The Video Head Impulse Test and Predictability of Vestibular Disorders
Authors
Arberg, Nicklas ; Jaurji, Mark
Term
5. Term (Master thesis)
Education
Publication year
2025
Submitted on
2025-12-19
Pages
28
Abstract
Background Vestibular disorders are common causes of dizziness and difficult to diagnose. The video head impulse test (vHIT) is a relatively new tool that measures the vestibulo-ocular reflex (VOR), which stabilizes vision during quick head turns. vHIT can assess each semicircular canal; in this study, we examined the horizontal canals. Methods People with vertigo and/or dizziness underwent vHIT using strict inclusion and exclusion criteria, standardized quality requirements, and clear definitions of abnormal VOR gain (the strength of the VOR) and pathological saccades (corrective eye movements). A vestibular diagnosis was assigned either on the day of vHIT or after more extensive vestibular testing. All vHIT reports were thoroughly reviewed, and markers of poor quality and artifacts were recorded for excluded tests. Results A total of 1119 participants met the inclusion criteria (mean age 59.2 years; 42.4% male). Of these, 52.0% received a vestibular diagnosis and 48.0% did not. We tested three interpretation rules: 1) low mean VOR gain plus pathological saccades, 2) low gain alone, and 3) pathological saccades alone. The results were: sensitivity 43.5%, 47.2%, and 54.3%; specificity 96.1%, 94.0%, and 84.0%; positive predictive value 92.3%, 89.6%, and 78.6%; negative predictive value 61.1%, 62.2%, and 62.9%. Overall agreement was 68.5–69.7%, and Cohen’s kappa indicated fair agreement. Performance varied substantially across specific diagnoses. Conclusion Combining gain and saccades was the most rational vHIT approach for predicting vestibular disease. When mean VOR gain is clearly reduced and saccades are uncertain, gain alone can be useful. vHIT should not be used as a general screening test for all vestibular disorders, but rather as a first-line vestibular test alongside others. It was a poor screening tool for benign paroxysmal positional vertigo (BPPV), Ménière’s disease, and vestibular schwannoma, but showed promising potential for vestibular neuritis. Notably, 40.5% of all vHIT examinations were of poor quality, highlighting the need for universal quality markers.
Baggrund Vestibulære lidelser er hyppige årsager til svimmelhed og kan være svære at diagnosticere. Video head impulse test (vHIT) er en relativt ny test, der måler den vestibulo-okulære refleks (VOR), som stabiliserer synet ved hurtige hovedbevægelser. vHIT kan i princippet teste alle seks buegange; i dette studie undersøgte vi de vandrette buegange. Metode Personer med svimmelhed og/eller vertigo gennemgik vHIT efter strenge inklusions- og eksklusionskriterier, standardiserede kvalitetskrav og klare definitioner af patologisk VOR-gain (styrken af VOR) og patologiske sakkader (korrigerende øjenbevægelser). En mulig vestibulær diagnose blev stillet enten samme dag som vHIT eller efter mere omfattende vestibulær udredning. Alle vHIT-rapporter blev grundigt gennemgået, og markører for lav kvalitet og artefakter blev registreret for udelukkede undersøgelser. Resultater I alt opfyldte 1119 personer inklusionskriterierne (gennemsnitsalder 59,2 år; 42,4% mænd). 52,0% fik en vestibulær diagnose, 48,0% gjorde ikke. Vi afprøvede tre fortolkningskriterier: 1) lavt gennemsnitligt VOR-gain kombineret med patologiske sakkader, 2) lavt gain alene, og 3) patologiske sakkader alene. Resultaterne var: følsomhed 43,5%, 47,2% og 54,3%; specificitet 96,1%, 94,0% og 84,0%; positiv prædiktiv værdi 92,3%, 89,6% og 78,6%; negativ prædiktiv værdi 61,1%, 62,2% og 62,9%. Den samlede overensstemmelse lå på 68,5–69,7%, og Cohen’s kappa indikerede fair overensstemmelse. Ydelsen varierede betydeligt mellem enkelt-diagnoser. Konklusion At kombinere gain og sakkader var den mest rationelle vHIT-tilgang til at forudsige vestibulær sygdom. Hvis det gennemsnitlige VOR-gain tydeligt er nedsat, kan gain alene være nyttigt, når sakkader er tvivlsomme. vHIT bør ikke bruges som generel screeningsmetode for alle vestibulære lidelser, men som en førstelinjetest blandt flere. Testen var en dårlig screening for benign paroksystisk positionel vertigo (BPPV), Ménières sygdom og vestibulært schwannom, men viste lovende resultater for neuritis vestibularis. Bemærk, at 40,5% af alle vHIT-undersøgelser var af lav kvalitet, hvilket understreger behovet for universelle kvalitetsmarkører.
[This apstract has been rewritten with the help of AI based on the project's original abstract]
Keywords