SEARCH

SEARCH BY CITATION

Keywords:

  • vestibular schwannoma;
  • gamma knife stereotactic radiosurgery;
  • hearing preservation;
  • radiotherapy dose

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

To evaluate the hearing preservation rate and to determine its prognostic factors after gamma knife (GK) stereotactic radiosurgery (SRS) in patients with vestibular schwannoma, the authors used a prospective study design to analyze these patients.

METHODS

Between December 1997 and January 2002, 25 patients with vestibular schwannoma with serviceable hearing were enrolled in the current study. The median tumor volume was 3.0 cc (0.16–9.1 cc). The prescription dose was 12.0 ± 0.7 gray at an isodose line of 49.8 ± 1.1%. The tumor control rate and complications were evaluated by focusing on hearing preservation and its prognostic factors.

RESULTS

Based on radiologic study, the tumor control rate was 92% during the median follow-up period of 45 months. The trigeminal and facial nerve preservation rates were 95% and 100%, respectively. Thirteen (52%) of the 25 patients preserved serviceable hearing and 9 (36%) patients retained their pre-GK G-R grade levels after GK SRS. However, 16 patients showed hearing deterioration > 20 dB within 3–6 months and this trend continued for 24 months after the treatment. The maximum radiotherapy dose delivered to the cochlear nucleus was the single, significant prognostic factor of hearing deterioration.

CONCLUSIONS

The authors concluded that a more sophisticated strategy to prevent hearing deterioration during the first 6 months post-GK SRS is necessary to improve long-term hearing preservation. Cancer 2005. © 2005 American Cancer Society.

The incidence of clinically apparent vestibular schwannomas is approximately 1:100,000 of the U.S. population.1 With the development of more sophisticated magnetic resonance imaging (MRI) scans, the chances of finding asymptomatic patients with vestibular schwannomas are increasing. Among the therapeutic modalities available for the management of patients with a newly diagnosed small vestibular schwannoma, the role of gamma knife (GK) stereotactic radiosurgery (SRS) has been emphasized in the context of an acceptable tumor control rate and low morbidity.

GK was used first in 1969 to treat vestibular schwannoma. Noren et al.2 reported on the long-term results of GK treatment in patients with vestibular schwannomas.2 However, high radiotherapy doses (25–35 gray [Gy] marginal dose) delivered to tumors resulted in 38% of patients with facial weakness after GK SRS despite their perfect long-term tumor control rate. Since then, dose reduction, the introduction of MRI scans, and more sophisticated dose planning have improved outcomes. In 2001, Flickinger et al.3 reported an excellent 5-year clinical tumor-control rate of 97.1% after GK SRS with a median marginal dose of 13 Gy. However, the hearing preservation rate remained at 71% as the complication rate of newly developed facial weakness decreased to 1.1%.3 Recently, Iwai et al.4 reported that useful hearing was preserved in only 10 (56%) of 18 patients after GK SRS although good tumor control rate (96%) and no development of facial palsy or trigeminal neuropathy was observed with a marginal dose of 12 Gy.

The number of patients with vestibular schwannomas treated by GK SRS has increased dramatically during the past decade, thanks to the improved outcome. However, the risk of hearing deterioration is still much higher than that of any other cranial nerve neuropathies after GK SRS. In the current study, the authors prospectively analyzed serviceable hearing outcome and its prognostic factors after GK SRS in patients with newly diagnosed vestibular schwannomas.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Study Design

Twenty-five patients with sporadic vestibular schwannoma with serviceable hearing who received GK SRS between December 1997 and January 2002 were enrolled in the current study. Serviceable hearing was defined as hearing of Gardner–Robertson (G-R) Grade I or II, which is < a 50-dB pure tone average (PTA), and as a speech discrimination score (SDS) > 50%. All enrolled patients met the following criteria: 1) a newly diagnosed unilateral vestibular schwannoma, 2) a tumor < 3 cm in maximum diameter, and 3) serviceable hearing on the involved side at enrollment. Patients with a residual or recurrent tumor after surgical resection or a vestibular schwannoma of neurofibromatosis type II were excluded. Patients were treated using a GK B-model (Elekta Ltd., Stockholm, Sweden) with a Leksell Gamma Plan (Elekta Ltd.) (version 5) at the Gamma Knife Center, Seoul National University Hospital (Seoul, Korea).

