Machine learning models help differentiate between causes of recurrent spontaneous vertigo

BACKGROUND

Vestibular migraine (VM) and Menière's disease (MD) are two common causes of recurrent spontaneous vertigo. Using history, video-nystagmography and audiovestibular tests, we developed machine learning models to separate these two disorders.

METHODS

We recruited patients with VM or MD from a neurology outpatient facility. One hundred features from six "feature subsets": history, acute video-nystagmography and four laboratory tests (video head impulse test, vestibular-evoked myogenic potentials, caloric testing and audiogram) were used. We applied ten machine learning algorithms to develop classification models. Modelling was performed using three "tiers" of data availability to simulate three clinical settings. "Tier 1" used all available data to simulate the neuro-otology clinic, "Tier 2" used only history, audiogram and caloric test data, representing the general neurology clinic, and "Tier 3" used history alone as occurs in primary care. Model performance was evaluated using tenfold cross-validation.

RESULTS

Data from 160 patients with VM and 114 with MD were used for model development. All models effectively separated the two disorders for all three tiers, with accuracies of 85.77-97.81%. The best performing algorithms (AdaBoost and Random Forest) yielded accuracies of 97.81% (95% CI 95.24-99.60), 94.53% (91.09-99.52%) and 92.34% (92.28-96.76%) for tiers 1, 2 and 3. The best feature subset combination was history, acute video-nystagmography, video head impulse test and caloric testing, and the best single feature subset was history.

CONCLUSIONS

Machine learning models can accurately differentiate between VM and MD and are promising tools to assist diagnosis by medical practitioners with diverse levels of expertise and resources.

Subjective visual horizontal correlates better with ocular than with cervical vestibular evoked myogenic potentials

OBJECTIVE

To examine the relationship between widely used otolith function tests: the Subjective Visual Horizontal (SVH) and Vestibular Evoked Myogenic Potentials (VEMP).

METHODS

A retrospective analysis was performed on 301 patients who underwent SVH, ocular and cervical VEMP (oVEMP and cVEMP) tests on the same day. Correlations between the mean SVH tilt and amplitude asymmetry ratios for bone-conducted (BC) oVEMP and air-conducted (AC) cVEMP were examined. Diagnoses included vestibular neuritis, stroke, vestibular migraine, Meniere's disease, sudden sensorineural hearing loss (SSNHL) and vestibular schwannoma.

RESULTS

SVH results were concordant with the oVEMP in 64% of cases and the cVEMP in 51%. Across all patients, SVH demonstrated a significant moderate correlation with BC oVEMP amplitude asymmetry ratios (r = 0.55, p < 0.001) and a weak correlation with AC cVEMP amplitude asymmetry ratios (r = 0.35, p < 0.001). A stronger correlation between SVH and oVEMPs was observed in patients with vestibular neuritis (r = 0.67, p < 0.001) and SSNHL (r = 0.76, p = 0.001).

CONCLUSIONS

SVH correlates better with oVEMP than cVEMP symmetry.

SIGNIFICANCE

This finding reinforces the hypothesis of a common utricular origin for both SVH and oVEMPs which is distinct from the saccular origin of cVEMPs.

Modern vestibular tests can accurately separate stroke and vestibular neuritis

OBJECTIVES

To separate posterior-circulation stroke (PCS) and vestibular-neuritis (VN) using quantitative vestibular tests.

METHODS

Patients were prospectively recruited from the emergency room within 72 h of presentation. Video-nystagmography (VNG), three-dimensional video head-impulse testing (vHIT), vestibular-evoked myogenic potentials (VEMPs), and subjective visual-horizontal (SVH) were performed.

RESULTS

There were 128 PCS and 134 VN patients. Common stroke-territories were: posterior-inferior cerebellar artery, basilar-perforators, multi-territory and anterior-inferior cerebellar artery (41.4%, 21.1%, 14.1%, 7.8%). VN included superior, inferior and pan-neuritis (53.3%, 4.2%, and 41.5%). Most VN and stroke patients presented with acute vestibular syndrome (96.6%, 61.7%). In VN, we recorded horizontal (98.5%) or vertical/torsional spontaneous nystagmus (1.5%) and in PCS, absent-nystagmus (53.9%), horizontal (32%) or vertical/torsional (14.1%) nystagmus. The mean slow-phase velocity of horizontal nystagmus was faster in VN than PCS (11.8 ± 7.2 and 5.2 ± 3.0°/s, p < 0.01). Ipsilesional horizontal-canal (HC) vHIT-gain was lower in VN than in stroke (0.47 ± 0.24, 0.92 ± 0.20, p < 0.001). Ipsilesional catch-up saccades occurred earlier, and their amplitude, prevalence, and velocity were greater in VN than PCS (p < 0.01). Ipsilesional SVH deviation > 2.5° occurred more often in VN than in stroke (97.6% and 24.3%, p < 0.01). Abnormal bone-conducted ocular-VEMP asymmetry ratio was more common in VN than PCS (50% and 14.4%, p < 0.01). Using the ten best discriminators (VNG, vHIT, SVH, and oVEMP metrics), VN was separated from PCS with a sensitivity of 92.9% and specificity of 89.8%. Adding VNG and vHIT to the bedside head-impulse-nystagmus-and-test-of-skew (HINTS) test enhanced sensitivity and specificity from 95.3% and 63.4% to 96.5% and 80.6%.

