The Association of Vestibular Schwannoma Volume With Facial Nerve Outcomes After Surgical Resection

Development and application of explainable artificial intelligence using machine learning classification for long-term facial nerve function after vestibular schwannoma surgery

PURPOSE

Vestibular schwannomas (VSs) represent the most common cerebellopontine angle tumors, posing a challenge in preserving facial nerve (FN) function during surgery. We employed the Extreme Gradient Boosting machine learning classifier to predict long-term FN outcomes (classified as House-Brackmann grades 1-2 for good outcomes and 3-6 for bad outcomes) after VS surgery.

METHODS

In a retrospective analysis of 256 patients, comprehensive pre-, intra-, and post-operative factors were examined. We applied the machine learning (ML) classifier Extreme Gradient Boosting (XGBoost) for the following binary classification: long-term good and bad FN outcome after VS surgery To enhance the interpretability of our model, we utilized an explainable artificial intelligence approach.

RESULTS

Short-term FN function (tau = 0.6) correlated with long-term FN function. The model exhibited an average accuracy of 0.83, a ROC AUC score of 0.91, and Matthew's correlation coefficient score of 0.62. The most influential feature, identified through SHapley Additive exPlanations (SHAP), was short-term FN function. Conversely, large tumor volume and absence of preoperative auditory brainstem responses were associated with unfavorable outcomes.

CONCLUSIONS

We introduce an effective ML model for classifying long-term FN outcomes following VS surgery. Short-term FN function was identified as the key predictor of long-term function. This model's excellent ability to differentiate bad and good outcomes makes it useful for evaluating patients and providing recommendations regarding FN dysfunction management.

Retrosigmoid versus middle fossa approach for hearing and facial nerve preservation in vestibular schwannoma surgery: A systematic review and comparative meta-analysis

BACKGROUND

Vestibular schwannomas (VS) are benign tumors arising from vestibular nerve's Schwann cells. Surgical resection via retrosigmoid (RS) or middle fossa (MF) is standard, but the optimal approach remains debated. This meta-analysis evaluated RS and MF approaches for VS management, emphasizing hearing preservation and Cranial nerve seven (CN VII) outcomes stratified by tumor size.

METHODS

Systematic searches across PubMed, Cochrane, Web of Science, and Embase identified relevant studies. Hearing and CN VII outcomes were gauged using the American Academy of Otolaryngology-Head and Neck Surgery, Gardner Robertson, and House-Brackmann scores.

RESULTS

Among 7228 patients, 56 % underwent RS and 44 % MF. For intracanalicular tumors, MF recorded 38 % hearing loss, compared to RS's 54 %. In small tumors (<1.5 cm), MF showed 41 % hearing loss, contrasting RS's lower 15 %. Medium-sized tumors (1.5 cm-2.9 cm) revealed 68 % hearing loss in MF and 55 % in RS. Large tumors (>3cm) were only reported in RS with a hearing loss rate of 62 %.

CONCLUSION

Conclusively, while MF may be preferable for intracanalicular tumors, RS demonstrated superior hearing preservation for small to medium-sized tumors. This research underlines the significance of stratified outcomes by tumor size, guiding surgical decisions and enhancing patient outcomes.

Temporal lobe injury with middle fossa approach to intracanalicular vestibular schwannomas: a systematic review

There are several surgical approaches for vestibular schwannoma (VS) resection. However, management has gradually shifted from microsurgical resection, toward surveillance and radiosurgery. One of the arguments against microsurgery via the middle fossa approach (MFA) is the risk of temporal lobe retraction injury or sequelae. Here, we sought to evaluate the incidence of temporal lobe retraction injury or sequela from a MFA via a systematic review of the existing literature. This systematic review was conducted according to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines. Relevant studies reporting temporal lobe injury or sequela during MFA for VS were identified. Data was aggregated and subsequently analyzed to evaluate the incidence of temporal lobe injury. 22 studies were included for statistical analysis, encompassing 1522 patients that underwent VS resection via MFA. The overall rate of temporal lobe sequelae from this approach was 0.7%. The rate of CSF leak was 5.9%. The rate of wound infection was 0.6%. Meningitis occurred in 1.6% of patients. With the MFA, 92% of patients had good facial outcomes, and 54.9% had hearing preservation. Our series and literature review support that temporal lobe retraction injury or sequelae is an infrequent complication from an MFA for intracanalicular VS resection.