Clinical parameters were assessed by neurologic evaluation, electroneurographies (ENOG), electronystagmographies (ENG), audiometry, and auditory brainstem evoked response (ABER). In addition, neuroimaging studies were performed before GK SRS and at 6-months intervals after the treatment for the first 2 years and then annually according to an individually designed protocol. Follow-up audiometry was analyzed using Kaplan–Meier survival plotting and by individual sequential plotting. Tumor volumes were measured in follow-up MRI scan images using the Osiris program (version 4.0; Unité d'Imagerie Numérique [Digital Imaging Unit]/Hôpitaux Universitaires de Genève [University Hospital of Geneva] (UIN/HUG), Geneva, Switzerland). Tumor control rate, complication development, and hearing preservation rates were evaluated up to the last follow-up.

Pretreatment Evaluation

Patients' clinical information was assessed using medical records and an electronic database. The median patient age was 47 years (range, 22–65 years), and the male-to-female ratio was18:7. ENOG, ENG, audiometry, and ABER were usually done within 1–3 months before treatment in patients treated with GK SRS, as baseline studies.

PTAs were calculated by averaging audiometric masked bone conduction responses at 500 Hz, 1000 Hz, and 2000 Hz. SDSs were recorded to establish a pretreatment audiometric grade based on the G-R grading scale, as previously described.5, 6 Thirteen patients were in G-R Grade I before GK SRS.

Pretreatment tumor volumes treated by GK SRS were determined using treatment planning software (Leksell GammaPlan, Elekta Ltd.). The median tumor volume of patients who received GK SRS was 3.0 cc (range, 0.16–9.1 cc).

Gamma Knife Stereotactic Radiosurgery

GK SRS using the GK B-model with a Gamma Plan (version 5) was received as a primary treatment by all patients. A Leksell stereotactic head frame was applied under local anesthesia with a frame shift to center the lesion as much as possible in the stereotactic space. T1-weighted three-dimensional multiplanar rapid-acquisition gradient echo MRI scan images were obtained before and after gadolinium enhancement to determine the target volume. Planning was performed on axial images supplemented by coronal and sagittal reconstructed images. The median prescription dose was 12 Gy (mean, 12.0 ± 0.7 Gy; range, 11–14 Gy) and the median prescription isodose line was 50% (mean, 49.8 ± 1.1%; range, 45–55%), with multiple-shot/high-conformality treatments according to previously published procedures.5, 7–9

The prescribed isodose ratio-to-tumor volume was derived by dividing the prescribed isodose volume by the tumor volume.5 The clinical characteristics of patients are summarized in Table 1. Radiotherapy doses delivered to the cochlea, vestibulocochlear nerve, and cochlear nucleus in the brainstem were determined from retrieved treatment back-up files.

Table 1. Characteristics of 25 Patients with Vestibular Schwannoma with Serviceable Hearing
CharacteristicsValues
  • M: male; F: female; yrs: years; mos: months; Gy: gray; ABER: auditory brainstem evoked response; ENG: electroneurography; G-R: Gardner–Robertson grade; PITV: prescribed isodose volume to tumor volume; pre-GK: pre-gamma knife; VFT: vestibular function test; MRI: magnetic resonance imaging.

  • a

    Volumes measured in cubic centimeters (cc).

  • b

    Prescription dose measured in grays (Gy).

  • c

    Follow-up time measured in months (mos).

Gender (M:F)18:7
Median age (yrs)47 (22–65)
Pre-GK hearing status 
G-R Grade I13
G-R Grade II12
Pre-GK ABER 
 Delayed24
 Normal1
Pre-GK VFT 
 Normal8
 Hypofunction17
Pre-GK ENG percentage (ipsilateral)93 ± 2 (86–96)
Pre-GK volumea3.0 (0.16–9.1)
< 0.55
0.5–14
1–2.994
3–5.993
6–109
Prescription doseb12.0 ± 0.7 (11–14)
Isodose line percentage49.8 ± 1.1 (45–5)
No. of shots10.6 ± 4.4 (4–19)
PITV ratio1.124 ± 0.156 (0.947–1.6)
Median follow-up (mos)c49 (audiometries)
 45 (MRI scan images)