CONCLUSION

Quantitative vestibular testing helps separate stroke from vestibular neuritis and, when used, could improve diagnostic accuracy in the emergency room.

Association of Lesion Topography with Functional Outcomes in Acute Ischemic Stroke Patients Considered for, or Receiving, Reperfusion Therapy: A Meta-Analysis

Background: The impact of lesion topography (LT), characterised by the Alberta Stroke Programme Early CT Score (ASPECTS), on outcomes after reperfusion therapy in acute ischemic stroke (AIS) is poorly elucidated. We investigated the prognostic accuracy of ASPECTS-based LT assessment and its association with clinical outcomes in AIS patients considered for reperfusion therapy or receiving intravenous thrombolysis (IVT), endovascular thrombectomy (EVT), or none or both. Methods: Studies were identified from PubMed with additional studies added from Google Scholar. The prevalence of individual ASPECTS regions will also be determined. The association of individual ASPECTS regions with the functional outcome at 90 days will be assessed using random-effects modelling for various cut-offs, such as 6, 7 and 8. The association of continuous ASPECTS with the functional outcome at 90 days will also be undertaken. Forest plots of odds ratios (ORs) will be generated. Results: A total of 25 studies have been included in the final analysis, encompassing 11,404 patients. Pooled estimates indicate that the highest prevalence rates were in cases involving the insula and lentiform nucleus. Subgroup analysis for ASPECTS < 6 (OR 6.10; 95% CI 2.50−14.90; p < 0.0001), ASPECTS < 7 (OR 4.58; 95% CI 1.18−17.86; p < 0.0001) and ASPECTS < 8 (OR 2.26; 95% CI 1.32−3.89; p < 0.0001) revealed a significant association with poor functional outcome at 90 days. Decreasing ASPECTS significantly increased the odds of poor functional outcomes at 90 days (SMD −1.15; 95% CI −1.77−−0.52; p < 0.0001). Conclusions: Our meta-analysis demonstrates that decreasing ASPECTS is significantly associated with poor functional outcomes. Individual ASPECTS regions associated with the highest odds of poor functional outcomes were identified. Future studies on the association of LT and clinical outcomes specific to EVT are required.

Video head impulse testing to differentiate vestibular neuritis from posterior circulation stroke in the emergency department: a prospective observational study

Background and aims

Vertigo is a common presentation to the emergency department (ED) with 5% of presentations due to posterior circulation stroke (PCS). Bedside investigations such as the head impulse test (HIT) are used to risk stratify patients, but interpretation is operator dependent. The video HIT (v-HIT) provides objective measurement of the vestibular-ocular-reflex (VOR) and may improve diagnostic accuracy in acute vestibular syndrome (AVS). We aimed to evaluate the use of v-HIT as an adjunct to clinical assessment to acutely differentiate vestibular neuritis (VN) from PCS.

Methods

133 patients with AVS were consecutively enrolled from the ED of our comprehensive stroke centre between 2018 and 2021. Patient assessment included a targeted vestibular history, HINTs examination (Head Impulse, Nystagmus and Test of Skew), v-HIT and MRI>48 hours after symptom onset. The HINTS/v-HIT findings were analysed and compared between VN, PCS and other cause AVS. Clinical course, v-HIT and MRI findings were used to determine diagnosis.

Results

Final diagnosis was VN in 40%, PCS 15%, migraine 16% and other cause AVS 29%. PCS patients were older than VN patients (mean age 68.5±10.6 vs 60.1±14.2 years, p=0.14) and had more cardiovascular risk factors (3 vs 2, p=0.002). Mean VOR gain was reduced (<0.8) in ipsilateral horizontal and (<0.7) anterior canals in VN but was normal in PCS, migraine and other cause AVS. V-HIT combined with HINTs was 89% sensitive and 96% specific for a diagnosis of VN.

Conclusions

V-HIT combined with HINTs is a reliable tool to exclude PCS in the ED.

