- Email: [email protected]
Functional Outcome of the Facial Nerve After Surgery for Vestibular Schwannoma: Prediction of Acceptable Long-Term Facial Nerve Function Based on Immediate Postoperative Facial Palsy
Original Article
Functional Outcome of the Facial Nerve After Surgery for Vestibular Schwannoma: Prediction of Acceptable Long-Term Facial Nerve Function Based on Immediate Postoperative Facial Palsy Seunghoon Lee1, Ho Jun Seol1, Kwan Park1, Jung-Il Lee1, Do-Hyun Nam1, Doo-Sik Kong1, Yang-Sun Cho2
OBJECTIVE: To examine the relationship between immediate postoperative facial palsy and long-term facial palsy and identify a treatment strategy for vestibular schwannoma considering long-term outcomes of facial nerve function and tumor control.
-
METHODS: Patients (N [ 385) who underwent surgery in a single institution were reviewed retrospectively; 12 patients with neurofibromatosis, 6 with preoperative radiosurgery, and 14 with multiple surgeries were excluded. The generalized estimating equation method was used to show the correlation between immediate and later postoperative facial palsy and to identify the cutoff grade of immediate postoperative facial palsy.
-
RESULTS: The tumor control rates for 1 year, 3 years, and 5 years were 88.7%, 83.9%, and 80.0%. Preservation of facial function above House-Brackmann (H-B) grades 1 and 2 was achieved in 47.9% of patients immediately postoperatively, in 50.1% after 1 month, and in 74.5% after >2 years. The immediate postoperative facial palsy grade showed a statistically significant relationship with the facial palsy grade on long-term follow-up (P < 0.001). H-B grade 3 immediate postoperative facial palsy was identified as the cutoff grade that showed the most significant relationship between the grade of immediate postoperative facial palsy and the grades above the cutoff (H-B grade 1e3) on long-term follow-up (P < 0.001).
-
CONCLUSIONS: H-B grade of immediate postoperative facial palsy can predict facial palsy at long-term follow-up.
-
Key words - Facial nerve preservation surgery - Facial palsy - Vestibular schwannoma Abbreviations and Acronyms GEE: Generalized estimating equation GKRS: Gamma Knife radiosurgery GTR: Gross total resection H-B: House-Brackmann MRI: Magnetic resonance imaging NTR: Near-total resection STR: Subtotal resection VS: Vestibular schwannoma
WORLD NEUROSURGERY 89: 215-222, MAY 2016
H-B grade 3 immediate postoperative facial palsy is the lowest tolerable grade that guarantees functional improvement on long-term follow-up. Planned facial nerve preservation surgery followed by radiosurgery is thought to be optimal treatment in patients with vestibular schwannoma for both tumor control and facial nerve function.
INTRODUCTION
M
icrosurgical complete resection has been the treatment of choice for vestibular schwannoma (VS), the most common intracranial schwannoma. The goals of surgery are to render the patient free from tumor growth and from mortality and neurologic morbidity caused by continuous uncontrolled tumor growth.1 Advances in microsurgical techniques have achieved complete tumor resection with low or no mortality and shifted our concerns from mortality toward neurologic morbidity. Facial palsy is one of the most serious morbidities because loss of facial expression is a functionally and psychologically debilitating condition.2,3 Reducing the high risk of facial palsy after surgery has become a major issue in treatment of VS. Maximal tumor resection with facial nerve preservation has been advocated by several surgeons,4 and many institutions reported excellent outcomes in tumor control and facial nerve preservation.5-8 Nevertheless, the incidence of postoperative facial palsy can be 30%e80%, especially for large tumors.9 Radiosurgery, such as Gamma Knife radiosurgery (GKRS), has decreased the recurrence rate in patients who had a partial tumor resection to save facial nerve function. Given the acceptable outcomes of
From the Departments of 1Neurosurgery-Head and Neck Surgery, and 2 Otorhinolaryngology-Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea To whom correspondence should be addressed: Ho Jun Seol, M.D., Ph.D. [E-mail: [email protected]] Citation: World Neurosurg. (2016) 89:215-222. http://dx.doi.org/10.1016/j.wneu.2016.01.038 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
215
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
radiosurgery for residual tumor after surgery, we have been hesitant to perform complete resection of a VS at the expense of facial nerve function. Several studies suggested that planned near-total resection (NTR) or subtotal resection (STR) along with postoperative radiosurgery for residual tumor allowed for the best combination of facial nerve function and tumor control.10 There have been no well-established measures except preoperative tumor size and surgical approach to predict long-term prognosis of facial palsy after VS surgery.11-17 We could not locate any previous studies examining the relationship between immediate postoperative facial palsy and long-term serial outcomes of facial palsy. In this study, we review our surgical series and analyze serial facial nerve functions to identify the most efficient treatment strategy for patients with VS based on long-term outcomes of facial nerve function as well as tumor control. MATERIALS AND METHODS This study was approved by the Samsung Medical Center institutional review board. During the period 1995e2013, 385 patients with VS were operated on at our institution. Medical records and images of patients were reviewed retrospectively. We included only patients with sporadic unilateral VS. There were 12 patients with neurofibromatosis type 2 who were excluded because of the fast and aggressive growth of the disease. Of the remaining 373 patients, we selected patients who were followed after being treated with microsurgery with or without radiotherapy or radiosurgery, including GKRS, for residual or recurrent tumor. To show the exact relationship between microsurgery and facial palsy, we excluded 6 patients who had microsurgery after preoperative radiosurgery and 14 patients who had multiple surgeries, including staged operations. One patient who died was excluded because multiple surgeries were involved and postoperative facial and hearing functions were not evaluated. Tumor control was assessed without 12 neurofibromatosis patients so as not to exaggerate the proportion of patients free of tumor progression (Figure 1). Follow-up at the outpatient clinic was conducted 1 month, 1 year, and 2 years after the surgery. Tumor size was measured using the greatest dimension on preoperative and postoperative axial magnetic resonance imaging (MRI). For a patient with a tumor exceeding a maximal diameter of 2.5 cm who is competent to general anesthesia, microsurgery is recommended first. However, radiosurgery is considered in cases of patients with low performance status or on the basis of patient preference. During an operation, we pursue a maximal resection without taking the risk of harming any vessels or nerves. Radiosurgery is provided for persistent residual tumors. The microsurgeries were performed by neurosurgeons and otorhinolaryngologists at our institution. Surgical approaches were determined based on patient age, tumor size and location, and initial neurologic manifestations. The extent of tumor resection, which was classified as gross total resection (GTR), NTR, or STR, was assessed by reviewing surgery records and comparing preoperative and postoperative MRI; postoperative MRI was obtained within 48 hours in most cases or within 1e3 months after surgery in other cases. NTR was defined as resection with <5 mm of a residual intracanalicular tumor or tumor adhering to the facial nerve or brainstem that was intentionally left behind and
216
www.SCIENCEDIRECT.com
Figure 1. Flowchart showing the numbers of included and excluded patients. There were 373 patients analyzed as tumor controls after excluding 12 patients with neurofibromatosis (NF type II). Facial palsy was assessed in 353 patients after further exclusion of 6 patients following preoperative radiosurgery and 14 patients who underwent multiple surgeries.