Posttreatment Evaluation

Patients were usually followed up at 1, 3, 6, 12, 18, and 24 months after GK SRS and then annually. The physical and neurologic examination was done at each visit. Trigeminal nerve function was assessed based on patients' perception of pain and on corneal reflex. Facial nerve function was assessed using the House–Brackmann grading scale, and corroborated using ENOG follow-up results.10

The tumor volumes of patients were measured in follow-up images using the Osiris (version 4.0, UIN/HGUG) program. The volume of a tumor can be computed by specifying an object in a series of regions of interest (ROIs) over a set of multiplanar images and by summating ROI areas in each plane. MRI scan images stored in the host computers were retrieved and connected to a Picture Archiving and Communication System. MRI scan images were then electronically transferred as a DICOM file for the determination of tumor volumes using the Osiris program. The error associated with repeat measurements of the tumor volume of the same MRI scan images was within ±2%.

Clinical assessments including neurologic examination, ENOG, ENG, audiometries, ABER, and neuroimaging studies were performed at 6-month intervals after GK SRS for the first 2 years and then annually using an individually designed protocol. All patients were assigned serial audiometries using the G-R grading scale, as previously described.1, 11 PTA and SDS were recorded to establish a posttreatment audiometric grade in the same way as in the pre-GK evaluation. After assessing post-GK hearing status during the follow-up, audiometries were treated as censored observations in two ways, either if serviceable hearing was preserved (either within G-R Grade I or II), or more strictly, if the G-R grade remained the same as the pre-GK G-R grade. Actuarial hearing preservation rates were determined using the Kaplan–Meier product-limit method. Simultaneously, we also evaluated hearing preservation outcomes by plotting sequential PTA scores for individual patients. To determine prognostic factors for hearing preservation, we performed univariate and multivariate analyses between the deteriorated (≥ 20 dB loss in PTA) and the preserved hearing groups (< 20 dB loss in PTA).

Patients' data were entered into a statistical spreadsheet, Statview (version 5.01, SAS Institute Inc., Cary, NC), and statistical analyses were performed using the same software. Paired and unpaired t tests were utilized for parametric comparisons, and the hearing preservation rates were established using the Kaplan–Meier product-limit method. Differences were assessed using the log-rank test, and statistical significance was accepted at P < 0.05. Multivariate analysis was performed using the Cox proportional hazard function, and again P < 0.05 was statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Tumor Control Rate

The median MRI scan follow-up period was 45 months (range, 22–75 months). Tumor size decreased after GK SRS in all but 2 patients who exhibited a slight increase in tumor size during the follow-up period (Table 2). However, no surgical resection was required to the last follow-up of 45 and 48 months. The tumor control rate in terms of tumor size increase was 92% after GK SRS, whereas the tumor control rates in terms of requiring further treatment was 100%.

Table 2. Hearing and Tumor Volume Change after GK SRS in 25 Patients with Vestibular Schwannoma with Serviceable Hearing
CaseDosePre-GKVolumeaInterim post-GKLast post-GKVolumeb
PTASDSG-RPTASDSG-RPTASDSG-R
  • GK SRS: gamma knife stereotactic radiosurgery; G-R: Gardner–Robertson grade; PTA: pure tone average (dB); SDS: speech discrimination score (%).

  • a

    Initial tumor volume (cc) before GK SRS.

  • b

    Tumor volume (cc) at the last follow-up after GK SRS. Bold type indicates a statistically significant increase in tumor volume during follow-up.

111.024100I3.1020100I2890I3.84
212.015100I0.7230100I26100I0.76
312.024100I3.5030100I30100I1.39
413.012100I0.1720100I20100I0.13
512.024100I3.1030100I5070II2.74
612.026100I0.162880I3088II0.02
712.020100I0.353290II3290II0.22
812.08100I9.003268II3268II2.72
912.04680I7.106048III6822III6.50
1012.010100I2.503438III3438III0.50
1112.022100I0.186252III6252III0.10
1212.02074I7.00620V620V3.65
1311.03886II3.805076II4260II5.08
1411.530100II7.203666II3666II3.14
1512.03882II0.804452II4452II0.36
1612.02290II3.005052II5052II1.92
1712.03276II3.102464II2464II2.80
1812.04296II7.00578III6224III4.34
1914.04870II0.606036III7020III0.50
2011.03294II0.914842III5646III0.36
2111.03650II7.605038II8096III5.0
2211.53690II9.105010III666III6.10
2312.03892II2.404034III5420III0.80
2413.03582II0.306028III5818III0.26
2512.03872II7.80602IV622IV4.74