A Portrait of Menière's Disease Using Contemporary Hearing and Balance Tests

OBJECTIVE

Menière's disease (MD) is characterized by recurrent vertigo and fluctuating aural symptoms. Diagnosis is straightforward in typical presentations, but a proportion of patients present with atypical symptoms. Our aim is to profile the array of symptoms patients may initially present with and to analyze the vestibular and audiological test results of patients with a diagnosis of MD.

DESIGN

A retrospective study of patient files.

SETTING

A tertiary, neuro-otology clinic Royal Prince Alfred Hospital, Sydney, Australia.

METHOD

We identified 375 patients. Their history, examination, vestibular-evoked myogenic potentials (VEMP), video head-impulse test, canal-paresis on caloric testing, subjective visual horizontal (SVH), electrocochleography, ictal nystagmus, and audiometry were assessed.

RESULTS

Atypical presenting symptoms were disequilibrium (n = 49), imbalance (n = 13), drop-attacks (n = 12), rocking vertigo (n = 2), and unexplained vomiting (n = 3), nonspontaneous vestibular symptoms in 21.6%, fluctuation of aural symptoms only (46%), and headaches (31.2%). Low velocity, interictal spontaneous-nystagmus in 13.3% and persistent positional-nystagmus in 12.5%. Nystagmus recorded ictally in 90 patients was mostly horizontal (93%) and of high velocity (48 ± 34°/s). Testing yielded abnormal caloric responses in 69.6% and abnormal video head impulse test 12.7%. Air-conducted cervical VEMPs were abnormal in 32.2% (mean asymmetry ratio [AR] 30.2 ± 46.5%) and bone-conducted ocular VEMPs abnormal in 8.8% (AR 11.2 ± 26.8%). Abnormal interictal SVH was in 30.6%, (ipsiversive n = 46 and contraversive n = 19). Mean pure-tone averages 50 dB ± 23.5 and 20 dB ± 13 for affected and unaffected ears.

CONCLUSION

Menière's disease has a distinctive history, but atypical presentations with normal vestibular function and hearing are a diagnostic challenge delaying treatment initiation.

Stroke Aetiology and Collateral Status in Acute Ischemic Stroke Patients Receiving Reperfusion Therapy-A Meta-Analysis

BACKGROUND

The interplay between collateral status and stroke aetiology may be crucial in the evaluation and management of acute ischemic stroke (AIS). Our understanding of this relationship and its level of association remains sub-optimal. This study sought to examine the association of pre-intervention collateral status with stroke aetiology, specifically large artery atherosclerosis (LAA) and cardio-embolism (CE), in AIS patients receiving reperfusion therapy, by performing a meta-analysis.

METHODS

Relevant search terms were explored on Medline/PubMed, Embase and Cochrane databases. Studies were included using the following inclusion criteria: (a) patients aged 18 or above; (b) AIS patients; (c) patients receiving reperfusion therapy; (d) total cohort size of >20, and (e) qualitative or quantitative assessment of pre-intervention collateral status on imaging using a grading scale. Random-effects meta-analysis was performed to investigate the association of aetiology with pre-intervention collateral status, and forest plots of risk ratio (RR) were generated.

RESULTS

A meta-analysis was conducted on seven studies, with a cumulative cohort of 1235 patients, to assess the association of pre-intervention collateral status with stroke aetiology. Patients with LAA were associated significantly with an increased rate of good collaterals (RR 1.24; 95% CI 1.04-1.50; p = 0.020, z = 2.33). Contrarily, CE aetiology was associated significantly with a decreased rate of good collaterals (RR 0.83; 95% CI 0.71-0.98; p = 0.027, z = -2.213).

CONCLUSIONS

This study demonstrates that, in AIS patients receiving reperfusion therapy, LAA and CE aetiologies are associated significantly with collateral status.