Wait-and-scan management in sporadic Koos grade 4 vestibular schwannomas: A longitudinal volumetric study

Background

Volumetric natural history studies specifically on large vestibular schwannomas (VSs), commonly classified as Koos grade 4, are lacking. The aim of the current study is to present the volumetric tumor evolution in sporadic Koos grade 4 VSs and possible predictors for tumor growth.

Methods

Volumetric tumor measurements and tumor evolution patterns from serial MRI studies were analyzed from selected consecutive patients with Koos grade 4 VS undergoing initial wait-and-scan management between January 2001 and July 2020. The significant volumetric threshold was defined as a change in volume of ≥10%.

Results

Among 215 tumors with a median size (IQR) of 2.7 cm3 (1.8-4.2), 147 tumors (68%) demonstrated growth and 75 tumors (35%) demonstrated shrinkage during follow-up. Growth-free survival rates (95% CI) at 1, 2, 5, and 10 years were 55% (48-61), 36% (29-42), 29% (23-36), and 28% (21-34), respectively and did not significantly differ in tumors> 20 mm (Chi-square = .40; P-value = .53). Four tumor evolution patterns (% of total) were observed: continued growth (60); initial growth then shrinkage (7); continued shrinkage (27); and stability (5). Good hearing (adjusted HR 2.21, 95% CI 1.48-3.30; P < .001) and peritumoral edema (adjusted HR 2.22, 95% CI 1.18-4.13; P = .01) at diagnosis were significantly associated with an increased likelihood of growth.

Conclusions

Koos grade 4 VSs show a wide variety in size and growth. Due to variable growth patterns, an initial wait-and-scan strategy with short scan intervals may be an acceptable option in selected tumors, if no significant clinical symptoms of mass effect that warrant treatment are present.

[Our approach to the treatment of vestibular schwannomas with arachnoid dissection of the facial nerve]

Preserving the function of the facial nerve is extremely important in surgery for vestibular schwannomas. Two methods of arachnoid dissection are described for resection of vestibular schwannoma via retrosigmoid approach (from the brain stem and internal auditory canal).

OBJECTIVE

To evaluate the results of arachnoid dissection of the facial nerve from internal auditory canal when resecting the vestibular schwannoma.

MATERIAL AND METHODS

We analyzed 61 patients with vestibular schwannomas. Patients were divided into 2 groups depending on surgical technique. We estimated facial nerve function before and after surgery, preoperative dimension of vestibular schwannoma and extent of resection. The influence of various factors on extent of resection and postoperative facial nerve function was studied.

RESULTS

Vestibular schwannoma resection from the brain stem was performed in 30 patients, arachnoid dissection - in 31 patients. There was no significant between-group difference. Gross total resection was performed in 78.7% of cases. Both techniques demonstrated similar results regarding extent of resection. Arachnoid dissection showed the advantage regarding facial nerve function immediately after surgery (p=0.012) and 6 months later (p<0.001). Normal facial nerve function in 6 months after arachnoid dissection was observed in 80.7% of patients. Preoperative dimension of tumor influenced facial nerve function in addition to technique of resection (p=0.001).

CONCLUSION

We identified the factors influencing facial nerve function after resection of vestibular schwannoma. Surgical technique was the most significant factor. These data expand and popularize arachnoid dissection in surgery of vestibular schwannomas.

Transcanal Transpromontorial Approaches to the Internal Auditory Canal: A Systematic Review

OBJECTIVE

Exclusive endoscopic (EETTA) and expanded (ExpTTA) transcanal transpromontorial approaches have shown promising results for treating internal auditory canal (IAC) lesions. We reviewed the literature to answer the question: "Do EETTA and ExpTTA achieve high rates of complete resection and low rates of complications in treating patients with IAC pathologies?"

DATA SOURCES

PubMed, EMBASE, Scopus, Web of Science, and Cochrane were searched.