Hearing Preservation Outcome

The median follow-up period for audiometry was 49 months (range, 29–77 months). During the follow-up period, 13 (52%) of the 25 patients had preserved serviceable hearing and 9 (36%) retained their pre-GK G-R grades (Table 2). According to Kaplan–Meier survival plots, the 5-year serviceable hearing preservation rate (i.e., remained at G-R Grade I or II) of patients with serviceable hearing after GK SRS was 46%. The 5-year hearing preservation rate (i.e., the rate remained at the same pre-GK G-R grade) of patients with serviceable hearing after GK SRS was 10%. The time period over which the 50% of patients with serviceable hearing before GK SRS retained serviceable hearing was 53 months, whereas the time period over which the 50% of the patients had the same G-R grade after GK SRS was 18 months (Fig. 1).

thumbnail image

Figure 1. Kaplan–Meier survival plot of hearing preservation in patients treated with gamma knife stereotactic radiosurgery (GK SRS). (A) Probability of maintenance of serviceable hearing (Gardner–Robertson [G-R] Grade I or II) after GK SRS. (B) Probability of hearing preservation with the same G-R grade as pre-GK.

Download figure to PowerPoint

Individual sequential audiometry plotting during the follow-up period revealed two different response groups. In 1 group of patients, hearing improved or remained within a < 20-dB deterioration, whereas in the other group of patients, hearing deteriorated by a ≥ 20 dB loss in PTA (Fig. 2). This individual sequential plotting showed that 16 patients had hearing deterioration within 3–6 months of GK SRS and this continued until 24 months after GK SRS (Fig. 3). In contrast, 9 patients showed no deterioration during the same follow-up period and this was maintained to the last follow-up (Fig. 4).

thumbnail image

Figure 2. Sequential plotting of pure tone average changes in patients treated with gamma knife stereotactic radiosurgery (GK SRS). Hearing deterioration post-GK SRS progressed over the first 2 years of follow-up.

Download figure to PowerPoint

thumbnail image

Figure 3. Sequential plotting of pure tone average (PTA) changes in patients with hearing deterioration (n = 16), i.e., the group of patients that experienced a PTA deterioration of > 20 dB during follow-up. The rectangular box indicates the period of rapid deterioration of hearing after GK SRS.

Download figure to PowerPoint

thumbnail image

Figure 4. Sequential plotting of pure tone average changes in patients with hearing preservation (n = 9). The rectangular box indicates the period of rapid deterioration of hearing after gamma knife stereotactic radiosurgery in the deteriorated hearing group. Relevant silent change was notified in the preserved hearing group over the same period.

Download figure to PowerPoint

The sequential plotting of hearing after GK SRS was also plotted versus tumor volume change as determined by the Osiris program for the two groups. Tumor volume changes after GK SRS were expressed as a percentile increase versus initial volume as measured in MRI scan images at planning. In patients who experienced significant hearing deterioration of ≥ 20 dB in PTA 3–24 months after treatment, transient tumor volume increases were found more frequently. However, no statistical difference was found between the deteriorated (≥ 20 dB loss in PTA) and the maintained hearing groups (< 20 dB loss in PTA) (P = 0.52).

Prognostic Factors of Hearing Preservation

We measured the radiotherapy dose delivered to the cochlea, vestibulocochlear nerve, and the cochlear nucleus in the brainstem (Table 3). Among the structures, the cochlear nucleus in the brainstem was significantly different in radiotherapy dose delivery between the deteriorated (≥ 20 dB loss in PTA) and the preserved hearing groups (< 20 dB loss in PTA). The mean maximum radiotherapy dose of total population delivered to the cochlear nucleus in the brainstem was 9.4 ± 4.6 Gy. Patients in the deteriorated hearing group received 11.1 ± 3.9 Gy and patients in the preserved hearing group received 6.9 ± 4.5 Gy (P = 0.0322). By univariate analysis, the single prognostic factor of hearing deterioration after GK SRS was a maximum radiotherapy dose to the cochlear nucleus of ≥ 10 Gy (P = 0.048). Other variables such as age, gender, pre-GK hearing grade, total GK dose, and transient volume increase failed to show significance as prognostic factors.