Clinical, oculographic and vestibular test characteristics of Ménière's disease

Seventy Ménière's disease (MD) patients with spontaneous vertigo (100%), unilateral aural fullness (57.1%), tinnitus (78.6%), and subjective hearing loss (75.7%) self-recorded nystagmus during their episodes of vertigo using portable video oculography goggles. All demonstrated ictal spontaneous nystagmus, horizontal in 94.3% (n = 66) and vertical in 5.7% (n = 4), with a mean slow-phase velocity (SPV) of 42.8 ± 31.1°/s (range 5.3-160.1). Direction reversal of spontaneous horizontal nystagmus was captured in 58.6%, within the same episode in 34.3%, and over different days in 24.3%. In 18.6%, we observed ipsiversive then contraversive nystagmus, and in 12.9% contraversive to ipsiversive direction reversal. Ictal nystagmus SPV (42.8 ± 31.1°/s) was significantly faster than interictal (1.4 ± 3.1°/s, p < 0.001, CI 34.277-48.776). Compared to age-matched healthy controls, interictal video head impulse test gains in MD ears were significantly lower, cumulative and first saccade (S1) amplitudes were significantly larger, and S1 peak velocities were significantly faster (p = 0.038/0.019/0.008/ < 0.001, CI 0.002-0.071/0.130-1.444/0.138-0.909/14.614-41.506). Audiometry showed asymmetrically increased thresholds in 100% of MD ears (n = 70). Significant caloric, air-conducted (AC) cervical vestibular-evoked myogenic potential (VEMP), and AC ocular VEMP asymmetries were found in 61.4, 37.9, and 44.4% of patients (MD ear reduced). Transtympanic electrocochleography tested in 36 ears (23 patients) showed 81.8% of MD ears had a positive result for hydrops (either a summating potential at 1/2 kHz < - 6 µV, or an SP/AP ratio > 40%). Using ictal nystagmus findings of SPV > 12°/s, and a caloric canal paresis > 25%, we correctly separated a diagnosis MD from Vestibular Migraine with a sensitivity and specificity of 95.7% and 85.1% (CI 0.89-0.97).

Early decompressive hemicraniectomy in thrombolyzed acute ischemic stroke patients from the international ENCHANTED trial

Decompressive hemicraniectomy (DHC) can improve outcomes for patients with severe forms of acute ischemic stroke (AIS), but the evidence is mainly derived from non-thrombolyzed patients. We aimed to determine the characteristics and outcomes of early DHC in thrombolyzed AIS participants of the international Enhanced Control of Hypertension and Thrombolysis Stroke Study (ENCHANTED). Post-hoc analyses of ENCHANTED, an international, partial-factorial, open, blinded outcome-assessed, controlled trial in 4557 thrombolysis-eligible AIS patients randomized to low- versus standard-dose intravenous alteplase (Arm A, n = 2350), intensive versus guideline-recommended blood pressure control (Arm B, n = 1280), or both (Arms A + B, n = 947). Logistic regression models were used to identify baseline variables associated with DHC, with inverse probability of treatment weights employed to eliminate baseline imbalances between those with and without DHC. Logistic regression was also used to determine associations of DHC and clinical outcomes of death/disability, major disability, and death (defined by scores 2-6, 3-5, and 6, respectively, on the modified Rankin scale) at 90 days post-randomization. There were 95 (2.1%) thrombolyzed AIS patients who underwent DHC, who were significantly younger, of non-Asian ethnicity, and more likely to have had prior lipid-lowering treatment and severe neurological impairment from large vessel occlusion than other patients. DHC patients were more likely to receive other management interventions and have poor functional outcomes than non-DHC patients, with no relation to different doses of intravenous alteplase. Compared to other thrombolyzed AIS patients, those who received DHC had a poor prognosis from more severe disease despite intensive in-hospital management.

Diagnostic and prognostic utility of computed tomography perfusion imaging in posterior circulation acute ischemic stroke: A systematic review and meta-analysis

BACKGROUND

Computed tomography perfusion (CTP) imaging could be useful in the diagnosis of posterior circulation stroke (PCS) and in identifying patients who are likely to experience favorable outcomes following reperfusion therapy. The current study sought to investigate the diagnostic and prognostic capability of CTP in acute ischemic PCS by performing a systematic review and meta-analysis.

METHODS

Medline/PubMed and the Cochrane Library were searched using the terms: "posterior circulation", "CT perfusion", "acute stroke", and "reperfusion therapy". The following studies were included: (1) patients aged 18 years or above; (2) patients diagnosed with PCS; and (3) studies with good methodological design. Pooled sensitivity (SENS), specificity (SPEC), and area under the curve (AUC), computed using the summary receiver operating characteristic (SROC) curves, were used to determine diagnostic/prognostic capability.

RESULTS

Out of 14 studies included, a meta-analysis investigating diagnostic accuracy of CTP was performed on nine studies. Meta-analysis demonstrated comparable diagnostic accuracy of CTP to non-contrast computed tomography (NCCT) (AUC_CTP : 0.90 [95% CI 0.87-0.92] vs. AUC_NCCT : 0.96 [95% CI 0.94-0.97]); however, with higher pooled sensitivity (SENS_CTP : 72% [95% CI 57%-83%] vs. SENS_NCCT : 25% [95% CI 17%-35%]) and lower specificity (SPEC_CTP : 90% [95% CI 83%-94%] vs. SPEC_NCCT : 96% [95% CI 95%-98%]) than NCCT. Meta-analysis to determine prognostic capability of CTP could not be performed.

CONCLUSIONS

CTP has limited diagnostic utility in acute ischemic PCS, albeit with superior diagnostic sensitivity and inferior diagnostic specificity to NCCT. Further prospective trials are required to validate the prognostic capability of CTP-derived parameters in PCS.