not seen on postoperative MRI. Cases with residual tumors with maximal thickness >5 mm were considered a STR.18-21 Facial nerve function was clinically assessed by the HouseBrackmann (H-B) scale.22 Facial palsy grading of patients was completed by 3 parties: operators, residents in the intensive care unit or ward, and nurses in the neurosurgical intensive care unit or ward. If there were any discrepancies in grading, the worst grade was taken. Facial nerve function was graded preoperatively, immediately postoperatively (promptly after patients are taken out of the operating room, extubated and fully awake), 1 month postoperatively, 1 year postoperatively, and 2 years postoperatively. At each outpatient clinic visit, operators and neurosurgical physician assistants did the grading. To evaluate hearing, the pure-tone average and speech discrimination score were measured. The pure-tone average was calculated as an average of pure-tone thresholds at 3 frequencies (500, 1000, and 2000 Hz). The speech discrimination score was defined as the percentage of words correctly repeated from a standardized list. With audiometric results, patients were categorized into serviceable hearing and nonserviceable hearing groups according to the Gardner-Robertson hearing scale.23 Serviceable hearing was defined as Gardner-Robertson hearing grades 1e2, and nonserviceable hearing was defined as a puretone average <50 dB or a speech discrimination score <50%. The comparison of facial palsy and tumor control in each group was conducted using the Kaplan-Meier method, Mann-Whitney test, log-rank test, and Cox proportional hazards regression model. To show the correlation between immediate postoperative facial palsy and facial palsy thereafter, the generalized estimating equation (GEE) method24 was used because facial palsy grades at
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.01.038
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
each follow-up visit in each patient had repetitive categorical values. With the GEE method, we also identified the cutoff grade of immediate postoperative facial palsy that led us to expect longterm functional improvement of the facial nerve. To identify this value, every H-B grade from 1 to 5 (none of the patients showed H-B grade 6 facial palsy) was assessed as a cutoff value and compared with the groups above and below it. The grade with the smallest P value after Bonferroni correction was identified as a cutoff grade. The Bonferroni method was used to prevent an inflated type I error. The result was considered statistically significant if the P value was < 0.05. All statistical analyses were performed using IBM SPSS Statistics version 22 (IBM Corporation, Armonk, New York, USA).
Table 1. Characteristics of Patients with Vestibular Schwannomas Characteristics
Number (%)
Total number
P Value
353
Sex Female/Male
199 (56.4)/154 (43.6)
Age (years) Mean
49
Range
21e83
Presenting symptoms
RESULTS Among the 353 enrolled patients, the number of female patients was 199, and the number of male patients was 154. The mean age of patients was 49.0 years (range, 21e83 years). The most common presenting symptom was hearing difficulty in 210 (56.3%) patients, followed by tinnitus in 172 (46.1%), dizziness or vertigo in 94 (25.2%), headache in 75 (20.1%), facial paresthesia in 65 (17.4%), and facial palsy in 19 (5.1%). There was no predominant tumor side (right, 164; left, 189). The preoperative mean tumor size was 30.2 mm (range, 5.0e60.0 mm). Tumors were categorized as large (>40 mm), medium (21e40 mm), or small (20 mm), with 70 (19.8%), 197 (55.8%), and 86 (24.4%) patients in the large, medium, and small groups. We used a lateral suboccipital retrosigmoid approach in 165 patients (75.6%), a translabyrinthine approach in 77 patients (22.9%), and a middle fossa approach in 4 patients (1.4%). All translabyrinthine and middle fossa approaches were performed in collaboration with otorhinolaryngologists. GTR was achieved in 183 (51.8%) patients, NTR was achieved in 64 (18.1%) patients, and STR was achieved in 106 patients (30.0%) (Table 1). After excluding 12 patients with neurofibromatosis type 2, we observed 55 cases of recurrence. Tumors in 6 patients recurred after preoperative radiosurgery, 14 patients had recurrent tumors and were operated on more than twice, and 35 patients experienced recurrence after microsurgery (NTR or STR) with radiotherapy or radiosurgery (Figure 2A). The tumor control rates for 1 year, 3 years, and 5 years were 88.7%, 83.9%, and 80.0%. By using the log-rank test, patients with small tumors achieved better tumor control than patients with medium or large tumors (P ¼ 0.018) (Figure 2B). Also, patients who had GTR achieved better tumor control compared with patients who had NTR or STR (P < 0.001) (Figure 2C). However, there were no statistically significant differences in tumor control according to preoperative tumor size and the extent of tumor resection on the Cox regression model analysis. Tumor controls for each microsurgical approach did not differ significantly on the logrank test (P ¼ 0.179) or the Cox regression model (P ¼ 0.807) (Table 2). Based on an audiometric test, hearing preservation was achieved within 1 year after surgery in 12 of 88 patients with preoperative serviceable hearing (13.6%). In 5 patients, delayed hearing loss resulted in nonserviceable hearing (preserved hearing for 6 months after surgery and then deterioration).
WORLD NEUROSURGERY 89: 215-222, MAY 2016
Hearing difficulty
210 (56.3)
Tinnitus
172 (46.1)
Dizziness or vertigo
94 (25.2)
Headache
75 (20.1)
Facial paresthesia
65 (17.4)
Facial palsy
19 (5.1)
Tumor side Right/left
164 (46.5)/189 (53.5)
Tumor size (mm)
0.018
Mean
30.2
Range
5.0e60.0
Large (>40 mm)
70 (19.8)
Medium (21e40 mm)
197 (55.8)
Small (20 mm)
86 (24.4)
Surgical approach
0.179
RS
267 (75.6)
TL
81 (22.9)
MF
5 (1.4) < 0.001
Extent of tumor resection GTR
183 (51.8)
NTR
64 (18.1)
STR
106 (30.0)
RS, retrosigmoid; TL, translabyrinthine; MF, middle fossa; GTR, gross total resection; NTR, near-total resection; STR, subtotal resection.
Preoperative and postoperative facial nerve function assessments are provided in Table 3. Facial functional preservation above an acceptable or useful facial function, H-B grade 1 and 2, was achieved in 47.9% of patients immediately after surgery, in 50.1% 1 month after surgery, and in 74.5% 2 years after surgery. Using the GEE method, the immediate postoperative facial palsy grade showed a statistically significant relationship with serial postoperative facial palsy grades (P < 0.001) after controlling for the extent of resection and tumor size. We identified H-B grade 3 as the cutoff that showed the most significant relationship
www.WORLDNEUROSURGERY.org
217
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
between the grade of immediate postoperative facial palsy and the grades above the cutoff grade (H-B grade 1e3) on long-term follow-up (P < 0.001). As few patients had H-B grade 5 and 6 postoperative facial palsy, the statistical analysis failed to produce significant results (Table 3). The number of patients in each facial palsy group is shown in Figure 3. The proportion of patients in H-B grade 1e3 increased over the first year after surgery and then remained steady. Comparisons of the percentages of H-B grade 1e3 patients in each group according to the extent of tumor resection (Figure 4A), tumor size (Figure 4B), and surgical approach (Figure 4C) did not show significant differences.
DISCUSSION Tumor Control We achieved 88.7%, 83.9%, and 80.0% tumor control at 1 year, 3 years, and 5 years postoperatively. We confirmed that maximal tumor resection was important in tumor control, as we observed significantly different tumor controls between the GTR and STR groups (P < 0.001). Tumor control for GTR was 91.9% and for NTR or STR was 92.1% when followed by adjuvant radiotherapy or radiosurgery. This result affirms that planned microsurgery followed by radiosurgery for residual tumors leads to functional
Figure 2. Tumor progression-free (recurrence-free) survival of 373 patients with vestibular schwannoma. (A) Overall tumor control was achieved in 85.3% of patients during a median follow-up period of 43.0 months (range, 0.4e228.5 months). (B) Tumor controls for each microsurgery group were classified by extent of tumor resection: 91.9%, gross total resection (GTR); 84.3%, near-total resection (NTR); and 75.2%, subtotal resection (STR). There was a significant long-term difference between GTR and STR (P < 0.001), but the differences between GTR and NTR (P ¼ 0.061) and between NTR and STR (P ¼ 0.091) were not significant. Controls for tumor size were 79.7% in the large tumor group, 83.3% in the medium tumor group, and 94.4% in the small tumor group. (C) There was a significant long-term difference between the small tumor group and the medium (P ¼ 0.019) or large (P ¼ 0.003) tumor groups, but not between medium and large (P ¼ 0.310) tumor groups.