Table 3. Radiotherapy Dose Delivered to the Cochlea, Vestibulocochlear Nerve, and Cochear Nucleus
CharacteristicsPreserved (PTA loss < 20 dB)aDeteriorated (PTA loss ≥ 20 dB)bTotal
  • a

    Preserved hearing group with PTA loss < 20 dB.

  • b

    Deteriorated hearing group with PTA loss ≥ 20 dB.

Cochlea   
Maximum (Gy)9.1 ± 4.47.8 ± 2.68.1 ± 3.1
Minimum (Gy)3.4 ± 1.33.8 ± 1.03.6 ± 1.2
Vestibulocochlear nerve   
Maximum (Gy)13.8 ± 5.014.7 ± 1.414.3 ± 3.3
Minimum (Gy)5.5 ± 4.46.8 ± 2.96.3 ± 3.6
Cochlear nucleus   
Maximum (Gy)6.9 ± 4.5a11.1 ± 3.9a9.4 ± 4.6
Minimum (Gy)3.8 ± 4.05.1 ± 2.14.6 ± 3.0

One noteworthy finding in relation to hearing preservation is that, although we were not able to perform a statistical analysis because of the small sample of patients, there was a transient volume increase in the intracanalicular tumor. We found that when the tumor size in the internal auditory canal (IAC) did not increase during the window period of 3–24 months after GK SRS, hearing was preserved (Fig. 5). In contrast, when the tumor size in the IAC increased during the window period of 3–24 months after GK SRS, hearing was not preserved (Fig. 6). However, if tumor size in the IAC showed no increase during 3–24 months, hearing did not deteriorate even in patients whose overall tumor volume had increased mainly in the cisternal portion (Fig. 7).

thumbnail image

Figure 5. A 44-year-old male patient with preserved hearing with tumor volume decrease in the intracanalicular region. This patient underwent gamma knife stereotactic radiosurgery (GK SRS) with a dose of 12 gray at the isodose line of 50%. The follow-up magnetic resonance imaging scan 6 months after GK SRS revealed an intracanalicular tumor that decreased in volume from 0.16 cc to 0.15 cc. With these changes, his audiometry scores changed from 26 dB/100% (A) to 26 dB/68% (B). The consecutive tumor volume and audiometry changes were 0.03 cc with 32 dB/82% at 3 years (C) and 0.02 cc with 30 dB/88% at 4 years (D) after GK SRS.

Download figure to PowerPoint

thumbnail image

Figure 6. A 52-year-old male patient with hearing deterioration with a transient tumor volume increase in the intracanalicular portion. This patient underwent gamma knife stereotactic radiosurgery (GK SRS) with a dose of 13 gray at the isodose line of 50% for vestibular schwannoma in the right side. The follow-up magnetic resonance imaging scan 1 year after GK SRS revealed an increase in tumor volume in the right internal auditory canal from 0.3 cc to 0.32 cc. With these changes, his hearing deteriorated from 35 dB/82% (A) to 62 dB/44% (B). Tumor volume and audiometry measured at 1.5 years after GK SRS were 0.31 cc and 60 dB/28% (C), and at 3 years after GK SRS were 0.26 cc and 58 dB/18% (D).

Download figure to PowerPoint

thumbnail image

Figure 7. A 22-year-old male patient had preserved hearing regardless of a transient tumor volume increase in the cisternal region without changes in the intracanalicular region. This patient underwent gamma knife stereotactic radiosurgery (GK SRS) with a dose of 11 gray at the isodose line of 50% for a vestibular schwannoma in the left side. The follow-up magnetic resonance imaging scan 6 months after GK SRS revealed an increase in tumor volume from 3.1 cc to 3.5 cc. However, there was no significant tumor volume change in the internal auditory canal although a central necrotic change was observed. With these changes, his audiometry scores were preserved to be serviceable from 24 dB/100% (A) to 20 dB/100% (B).

