How Many Electromyography Channels Do We Need for Facial Nerve Monitoring?

An Artificial Neural Network Model for Predicting Postoperative Facial Nerve Outcomes After Vestibular Schwannoma Surgery

BACKGROUND AND OBJECTIVES

The emergence of machine learning models has significantly improved the accuracy of surgical outcome predictions. This study aims to develop and validate an artificial neural network (ANN) model for predicting facial nerve (FN) outcomes after vestibular schwannoma (VS) surgery using the proximal-to-distal amplitude ratio (P/D) along with clinical variables.

METHODS

This retrospective study included 71 patients who underwent VS resection between 2018 and 2022. At the end of surgery, the FN was stimulated at the brainstem (proximal) and internal acoustic meatus (distal) and the P/D was calculated. Postoperative FN function was assessed using the House-Brackmann grading system at discharge (short-term) and after 9-12 months (long-term). House-Brackmann grades I-II were considered good outcome, whereas grades III-VI were considered fair/poor. An ANN model was constructed, and the performance of the model was evaluated using the area under the ROC curve for internal validation and accuracy, sensitivity, specificity, and positive and negative predictive values for external validation.

RESULTS

The short-term FN outcome was grades I-II in 57.7% and grades III-VI in 42.3% of patients. Initially, a model using P/D had an area under the curve of 0.906 (internal validation) and an accuracy of 89.1% (95% CI: 68.3%-98.8%) (external validation) for predicting good vs fair/poor short-term FN outcomes. The model was then refined to include only muscles with a P/D with a proximal latency between 6 and 8 ms. This improved the accuracy to 100% (95% CI: 79%-100%). Integrating clinical variables (patient's age, tumor size, and preoperative HB grade) in addition to P/D into the model did not significantly improve the predative value. A model was then created to predict the long-term FN outcome using P/D with latencies between 6 and 8 ms and had an accuracy of 90.9% (95% CI: 58.7%-99.8%).

CONCLUSION

ANN models incorporating P/D can be a valuable tool for predicting FN outcomes after VS surgery. Refining the model to include P/D with latencies between 6 and 8 ms further improves the model's prediction. A user-friendly interface is provided to facilitate the implementation of this model.

EMG monitoring

While most neurophysiologists are familiar with electromyography (EMG) for the purpose of clinical diagnostics, this technique also has a long tradition for neuro-monitoring. In short, intra-operative use of EMG can be divided into stimulated EMG on the one hand and monitoring of the free-running EMG and its spontaneous activity on the other hand. Methods for utilization of stimulated EMG are covered elsewhere in this book. This chapter focuses on the monitoring of spontaneous, or, more correctly, "surgically evoked" EMG. The history and underlying physiologic principles of intra-operative EMG are covered in this chapter as well as a detailed overview of the methodology. Building up from the basis of this background, we describe examples of how EMG can be used to help in intra-operative detection of adverse events and also in the prediction of postoperative outcomes. In the opinion of the authors, EMG should not be used as a "standalone" technique in contemporary neuro-monitoring. Most of its significant potential may be realized when it is used in a complementary way together with other modalities, mainly motor evoked potentials. Bearing this in mind, we sketch out how such a complementary setup may be used for improved neuro-monitoring.

Best Practices in Facial Nerve Monitoring

OBJECTIVES/HYPOTHESIS

Facial nerve monitoring (FNM) has evolved into a widely used adjunct for many surgical procedures along the course of the facial nerve. Even though majority opinion holds that FNM reduces the incidence of iatrogenic nerve injury, there are few if any studies yielding high-level evidence and no practice guidelines on which clinicians can rely. Instead, a review of the literature and medicolegal cases reveals significant variations in methodology, training, and clinical indications.

STUDY DESIGN

Literature review and expert opinion.

METHODS

Given the lack of standard references to serve as a resource for FNM, we assembled a multidisciplinary group of experts representing more than a century of combined monitoring experience to synthesize the literature and provide a rational basis to improve the quality of patient care during FNM.