218
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.01.038
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
Table 2. Tumor Control in Groups Classified by Extent of Tumor Resection, Preoperative Tumor Size, and Microsurgical Approach Controlled (%)
Recurred (%)
HR (95% CI)
P Value
Extent of tumor resection
0.171
GTR
171 (91.9)
15 (8.1)
1.800 (0.883e3.668)
0.105
NTR
59 (84.3)
11 (15.7)
0.875 (0.434e1.763)
0.709
STR
88 (75.2)
29 (24.8)
59 (79.7)
11 (20.3)
0.798 (0.221e2.875)
0.730
175 (83.3)
22 (16.7)
0.613 (0.182e2.066)
0.430
84 (94.4)
2 (5.6)
Preoperative tumor size
0.562
Large Medium Small Microsurgical approach
0.807 $
RS
235 (83.0)
48 (17.0)
TL
79 (92.9)
6 (7.1)
1.138 (0.435e2.977)
0.792
MF
4 (80)
1 (20.0)
2.270 (0.190e27.135)
0.517
HR, hazard ratio; CI, confidence interval; GTR, gross total resection; NTR, near-total resection; STR, subtotal resection; RS, retrosigmoid; TL, translabyrinthine; MF, middle fossa.
preservation and tumor control. Small tumors had higher tumor control than medium (P ¼ 0.019) or large (P ¼ 0.003) tumors as in other studies; however, multivariate analysis was unable to confirm that both the extent of tumor resection and the preoperative tumor size affected tumor control. Facial Nerve Preservation Facial nerve function is a critical element in quality of life. Facial palsy is a debilitating condition that may lead to difficulties in several areas of daily life, including speaking, eating, and blinking (which may lead to dry eyes and corneal damage).2 Facial palsy is also a movement disorder and an emotional and communication disorder. Many patients with facial palsy experience social isolation and develop depression because of their difficulty in emotionally connecting or communicating with others.3 Previous studies addressed efforts to preserve facial nerve function during resection of VS. Samii et al.25 reported that after 2
weeks of surgery, excellent facial nerve function (H-B grades 1 and 2) was present in 59% of patients, and good function (H-B grade 3) was present in 16% of patients. Several other institutions achieved facial nerve preservation (H-B grades 1 and 2) in 25%e57% of patients.5-8,26 We produced comparable results, with 76.5% of patients showing H-B grades 1e3 and 50.1% of patients showing H-B grades 1 and 2 at 1 month after surgery. When we analyzed facial palsy, we did not include patients with >1 surgery. As the number of patients with >1 surgery was small, there would not have much influence on the results. Also, most patients had radiotherapy or multiple radiosurgeries as well as surgeries, each of which affects facial nerve function and hinders an accurate assessment of changes in facial nerve function. Several studies reported data regarding the contributing factors to postoperative facial nerve function.11,12 We also investigated the relevance of tumor size, the extent of tumor resection, and surgical approach with facial palsy. Although we could not identify
Table 3. Grade of Facial Nerve Function and Preoperative and Postoperative Status Preoperative
Postoperative, Immediate
Postoperative, 1 Month
Postoperative, 1 Year
Postoperative, ‡2 Years
As a Cutoff Value
1
305
100
104
139
113
P ¼ 0.220
2
40
69
73
91
74
P ¼ 0.335
3
5
92
93
44
34
P < 0.001
4
3
82
79
36
29
P ¼ 0.815
5
0
10
4
1
1
6
0
0
0
0
0
H-B Grade
H-B, House-Brackmann.
WORLD NEUROSURGERY 89: 215-222, MAY 2016
www.WORLDNEUROSURGERY.org
219
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
Figure 3. Numbers of patients in each facial palsy group on serial follow-up. The light gray bar shows the number of patients in House-Brackmann grades 1, 2, and 3 (Gr 1e3), and the dark gray bar shows the number of patients in House-Brackmann grades 4, 5, and 6 (Gr 4e6). Most improvement in facial nerve function occurs during the first year after surgery. imm, immediate.
any statistically significant relationship among them, we were able to confirm a similar pattern of worse facial function being expected with large tumors. We did not analyze the effects of preoperative facial nerve function status on postoperative facial palsy, as our objective of this study was to investigate the relationship between postoperative and long-term facial palsy. We found that, after controlling for the extent of resection and preoperative tumor size, the immediate postoperative facial palsy grade was significantly correlated with the postoperative facial palsy grade on serial follow-up (P < 0.001). We used the GEE method to investigate serial facial palsy outcomes, which are repetitive categorical values. The GEE method is used to model correlated data from longitudinal or repeated measures studies and clustered or multilevel studies. In this study, we used the repeated and consecutive measure (H-B grade at each encounter) in the same subject (patient) to find out that the immediate postoperative HB grade was correlated with H-B grades afterward. With this method, we were able to show how patients recovered from facial palsy as time passed and the relevance between facial palsy immediately after surgery and after a long period, quantitatively
Figure 4. Serial percentages of patients in House-Brackmann (H-B) grades 1, 2, and 3 (Gr 1e3). There were no significant differences of the percentages in each group. (A) Groups according to the extent of tumor resection (gross total resection [GTR] vs. near-total resection [NTR], P ¼ 0.841; NTR vs. subtotal resection [STR], P ¼ 0.548; GTR vs. STR, P ¼ 0.690). (B) Groups according to preoperative tumor size (large vs. medium, P ¼ 0.095; medium vs. small, P ¼ 0.421; large vs. small, P ¼ 0.056). (C) Groups according to surgical approach (retrosigmoid [RS] vs. translabyrinthine [TL], P ¼ 0.310; TL vs. middle fossa [MF], P ¼ 0.690; RS vs. MF, P ¼ 0.690). imm, immediate.
220
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.01.038
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
and statistically. We also applied this method to identify the cutoff grade for facial palsy that guarantees improvement of function to at least H-B grade 3 on long-term follow-up. The serial ratios of good functional H-B grade to poor functional H-B grade showed the recovery time for damaged facial nerve. Improvement of facial nerve function mostly occurred in the first year after surgery. A similar pattern was shown when we depicted the ratio of H-B grades 1 and 2 to grades 3e6. This information will help us to explain the prognosis of facial palsy after surgery to patients and their families. In addition, while attempting to achieve total tumor resection during surgery, the aim should be to achieve better than H-B grade 3 immediate postoperative facial palsy. Treatment Strategy for VS The optimal treatment strategy for VS achieves both tumor control and functional preservation of cranial nerves, especially the cochlear and facial nerves. In most cases, cochlear function is already debilitated to nonserviceable hearing status. Advances in radiosurgery have allowed surgeons to place more emphasis on functional preservation of the facial and cochlear nerves. Planned facial nerve preservation surgery followed by radiosurgery on the residual tumor has been established.26 Using this strategy, 1 group did not open the internal auditory canal and resect the tumor inside, which is suitable for GKRS; they achieved a 95% facial preservation rate (H-B grade 1 and 2) at the last follow-up and 92%, 86%, and 86% tumor control rate at 3, 5, and 10 years after surgery.26
REFERENCES 1. Thomassin JM, Pellet W, Epron JP, Braccini F, Roche PH. Recurrent acoustic neurinoma after complete surgical resection [in French]. Ann Otolaryngol Chir Cervicofac. 2001;118:3-10. 2. Cross T, Sheard CE, Garrud P, Nikolopoulos TP, O’Donoghue GM. Impact of facial paralysis on patients with acoustic neuroma. Laryngoscope. 2000;110:1539-1542.