Download figure to PowerPoint

Functional Outcomes Other than Hearing

Newly developing facial nerve or trigeminal nerve dysfunction was absent in all patients, whereas preexisting trigeminal nerve dysfunction was aggravated in one patient. The trigeminal nerve preservation rate after GK SRS was 95%. One patient aggravated a preexisting hemifacial spasm during the follow-up and underwent microvascular decompression surgery for symptomatic relief 14 months after GK SRS. No ENOG change was detected in the other patients. Follow-up ABER showed that delayed latencies of III and V waves were normalized in 3 of 24 patients with delayed latency in pre-GK SRS study. In follow-up ENG, vestibular function improved in three patients and was aggravated in in three patients. Hydrocephalus did not develop during the follow-up period. No acute or chronic toxicity induced by GK SRS was detected, and no secondary malignancy developed in the radiotherapy field of GK SRS-treated patients during the follow-up.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Hearing Preservation after Gamma Knife Stereotactic Radiosurgery for Vestibular Schwannoma

According to the 1991 National Institute of Health Consensus Statement, an estimated 2000–3000 new cases of unilateral vestibular schwannoma are diagnosed in the United States each year—an incidence of approximately 1 per 100,000 per annum.1 Several management options are available for these patients, i.e., close observation, radiosurgery, or microsurgical resection. Among them, the number of patients treated with GK SRS has increased disproportionately over the past decade, thanks to its low morbidity and high tumor control rate. In modern series of GK SRS for vestibular schwannomas, facial nerve preservation rates have been reported to range from 95% to 100% and the rates of hearing loss after GK SRS have also declined.5, 11–16 The increased precision of stereotactic imaging, targeting, and upgraded dose-planning programs providing means of lowering the prescription dose may be responsible for this trend. However, the hearing preservation rate is still not satisfactory compared with the results of other cranial nerve preservation modalities. Moreover, disparate hearing preservation results have been reported after GK SRS, because of the different hearing preservation criteria, follow-up times, and radiotherapy doses used. Prasad et al.17 treated patients using 13.2 ± 2.2 Gy for the tumor margins. They reported an 89% tumor control rate in patients who had received previous tumor resections, and a 94 % rate in patients who had not. Hearing was preserved in 40% of patients whereas new development of facial weakness and trigeminal neuropathy were found in 2% and 3% of patients, respectively. Recently, Iwai et al.4 reported on the results of low-dose GK SRS for the treatment of patients with vestibular schwannomas. The median marginal dose was 12 Gy. The clinical tumor growth control (without tumor resection) rate was 96% and the 5-year tumor growth control rate was 92%. Hearing was preserved or improved in 68% of patients and useful hearing was preserved in 56% of patients. No case of new facial palsy or trigeminal neuropathy occurred after GK SRS.

In the current study, patients treated with GK SRS received 12 Gy of the median prescription dose at the 50% isodose line. This dose profile is one of the lowest practiced in GK SRS for the treatment of vestibular schwannomas. Although long-term follow-up is mandatory to validate an optimal tumor suppression effect with these lower dose profiles, the interim tumor control rate results of the current study are comparable to those of previous reports where higher prescription doses have been applied. However, the preservation of hearing after GK SRS is unsatisfactory as stated by many previous authors. In our study, 13 patients had G-R Grade I and 12 had G-R Grade II before GK SRS. During the follow-up, 13 (52%) of the 25 patients retained serviceable hearing, whereas only 9 (36%) patients had the same G-R grade after GK SRS. Therefore, almost one-half of the patients treated lost serviceable hearing during the follow-up period. Furthermore, only one-third of the GK SRS-treated patients maintained their pre-GK status hearing level. Taking into consideration the low complication rate of other cranial nerve palsies such as trigeminal or facial nerve palsy (5% and 0%, respectively), the hearing preservation rate after GK SRS using our current protocol is unexpectedly poor and difficult to explain.

Hypothetical Causes of Hearing Deterioration after Gamma Knife Stereotactic Radiosurgery

The reason for the hearing deterioration after GK SRS in patients with vestibular schwannoma has not been documented clearly. However, there are several hypothetical causes of hearing deterioration proposed in the literature.

