Life Expectancy After Diagnosis of a Vestibular Schwannoma in Patients 70 Years and Older

Key Points

Question  What is the survival after diagnosis of a vestibular schwannoma (VS) in Danish patients 70 years and older, treated with either observation or surgery, and is the mean survival different from the age-matched background population in Denmark?

Findings  This cohort study of 624 patients found no significant difference in the life span between patients with VS, whether conservatively or surgically managed, and the matched general population.

Meaning  The findings show that advanced age among patients with VS is not a contraindication itself for cases where surgery is considered, and untreated tumor is not significantly associated with life span.

Importance  Over the past decades, the number of patients, especially in the older adult patient group, diagnosed with vestibular schwannoma (VS) has increased. Assuming that older adult patients have more comorbidities, a longer recovery period after surgery, a higher rate of surgical complications, and a higher mortality rate after VS surgery, a treatment strategy for this group of patients is warranted, based on clinical evidence on postsurgical survival.

Objective  To evaluate the survival after diagnosis of a VS in patients 70 years and older, treated with either observation or surgery, and to compare these findings with the life span of an age-matched background population in Denmark.

Design, Setting, and Participants  This was a retrospective cohort study of 624 patients 70 years and older diagnosed with VS in Denmark from 1976 to 2016. Since 1976, all patients with a VS have been registered in a national database, which contains 3637 patients. Of the included patients in this study, 477 were treated conservatively with the “wait-and-scan” strategy, and 147 were treated surgically with removal of the tumor. The survival of the patients was compared with a matched background population in Denmark. Data analysis was performed from January 1976 to January 2017.

Exposures  Surgery, radiotherapy, or none.

Main Outcomes and Measures  The main outcome was survival among the patients and compared with the matched background population.

Results  A total of 624 patients were included (317 female patients [50.8%] and 307 male patients [49.2%]). The mean (SD) survival in the observed patients was 9.2 (4.7) years after diagnosis, whereas for the background population, the expected survival was 11 years from the mean age at diagnosis. For the surgically treated patients, the mean (SD) survival was 11.8 (6.6) years, and expected survival was 11 years for the matched background population. The mean (SD) survival was 10.7 (5.5) years in female patients and 8.9 (5.0) years in male patients. There was no significant difference in survival between treatment modalities, irrespective of tumor size.

Conclusions and Relevance  In this cohort study, survival after diagnosis of a VS in patients 70 years and older was similar in the surgical group compared with the age-matched background population. In the wait-and-scan group, the survival after diagnosis was marginally shorter, which may be associated with increased comorbidity.

Over the past 2 decades particularly, there has been an enormous increase in the number of patients diagnosed with vestibular schwannoma (VS) in Denmark.¹ In the same period, the mean size of the tumors has decreased, and the mean age at diagnosis has increased in both sexes, with slightly more men being diagnosed in recent years and women having higher occurrence of extrameatal tumors and hence statistically significant larger tumors.² Consequently, there is a growing group of older adult patients with VS, which warrants an optimal treatment strategy.^3,4 Recent multicenter data have provided more insight in growth patterns in observed VS, reporting 5-year growth-free survival rates of 32% for intrameatal VS and 22% for extrameatal VS, although these numbers do not focus on the older adults with VS.⁵

Since magnetic resonance imaging (MRI) scans have become widely accessible, it has become possible to observe the natural history of the tumor behavior, and it has been shown that only about 30% of the tumors grow during an observation period.⁶ The most recent publication from our treatment center showed no growth of intrameatal VS in 75% of the cases and no growth of extrameatal VS in nearly 60% of the cases during 10 years of observation after diagnosis.⁷