Clinical, oculographic, and vestibular test characteristics of vestibular migraine

BACKGROUND

We characterise the history, vestibular tests, ictal and interictal nystagmus in vestibular migraine.

METHOD

We present our observations on 101 adult-patients presenting to an outpatient facility with recurrent spontaneous and/or positional vertigo whose final diagnosis was vestibular migraine (n = 27) or probable vestibular migraine (n = 74). Ictal and interictal video-oculography, caloric and video head impulse tests, vestibular-evoked myogenic potentials and audiometry were performed.

RESULTS

Common presenting symptoms were headache (81.2%), spinning vertigo (72.3%), Mal de Débarquement (58.4%), and motion sensitivity (30.7%). With fixation denied, ictal and interictal spontaneous nystagmus was observed in 71.3 and 14.9%, and purely positional nystagmus in 25.8 and 55.4%. Spontaneous ictal nystagmus was horizontal in 49.5%, and vertical in 21.8%. Ictal spontaneous and positional nystagmus velocities were 5.3 ± 9.0°/s (range 0.0-57.4), and 10.4 ± 5.8°/s (0.0-99.9). Interictal spontaneous and positional nystagmus velocities were <3°/s in 91.8 and 23.3%. Nystagmus velocities were significantly higher when ictal (p < 0.001/confidence interval: 2.908‒6.733, p < 0.001/confidence interval: 5.308‒10.085). Normal lateral video head impulse test gains were found in 97.8% (mean gain 0.95 ± 0.12) and symmetric caloric results in 84.2% (mean canal paresis 7.0 ± 23.3%). Air- and bone-conducted cervical-vestibular-evoked myogenic potential amplitudes were symmetric in 88.4 and 93.4% (mean corrected amplitude 1.6 ± 0.7, 1.6 ± 0.8) with mean asymmetry ratios of 13.0 and 9.0%. Air- and bone-conducted ocular-vestibular-evoked myogenic potentials were symmetric in 67.7 and 97.2% (mean amplitude 9.2 ± 6.4 and 20.3 ± 12.8 µV) with mean asymmetry ratios of 15.7 and 9.9%. Audiometry was age consistent and symmetric in 85.5%.

CONCLUSION

Vestibular migraine is characterised by low velocity ictal spontaneous nystagmus, which can be horizontal, vertical, or torsional, and normal audiovestibular test results.

Brain imaging abnormalities and outcome after acute ischaemic stroke: the ENCHANTED trial

OBJECTIVE

To test the hypothesis that imaging signs of 'brain frailty' and acute ischaemia predict clinical outcomes and symptomatic intracranial haemorrhage (sICH) after thrombolysis for acute ischaemic stroke (AIS) in the alteplase dose arm of ENhanced Control of Hypertension ANd Thrombolysis strokE stuDy (ENCHANTED).

METHODS

Blinded assessors coded baseline images for acute ischaemic signs (presence, extent, swelling and attenuation of acute lesions; and hyperattenuated arteries) and pre-existing changes (atrophy, leucoaraiosis and old ischaemic lesions). Logistic regression models assessed associations between imaging features and death at 7 and 90 days; good recovery (modified Rankin Scale scores 0-2 at 90 days) and sICH. Data are reported with adjusted ORs and 95% CIs.

RESULTS

2916 patients (67±13 years, National Institutes of Health Stroke Scale 8 (5-14)) were included. Visible ischaemic lesions, severe hypoattenuation, large ischaemic lesion, swelling and hyperattenuated arteries were associated with 7-day death (OR (95% CI): 1.52 (1.06 to 2.18); 1.51 (1.01 to 2.18); 2.67 (1.52 to 4.71); 1.49 (1.03 to 2.14) and 2.17 (1.48 to 3.18)) and inversely with good outcome. Severe atrophy was inversely associated with 7-day death (0.52 (0.29 to 0.96)). Atrophy (1.52 (1.08 to 2.15)) and severe leucoaraiosis (1.74 (1.20 to 2.54)) were associated with 90-day death. Hyperattenuated arteries were associated with sICH (1.71 (1.01 to 2.89)). No imaging features modified the effect of alteplase dose.

CONCLUSIONS

Non-expert-defined brain imaging signs of brain frailty and acute ischaemia contribute to the prognosis of thrombolysis-treated AIS patients for sICH and mortality. However, these imaging features showed no interaction with alteplase dose.

Combined utility of blood glucose and white blood cell in predicting outcome after acute ischemic stroke: The ENCHANTED trial

BACKGROUND

As hyperglycemia and leukocytosis individually predict poor outcome in acute ischemic stroke (AIS), we aimed to determine the significance of their combination on functional outcome and symptomatic intracerebral hemorrhage (sICH) among participants of the Enhanced Control of Hypertension and Thrombolysis Stroke Study (ENCHANTED).