However, tumor regrowth may require additional surgery. We achieved 75.2% tumor control in the STR group, which was significantly different from the GTR and NTR groups. In a previous report, the chance of salvage surgery increased at the time of transient expansion when residual tumors were large, despite plans for surgery with subsequent GKRS.27 We could achieve comparable results in patients with VS with regard to both tumor control and functional preservation of facial nerve with this strategy. Maximal tumor resection followed by radiosurgery is critical and necessary as a planned process for tumor control, but it should not be performed to the degree that may harm the facial nerve or result in H-B grade 3 and higher immediate postoperative facial palsy. CONCLUSIONS Many functional preservation results have been reported after microsurgery or radiosurgery, but there are few guidelines to help surgeons set practical goals for maintaining acceptable facial nerve function after VS surgery. We have shown that immediate postoperative facial palsy reflects long-term facial nerve function. Specifically, immediate postoperative facial palsy of HB grade 3 is the lowest tolerable grade that would guarantee long-term functional improvement. In an era of advanced techniques for microsurgery and radiosurgery, we achieved high tumor control and expanded knowledge of functional preservation.
8. Zhang X, Fei Z, Chen YJ, Fu LA, Zhang JN, Liu WP, et al. Facial nerve function after excision of large acoustic neuromas via the suboccipital retrosigmoid approach. J Clin Neurosci. 2005;12: 405-408. 9. Arriaga MA, Chen DA. Facial function in hearing preservation acoustic neuroma surgery. Arch Otolaryngol Head Neck Surg. 2001;127:543-546.
3. Dobel C, Miltner WH, Witte OW, Volk GF, Guntinas-Lichius O. Emotional impact of facial palsy [in German]. Laryngorhinootologie. 2013;92:9-23.
10. van de Langenberg R, Hanssens PE, van Overbeeke JJ, Verheul JB, Nelemans PJ, de Bondt BJ, et al. Management of large vestibular schwannoma. Part I. Planned subtotal resection followed by Gamma Knife surgery: radiological and clinical aspects. J Neurosurg. 2011;115:875-884.
4. Samii M, Matthies C. Management of 1000 vestibular schwannomas (acoustic neuromas): surgical management and results with an emphasis on complications and how to avoid them. Neurosurgery. 1997;40:11-21 [discussion: 21-23].
11. Carlstrom LP, Copeland WR 3rd, Neff BA, Castner ML, Driscoll CL, Link MJ. Incidence and risk factors of delayed facial palsy after vestibular schwannoma resection. Neurosurgery. 2016;78: 251-255.
5. Jung S, Kang SS, Kim TS, Kim HJ, Jeong SK, Kim SC, et al. Current surgical results of retrosigmoid approach in extralarge vestibular schwannomas. Surg Neurol. 2000;53:370-377 [discussion: 377-378].
12. Falcioni M, Fois P, Taibah A, Sanna M. Facial nerve function after vestibular schwannoma surgery. J Neurosurg. 2011;115:820-826.
6. Samii M, Gerganov VM, Samii A. Functional outcome after complete surgical removal of giant vestibular schwannomas. J Neurosurg. 2010;112: 860-867. 7. Wanibuchi M, Fukushima T, McElveen JT Jr, Friedman AH. Hearing preservation in surgery for large vestibular schwannomas. J Neurosurg. 2009; 111:845-854.
15. Sterkers JM, Morrison GA, Sterkers O, ElDine MM. Preservation of facial, cochlear, and other nerve functions in acoustic neuroma treatment. Otolaryngol Head Neck Surg. 1994;110:146-155. 16. Wiet RJ, Mamikoglu B, Odom L, Hoistad DL. Longterm results of the first 500 cases of acoustic neuroma surgery. Otolaryngol Head Neck Surg. 2001;124:645-651. 17. Zaouche S, Ionescu E, Dubreuil C, Ferber-Viart C. Pre- and intraoperative predictive factors of facial palsy in vestibular schwannoma surgery. Acta Otolaryngol. 2005;125:363-369. 18. Anaizi AN, Gantwerker EA, Pensak ML, Theodosopoulos PV. Facial nerve preservation surgery for koos grade 3 and 4 vestibular schwannomas. Neurosurgery. 2014;75:671-675 [discussion: 676-677; quiz 677]. 19. El-Kashlan HK, Zeitoun H, Arts HA, Hoff JT, Telian SA. Recurrence of acoustic neuroma after incomplete resection. Am J Otol. 2000;21:389-392.
13. Fenton JE, Chin RY, Fagan PA, Sterkers O, Sterkers JM. Predictive factors of long-term facial nerve function after vestibular schwannoma surgery. Otol Neurotol. 2002;23:388-392.
20. Kemink JL, Langman AW, Niparko JK, Graham MD. Operative management of acoustic neuromas: the priority of neurologic function over complete resection. Otolaryngol Head Neck Surg. 1991;104:96-99.
14. Ikeda M, Abiko Y, Kukimoto N, Omori H, Nakazato H, Ikeda K. Clinical factors that influence the prognosis of facial nerve paralysis and the magnitudes of influence. Laryngoscope. 2005; 115:855-860.
21. Seol HJ, Kim CH, Park CK, Kim CH, Kim DG, Chung YS, et al. Optimal extent of resection in vestibular schwannoma surgery: relationship to recurrence and facial nerve preservation. Neurol Med Chir (Tokyo). 2006;46:176-180 [discussion: 180-181].
WORLD NEUROSURGERY 89: 215-222, MAY 2016
www.WORLDNEUROSURGERY.org
221
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
22. House JW. Facial nerve grading systems. Laryngoscope. 1983;93:1056-1069. 23. Gardner G, Robertson JH. Hearing preservation in unilateral acoustic neuroma surgery. Ann Otol Rhinol Laryngol. 1988;97:55-66. 24. Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika. 1986; 73:13-22. 25. Samii M, Gerganov V, Samii A. Improved preservation of hearing and facial nerve function in vestibular schwannoma surgery via the retrosigmoid approach in a series of 200 patients. J Neurosurg. 2006;105:527-535.
222
www.SCIENCEDIRECT.com
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
26. Iwai Y, Ishibashi K, Watanabe Y, Uemura G, Yamanaka K. Functional preservation after planned partial resection followed by Gamma Knife radiosurgery for large vestibular schwannomas. World Neurosurg. 2015;84:292-300. 27. Pan HC, Sheehan J, Sheu ML, Chiu WT, Yang DY. Intracapsular decompression or radical resection followed by Gamma Knife surgery for patients harboring a large vestibular schwannoma. J Neurosurg. 2012;117(suppl):69-77.
commercial or financial relationships that could be construed as a potential conflict of interest. Received 1 September 2015; accepted 19 January 2016 Citation: World Neurosurg. (2016) 89:215-222. http://dx.doi.org/10.1016/j.wneu.2016.01.038 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.01.038
Functional Outcome of the Facial Nerve After Surgery for Vestibular Schwannoma: Prediction of Acceptable Long-Term Facial Nerve Function Based on Immediate Postoperative Facial Palsy Seunghoon Lee1, Ho Jun Seol1, Kwan Park1, Jung-Il Lee1, Do-Hyun Nam1, Doo-Sik Kong1, Yang-Sun Cho2
OBJECTIVE: To examine the relationship between immediate postoperative facial palsy and long-term facial palsy and identify a treatment strategy for vestibular schwannoma considering long-term outcomes of facial nerve function and tumor control.