REVIEW METHODS

Studies reporting EETTA/ExpTTA for IAC pathologies were included. Indications and techniques were discussed and meta-analyzed rates of outcomes and complications were obtained with random-effect model meta-analyses.

RESULTS

We included 16 studies comprising 173 patients, all with non-serviceable hearing. Baseline FN function was mostly House-Brackmann-I (96.5%; 95% CI: 94.9-98.1%). Most lesions were vestibular/cochlear schwannomas (98.3%; 95% CI: 96.7-99.8%) of Koos-I (45.9%; 95% CI: 41.3-50.3%) or II (47.1%; 95% CI: 43-51.1%). EETTA was performed in 101 patients (58.4%; 95% CI: 52.4-64.3%) and ExpTTA in 72 (41.6%; 95% CI: 35.6-47.6%), achieving gross-total resection in all cases. Transient complications occurred in 30 patients (17.3%; 95% CI: 13.9-20.5%), with meta-analyzed rates of 9% (95% CI: 4-15%), comprising FN palsy with spontaneous resolution (10.4%; 95% CI: 7.7-13.1%). Persistent complications occurred in 34 patients (19.6%; 95% CI: 17.1-22.2%), with meta-analyzed rates of 12% (95% CI: 7-19%), comprising persistent FN palsy in 22 patients (12.7%; 95% CI: 10.2-15.2%). Mean follow-up was 16 months (range, 1-69; 95% CI: 14.7-17.4). Post-surgery FN function was stable in 131 patients (75.8%; 95% CI: 72.1-79.5%), worsened in 38 (21.9%; 95% CI: 18.8-25%), and improved in 4 (2.3%; 95% CI: 0.7-3.9%), with meta-analyzed rates of improved/stable response of 84% (95% CI: 76-90%).

CONCLUSION

Transpromontorial approaches offer newer routes for IAC surgery, but their restricted indications and unfavorable FN outcomes currently limit their use. Laryngoscope, 133:2856-2867, 2023.

Comparing Linear and Volumetric Tumor Measurements During Observation of Growing Sporadic Vestibular Schwannomas

OBJECTIVE

To assess the correlation between linear and volumetric changes in vestibular schwannomas (VS).

STUDY DESIGN

Retrospective imaging review was performed on patients diagnosed with sporadic VS from 2000 to 2019 who demonstrated linear growth on observation with serial magnetic resonance imaging (MRI).

SETTING

Two large tertiary care centers.

METHODS

Changes in diameter on serial MRI scans, measured by 1995 American Academy of Otolaryngology-Head and Neck Surgery guidelines, were compared to changes in volume, calculated by segmentation.

RESULTS

Ninety-two patients had VS confined to the internal auditory canal (IAC) with 236 MRIs analyzed, and 108 patients had VS involving the cerebellopontine angle (CPA) with 193 MRIs analyzed. The Spearman rank correlation coefficients between changes in diameter and volume for IAC and CPA tumors were 0.43 (p < .001) and 0.65 (p < .001), respectively. Linear diameter increases of 1 to <2 mm corresponded to a median volume change of 32% (interquartile range [IQR]: 6%-86%) for IAC tumors, compared to 23% (IQR: 13%-40%) for CPA tumors. Linear diameter increases of 2 to <3 mm (ie, the minimum linear diameter change classically considered "true growth") corresponded to a median volume change of 42% (IQR: 23%-100%) and 47% (IQR: 26%-69%) for IAC and CPA tumors, respectively.

CONCLUSION

Changes in linear diameter significantly correlated with changes in volume for IAC and CPA tumors, although diameter changes that did not meet the definition of linear growth (<2 mm) had corresponding median volume changes in excess of 20% for both IAC and CPA tumors.

Prediction of Vestibular Schwannoma Surgical Outcome Using Deep Neural Network

OBJECTIVE

To compare shallow machine learning models and deep neural network (DNN) model in prediction of vestibular schwannoma (VS) surgical outcome.