METHODS

Post-hoc analyzes of the full ENCHANTED cohort, an international, multicenter, quasi-factorial, randomized, open, blinded outcome-assessed trial of low- versus standard-dose intravenous alteplase and early intensive versus standard blood pressure (BaP) lowering treatment in 4557 thrombolysis-eligible and treated AIS patients. Patients were divided into four groups according to baseline blood glucose and white blood cells (WBC) levels: A (normal glucose + WBC), B (hyperglycemia + normal WBC), C (normal glucose + high WBC), and D (hyperglycemia + high WBC). Logistic regression models were used to determine associations of each group and poor functional outcome (modified Rankin scale scores 2-6) at 90 days and sICH within 48 h, adjusted for confounders. Quality of model fit was examined with Akaike information classification (AIC), Bayesian information classification (BIC), and likelihood ratio test.

RESULTS

Of 4181 AIS patients included in analyzes, and with group A as the reference, an increasing odds of poor functional outcome was evident across groups B (odds ratio [OR] 1.38, 95 % confidence interval [CI] 1.17-1.63), C (OR 1.26, 95 %CI 0.99-1.60), and D (OR 2.26, 95 %CI 1.79-2.85) (P trend <0.001). Group D patients also had a higher rate of sICH (P trend <0.05). The model fit with the combination of blood glucose and WBC was better than models of their individual components.

CONCLUSIONS

Using a combination of blood glucose and WBC provides strong prognostic significance than either alone in thrombolyzed AIS patients.

Vestibular migraine presenting with acute peripheral vestibulopathy: Clinical, oculographic and vestibular test profiles

To describe clinical, oculographic and vestibular test profiles in patients with vestibular migraine (VM) who presented with acute peripheral vestibulopathy. VM was diagnosed according to Bárány Society or Neuhauser criteria. Neuro-otological examination, video-head impulse tests (v-HIT), cervical and ocular vestibular-evoked myogenic potentials (cVEMP/oVEMP), subjective visual horizontal (SVH) and audiometry were undertaken. Ten patients presented with prolonged vertigo. All had primary position unidirectional horizontal spontaneous nystagmus (mean slow-phase velocity 9.6 ± 7.0°). Horizontal canal vestibulo-ocular reflex was reduced in all (mean gain 0.54 ± 0.2) with refixation saccades (cumulative amplitude 6.4 ± 3.2°). Abnormality rates for cVEMP, oVEMP and SVH were 30%, 80%, 78%, respectively. Magnetic resonance imaging brain was normal in all patients. Patients were followed up over 6 months to 8 years with no change in the final diagnosis. VM can rarely present as an acute peripheral vestibulopathy with findings that mimic vestibular neuritis and should be considered in the differential diagnosis of acute prolonged vertigo.

Separating posterior-circulation stroke from vestibular neuritis with quantitative vestibular testing

OBJECTIVE

To separate vestibular neuritis (VN) from posteriorcirculation stroke (PCS) using quantitative tests of canal and otolith function.

METHODS

Video Head-Impulse tests (vHIT) were used to assess all three semicircular canal pairs; vestibulo-ocular reflex (VOR) gain and saccade metrics were examined. Cervical and ocular-Vestibular-Evoked Myogenic Potentials (c- and oVEMP) and Subjective Visual Horizontal (SVH) were used to assess otolith function.

RESULTS

For controls (n = 40), PCS (n = 22), and VN (n = 22), mean horizontal-canal VOR-gains were 0.96 ± 0.1, 0.85 ± 0.3 and 0.40 ± 0.2, refixation-saccade prevalence was 71.9 ± 41, 90.7 ± 57, 209.2 ± 62 per 100 impulses and cumulative-saccade amplitudes were 0.9 ± 0.4°, 2.4 ± 2.2°, 8.0 ± 3.5°. Abnormality-rates for cVEMP, oVEMP and SVH were 38%, 9%, 72% for PCS, and 43%, 50%, 91% for VN. A gain ≤0.68, refixation-saccade prevalence of ≥135% and cumulative-saccade amplitudes ≥5.3° separated VN from PCS with sensitivities of 95.5%, 95.5%, and 81.8%, and specificities of 68.2%, 86.4% and 95.5%. VOR-gain and saccade prevalence when combined, separated VN from PCS with a sensitivity and specificity of 90.9%. Abnormal oVEMP asymmetry-ratios were of low sensitivity (50%) but high specificity (90.9%) for separating VN from PCS.

CONCLUSION

vHIT provided the best separation of VN from PCS. VOR-gain, refixation-saccade prevalence and amplitude were effective discriminators of VN from PCS.

SIGNIFICANCE

vHIT and oVEMP could assist early identification of the aetiology of Acute Vestibular Syndrome in the Emergency Room.

084 Separating stroke from vestibular neuritis by vestibular function test parameters

Introduction

Vestibular neuritis (VN) and posterior circulation stroke (PCS) are the commonest causes of acute vestibular syndrome (AVS). We aim to identify discriminators of VN from PCS by testing all five vestibular end-organs in patients presenting with AVS.

Methods

Three-dimensional video-head impulse test (v-HIT), cervical and ocular-vestibular evoked myogenic potentials (c-and oVEMP) and subjective visual horizontal (SVH) tests were performed in 22 patients with VN and 22 with PCS. Ipsilesional horizontal, anterior and posterior canal (HC, AC, PC) v-HIT gain and first catch-up saccade characteristics, VEMP amplitude asymmetry-ratios were compared.

Results

All VN and 6 PCS patients had positive clinical HIT. Mean time to testing was 4.7 days for VN, 7.0 days for PCS. VN mean ipsilesional HC and AC first saccade amplitude was larger, peak-velocities faster and onset latencies earlier compared to PCS (p<0.05). No significant difference between VN and PCS in first saccade characteristics was found in PC. Ipsilesional first saccade amplitude, peak-velocity and duration were significantly different between PCS and controls for all canals (p<0.05). A gain <0.68 and first saccade amplitudes >2.2°separated VN from PCS with sensitivities of 95.5% and 86.4% and specificities of 72.7% and 63.6%. First saccade amplitude of >0.91°identified PCS from controls with sensitivity of 68.2% and specificity of 70%. Abnormality rates for AC cVEMP, BC oVEMP and SVH were 42.9%, 50% and 91% for VN and 38.1%, 9% 72% for PCS.

Conclusion

v-HIT gain and catch-up saccade metrics are useful separators of VN from PCS. Detailed saccade analysis complements existing vestibular tests.

Clinical and laboratory factors related to acute isolated vertigo or dizziness and cerebral infarction

OBJECTIVE

To clarify the relationship of clinical factors with isolated vertigo or dizziness of cerebrovascular origin.

METHODS

Clinical data of patients admitted in East Hospital from Jan. 2015 to Apr. 2016, whose complaint were acute vertigo or dizziness were retrospectively collected. All patients arrived at the emergency department within 24 hr of symptom onset, had no acute ischemic lesion first CT and NIHSS score of 0. Patients were divided into cerebral infarction group and noncerebral infarction group according to subsequent cerebral imaging results and clinical and laboratory factors related to cerebral infarction were analyzed.

RESULT

51.6% of patients were female (n = 141). 46 patients (16.8%) were diagnosed with acute cerebral infarction. Baseline demographic data of the two groups was not significantly different. Univariate analysis found that history of smoking (p = 0.009), headache (p = 0.028), unsteadiness (p = 0.009), neuron specific enolase (p = 0.001), and vertebral artery abnormalities found on imaging (p = 0.009) were the significant difference between two groups. Increased neuron specific enolase (p = 0.005) and an abnormal vertebral artery (p = 0.044) were significant on multivariate analysis.

CONCLUSIONS

16.8% of acute isolated vertigo or dizziness presentations were diagnosed with acute cerebral infarction. Increased serum neuron specific enolase and vertebral artery abnormalities were the strongest indicators of acute cerebral infarction.

Frequency, Aetiology, and Outcome of Small Cerebellar Infarction

BACKGROUND AND PURPOSE

Strokes due to small (<2 cm) cerebellar infarction are under-recognised, and their profile and aetiology have not been well characterised. We aimed to determine the frequency, clinical features, aetiology, and outcome of small as compared to large cerebellar infarction.

METHODS

This study is a retrospective analysis of clinical and imaging features of a prospectively assessed series of 108 consecutive patients with acute cerebellar infarction admitted to Liverpool Hospital, Sydney, NSW, Australia, during 2011-2015.

RESULTS

The mean age of the patients was 67 years, and 33 (31%) had small cerebellar infarction. Compared to large cerebellar infarction, those with small cerebellar infarction had a comparable distribution of vascular risk factors but significantly less nausea and vomiting, gait disturbance, limb ataxia, and dysarthria. The posterior (n = 22, 67%) lobe was most commonly affected, followed by the anterior (n = 9, 27%) and flocculonodular (n = 2) lobes. Dizziness, limb ataxia, and nystagmus were significantly more common in patients with anterior lobe infarction. Vertebrobasilar disease was the presumed aetiology in 40 patients (37%), and was less commonly seen in small as compared to large cerebellar infarction. Cardioembolism affected 37% of the patients, irrespective of the size or topography of the cerebellar infarction, and there was no relation of supratentorial white matter lucencies (WMLs) to the size of cerebellar infarction. At 3 months, 65% of the patients were functionally independent (according to modified Rankin Scale scores of 0-2), and having a poor outcome was significantly related to moderate-to-severe supratentorial WML and large cerebellar infarction.