First, direct radiotherapy damage to the hearing apparatus might be one of the causes. Historically, cranial nerve palsies caused by GK SRS have been reduced by reducing the radiotherapy dose delivered to the tumor.1–4, 8, 16–24 Dose reduction is associated with a marked improvement in the preservation of trigeminal or facial nerve function. However, at the level of radiotherapy at which other cranial nerve preservation rates, such as facial and trigeminal nerves, were > 90%, one-half of the patients still lost hearing after GK SRS. The authors measured the radiotherapy dose delivered to the cochlea, the vestibulocochlear nerves, and the cochlear nucleus in the brainstem. As the maximum radiotherapy dose delivered to the cochlear nucleus in the brainstem > 10 Gy was the significant prognostic factor for hearing deterioration after GK SRS, the cochlear nucleus may be 1 of the sensitive structures to be damaged by radiotherapy. Linskey et al.21 retrospectively quantified the dose of radiotherapy received by the inner, middle, and external ear structures of 18 patients with vestibular schwannoma treated with GK SRS. They reported that the basal turn near the modiolus and its inferior portion in the cochlea as well as the amputated ends of the lateral and posterior semicircular canals in the vestibular labyrinth are the most susceptible areas, with doses > 12 Gy detected in 10.8%, 14.8%, 7.4%, and 5.1% of patients, respectively. They suggested that such an unintentional delivery of high doses to the stria vascularis, the sensory neuroepithelium of the inner ear organs, and/or their ganglia might play a role in the development of post-GK tinnitus. There are also several reports about hearing deterioration after conventional radiotherapy. These higher radiotherapy exposures of the hearing apparatus near the tumor, including the cochlea, vestibulocochlear nerve, or cochlear nucleus, might explain, in part, hearing deterioration after GK SRS.

Second, hearing deterioration after GK SRS was ascribed to changes in the blood supply to the cochlea or cochlear nerve. Vascular changes such as endothelial proliferation and hyalinization of small and medium-sized vessels are well known phenomena in arteriovenous malformations treated with GK SRS.25–27 Radiotherapy effects on blood vessels include the hyalinization of arterioles and myointimal cell injury response in larger arterioles and small arteries.28 Myointimal cells respond to this injury by proliferating. Both myointimal cell proliferation and mural hyalinization result in small artery and arteriolar occlusion.28 Potential pathologic results include tumor cell ischemia and hypoxemic cell death with subsequent cell loss. The clinical corollary is loss of central tumor contrast enhancement and delayed tumor shrinkage once ischemic cell loss predominates over ischemic cell swelling. Such changes are prominent 3–24 months after GK SRS, which may account for the specific time window of hearing deterioration after GK SRS. However, the process of devascularizaton caused byendothelial proliferation or hyalinization of the vessels after GK SRS is a common effect seen in most patients treated with GK SRS. Therefore, the devascularization effect induced by GK SRS has limited value as an explanation that approximately 50% of the patients treated with GK SRS experienced hearing deterioration after GK SRS.

Third, there may be adhesion between perineural tissues and tumors after SRS.29, 30 Couldwell and Mohan31 reported a patient who underwent surgical resection for a tumor that increased to cause significant brainstem compression 3 months after a course of fractionated stereotactic radiotherapy with 5 treatments of 4 Gy to the 90% isodose line over a 3-week period. MRI scans revealed a change in the enhancement pattern of the lesion, with a decrease in central contrast enhancement, associated with an overall increase in the size of the lesion and the mass effect on the brainstem compared with the pretreatment study. They found that at surgery, the tumor was soft with central necrosis and little vascularity, and that the facial nerve was very adherent to the perimeter of the tumor. Lee et al.29 reported four patients who underwent microsurgical resection of vestibular schwannoma after primary stereotactic radiotherapy including GK SRS or fractionated stereotactic radiotherapy. They found markedly variable fibrosis outside and within the tumor bed, which complicated microsurgical dissection. This could also be a possible reason for hearing deterioration after GK SRS. If this were the case, however, the hearing deterioration would progress gradually with time elapsed after GK SRS rather than show an abrupt deterioration 3 months–2 years after treatment and then stabilize.

