During human locomotion, vestibular feedback control is fundamental for maintaining dynamic stability and adapting the gait pattern to external circumstances. Within the supraspinal locomotor network, the cerebellum represents the key site for the integration of vestibular feedback information. The cerebellum is further important for the fine-tuning and coordination of limb movements during walking. The aim of this review article is to highlight the shared structural and functional sensorimotor principles in vestibular and cerebellar locomotion control. Vestibular feedback for the maintenance of dynamic stability is integrated into the locomotor pattern via midline, caudal cerebellar structures (vermis, flocculonodular lobe). Hemispheric regions of the cerebellum facilitate feed-forward control of multi-joint coordination and higher locomotor functions. Characteristic features of the gait disorder in patients with vestibular deficits or cerebellar ataxia are increased levels of spatiotemporal gait variability in the fore-aft and the medio-lateral gait dimension. In the fore-aft dimension, pathologic increases of gait fluctuations critically depend on the locomotion speed and predominantly manifest during slow walking velocities. This feature is associated with an increased risk of falls in both patients with vestibular hypofunction as well as patients with cerebellar ataxia. Pharmacological approaches for the treatment of vestibular or cerebellar gait ataxia are currently not available. However, new promising options are currently tested in randomized, controlled trials (fampridine/FACEG; acetyl-DL-leucine/ALCAT).

Background Diagnostic classification of central vs. peripheral etiologies in acute vestibular disorders remains a challenge in the emergency setting. Novel machine-learning methods may help to support diagnostic decisions. In the current study, we tested the performance of standard and machine-learning approaches in the classification of consecutive patients with acute central or peripheral vestibular disorders. Methods 40 Patients with vestibular stroke (19 with and 21 without acute vestibular syndrome (AVS), defined by the presence of spontaneous nystagmus) and 68 patients with peripheral AVS due to vestibular neuritis were recruited in the emergency department, in the context of the prospective EMVERT trial (EMergency VERTigo). All patients received a standardized neuro-otological examination including videooculography and posturography in the acute symptomatic stage and an MRI within 7 days after symptom onset. Diagnostic performance of state-of-the-art scores, such as HINTS (Head Impulse, gazeevoked Nystagmus, Test of Skew) and ABCD 2 (Age, Blood, Clinical features, Duration, Diabetes), for the differentiation of vestibular stroke vs. peripheral AVS was compared to various machine-learning approaches: (i) linear logistic regression (LR), (ii) non-linear random forest (RF), (iii) artificial neural network, and (iv) geometric deep learning (Single/MultiGMC). A prospective classification was simulated by tenfold cross-validation. We analyzed whether machine-estimated feature importances correlate with clinical experience. Results Machine-learning methods (e.g., MultiGMC) outperform univariate scores, such as HINTS or ABCD 2 , for differentiation of all vestibular strokes vs. peripheral AVS (MultiGMC area-under-the-curve (AUC): 0.96 vs. HINTS/ABCD 2 AUC: 0.71/0.58). HINTS performed similarly to MultiGMC for vestibular stroke with AVS (AUC: 0.86), but more poorly for vestibular stroke without AVS (AUC: 0.54). Machine-learning models learn to put different weights on particular features, each of which is relevant from a clinical viewpoint. Established non-linear machine-learning methods like RF and linear methods like LR are less powerful classification models (AUC: 0.89 vs. 0.62). Conclusions Established clinical scores (such as HINTS) provide a valuable baseline assessment for stroke detection in acute vestibular syndromes. In addition, machine-learning methods may have the potential to increase sensitivity and selectivity in the establishment of a correct diagnosis. Keywords Acute vestibular syndrome • HINTS • Machine-learning • MRI • Vestibular neuritis • Vestibular stroke Abbreviations ABCD 2 Age, blood pressure, clinical features, duration, diabetes ANN Artificial neural network AUC Area-under-the-curve AVS Acute vestibular syndrome CVRF Cardiovascular risk factors DT Decision tree DWI Diffusion weighted images ED Emergency department EMVERT EMergency VERTigo FLAIR Fluid attenuated inversion recovery GMC Geometric matrix completion HINTS Head impulse, gaze-evoked nystagmus, test of skew Seyed-Ahmad Ahmadi an...

No abstract

Background and purpose Acute vestibular symptoms have a profound impact on patients’ well‐being. In this study, health‐related quality of life (HRQoL) and functional impairment were investigated prospectively in patients with different peripheral and central vestibular disorders during the acute symptomatic stage to decipher the most relevant underlying factors. Methods In all, 175 patients with acute vestibular disorders were categorized as central vestibular (CV, n = 40), peripheral vestibular (PV, n = 68) and episodic vestibular disorders (EV, n = 67). All patients completed scores to quantify generic HRQoL (European Quality of Life Score Five Dimensions Five Levels, EQ‐5D‐5L) and disease‐specific HRQoL (Dizziness Handicap Inventory, DHI). Vestibular‐ocular motor signs were assessed by video‐oculography, vestibular‐spinal control by posturography and verticality perception by measurement of subjective visual vertical. Results Patients with PV had a poorer HRQoL compared to patients with CV and EV (EQ‐5D‐5L/DHI: PV, 0.53 ± 0.31/56.1 ± 19.7; CV, 0.66 ± 0.28/43.3 ± 24.0; EV, 0.75 ± 0.24/46.7 ± 21.4). After adjusting for age, gender, cardiovascular risk factors and non‐vestibular brainstem/cerebellar dysfunction patients with PV persisted to have poorer generic and disease‐specific HRQoL (EQ‐5D‐5L −0.17, DHI +11.2) than patients with CV. Horizontal spontaneous nystagmus was a highly relevant factor for subgroup differences in EQ‐5D‐5L and DHI, whilst vertical spontaneous nystagmus, subjective visual vertical and sway path were not. EQ‐5D‐5L decreased significantly with more intense horizontal subjective visual vertical in CV (rho = −0.57) and PV (rho = −0.5) but not EV (rho = −0.13). Conclusions Patients with PV have the highest functional impairment of all patients with acute vestibular disorders. Vestibular‐ocular motor disturbance in the yaw plane has more impact than vestibular‐spinal or vestibular‐perceptive asymmetry in the roll and pitch plane, suggesting that horizontal visual stability is the most critical for HRQoL.