This, together with other studies showing conservative management to be the most beneficial treatment considering life quality and acceptable tumor control in regard to symptoms, eg, hearing impairment and tinnitus, has led to the acceptance of a “wait-and-scan” treatment for small and some medium-sized tumors when managing the treatment of patients with VS. Especially in older adult patients, this is the preferred treatment because of the lower expected residual life span, multiple comorbidities, and higher risk of surgical complications.^3,8-10 That being said, other studies on VS in older adult patients have concluded that surgery is safe in patients older than 70 years, regarding surgical complications specific to VS removal, and reported no higher risk of stroke, bleeding, cardiac events, or mortality.¹¹ Furthermore, studies have underlined that comorbidities and dizziness are prevalent in older adult patients, yet no significant differences exist after surgery in hard parameters, such as cranial nerve damage or general surgical complications.^12-14 The aim of this study was to evaluate the postdiagnostic survival of conservatively and surgically treated patients 70 years and older and to compare these findings with the expected life span of an age-matched population in Denmark.

Since 1976, all patients with a diagnosed sporadic (unilateral) VS have been registered in a national database. Those data include, among others, information on age, sex, tumor size, localization, operation date, date of death, and hence, survival time. It includes 3637 patients. During the period from January 1976 to January 2016, 631 patients 70 years and older have been diagnosed with a unilateral VS. A total of 7 patients had missing data on tumor size and were excluded from this study. Only 10 patients in the study were treated with radiotherapy, 3 of whom had previous VS surgery, which is why this group was excluded from the study. To update the length of the survival after diagnosis, the data of all the included patients were updated through 2022. The Danish Data Protection Agency institutional review board approved the collection of patient data for the national VS database. Oral informed consent was obtained from each patient.

Based on the treatment they received, the patients were divided into surgical and nonsurgical groups, and the tumors were classified as either pure intrameatal tumors or as intrameatal and extrameatal tumors extending into the cerebellopontine angle. The purely intrameatal tumors have no size indication. Tumors extending into the cerebellopontine angle were determined by linear measurements by the largest extrameatal diameter only and subdivided into small (1-10 mm), medium (11-20 mm), moderately large (21-30 mm), large (31-40 mm), and giant (>40 mm), according to the consensus meeting in Japan in 2003.¹⁵

The mean survival time after diagnosis was calculated separately for the 2 groups in both men and women. These results were compared with the overall population, defined as the control group, by using reports from the Danish Statistical Institute,¹⁶ from which the expected survival in the background population was acquired.

Statistical analyses were performed in RStudio, version 2022.12.0 (R Foundation for Statistical Computing). Survival analyses were carried out using the packages survminer and/or survival. Survival probability estimates were calculated using Kaplan-Meier curves and compared with log-rank tests. Univariable analyses were performed with the Cox regression analysis method. The Pearson correlation coefficient was computed using the ggpubr package to assess the linear association between number of diagnosed and observed VS and the corresponding year. However, data were not found to be normally distributed (confirmed by a Shapiro-Wilk normality test), which is why a Spearman correlation coefficient calculation was performed as well. The test was computed using the ggpubr package as a rank-based measure of correlation. All P values were 2-sided with a significance level of .05.

A total of 624 patients were included (317 female patients [50.8%] and 307 male patients [49.2%]). The overall mean (SD) survival for both male and female patients was 9.2 (4.7) years in the observation group and 11.8 (6.6) years in the surgery group. Generally, the mean (SD) survival was 10.7 (5.5) years in female patients and 8.9 (5.0) years in males. In the 40-year period from 1976 to 2016, the annual number of diagnosed VS among older adult patients in Denmark had increased from 5 to 228 per 5-year periods. Due to the easier access to MRI scans, the treatment strategy has changed since the 1990s to more wait-and-scan treatment, especially in older adult patients. In fact, the annual percentage of the wait-and-scan treatment has increased from 3% in the first 5-year period to 89% in the last 5-year period from 2011 to 2016. Figure 1 shows a steadily increasing correlation, with coefficients of 0.86 (95% CI, 0.74-0.92) for diagnosed cases (Figure 1A) and 0.89 (95% CI, 0.74-0.92) for observed cases (Figure 1B). The Spearman correlation coefficients of 0.94 (P < .001; diagnosed cases), and 1.00 (P < .001; observed cases) indicate a strong correlation between number of diagnosed VS and increasing year of observation, from 1 in 1976 to 58 in 2015, as well as the number of observed VS and years, factoring in deaths over the years. Of the 624 patients, 373 died during the 40-year period.

