Prospective analysis of video head impulse tests in patients with acute posterior circulation stroke

Separation of stroke from vestibular neuritis using the video head impulse test: machine learning models versus expert clinicians

BACKGROUND

Acute vestibular syndrome usually represents either vestibular neuritis (VN), an innocuous viral illness, or posterior circulation stroke (PCS), a potentially life-threatening event. The video head impulse test (VHIT) is a quantitative measure of the vestibulo-ocular reflex that can distinguish between these two diagnoses. It can be rapidly performed at the bedside by any trained healthcare professional but requires interpretation by an expert clinician. We developed machine learning models to differentiate between PCS and VN using only the VHIT.

METHODS

We trained machine learning classification models using unedited head- and eye-velocity data from acute VHIT performed in an Emergency Room on patients presenting with acute vestibular syndrome and whose final diagnosis was VN or PCS. The models were validated using an independent test dataset collected at a second institution. We compared the performance of the models against expert clinicians as well as a widely used VHIT metric: the gain cutoff value.

RESULTS

The training and test datasets comprised 252 and 49 patients, respectively. In the test dataset, the best machine learning model identified VN with 87.8% (95% CI 77.6%-95.9%) accuracy. Model performance was not significantly different (p = 0.56) from that of blinded expert clinicians who achieved 85.7% accuracy (75.5%-93.9%) and was superior (p = 0.01) to that of the optimal gain cutoff value (75.5% accuracy (63.8%-85.7%)).

CONCLUSION

Machine learning models can effectively differentiate PCS from VN using only VHIT data, with comparable accuracy to expert clinicians. They hold promise as a tool to assist Emergency Room clinicians evaluating patients with acute vestibular syndrome.

Video head impulse test in subacute and chronic stroke survivors: new perspectives for implementation of assessment in rehabilitation

INTRODUCTION

The Video Head Impulse Test (vHIT) is a safe and reliable assessment of peripheral vestibular function. Many studies tested its accuracy in clinical settings for differential diagnosis and quantification of the vestibulo-oculomotor reflex (VOR) in various disorders. However, the results of its application after lesions of the CNS are discordant and have never been studied in rehabilitation. This study aims to assess the VOR performance in a sample of stroke survivors.

METHODS

This is a cross-sectional study on 36 subacute and chronic stroke survivors; only persons with first-ever stroke and able to walk independently, even with supervision, were included. We performed VOR assessments for each semicircular canal by vHIT and balance assessments by the Berg Balance Scale and the MiniBESTest scale.

RESULTS

Two hundred and sixteen semicircular canals were assessed using the Head Impulse paradigm (in both the vertical and horizontal planes), while 72 semicircular canals were assessed using the Suppressed Head Impulse paradigm (horizontal plane). There was a high prevalence of participants with dysfunctional canals, particularly for the left anterior and right posterior canals, which were each prevalent in more than one-third of our sample. Furthermore, 16 persons showed an isolated canal dysfunction. The mean VOR gain for the vertical canals had confidence intervals out of the normal values (0.74-0.91 right anterior; 0.74-0.82 right posterior; 0.73-0.87 left anterior).

CONCLUSION

Our findings suggest that peripheral vestibular function may be impaired in people with stroke; a systematic assessment in a rehabilitation setting could allow a more personalized and patient-centred approach.

Application of a Video Head Impulse Test in the Diagnosis of Vestibular Neuritis

In patients presenting in the emergency department with acute vertigo, a rapid and accurate differential diagnosis is crucial, as posterior circulation strokes can mimic acute vestibular losses, leading to inappropriate treatment. The diagnosis of vestibular neuritis is made based on the clinical manifestation and a bedside otoneurological assessment. In the clinical examination, an evaluation of the vestibulo-ocular reflex is the key element; however, the accuracy of the bedside head impulse test depends on the clinician's experience. Thus, new diagnostic methods are needed to objectify and facilitate such rapid vestibular evaluations. The aim of our paper is to provide a comprehensive review of the video head impulse test's application in the diagnosis of vestibular neuritis. Numerous studies have reported advantages that make this method helpful in detailed otoneurological evaluations; in contrast to the bedside head impulse test, it enables an analysis of all six semicircular canals function and records the covert corrective saccades, which are invisible to the naked eye. As a portable and easy diagnostic tool, it is known to improve the diagnostic accuracy in patients with acute vertigo presenting in the emergency department. Moreover, as it evaluates the vestibulo-ocular reflex across different frequencies, as compared to caloric tests, it can be used as an additional test that is complementary to videonystagmography. Recently, several papers have described the application of the video head impulse test in follow-up and recovery evaluations in patients with vestibular neuritis.

Video head impulse test in stroke: a review of published studies

Accurate and timely diagnosis of posterior circulation stroke in patients with acute dizziness is a challenge that can lead to misdiagnosis and significant harm. The present review sought to identify and describe published research on the clinical application of vHIT in posterior circulation stroke. vHIT, a portable device, has gained prominence in evaluating peripheral vestibular disorders and offers potential applications in diagnosing neurological disorders, particularly posterior circulation stroke. Several studies have shown that vHIT can differentiate between stroke and vestibular neuritis based on VOR gain values, with high sensitivity and specificity. The manuscript also discusses vHIT's performance in differentiating between types of posterior circulation stroke, such as PICA, AICA, and SCA strokes. While vHIT has demonstrated promise, the review emphasizes the need for further research to validate its use as a tool to rule out stroke in acute dizziness patients in the emergency department. In conclusion, the manuscript underscores the potential of vHIT as a valuable addition to the diagnostic arsenal for acute dizziness, particularly in the context of posterior circulation stroke. It calls for further research and wider adoption of vHIT in clinical settings to improve patient care and reduce unnecessary costs associated with misdiagnoses.