RESULTS

Over the years, two models of monitoring have become well-established: 1) monitoring by the surgeon using a stand-alone device that provides auditory feedback of facial electromyography directly to the surgeon, and 2) a team, typically consisting of surgeon, technologist, and interpreting neurophysiologist. Regardless of the setting and the number of people involved, the reliability of monitoring depends on the integration of proper technical performance, accurate interpretation of responses, and their timely application to the surgical procedure. We describe critical steps in the technical set-up and provide a basis for context-appropriate interpretation and troubleshooting of recorded signals.

CONCLUSIONS

We trust this initial attempt to describe best practices will serve as a basis for improving the quality of patient care while reducing inappropriate variations.

LEVEL OF EVIDENCE

4 Laryngoscope, 131:S1-S42, 2021.

Optimizing Location of Subdermal Recording Electrodes for Intraoperative Facial Nerve Monitoring

OBJECTIVE

The purpose of this study is to determine if different facial muscle groups demonstrate different responses to facial nerve stimulation, the results of which could potentially improve intraoperative facial nerve monitoring (IOFNM).

METHODS

IOFNM data were prospectively collected from patients undergoing cochlear implantation. At different stages of nerve exposure, three sites were stimulated using a monopolar pulse. Peak electromyography (EMG) amplitude (μV) in four muscle groups innervated by four different branches of the facial nerve (frontalis-temporal, inferior orbicularis oculi-zygomatic, superior oribularis oris-buccal, and mentalis-marginal mandibular) were recorded.

RESULTS

A total of 279 peak EMG amplitudes were recorded in 93 patients. At all three stimulating sites, the zygomatic branch mean peak EMG amplitudes were statistically greater than those of the temporal, buccal, and marginal mandibular branches (P < .05). At stimulating Site C, the marginal mandibular branch mean peak EMG was stronger than the temporal or buccal branches (P < .05). Of the 279 stimulations, the zygomatic branch demonstrated the highest amplitude in 128 (45.9%) trials, followed by the marginal mandibular branch (22.2%).

CONCLUSIONS

When utilized, IOFNM should be performed with at least two electrodes, one of which is placed in the orbicularis oculi muscles and the other in the mentalis muscle. However, there is wide variability between patients. As such, in cases of suspected variant nerve anatomy or increased risk of injury (intradural procedures), surgeons should consider using more than two recording electrodes, with at least one in the orbicularis oculi muscle.

LEVEL OF EVIDENCE

3 Laryngoscope, 131:E2329-E2334, 2021.

Facial Nerve EMG: Low-Tech Monitoring with a Stopwatch

OBJECTIVE

 The quantity of A-trains, a high-frequency pattern of free-running facial nerve electromyography, is correlated with the risk for postoperative high-grade facial nerve paresis. This correlation has been confirmed by automated analysis with dedicated algorithms and by visual offline analysis but not by audiovisual real-time analysis.

METHODS

 An investigator was presented with 29 complete data sets measured during actual surgeries in real time and without breaks in a random order. Data were presented either strictly via loudspeaker (audio) or simultaneously by loudspeaker and computer screen (audiovisual). Visible and/or audible A-train activity was then quantified by the investigator with the computerized equivalent of a stopwatch. The same data were also analyzed with quantification of A-trains by automated algorithms.

RESULTS

 Automated (auto) traintime (TT), known to be a small, yet highly representative fraction of overall A-train activity, ranged from 0.01 to 10.86 s (median: 0.58 s). In contrast, audio-TT ranged from 0 to 1,357.44 s (median: 29.69 s), and audiovisual-TT ranged from 0 to 786.57 s (median: 46.19 s). All three modalities were correlated to each other in a highly significant way. Likewise, all three modalities correlated significantly with the extent of postoperative facial paresis. As a rule of thumb, patients with visible/audible A-train activity < 1 minute presented with a more favorable clinical outcome than patients with > 1 minute of A-train activity.