METHODS

One hundred eighty-eight patients with VS were included; all underwent suboccipital retrosigmoid sinus approach, and preoperative magnetic resonance imaging recorded a series of patient characteristics. Degree of tumor resection was collected during surgery, and facial nerve function was evaluated on the eighth day after surgery. Potential predictors of VS surgical outcome were obtained by univariate analysis, including tumor diameter, tumor volume, tumor surface area, brain tissue edema, tumor property, and tumor shape. This study proposes a DNN framework to predict the prognosis of VS surgical outcomes based on potential predictors and compares it with a series of classic machine learning algorithms including logistic regression.

RESULTS

The results showed that 3 predictors of tumor diameter, tumor volume, and tumor surface area were the most important prognostic factors for VS surgical outcomes, followed by tumor shape, while brain tissue edema and tumor property were the least influential. Different from shallow machine learning models, such as logistic regression with average performance (area under the curve: 0.8263; accuracy: 81.38%), the proposed DNN shows better performance, where area under the curve and accuracy were 0.8723 and 85.64%, respectively.

CONCLUSIONS

Based on potential risk factors, DNN can be exploited to achieve preoperative automatic assessment of VS surgical outcomes, and its performance is significantly better than other methods. It is therefore highly warranted to continue to investigate their utility as complementary clinical tools in predicting surgical outcomes preoperatively.

Facial nerve outcome score: a new score to predict long-term facial nerve function after vestibular schwannoma surgery

Introduction

Patients' quality of life (QoL), facial nerve (FN), and cochlear nerve (CN) (if conserved) functions should be pursued as final outcomes of vestibular schwannoma (VS) surgery. In regard to FN function, different morphologic and neurophysiological factors have been related to postoperative outcomes. The aim of the current retrospective study was to investigate the impact of these factors on the short- and long-term FN function after VS resection. The combination of preoperative and intraoperative factors resulted in designing and validating a multiparametric score to predict short- and long-term FN function.

Methods

A single-center retrospective analysis was performed for patients harboring non-syndromic VS who underwent surgical resection in the period 2015-2020. A minimum follow-up period of 12 months was considered among the inclusion criteria. Morphological tumor characteristics, intraoperative neurophysiological parameters, and postoperative clinical factors, namely, House-Brackmann (HB) scale, were retrieved in the study. A statistical analysis was conducted to investigate any relationships with FN outcome and to assess the reliability of the score.

Results

Seventy-two patients with solitary primary VS were treated in the period of the study. A total of 59.8% of patients showed an HB value < 3 in the immediate postoperative period (T1), reaching to 76.4% at the last follow-up evaluation. A multiparametric score, Facial Nerve Outcome Score (FNOS), was built. The totality of patients with FNOS grade A showed an HB value < 3 at 12 months, decreasing to 70% for those with FNOS grade B, whereas 100% of patients with FNOS grade C showed an HB value ≥ 3. The ordinal logistic regression showed three times increasing probability to see an HB value ≥ 3 at 3-month follow-up for each worsening point in FNOS score [Exp(B), 2,999; p < 0.001] that was even more probable [Exp(B), 5.486; p < 0.001] at 12 months.

Conclusion

The FNOS score resulted to be a reliable score, showing high associations with FN function both at short- and long-term follow-up. Although multicenter studies would be able to increase its reproducibility, it could be used to predict the FN damage after surgery and the potential of restoring its function on the long-term period.

How to Precisely Open the Internal Auditory Canal for Resection of Vestibular Schwannoma via the Retrosigmoid Approach

Objective

The aim of this study was to investigate how to precisely expose the intrameatal portion of vestibular schwannomas (VSs) without damaging the labyrinth.

Methods

This was a retrospective study of patients who had undergone retrosigmoid resection of a VS in our institution from April 2018 to December 2021. The patients were divided into microsurgery (MS) and navigation endoscopic-assisted (combined surgery, CS) groups and the effects of image guidance and endoscopy evaluated. The tumors in the CS group were then divided into medial and lateral types by fusion imaging and the differences between the two types analyzed.