The other possible explanation for hearing deterioration might be the transient volume change of the tumor located in the IAC after GK SRS.30, 32 It is well known that there are transient volume increases with signal change in the center and peripheral rim enhancement of the tumor on follow-up MRI scan images after GK SRS.30, 32 If the IAC portion of the tumor is specifically increased, the compression effect on the cochlear nerve in the IAC would be more aggravated. In our study, we closely examined the relation between tumor volume change in the IAC and hearing deterioration after GK SRS and demonstrated a few illustrative cases. Thus, a transient volume change of the tumor located in the IAC after GK SRS may be one of the causes for hearing deterioration.

It should be added that a single reason may not explain the issue of hearing lose during follow-up after GK SRS. Multiple factors occurring after GK SRS may contribute.

Proposal to Reduce Hearing Deterioration after Gamma Knife Stereotactic Radiosurgery

Historically, the utmost effort has been made to reduce the radiotherapy dose to decrease the complications caused by GK SRS in the treatment of vestibular schwannomas.2, 3, 8, 18, 19 The radiotherapy dose delivered to the surrounding normal structure, the vascular changes caused by the radiotherapy dose, the peritumoral pathologic changes, and the possibility that transient tumor volume increases after GK SRS might all be reduced by decreasing the dose delivered to the tumor. Conversely, decreasing the tumor dose too much also causes tumor regrowth after treatment. The dose profile applied in the current study is one of the lowest reported in the literature. With 12 Gy at the 50% isodose line in GK SRS, the tumor control rate was comparable to previous reports. To validate the merits of further reducing the radiotherapy dose to patients with vestibular schwannomas with measurable hearing, long-term follow-up should be mandatory to monitor any adverse effects on tumor control. In addition, more sophisticated radiosurgical planning for reduction of the dose delivered to the critical structures such as the cochlear nucleus is also required for hearing preservation.

Another way to preserve hearing after radiosurgery is to deliver fractionated doses of radiotherapy. The use of conventionally fractionated or hypofractionated SRS is theoretically beneficial in preventing late tissue damage by delivering smaller doses per fraction. Thus, it is possible that a greater likelihood of hearing preservation is associated with stereotactic dose fractionation not exceeding a conventional 2-Gy dose per fraction.5 Andrews et al.5 demonstrated the superiority of this kind of strategy in preserving hearing function, although the used treatment modality was stereotactic radiotherapy.

Another option to consider in the prevention of the adverse effects caused by GK SRS using the current treatment protocol is the application of steroids during the period when the tumor size increases transiently after treatment.33–37 Because of their antiinflammatory and membrane-stabilizing effects, steroids are used frequently in the management of patients with peritumoral brain edema or transient tumor swelling in various brain tumors including intraaxial or extraaxial tumors after radiosurgery.38–40 Several studies have found that hearing is improved after administering steroids to patients who have experienced a hearing deterioration after SRS.33, 37 If steroids can attenuate the adverse effects caused by radiotherapy on the normal neurovascular structures in the tumor proximity, the preventive use of steroids over a specific time period, i.e., 3–6 months after treatment, might reduce the rate of hearing deterioration in patients treated with GK SRS. However, because steroids can cause serious side effects, their use should be addressed with caution and should be limited in duration with close monitoring.

Although the authors prospectively analyzed the outcomes of patients who received GK SRS, the study has some limitations in terms of the interpretation of results. First, although total tumor volume was measured by using the Osiris program, the tumor volume change in the IAC could not be evaluated precisely because of the too-small tumors. Thus, we had to estimate the size change of the tumor by comparing the magnified IAC MRI scan images of patients before and after GK SRS. Second, the follow-up period of 45 months was relatively short. A longer follow-up period is still needed to yield convincing long-term tumor control rates with such a low marginal dose of 12 Gy, because vestibular schwannoma has a very long natural history that most studies describe the growth rates of vestibular schwannomas in millimeters per year.41–45

Despite an acceptable tumor control rate (92%) and high rates of preservation (> 95%) of cranial nerves, only one-half of the patients had serviceable hearing after GK SRS using the current low-dose protocol. Patients who experienced hearing deterioration within 6 months of GK SRS were vulnerable to failure of hearing preservation. Dose reduction to the cochlear nucleus and a more sophisticated strategy to prevent hearing deterioration during the first 6 months after GK SRS is necessary to improve long-term hearing preservation.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES