Article
Comparing direct nerve stimulation and corticobulbar MEPs in their diagnostic power to predict facial nerve palsy in vestibular schwannoma surgery
Vergleich direkter Nervenstimulation und kortikobulbärer motorisch-evozierter Potentiale hinsichtlich ihres prädikativen Werts für Fazialisparesen nach Vestibularisschwannom-Operationen
Search Medline for
Authors
-
Tobias Greve
- Klinikum der Ludwig-Maximilians-Universität München, Neurochirurgische Klinik und Poliklinik, München, Deutschland
-
Liang Wang
- Beijing Tiantan Hospital, Abteilung für Neurochirurgie, Peking, China
-
Jörg-Christian Tonn
- Klinikum der Ludwig-Maximilians-Universität München, Neurochirurgische Klinik und Poliklinik, München, Deutschland
-
Christian Schichor
- Klinikum der Ludwig-Maximilians-Universität München, Neurochirurgische Klinik und Poliklinik, München, Deutschland
-
Andrea Szelényi
- Klinikum der Ludwig-Maximilians-Universität München, Neurochirurgische Klinik und Poliklinik, München, Deutschland
| Published: | June 26, 2020 |
|---|
Outline
Text
Objective: We compared two IONM modalities for the precise prediction of facial nerve outcome in vestibular schwannoma surgery (VS): direct nerve stimulation (DNS) and facial muscle corticobulbar motor evoked potentials (FMcoMEP).
Methods: Data of 30 patients (45 ± 13.6 years, 14 f) were retrospectively analyzed. Monitored muscles were orbicularis oculi, orbicularis oris and mentalis. For DNS of the facial nerve, the proximal-to-distal amplitude ratio at the end of tumor resection was calculated (DNS_pdAMP). FMcoMEPs were elicited with close-to-threshold stimulation parameters and stimulation intensity was adjusted if amplitudes decreased. The baseline-to-final stimulation intensity increase (MEP_bfSI_inc) was calculated. To assess the predictive power of DNS_pdAMP and MEP_bfSI_inc, facial nerve function (FaF) on day 1 (D1) and day 7 (D7) was dichotomized between marked deterioration with impaired eye closure (HB increase ≥ 1 and absolute HB score ≥ 4) and mild deterioration without impaired eye closure (HB increase ≤ 2 and absolute HB score ≤ 3).
Results: On D1, ROC analysis revealed the highest area under the curve (AUC) for MEP_bfSI_inc, followed by DNS_pdAMP. On D7, DNS_pdAMP showed the highest AUC, followed by MEP_bfSI_inc. At 3M, MEP_bfSI_inc showed a much higher AUC than DNS_pdAMP. Using the Youden’s index, optimal cutoff values for DNS_pdAMP and MEP_bfSI_inc were determined for D1, D7 and 3M. With these improved cutoffs, the false-negative rate was equal for D1 and D7 but 50% higher at 3M for DNS_pdAMP (Table 1 [Tab. 1]).
Conclusion: Ratios in DNS_pdAMP > 0.3 resp. of MEP_bfMT_inc > 20% reliably predict short termfacial paresis. On a long-term basis, only FMcoMEPs seem to provide acceptable predictive value. DNS stimulation induces synchronized activation of fast conducting larger axons, which might result in overestimating axonal conductivity. On the contrary, near threshold TES only partially activates all axons within a nerve. This might result in underestimating peripheral axonal conductivity. This explains the lower predictive value for DNS at 3M. MEPs can be performed throughout the tumor removal, whereas DNS can only be intermittently performed.
