J, Friedman AH, Adamson C. Vestibular schwannomas in the modern era: epidemiology, treatment trends, and disparities in management. J. Neurosurg 2013;119:121–130.
Laryngoscope 00: Month 2015
6
2. Patel J, Vasan R, van Loveren H, Downes K, Agazzi S. The changing face of acoustic neuroma management in the USA: analysis of the 1998 and 2008 patient surveys from the acoustic neuroma association. Br J Neurosurg 2014;28:20–24. 3. Chen DA. Acoustic neuroma in a private neurotology practice: trends in demographics and practice patterns. Laryngoscope 2007;117:2003–2012. 4. Tan M, Myrie OA, Lin FR, et al. Trends in the management of vestibular schwannomas at Johns Hopkins 1997–2007. Laryngoscope 2010;120: 144–149. 5. Godefroy WP, Kaptein AA, Vogel JJ, van der Mey AGL. Conservative treatment of vestibular schwannoma: a follow-up study on clinical and quality-of-life outcome. Otol. Neurotol 2009;30:968–974. 6. Stangerup S-E, Caye-Thomasen P. Epidemiology and natural history of vestibular schwannomas. Otolaryngol Clin North Am 2012;45:257–268, vii. 7. Yoshimoto Y. Systematic review of the natural history of vestibular schwannoma. J Neurosurg 2005;103:59–63. 8. Stangerup S-E, Caye-Thomasen P, Tos M, Thomsen J. Change in hearing during “wait and scan” management of patients with vestibular schwannoma. J Laryngol Otol 2008;122:673–681. 9. Stangerup S-E, Thomsen J, Tos M, Caye-Thomasen P. Long-term hearing preservation in vestibular schwannoma. Otol Neurotol 2010;31:271–275. 10. Gurgel RK, Jackler RK, Dobie RA, Popelka GR. A new standardized format for reporting hearing outcome in clinical trials. Otolaryngol Head Neck Surg 2012;147:803–807. 11. Smouha EE, Yoo M, Mohr K, Davis RP. Conservative management of acoustic neuroma: a meta-analysis and proposed treatment algorithm. Laryngoscope 2005;115:450–454. 12. Timmer FCA, Artz JCJM, Beynon AJ, et al. Prediction of vestibular schwannoma growth: a novel rule based on clinical symptomatology. Ann Otol Rhinol Laryngol 2011;120:807–813. 13. Mohyuddin A, Vokurka EA, Evans DGR, Ramsden RT, Jackson A. Is clinical growth index a reliable predictor of tumour growth in vestibular schwannomas? Clin Otolaryngol Allied Sci 2003;28:85–90. 14. Herwadker A, Vokurka EA, Evans DGR, Ramsden RT, Jackson A. Size and growth rate of sporadic vestibular schwannoma: predictive value of information available at presentation. Otol Neurotol 2005;26:86–92. 15. Flint D, Fagan P, Panarese A. Conservative management of sporadic unilateral acoustic neuromas. J Laryngol Otol 2005;119:424–428. 16. Nutik SL, Babb MJ. Determinants of tumor size and growth in vestibular schwannomas. J Neurosurg 2001;94:922–926. 17. Agrawal Y, Clark JH, Limb CJ, Niparko JK, Francis HW. Predictors of vestibular schwannoma growth and clinical implications. Otol Neurotol 2010;31:807–812. 18. Van Gompel JJ, Patel J, Danner C, et al. Acoustic neuroma observation associated with an increase in symptomatic tinnitus: results of the 2007–2008 Acoustic Neuroma Association survey. J Neurosurg 2013;119: 864–868. 19. Malhotra PS, Sharma P, Fishman MA, et al. Clinical, radiographic, and audiometric predictors in conservative management of vestibular schwannoma. Otol Neurotol 2009;30:507–514. 20. Artz JCJM, Timmer FCA, Mulder JJS, Cremers CWRJ, Graamans K. Predictors of future growth of sporadic vestibular schwannomas obtained by history and radiologic assessment of the tumor. Eur Arch Otorhinolaryngol 2009;266:641–646. 21. Fayad JN, Semaan MT, Lin J, Berliner KI, Brackmann DE. Conservative Management of vestibular schwannoma: expectations based on the length of the observation period. Otol Neurotol 2014;35:1258–1265. doi: 10.1097/MAO.0000000000000285. 22. Hajioff D, Raut VV, Walsh RM, et al. Conservative management of vestibular schwannomas: third review of a 10-year prospective study. Clin Otolaryngol 2008;33:255–259. 23. Fiirgaard B, Pedersen CB, Lundorf E. The size of acoustic neuromas: CT and MRI. Neuroradiology 1997;39:599–601.
Jethanamest et al.: Conservative Management of VS