CONCLUSION

 Detection and even quantification of A-trains is technically possible not only with intraoperative automated real-time calculation or postoperative visual offline analysis, but also with very basic monitoring equipment and real-time good quality audiovisual analysis. However, the investigator found audiovisual real-time-analysis to be very demanding; thus tools for automated quantification can be very helpful in this respect.

A-train clusters and the intermedius nerve in vestibular schwannoma patients

OBJECTIVE

EMG "A-train" activity correlates with postoperative facial palsy after vestibular schwannoma (VS) surgery. An intermedius nerve separate from the facial nerve increases A-trains without significant impact on function. We investigate occurrence of A-train "clusters", A-trains over a majority of channels within a short time frame.

METHODS

Data from 217 patients with first surgery for VS were evaluated retrospectively. Continuous EMG recorded with 9 channels was evaluated for A-train patterns. "Clusters" of A-trains were identified, i.e. A-trains within 3 seconds over a majority of channels. Relation to a separate intermedius, tumor size and facial palsy was evaluated.

RESULTS

Correlations between A-trains and postoperative facial palsy were higher in patients without separate intermedius (r = 0.562 versus r = 0.194). Clusters were identified in 107 patients (49.3%), separate intermedius in 109 (50.2%), with significant association of both (p < 0.001, Chi-Square test). Excluding clusters slightly increased correlation of A-trains to facial nerve function.

CONCLUSIONS

A-train clusters have limited relevance for predicting postoperative paresis. However, they should be regarded as warning signs, suggesting the presence of a separate intermedius nerve.

SIGNIFICANCE

A-train "clusters" are a sign of hyperactivity of the facial nerve due to a separate intermedius nerve and may confound intraoperative monitoring during VS surgery.

The intermedius nerve as a confounding variable for monitoring of the free-running electromyogram

OBJECTIVE

A-trains, a facial nerve EMG-pattern, are correlated with postoperative functional impairment. However, an unknown confounder is suspected to cause false positive monitoring results. The intermedius nerve contains motor fibers targeting lower facial muscles; their significance for facial nerve monitoring is yet unknown.

METHODS

Intraoperative videotapes and free-running 9-channel facial nerve EMG assessed from 87 patients undergoing surgery for vestibular schwannoma were evaluated, and presence/absence of an identifiable intermedius nerve was determined. The prognostic value of train time, a quantitative measure for A-train activity, was evaluated for both the groups with and without an identifiable intermedius nerve.

RESULTS

Correlation between traintime and outcome (Spearman's Rho) rose to 0.73 (p<0.001) when only patients without an identified intermedius nerve were considered, and fell to 0.43 (p<0.05) with the other patient group. This difference was statistically significant (p=0.036), was more prominent in the channels monitoring perioral facial muscles, and resulted from additional A-train activity in patients with an identifiable intermedius nerve.

CONCLUSIONS

A separate intermedius nerve may be more prone to manipulation, leading to A-train activity without clinical correlate, thus causing false positive monitoring results.

SIGNIFICANCE

For interpretation of the free-running EMG, the intermedius nerve needs to be taken into account as a confounder.

The relationship between nervus intermedius anatomy, ultrastructure, electrophysiology, and clinical function. Usefulness in cerebellopontine microsurgery

BACKGROUND

Although previous studies have described the clinical features of the nervus intermedius (NI), no attempt has yet been made to describe the relationship between the ultrastructural and electrophysiological characteristics of the nervus intermedius and its motor competence.

OBJECTIVE

In this study, we analyzed the intraoperative electrophysiological response obtained during vestibular schwannoma surgery. The ultrastructure was studied using electron microscopy.

MATERIALS AND METHODS

Thirty-six consecutive patients underwent microsurgery for vestibular schwannoma with cerebellopontine angle tumors. The patients were extensively monitored intraoperatively. Selective stimulation of the nervus intermedius was attempted in all cases. The patients were then examined postoperatively and followed for a minimum of 1 year. Forty-three isolated human brainstems were analyzed to collect the ultrastructural NI data.