-
METHODS: Patients (N [ 385) who underwent surgery in a single institution were reviewed retrospectively; 12 patients with neurofibromatosis, 6 with preoperative radiosurgery, and 14 with multiple surgeries were excluded. The generalized estimating equation method was used to show the correlation between immediate and later postoperative facial palsy and to identify the cutoff grade of immediate postoperative facial palsy.
-
RESULTS: The tumor control rates for 1 year, 3 years, and 5 years were 88.7%, 83.9%, and 80.0%. Preservation of facial function above House-Brackmann (H-B) grades 1 and 2 was achieved in 47.9% of patients immediately postoperatively, in 50.1% after 1 month, and in 74.5% after >2 years. The immediate postoperative facial palsy grade showed a statistically significant relationship with the facial palsy grade on long-term follow-up (P < 0.001). H-B grade 3 immediate postoperative facial palsy was identified as the cutoff grade that showed the most significant relationship between the grade of immediate postoperative facial palsy and the grades above the cutoff (H-B grade 1e3) on long-term follow-up (P < 0.001).
-
CONCLUSIONS: H-B grade of immediate postoperative facial palsy can predict facial palsy at long-term follow-up.
-
Key words - Facial nerve preservation surgery - Facial palsy - Vestibular schwannoma Abbreviations and Acronyms GEE: Generalized estimating equation GKRS: Gamma Knife radiosurgery GTR: Gross total resection H-B: House-Brackmann MRI: Magnetic resonance imaging NTR: Near-total resection STR: Subtotal resection VS: Vestibular schwannoma
WORLD NEUROSURGERY 89: 215-222, MAY 2016
H-B grade 3 immediate postoperative facial palsy is the lowest tolerable grade that guarantees functional improvement on long-term follow-up. Planned facial nerve preservation surgery followed by radiosurgery is thought to be optimal treatment in patients with vestibular schwannoma for both tumor control and facial nerve function.
INTRODUCTION
M
icrosurgical complete resection has been the treatment of choice for vestibular schwannoma (VS), the most common intracranial schwannoma. The goals of surgery are to render the patient free from tumor growth and from mortality and neurologic morbidity caused by continuous uncontrolled tumor growth.1 Advances in microsurgical techniques have achieved complete tumor resection with low or no mortality and shifted our concerns from mortality toward neurologic morbidity. Facial palsy is one of the most serious morbidities because loss of facial expression is a functionally and psychologically debilitating condition.2,3 Reducing the high risk of facial palsy after surgery has become a major issue in treatment of VS. Maximal tumor resection with facial nerve preservation has been advocated by several surgeons,4 and many institutions reported excellent outcomes in tumor control and facial nerve preservation.5-8 Nevertheless, the incidence of postoperative facial palsy can be 30%e80%, especially for large tumors.9 Radiosurgery, such as Gamma Knife radiosurgery (GKRS), has decreased the recurrence rate in patients who had a partial tumor resection to save facial nerve function. Given the acceptable outcomes of
From the Departments of 1Neurosurgery-Head and Neck Surgery, and 2 Otorhinolaryngology-Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea To whom correspondence should be addressed: Ho Jun Seol, M.D., Ph.D. [E-mail: [email protected]] Citation: World Neurosurg. (2016) 89:215-222. http://dx.doi.org/10.1016/j.wneu.2016.01.038 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
215
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
radiosurgery for residual tumor after surgery, we have been hesitant to perform complete resection of a VS at the expense of facial nerve function. Several studies suggested that planned near-total resection (NTR) or subtotal resection (STR) along with postoperative radiosurgery for residual tumor allowed for the best combination of facial nerve function and tumor control.10 There have been no well-established measures except preoperative tumor size and surgical approach to predict long-term prognosis of facial palsy after VS surgery.11-17 We could not locate any previous studies examining the relationship between immediate postoperative facial palsy and long-term serial outcomes of facial palsy. In this study, we review our surgical series and analyze serial facial nerve functions to identify the most efficient treatment strategy for patients with VS based on long-term outcomes of facial nerve function as well as tumor control. MATERIALS AND METHODS This study was approved by the Samsung Medical Center institutional review board. During the period 1995e2013, 385 patients with VS were operated on at our institution. Medical records and images of patients were reviewed retrospectively. We included only patients with sporadic unilateral VS. There were 12 patients with neurofibromatosis type 2 who were excluded because of the fast and aggressive growth of the disease. Of the remaining 373 patients, we selected patients who were followed after being treated with microsurgery with or without radiotherapy or radiosurgery, including GKRS, for residual or recurrent tumor. To show the exact relationship between microsurgery and facial palsy, we excluded 6 patients who had microsurgery after preoperative radiosurgery and 14 patients who had multiple surgeries, including staged operations. One patient who died was excluded because multiple surgeries were involved and postoperative facial and hearing functions were not evaluated. Tumor control was assessed without 12 neurofibromatosis patients so as not to exaggerate the proportion of patients free of tumor progression (Figure 1). Follow-up at the outpatient clinic was conducted 1 month, 1 year, and 2 years after the surgery. Tumor size was measured using the greatest dimension on preoperative and postoperative axial magnetic resonance imaging (MRI). For a patient with a tumor exceeding a maximal diameter of 2.5 cm who is competent to general anesthesia, microsurgery is recommended first. However, radiosurgery is considered in cases of patients with low performance status or on the basis of patient preference. During an operation, we pursue a maximal resection without taking the risk of harming any vessels or nerves. Radiosurgery is provided for persistent residual tumors. The microsurgeries were performed by neurosurgeons and otorhinolaryngologists at our institution. Surgical approaches were determined based on patient age, tumor size and location, and initial neurologic manifestations. The extent of tumor resection, which was classified as gross total resection (GTR), NTR, or STR, was assessed by reviewing surgery records and comparing preoperative and postoperative MRI; postoperative MRI was obtained within 48 hours in most cases or within 1e3 months after surgery in other cases. NTR was defined as resection with <5 mm of a residual intracanalicular tumor or tumor adhering to the facial nerve or brainstem that was intentionally left behind and
216
www.SCIENCEDIRECT.com
Figure 1. Flowchart showing the numbers of included and excluded patients. There were 373 patients analyzed as tumor controls after excluding 12 patients with neurofibromatosis (NF type II). Facial palsy was assessed in 353 patients after further exclusion of 6 patients following preoperative radiosurgery and 14 patients who underwent multiple surgeries.