Of the included patients, 477 (76.4%) were treated nonsurgically—223 female and 254 male patients. Overall the mean (SD) survival was 9.2 (4.7) years. In the female group, the mean age at the time of diagnosis was 75.0 years, 98 tumors (43.9%) were purely intrameatal, and the mean tumor size of the extrameatal tumors was 10.6 mm. In the male group, the mean age at the time of diagnosis was 74.5 years, 112 tumors (44.1%) were purely intrameatal, and the mean tumor size of the extrameatal tumors was 11.5 mm (Table).

For female patients, the mean survival was 10.0 years after diagnosis; for the background female population, the survival was 11.9 years. The mean survival after diagnosis in the male patients was 8.6 years; for the background male population, the survival was 10.1 years (Table).

In this group, 147 patients, 94 female and 53 male patients, were treated surgically, either just after diagnosis or, because of tumor growth, after some period of observation. Most of the surgical procedures were translabyrinthine resections (127 of 147 [86.4%]), except in cases where hearing preservation was an option, in which retrolabyrinthine procedures were performed. In the beginning of the study, some of the cases were also suboccipital (13 of 147 [8.8%]). Other approaches included retrosigmoid (6 of 147 [4.1%]) and middle cranial fossa (1 of 147 [0.7%]). In a few patients, intraoperative frozen-section diagnosis revealed a meningioma. In the female group, 1 tumor (1%) was purely intrameatal. The mean tumor size of the extrameatal tumors was 26.3 mm, and the mean (SD) age at the time of diagnosis was 72.7 (2.6) years. The mean (SD) survival was 12.6 (6.4) years after diagnosis; for the background female population, the survival was 12.2 years (Table). In the male group, 2 tumors (4%) were purely intrameatal. The mean tumor size of the extrameatal tumors was 23.7 mm. The mean (SD) age at the time of diagnosis was 73.7 (3.6) years. The mean (SD) survival was 10.4 (6.8) years after diagnosis. For the background male population, the survival was 9.4 years (Table). The operations were mostly performed in patients in their early 70s, with survival lengths of up to 27 years, with the oldest patient, aged 86 years, surviving 4 years after surgery, and demonstrating an expected tendency of decreased survival time and number of patients undergoing surgery as the patients get older.

The survival comparison of both groups was calculated and depicted by Kaplan-Meier curves, and no significant differences between the length of survival after diagnosis in the observed and the surgically treated patients were present (hazard ratio [HR], 0.86; 95% CI, 0.69-1.09; P = .20). Among the observed patients, there was a 5-year survival rate of 81.6% (95% CI, 75.6%-88.1%). Compared with the surgically treated patients, the 5-year survival rate was 81.1% (95% CI, 77.7%-84.7%), showing almost no difference in survival between the 2 treatment modalities (Figure 2A). The survival in association with tumor size was determined by dividing the tumors into 3 clinically meaningful groups: intrameatal, small to medium-sized tumors (1-20 mm), and moderately large tumors and above (>20 mm). Overall, there was no difference between the 3 groups. When looking at the surgically treated patients with a small to medium-sized tumor, the HR was 1.11 (95% CI, 0.15-8.12), P = .92, and 1.57 (95% CI, 0.22-11.35), P = .65, for tumors greater than 20 mm (Figure 2B), compared with intrameatal tumors, showing an almost similar survival probability between the different tumor sizes. Compared with the observed patients with a small to medium-sized tumor, the HR was 1.25 (95% CI, 0.96-1.62), P = .10, and 1.29 (95% CI, 0.74-2.24), P = .36, for tumors greater than 20 mm (Figure 2C) compared with the intrameatal tumors. Among the observed patients with small to medium-sized tumors, there was a tendency of lower, but not significant, survival probability, and with the moderately large to giant tumors, there seemed to be a decline in survival probability 5 years after diagnosis, but this was represented by a low number of patients.