Results

Data of 84 patients were analyzed. Residual tumor was detected by postoperative MRI at the fundus of the internal auditory canal in 5 of the 31 patients in the MS group and 1 of the 53 in the CS group. The labyrinth was damaged in four patients in the MS group but was not damaged in any of the CS group patients. The CS group included 29 lateral type and 24 medial type schwannomas. Endoscopic-assisted resection of residual tumor in the IAC was performed significantly more often on medial than on lateral tumors.

Conclusion

Navigation and endoscopy are useful in assisting the exposure of the intrameatal portion of VSs. Preoperative MRI/CT fusion imaging is helpful in preoperative evaluation and surgical planning in patients undergoing VS surgery. Tumors of the medial type require endoscopic assistance for resection.

Endoscope-Assisted Retrosigmoid Approach for Vestibular Schwannomas With Intracanalicular Extensions: Facial Nerve Outcomes

Objective

To explore the role of neuroendoscope assistance during surgical resection of the intracanalicular portion of vestibular schwannomas via the retrosigmoid approach and the subsequent early facial nerve outcomes.

Methods

Patients of vestibular schwannoma with intracanalicular extensions undergoing retrosigmoid dissection at a single institution were retrospectively analyzed in this study. Several surgical techniques were applied to ensure maximal and safe removal of tumors. Tumors extending less than 10 mm into the internal acoustic canal (IAC) were classified as Grade A, while those extending over 10 mm into IAC were taken as Grade B. Neuroendoscope was applied at the end of microscopic phase to search for potential remnants for Grade B tumors. Absolute tumor extension was defined and measured. House and Brackmann (HB) scale was used to evaluate immediate CN VII outcomes.

Results

Of the 61 patients, there were 38 females and 23 males. A total of 18 (29.51%) cases were Koos Grade II, 12 (19.67%) cases Koos Grade III, and 31 (50.82%) cases Koos Grade IV. There were 38 cases (62.30%) of Grade A and 23 cases (37.70%) of Grade B. Gross total resection was achieved in 60 cases (98.36%). Four cases of intracanalicular remnants were detected and completely removed under endoscopic visualizations. There was a significantly higher proportion (17%, p = 0.02) of intracanalicular remnants in Grade B than Grade A. CN VII and VIII were anatomically preserved in all cases. A total of 55 cases (90.16%) retained good (HB Grades 1 and 2) facial nerve outcomes.

Conclusions

In Grade B vestibular schwannomas, after maximal microsurgical removal, endoscopic evaluation of the intracanalicular portion revealed residual tumors in 17% of the patients. Hence endoscopic evaluation of the potential intracanalicular remnants for tumor extending over 10 mm within IAC (Grade B) is recommended.

Facial Nerve Length Influence on Vestibular Schwannoma Microsurgery Outcomes

OBJECTIVE

Facial nerve (FN) function preservation is the primary goal during vestibular schwannoma (VS) resection. Many factors are linked to postoperative FN outcomes. In the present study, we evaluated the association between FN length and VS surgical outcomes.

METHODS

We included 70 consecutive patients who had undergone VS microsurgery between October 2019 and November 2020. The clinical data were prospectively obtained from the patients. The relative FN (rFN) length was obtained by subtracting the contralateral FN length from the ipsilateral FN length as measured using DSI Studio software (available at: http://dsi-studio.labsolver.org/).

RESULTS

The postoperative FN function was House-Brackmann grade I in 47 of the 70 patients (67.1%), grade II in 10 (14.3%), and grade III in 13 (18.6%). Gross total resection (GTR) was performed in 61 patients (87.1%). A residual tumor was retained to preserve FN function in 9 of the 70 patients (12.9%), and rFN length was measured (mean diameter, 20.8 mm; range, 2.5-51.5]). On multivariate analysis, the rFN length was significantly associated with the extent of tumor resection. The receiver operating characteristic curve indicated that the cutoff value for rFN length to predict for intraoperative near total resection versus GTR was 36.6 mm, with a specificity and sensitivity of 93.4% and 88.9%, respectively.

CONCLUSIONS

The rFN length is important for predicting surgical outcomes. An rFN length >36.6 mm might indicate difficulty in achieving GTR with preservation of FN function. Therefore, the rFN length could become an objective indicator for neurosurgeons to predict the difficulty of GTR to preserve FN function.