not seen on postoperative MRI. Cases with residual tumors with maximal thickness >5 mm were considered a STR.18-21 Facial nerve function was clinically assessed by the HouseBrackmann (H-B) scale.22 Facial palsy grading of patients was completed by 3 parties: operators, residents in the intensive care unit or ward, and nurses in the neurosurgical intensive care unit or ward. If there were any discrepancies in grading, the worst grade was taken. Facial nerve function was graded preoperatively, immediately postoperatively (promptly after patients are taken out of the operating room, extubated and fully awake), 1 month postoperatively, 1 year postoperatively, and 2 years postoperatively. At each outpatient clinic visit, operators and neurosurgical physician assistants did the grading. To evaluate hearing, the pure-tone average and speech discrimination score were measured. The pure-tone average was calculated as an average of pure-tone thresholds at 3 frequencies (500, 1000, and 2000 Hz). The speech discrimination score was defined as the percentage of words correctly repeated from a standardized list. With audiometric results, patients were categorized into serviceable hearing and nonserviceable hearing groups according to the Gardner-Robertson hearing scale.23 Serviceable hearing was defined as Gardner-Robertson hearing grades 1e2, and nonserviceable hearing was defined as a puretone average <50 dB or a speech discrimination score <50%. The comparison of facial palsy and tumor control in each group was conducted using the Kaplan-Meier method, Mann-Whitney test, log-rank test, and Cox proportional hazards regression model. To show the correlation between immediate postoperative facial palsy and facial palsy thereafter, the generalized estimating equation (GEE) method24 was used because facial palsy grades at
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.01.038
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
each follow-up visit in each patient had repetitive categorical values. With the GEE method, we also identified the cutoff grade of immediate postoperative facial palsy that led us to expect longterm functional improvement of the facial nerve. To identify this value, every H-B grade from 1 to 5 (none of the patients showed H-B grade 6 facial palsy) was assessed as a cutoff value and compared with the groups above and below it. The grade with the smallest P value after Bonferroni correction was identified as a cutoff grade. The Bonferroni method was used to prevent an inflated type I error. The result was considered statistically significant if the P value was < 0.05. All statistical analyses were performed using IBM SPSS Statistics version 22 (IBM Corporation, Armonk, New York, USA).
Table 1. Characteristics of Patients with Vestibular Schwannomas Characteristics
Number (%)
Total number
P Value
353
Sex Female/Male
199 (56.4)/154 (43.6)
Age (years) Mean
49
Range
21e83
Presenting symptoms
RESULTS Among the 353 enrolled patients, the number of female patients was 199, and the number of male patients was 154. The mean age of patients was 49.0 years (range, 21e83 years). The most common presenting symptom was hearing difficulty in 210 (56.3%) patients, followed by tinnitus in 172 (46.1%), dizziness or vertigo in 94 (25.2%), headache in 75 (20.1%), facial paresthesia in 65 (17.4%), and facial palsy in 19 (5.1%). There was no predominant tumor side (right, 164; left, 189). The preoperative mean tumor size was 30.2 mm (range, 5.0e60.0 mm). Tumors were categorized as large (>40 mm), medium (21e40 mm), or small (20 mm), with 70 (19.8%), 197 (55.8%), and 86 (24.4%) patients in the large, medium, and small groups. We used a lateral suboccipital retrosigmoid approach in 165 patients (75.6%), a translabyrinthine approach in 77 patients (22.9%), and a middle fossa approach in 4 patients (1.4%). All translabyrinthine and middle fossa approaches were performed in collaboration with otorhinolaryngologists. GTR was achieved in 183 (51.8%) patients, NTR was achieved in 64 (18.1%) patients, and STR was achieved in 106 patients (30.0%) (Table 1). After excluding 12 patients with neurofibromatosis type 2, we observed 55 cases of recurrence. Tumors in 6 patients recurred after preoperative radiosurgery, 14 patients had recurrent tumors and were operated on more than twice, and 35 patients experienced recurrence after microsurgery (NTR or STR) with radiotherapy or radiosurgery (Figure 2A). The tumor control rates for 1 year, 3 years, and 5 years were 88.7%, 83.9%, and 80.0%. By using the log-rank test, patients with small tumors achieved better tumor control than patients with medium or large tumors (P ¼ 0.018) (Figure 2B). Also, patients who had GTR achieved better tumor control compared with patients who had NTR or STR (P < 0.001) (Figure 2C). However, there were no statistically significant differences in tumor control according to preoperative tumor size and the extent of tumor resection on the Cox regression model analysis. Tumor controls for each microsurgical approach did not differ significantly on the logrank test (P ¼ 0.179) or the Cox regression model (P ¼ 0.807) (Table 2). Based on an audiometric test, hearing preservation was achieved within 1 year after surgery in 12 of 88 patients with preoperative serviceable hearing (13.6%). In 5 patients, delayed hearing loss resulted in nonserviceable hearing (preserved hearing for 6 months after surgery and then deterioration).
WORLD NEUROSURGERY 89: 215-222, MAY 2016
Hearing difficulty
210 (56.3)
Tinnitus
172 (46.1)
Dizziness or vertigo
94 (25.2)
Headache
75 (20.1)
Facial paresthesia
65 (17.4)
Facial palsy
19 (5.1)
Tumor side Right/left
164 (46.5)/189 (53.5)
Tumor size (mm)
0.018
Mean
30.2
Range
5.0e60.0
Large (>40 mm)
70 (19.8)
Medium (21e40 mm)
197 (55.8)
Small (20 mm)
86 (24.4)
Surgical approach
0.179
RS
267 (75.6)
TL
81 (22.9)
MF
5 (1.4) < 0.001
Extent of tumor resection GTR
183 (51.8)
NTR
64 (18.1)
STR
106 (30.0)
RS, retrosigmoid; TL, translabyrinthine; MF, middle fossa; GTR, gross total resection; NTR, near-total resection; STR, subtotal resection.
Preoperative and postoperative facial nerve function assessments are provided in Table 3. Facial functional preservation above an acceptable or useful facial function, H-B grade 1 and 2, was achieved in 47.9% of patients immediately after surgery, in 50.1% 1 month after surgery, and in 74.5% 2 years after surgery. Using the GEE method, the immediate postoperative facial palsy grade showed a statistically significant relationship with serial postoperative facial palsy grades (P < 0.001) after controlling for the extent of resection and tumor size. We identified H-B grade 3 as the cutoff that showed the most significant relationship
www.WORLDNEUROSURGERY.org
217
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
between the grade of immediate postoperative facial palsy and the grades above the cutoff grade (H-B grade 1e3) on long-term follow-up (P < 0.001). As few patients had H-B grade 5 and 6 postoperative facial palsy, the statistical analysis failed to produce significant results (Table 3). The number of patients in each facial palsy group is shown in Figure 3. The proportion of patients in H-B grade 1e3 increased over the first year after surgery and then remained steady. Comparisons of the percentages of H-B grade 1e3 patients in each group according to the extent of tumor resection (Figure 4A), tumor size (Figure 4B), and surgical approach (Figure 4C) did not show significant differences.
DISCUSSION Tumor Control We achieved 88.7%, 83.9%, and 80.0% tumor control at 1 year, 3 years, and 5 years postoperatively. We confirmed that maximal tumor resection was important in tumor control, as we observed significantly different tumor controls between the GTR and STR groups (P < 0.001). Tumor control for GTR was 91.9% and for NTR or STR was 92.1% when followed by adjuvant radiotherapy or radiosurgery. This result affirms that planned microsurgery followed by radiosurgery for residual tumors leads to functional
Figure 2. Tumor progression-free (recurrence-free) survival of 373 patients with vestibular schwannoma. (A) Overall tumor control was achieved in 85.3% of patients during a median follow-up period of 43.0 months (range, 0.4e228.5 months). (B) Tumor controls for each microsurgery group were classified by extent of tumor resection: 91.9%, gross total resection (GTR); 84.3%, near-total resection (NTR); and 75.2%, subtotal resection (STR). There was a significant long-term difference between GTR and STR (P < 0.001), but the differences between GTR and NTR (P ¼ 0.061) and between NTR and STR (P ¼ 0.091) were not significant. Controls for tumor size were 79.7% in the large tumor group, 83.3% in the medium tumor group, and 94.4% in the small tumor group. (C) There was a significant long-term difference between the small tumor group and the medium (P ¼ 0.019) or large (P ¼ 0.003) tumor groups, but not between medium and large (P ¼ 0.310) tumor groups.