Figure 3 shows a scatterplot representing the survival for both groups in association with tumor size. When looking at the intrameatal tumors, only 3 of them were surgically removed and are not visible on the plot due to the larger amount of observed intrameatal tumors. The patients with an intrameatal tumor (n = 210), who were observed, had a mean (SD) survival of 8.8 (4.4) years after diagnosis. Among the patients with an extrameatal tumor, 267 were observed, and 144 tumors were surgically removed.

Figure 4 compares the estimated life expectancy for men and women at each year, using data based on the Danish population since 1901 derived from cross-sectional studies from the Danish Statistical Institute,¹⁶ and it shows an increase in the expected life span. For example, for a woman aged 70 years in 1976, the expected survival was 13.8 years, compared with 16.8 years if she was aged 70 years in 2015. For the male population, the corresponding survival was 10.8 years and 14.5 years, respectively. In the Table, the mean survival in the observed group and the surgically treated group is compared with the corresponding survival in the age-matched background population.

The aim of this work was to evaluate the outcomes from watchful waiting and surgery in patients 70 years and older indirectly by investigating survival after diagnosis in conservatively and surgically managed patients with VS and furthermore comparing these with an age-matched background population. Overall, the choice of treatment depends on tumor size, risk of tumor growth, symptoms, and, as noted, age and risk of surgical complications. Should the patients continue the wait-and-scan option due to greater risk of surgical complications in this age group? Or should they be offered lateral skull base surgery in case there are no or few comorbidities? Or should older adult patients with VS receive radiotherapy? When presented with the options, studies show that older adult patients with VS more often opt for observation, but the studies rarely focus solely on older adults with VS¹⁷ despite age being commonly mentioned as an important parameter for individualized treatment and VS management.¹⁴ This study presents a large group of older adult patients with VS undergoing either observation or surgery and comparing both with the general age-matched population. Overall, the survival after diagnosis of a VS in patients 70 years and older was similar between the surgically treated group and the age-matched background population. We would obviously not argue that surgery itself is equal with increasing survival, as this would not be sensible. In the observed/wait-and-scan group, the survival after diagnosis tended to be shorter. This may be associated with increased comorbidity in this group, which could explain why potential surgery was avoided. The cause of death in these patients was unfortunately not reported, meaning that the patient principally could have died from an accident or natural causes not practically related to their VS. Medical history was also not present in the older data material, and so unfortunately this data set is unable to stratify comorbidities in the surgical and observed groups as well as the comparison with the general population. In both groups, the life span, indeed, seems to be comparable with the life span of the background population. Looking at data from the Danish Statistical Institute, the survival also depends on the year of birth. According to the statistics, the mean survival of a 70-year-old person in 1976 had changed markedly until 2016, which may be due to improved health care systems and medical treatments. These differences have not been taken into consideration in this study, although the depicted mean survival curves have been presented as an average over the 40-year span. This indicates that concerns regarding age irrespective of treatment modality possibly should not be a major limiting factor. Additionally, nationally, no obvious differences were seen, as the population was quite homogeneous.