CONCLUSIONS

Small cerebellar infarction accounted for one-third of the ischaemic strokes in this location, most often involved the posterior lobe, causing fewer clinical features, and had a better clinical outcome than large cerebellar infarction. Patients with small cerebellar infarction require appropriate vascular management including investigation for a cardioembolic source.

Cerebellar Infarction and Factors Associated with Delayed Presentation and Misdiagnosis

Background and Purpose: The diagnosis of cerebellar infarction (CBI) is often challenging due to non-specific or subtle presenting symptoms and signs. We aimed to determine whether a common syndromic cluster of symptoms, signs or vascular risk factors were associated with delayed presentation or misdiagnosis to an Emergency Department (ED). The degree of misdiagnosis between ED and neurology physicians and the influence of delayed presentation or misdiagnosis on outcome were also investigated. Methods: A prospective study of CBI patients at a large tertiary-referral hospital with a comprehensive stroke service. Data are reported with OR and 95% CIs. Results: Of 115 consecutive CBI patients (mean age ± SD 66 ± 14 years, 51% male), infarction was isolated to the cerebellum in 46%; the remainder had additional vascular territory involvement (‘mixed CBI'). Most patients (n = 79, 69%) had a mild stroke (National Institute of Health Stroke Scale score ≤4), and tended to present late to ED (>4.5 h; p = 0.05). Dysarthria (OR 3.9, 95% CI 1.6-9.6, p = 0.003) and prior history of atrial fibrillation (AF; OR 3.0, 95% CI 1.02-9.1, p = 0.047) predicted early presentation (<4.5 h; in 52%). Neurological signs (as determined by neurology physicians) were more commonly absent in patients with isolated CBI (OR 4.0, 95% CI 1.2-13.3, p = 0.03) who were also less likely to receive acute stroke therapy (p = 0.03). ED physicians detected fewer neurological signs than neurology physicians (mean 1 vs. 2 signs, p < 0.001), and 34% of CBI patients were misdiagnosed, with peripheral vestibulopathy being the most common alternative diagnosis. Nausea and vomiting (OR 2.3, 95% CI 1.01-5.5, p = 0.046), absence of neurological signs as determined by ED physicians (OR 3.5, 95% CI 1.5-8.0, p = 0.003) and isolated CBI (OR 2.2, 95% CI 1.01-4.8, p = 0.047) correlated with misdiagnosis. Vascular territory involvement did not correlate with time to presentation or misdiagnosis. At 3 months, 65% of patients were functionally independent (modified Rankin Scale (mRS) score 0-2). History of hypertension (p = 0.008), AF (p = 0.012), mixed CBI (p = 0.004) and in-hospital stroke-related complications (p < 0.001) were associated with patients having a poor outcome (mRS ≥3). At 3 months, mortality was 16%, and AF was the only predictor of death (OR 3.2, 95% CI 1.1-8.9, p = 0.03). Late presentation to ED and misdiagnosis did not significantly influence 3-month functional outcome. Conclusions: Late ED presentation and misdiagnosis are common for CBI. Timely diagnosis of CBI may increase opportunity for acute stroke therapies and reduce risk of stroke-related complications.

Reversible cerebral vasoconstriction syndrome following indomethacin

BACKGROUND

Reversible cerebral vasoconstriction syndrome (RCVS) is characterized by severe thunderclap headaches and transient segmental cerebral arterial vasoconstriction. Precipitating factors, including the postpartum state and exposure to vasoactive substances are identified in approximately 50% of cases. Non-steroidal anti-inflammatory drugs have rarely been associated with RCVS.

CASE DESCRIPTION

We report a case of a 51-year-old female with RCVS after administration of indomethacin given to relieve pain caused by renal colic. Cerebral imaging showed non-aneurysmal cortical subarachnoid hemorrhage, and formal angiography demonstrated widespread multifocal segmental narrowing of medium-sized cerebral arteries. These changes resolved on repeat angiography at 3 weeks.

DISCUSSION

Indomethacin is a commonly used drug for treatment of certain primary headache disorders. To date, its mechanism of action remains unclear. A well described side effect of indomethacin is headache, which may be secondary to its vasoconstrictive effects. In our case, we postulate indomethacin, either alone or in combination with emotional stress from pain, triggered or exacerbated an underlying predisposition to RCVS.