218
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.01.038
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
Table 2. Tumor Control in Groups Classified by Extent of Tumor Resection, Preoperative Tumor Size, and Microsurgical Approach Controlled (%)
Recurred (%)
HR (95% CI)
P Value
Extent of tumor resection
0.171
GTR
171 (91.9)
15 (8.1)
1.800 (0.883e3.668)
0.105
NTR
59 (84.3)
11 (15.7)
0.875 (0.434e1.763)
0.709
STR
88 (75.2)
29 (24.8)
59 (79.7)
11 (20.3)
0.798 (0.221e2.875)
0.730
175 (83.3)
22 (16.7)
0.613 (0.182e2.066)
0.430
84 (94.4)
2 (5.6)
Preoperative tumor size
0.562
Large Medium Small Microsurgical approach
0.807 $
RS
235 (83.0)
48 (17.0)
TL
79 (92.9)
6 (7.1)
1.138 (0.435e2.977)
0.792
MF
4 (80)
1 (20.0)
2.270 (0.190e27.135)
0.517
HR, hazard ratio; CI, confidence interval; GTR, gross total resection; NTR, near-total resection; STR, subtotal resection; RS, retrosigmoid; TL, translabyrinthine; MF, middle fossa.
preservation and tumor control. Small tumors had higher tumor control than medium (P ¼ 0.019) or large (P ¼ 0.003) tumors as in other studies; however, multivariate analysis was unable to confirm that both the extent of tumor resection and the preoperative tumor size affected tumor control. Facial Nerve Preservation Facial nerve function is a critical element in quality of life. Facial palsy is a debilitating condition that may lead to difficulties in several areas of daily life, including speaking, eating, and blinking (which may lead to dry eyes and corneal damage).2 Facial palsy is also a movement disorder and an emotional and communication disorder. Many patients with facial palsy experience social isolation and develop depression because of their difficulty in emotionally connecting or communicating with others.3 Previous studies addressed efforts to preserve facial nerve function during resection of VS. Samii et al.25 reported that after 2
weeks of surgery, excellent facial nerve function (H-B grades 1 and 2) was present in 59% of patients, and good function (H-B grade 3) was present in 16% of patients. Several other institutions achieved facial nerve preservation (H-B grades 1 and 2) in 25%e57% of patients.5-8,26 We produced comparable results, with 76.5% of patients showing H-B grades 1e3 and 50.1% of patients showing H-B grades 1 and 2 at 1 month after surgery. When we analyzed facial palsy, we did not include patients with >1 surgery. As the number of patients with >1 surgery was small, there would not have much influence on the results. Also, most patients had radiotherapy or multiple radiosurgeries as well as surgeries, each of which affects facial nerve function and hinders an accurate assessment of changes in facial nerve function. Several studies reported data regarding the contributing factors to postoperative facial nerve function.11,12 We also investigated the relevance of tumor size, the extent of tumor resection, and surgical approach with facial palsy. Although we could not identify
Table 3. Grade of Facial Nerve Function and Preoperative and Postoperative Status Preoperative
Postoperative, Immediate
Postoperative, 1 Month
Postoperative, 1 Year
Postoperative, ‡2 Years
As a Cutoff Value
1
305
100
104
139
113
P ¼ 0.220
2
40
69
73
91
74
P ¼ 0.335
3
5
92
93
44
34
P < 0.001
4
3
82
79
36
29
P ¼ 0.815
5
0
10
4
1
1
6
0
0
0
0
0
H-B Grade
H-B, House-Brackmann.
WORLD NEUROSURGERY 89: 215-222, MAY 2016
www.WORLDNEUROSURGERY.org
219
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
Figure 3. Numbers of patients in each facial palsy group on serial follow-up. The light gray bar shows the number of patients in House-Brackmann grades 1, 2, and 3 (Gr 1e3), and the dark gray bar shows the number of patients in House-Brackmann grades 4, 5, and 6 (Gr 4e6). Most improvement in facial nerve function occurs during the first year after surgery. imm, immediate.
any statistically significant relationship among them, we were able to confirm a similar pattern of worse facial function being expected with large tumors. We did not analyze the effects of preoperative facial nerve function status on postoperative facial palsy, as our objective of this study was to investigate the relationship between postoperative and long-term facial palsy. We found that, after controlling for the extent of resection and preoperative tumor size, the immediate postoperative facial palsy grade was significantly correlated with the postoperative facial palsy grade on serial follow-up (P < 0.001). We used the GEE method to investigate serial facial palsy outcomes, which are repetitive categorical values. The GEE method is used to model correlated data from longitudinal or repeated measures studies and clustered or multilevel studies. In this study, we used the repeated and consecutive measure (H-B grade at each encounter) in the same subject (patient) to find out that the immediate postoperative HB grade was correlated with H-B grades afterward. With this method, we were able to show how patients recovered from facial palsy as time passed and the relevance between facial palsy immediately after surgery and after a long period, quantitatively
Figure 4. Serial percentages of patients in House-Brackmann (H-B) grades 1, 2, and 3 (Gr 1e3). There were no significant differences of the percentages in each group. (A) Groups according to the extent of tumor resection (gross total resection [GTR] vs. near-total resection [NTR], P ¼ 0.841; NTR vs. subtotal resection [STR], P ¼ 0.548; GTR vs. STR, P ¼ 0.690). (B) Groups according to preoperative tumor size (large vs. medium, P ¼ 0.095; medium vs. small, P ¼ 0.421; large vs. small, P ¼ 0.056). (C) Groups according to surgical approach (retrosigmoid [RS] vs. translabyrinthine [TL], P ¼ 0.310; TL vs. middle fossa [MF], P ¼ 0.690; RS vs. MF, P ¼ 0.690). imm, immediate.
220
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.01.038
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
and statistically. We also applied this method to identify the cutoff grade for facial palsy that guarantees improvement of function to at least H-B grade 3 on long-term follow-up. The serial ratios of good functional H-B grade to poor functional H-B grade showed the recovery time for damaged facial nerve. Improvement of facial nerve function mostly occurred in the first year after surgery. A similar pattern was shown when we depicted the ratio of H-B grades 1 and 2 to grades 3e6. This information will help us to explain the prognosis of facial palsy after surgery to patients and their families. In addition, while attempting to achieve total tumor resection during surgery, the aim should be to achieve better than H-B grade 3 immediate postoperative facial palsy. Treatment Strategy for VS The optimal treatment strategy for VS achieves both tumor control and functional preservation of cranial nerves, especially the cochlear and facial nerves. In most cases, cochlear function is already debilitated to nonserviceable hearing status. Advances in radiosurgery have allowed surgeons to place more emphasis on functional preservation of the facial and cochlear nerves. Planned facial nerve preservation surgery followed by radiosurgery on the residual tumor has been established.26 Using this strategy, 1 group did not open the internal auditory canal and resect the tumor inside, which is suitable for GKRS; they achieved a 95% facial preservation rate (H-B grade 1 and 2) at the last follow-up and 92%, 86%, and 86% tumor control rate at 3, 5, and 10 years after surgery.26
REFERENCES 1. Thomassin JM, Pellet W, Epron JP, Braccini F, Roche PH. Recurrent acoustic neurinoma after complete surgical resection [in French]. Ann Otolaryngol Chir Cervicofac. 2001;118:3-10. 2. Cross T, Sheard CE, Garrud P, Nikolopoulos TP, O’Donoghue GM. Impact of facial paralysis on patients with acoustic neuroma. Laryngoscope. 2000;110:1539-1542.