Other studies on VS in older adult patients have concluded similarly that surgery is safe in patients older than 70 years, regarding surgical complications specific to VS removal, and report no higher risk of stroke, bleeding, cardiac events, or mortality.¹¹ This is contrasted to the previous perception of VS surgery and outcome in the early years of the study period and before the 1970s, where surgery carried significant morbidity, ie, postoperative facial palsy in the majority of cases and a mortality rate of 20%.¹⁸ Historically, the type of operation has changed from primarily translabyrinthine to suboccipital in the beginning of the reported period, where the tumors were bigger and most of the ears were deaf, to include retrolabyrinthine and a few retrosigmoidal in the later period, when the tumors were diagnosed earlier and were smaller. The middle fossa approach, usually preferred in small intrameatal tumors, was used once, since most of these were nongrowing and many of the patients had stable hearing with 100% discrimination score. Studies also comment that radiotherapy often is a consideration in this group.¹² While stereotactic radiosurgery (SRS) worldwide may be a more frequently used treatment modality in the older adult population with increased complication risk, it was not a popular treatment modality in Denmark in the study period. Only 10 patients with a growing VS in Denmark (3 of whom had previous VS surgery) were treated with radiation treatment in the study period, and the group was, for that reason, excluded from the study; however, it would be logical to count in the SRS group as well. Studies have reported SRS to be an appealing option due to its capability to diminish tumor growth, minimal threat to the facial nerve, and lower risk of hearing loss,⁴ and underline furthermore that comorbidities and dizziness are prevalent in older adult patients, yet no significant differences exist after surgery in hard parameters, such as facial nerve weakness, hearing impairment, or general surgical complications.^4,12,13 Furthermore, regarding radiotherapy, gamma knife is not an option in Copenhagen, and the aforementioned patients received conventional fractionated or hypofractionated stereotactic radiotherapy, which timewise obviously is inferior to the onetime gamma knife treatment. Hence, understandably, this treatment modality is gaining increasing popularity even though compelling results are still lacking regarding superiority, eg, in long-term tumor control, as no randomized clinical trials have been performed.¹⁹

Successful management as well as life quality for the patient must be taken into consideration when choosing the right treatment strategy in older adult patients. If the patient is being considered to undergo surgery, the patient is evaluated against life expectancy and number of comorbidities, including vascular disease, diabetes, smoking, and alcohol use, to predict the postoperative mortality. If the patient has numerous health issues, the risk of undergoing surgery must be weighed against the beneficial outcome of the surgery.¹¹ Studies have demonstrated that older adult patients are more likely to report imbalance and poorer coordination preoperatively, which after VS surgery persists long term in a higher degree compared with matched postsurgical patients with VS younger than 70 years.²⁰ Poor contralateral hearing or persisting vestibular problems are also important factors to keep in mind when choosing the right treatment for the patient.

Monitoring of patients with MRI scans is cheap and feasible, provided that the patients follow an optimal follow-up scheme.⁷ However, there are also disadvantages in choosing conservative treatment, such as the psychological aspect, as some patients might find it stressful and anxiety-producing knowing they have a tumor that could be growing and compressing their brainstem.

The limitation of this study is mainly the lack of an SRS group. Unfortunately, there were not enough data or patients who received radiotherapy to present any statistical analysis or sensible results, which is why this group was excluded from the study. Furthermore, no medical history was recorded concerning cause of death as well as comorbidity (eg, diabetes, hypertension, and other cardiovascular disease) that may have contributed to the patient’s morbidity. Last but not least, the difference in survival and 95% CI between the patients with VS and the background population was not statistically comparable. The data from the Danish Statistical Institute are based on a type of cross-sectional data, which are incompatible in terms of a direct comparison and statistical analysis with the current cohort study.

In this cohort study, findings show a small and clinically meaningless difference in the life span between patients with VS, either conservatively or surgically managed, and the matched general population. These results suggest that age alone should not be the sole consideration when contemplating surgical options for the treatment of VS.

Accepted for Publication: September 10, 2023.

Published Online: October 26, 2023. doi:10.1001/jamaoto.2023.3485

Corresponding Author: Nicole Mistarz, BM, Department of Otorhinolaryngology, Head & Neck Surgery and Audiology, Rigshospitalet, Inge Lehmanns Vej 8, Copenhagen Ø 2100, Denmark (n-mistarz@hotmail.com).

Author Contributions: Drs Reznitsky and Stangerup had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Mistarz, Reznitsky, Stangerup, Cayé-Thomasen.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Mistarz, Stangerup, Høstmark.

Critical review of the manuscript for important intellectual content: Mistarz, Reznitsky, Stangerup, Cayé-Thomasen, Jakobsen.

Statistical analysis: Mistarz, Jakobsen.

Administrative, technical, or material support: Reznitsky, Stangerup, Høstmark.

Supervision: Reznitsky, Stangerup, Cayé-Thomasen.

Conflict of Interest Disclosures: None reported.

Data Sharing Statement: See the Supplement.