RESULTS

We found a correlation between the NI motor responses in the perinasal and perioral regions and the ultrastructure characteristics, with few (0.5 %) but large myelinated motor fibers (diameters >12 μm). Both characteristics are consistent with the clinical observation of transient weakness of the levator anguli oris muscle. These observations indicate a relationship between the intraoperative electrophysiological identification of the NI nervus intermedius and its clinical and ultrastructural characteristics.

CONCLUSIONS

Identifying the NI in the deformed anatomy of tumors could provide a fixed landmark during cerebellopontine surgery and help prevent damage of the facial nerve.

A-trains for intraoperative monitoring in patients with recurrent vestibular schwannoma

BACKGROUND

Second surgery of recurrent vestibular schwannoma (VS) after previous surgery, stereotactic radiosurgery (SR) or fractionated radiotherapy (FR) carries an increased risk for deterioration of facial nerve function, e.g., due to adhesions, underlining the need for intraoperative monitoring. Facial “Atrain” EMG activity (“traintime”) correlates with the degree of postoperative facial palsy. Studies investigating A-trains in VS patients with previous surgery, SR or FR are missing. We therefore investigated the value of A-train monitoring in patients undergoing second surgery for VS.

METHOD

Intraoperative EMG data from patients who underwent second surgery for VS after previous surgery, SR and/or FR at our institution between 2006 and 2012 were retrospectively analyzed. Ten patients were selected (5 male): Seven had previous SR/RT and MS, three previous surgery only. Traintime values and distribution was compared to published thresholds and to 77 patients who underwent first surgery for VS during the same time period.

RESULTS

A-trains were recorded early after opening of the dura, before facial nerve preparation. Mean traintime was 46.9 s (18.51 s – 80.82 s) in patients with previous SR/RT. In patients with previous MS only, traintime was 0.06 s, 0.99 s and 22.46 s. Compared to the literature, traintime was higher than expected in six patients (four with previous SR/RT, two without), respectively seven compared to the 77 patients with first surgery (5 SR/RT). Seven patients with previous SR/RT and none with previous surgery showed diffuse A-train distributions without significant percentages in single channels, compared to 60 of 77 patients with first surgery (p <0.02).

CONCLUSIONS

Especially SR/RT, but also previous surgery seems to induce changes in the facial nerve leading to hyperexcitability and exceedingly high traintime values. Based on these findings, A-train monitoring in this specific patient group should be interpreted with caution.

Vestibular schwannoma surgical treatment

Neurosurgical intervention remains the main step in the effective management of vestibular schwannomas. Extensive studies on vestibular schwannoma treatment have placed emphasis on preserving quality of life and neurological functions, particularly of the facial and vestibulocochlear nerves. Facial nerve preservation and hearing preservation have been achieved by significant advances in skull base microsurgical techniques and intraoperative neuromonitoring. Diffusion tensor imaging is a powerful and accurate method for preoperatively identifying the facial nerve in relation to vestibular schwannomas. Endoscopy offers excellent illumination of the anatomical structures and provides panoramic vision inside the surgical area. In this report, we focused on facial nerve and vestibulocochlear nerve preservation and analyzed the major techniques used for identifying the nerve-tumor relationship.

Controlling a human-computer interface system with a novel classification method that uses electrooculography signals

Electrooculography (EOG) signals can be used to control human-computer interface (HCI) systems, if properly classified. The ability to measure and process these signals may help HCI users to overcome many of the physical limitations and inconveniences in daily life. However, there are currently no effective multidirectional classification methods for monitoring eye movements. Here, we describe a classification method used in a wireless EOG-based HCI device for detecting eye movements in eight directions. This device includes wireless EOG signal acquisition components, wet electrodes and an EOG signal classification algorithm. The EOG classification algorithm is based on extracting features from the electrical signals corresponding to eight directions of eye movement (up, down, left, right, up-left, down-left, up-right, and down-right) and blinking. The recognition and processing of these eight different features were achieved in real-life conditions, demonstrating that this device can reliably measure the features of EOG signals. This system and its classification procedure provide an effective method for identifying eye movements. Additionally, it may be applied to study eye functions in real-life conditions in the near future.