However, tumor regrowth may require additional surgery. We achieved 75.2% tumor control in the STR group, which was significantly different from the GTR and NTR groups. In a previous report, the chance of salvage surgery increased at the time of transient expansion when residual tumors were large, despite plans for surgery with subsequent GKRS.27 We could achieve comparable results in patients with VS with regard to both tumor control and functional preservation of facial nerve with this strategy. Maximal tumor resection followed by radiosurgery is critical and necessary as a planned process for tumor control, but it should not be performed to the degree that may harm the facial nerve or result in H-B grade 3 and higher immediate postoperative facial palsy. CONCLUSIONS Many functional preservation results have been reported after microsurgery or radiosurgery, but there are few guidelines to help surgeons set practical goals for maintaining acceptable facial nerve function after VS surgery. We have shown that immediate postoperative facial palsy reflects long-term facial nerve function. Specifically, immediate postoperative facial palsy of HB grade 3 is the lowest tolerable grade that would guarantee long-term functional improvement. In an era of advanced techniques for microsurgery and radiosurgery, we achieved high tumor control and expanded knowledge of functional preservation.
8. Zhang X, Fei Z, Chen YJ, Fu LA, Zhang JN, Liu WP, et al. Facial nerve function after excision of large acoustic neuromas via the suboccipital retrosigmoid approach. J Clin Neurosci. 2005;12: 405-408. 9. Arriaga MA, Chen DA. Facial function in hearing preservation acoustic neuroma surgery. Arch Otolaryngol Head Neck Surg. 2001;127:543-546.
3. Dobel C, Miltner WH, Witte OW, Volk GF, Guntinas-Lichius O. Emotional impact of facial palsy [in German]. Laryngorhinootologie. 2013;92:9-23.
10. van de Langenberg R, Hanssens PE, van Overbeeke JJ, Verheul JB, Nelemans PJ, de Bondt BJ, et al. Management of large vestibular schwannoma. Part I. Planned subtotal resection followed by Gamma Knife surgery: radiological and clinical aspects. J Neurosurg. 2011;115:875-884.
4. Samii M, Matthies C. Management of 1000 vestibular schwannomas (acoustic neuromas): surgical management and results with an emphasis on complications and how to avoid them. Neurosurgery. 1997;40:11-21 [discussion: 21-23].
11. Carlstrom LP, Copeland WR 3rd, Neff BA, Castner ML, Driscoll CL, Link MJ. Incidence and risk factors of delayed facial palsy after vestibular schwannoma resection. Neurosurgery. 2016;78: 251-255.
5. Jung S, Kang SS, Kim TS, Kim HJ, Jeong SK, Kim SC, et al. Current surgical results of retrosigmoid approach in extralarge vestibular schwannomas. Surg Neurol. 2000;53:370-377 [discussion: 377-378].
12. Falcioni M, Fois P, Taibah A, Sanna M. Facial nerve function after vestibular schwannoma surgery. J Neurosurg. 2011;115:820-826.
6. Samii M, Gerganov VM, Samii A. Functional outcome after complete surgical removal of giant vestibular schwannomas. J Neurosurg. 2010;112: 860-867. 7. Wanibuchi M, Fukushima T, McElveen JT Jr, Friedman AH. Hearing preservation in surgery for large vestibular schwannomas. J Neurosurg. 2009; 111:845-854.
15. Sterkers JM, Morrison GA, Sterkers O, ElDine MM. Preservation of facial, cochlear, and other nerve functions in acoustic neuroma treatment. Otolaryngol Head Neck Surg. 1994;110:146-155. 16. Wiet RJ, Mamikoglu B, Odom L, Hoistad DL. Longterm results of the first 500 cases of acoustic neuroma surgery. Otolaryngol Head Neck Surg. 2001;124:645-651. 17. Zaouche S, Ionescu E, Dubreuil C, Ferber-Viart C. Pre- and intraoperative predictive factors of facial palsy in vestibular schwannoma surgery. Acta Otolaryngol. 2005;125:363-369. 18. Anaizi AN, Gantwerker EA, Pensak ML, Theodosopoulos PV. Facial nerve preservation surgery for koos grade 3 and 4 vestibular schwannomas. Neurosurgery. 2014;75:671-675 [discussion: 676-677; quiz 677]. 19. El-Kashlan HK, Zeitoun H, Arts HA, Hoff JT, Telian SA. Recurrence of acoustic neuroma after incomplete resection. Am J Otol. 2000;21:389-392.
13. Fenton JE, Chin RY, Fagan PA, Sterkers O, Sterkers JM. Predictive factors of long-term facial nerve function after vestibular schwannoma surgery. Otol Neurotol. 2002;23:388-392.
20. Kemink JL, Langman AW, Niparko JK, Graham MD. Operative management of acoustic neuromas: the priority of neurologic function over complete resection. Otolaryngol Head Neck Surg. 1991;104:96-99.
14. Ikeda M, Abiko Y, Kukimoto N, Omori H, Nakazato H, Ikeda K. Clinical factors that influence the prognosis of facial nerve paralysis and the magnitudes of influence. Laryngoscope. 2005; 115:855-860.
21. Seol HJ, Kim CH, Park CK, Kim CH, Kim DG, Chung YS, et al. Optimal extent of resection in vestibular schwannoma surgery: relationship to recurrence and facial nerve preservation. Neurol Med Chir (Tokyo). 2006;46:176-180 [discussion: 180-181].
WORLD NEUROSURGERY 89: 215-222, MAY 2016
www.WORLDNEUROSURGERY.org
221
ORIGINAL ARTICLE SEUNGHOON LEE ET AL.
22. House JW. Facial nerve grading systems. Laryngoscope. 1983;93:1056-1069. 23. Gardner G, Robertson JH. Hearing preservation in unilateral acoustic neuroma surgery. Ann Otol Rhinol Laryngol. 1988;97:55-66. 24. Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika. 1986; 73:13-22. 25. Samii M, Gerganov V, Samii A. Improved preservation of hearing and facial nerve function in vestibular schwannoma surgery via the retrosigmoid approach in a series of 200 patients. J Neurosurg. 2006;105:527-535.
222
www.SCIENCEDIRECT.com
FACIAL NERVE FUNCTION AFTER VESTIBULAR SCHWANNOMA SURGERY
26. Iwai Y, Ishibashi K, Watanabe Y, Uemura G, Yamanaka K. Functional preservation after planned partial resection followed by Gamma Knife radiosurgery for large vestibular schwannomas. World Neurosurg. 2015;84:292-300. 27. Pan HC, Sheehan J, Sheu ML, Chiu WT, Yang DY. Intracapsular decompression or radical resection followed by Gamma Knife surgery for patients harboring a large vestibular schwannoma. J Neurosurg. 2012;117(suppl):69-77.
commercial or financial relationships that could be construed as a potential conflict of interest. Received 1 September 2015; accepted 19 January 2016 Citation: World Neurosurg. (2016) 89:215-222. http://dx.doi.org/10.1016/j.wneu.2016.01.038 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2016 Elsevier Inc. All rights reserved.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2016.01.038