Acute Video-oculography for Vertigo in Emergency Rooms for Rapid Triage (AVERT) (AVERT)

AVERT is a randomized controlled trial comparing video-oculography (VOG)-guided care to standard care to assess accuracy of diagnoses and initial management decisions for emergency department (ED) patients with a chief symptom of vertigo or dizziness suspected to be of vestibular cause. The trial will test the hypothesis that VOG-guided rapid triage (VRT) will accurately, safely, and efficiently differentiate peripheral from central vestibular disorders in ED patients presenting acute vertigo or dizziness, and that doing so has the potential to improve post-treatment clinical outcomes for these patients.

Study Overview

• Status: Completed
• Conditions: Vertigo; Dizziness
• Intervention / Treatment: Device: VRT Care

Detailed Description

AVERT is a multicenter phase 2 clinical trial comparing a novel diagnostic strategy (VRT) to standard ED diagnostic care at three performance sites. The Specific Aims are to assess diagnostic accuracy, diagnostic workup costs, and estimate the short-term impact of correct diagnosis in anticipation of a larger, definitive phase 3 trial. Adult ED patients with a chief symptom of vertigo, dizziness, or unsteadiness, new or clearly worse in the previous 30 days, will undergo on-site vestibular function tests by trained research personnel using a portable, quantitative VOG recording device. Research personnel will also record a focused symptom history and bedside hearing tests. Eligible patients with at least one pathologic vestibular eye movement finding or pathologic ataxia will be randomized to VRT or standard ED care. Patients eligible for pre-randomization testing but excluded from randomization will be slated for the Observational Arm of the study and will undergo limited 1-month and 6-month phone follow-up. The VRT arm relies on an automated algorithm to interpret VOG results, thereby determining a patient-specific clinical care pathway. For safety, all VRT-arm study participants will undergo stroke protocol MRI before release. All randomized participants will undergo confirmatory testing at one week, including vestibular specialist exam and 1.5 or 3-Tesla research MRI combining stroke and internal auditory canal protocols. All randomized participants will also undergo 1-month and 6-month phone follow-up and medical record review to confirm diagnoses. Clinical findings, ED diagnoses, diagnostic resource utilization, treatments applied, and clinical events during follow-up will be recorded. A multidisciplinary, masked expert panel will adjudicate final diagnoses.

• Study Type: Interventional
• Enrollment (Actual): 130
• Phase: Phase 2

Contacts and Locations

Study Locations

• United States
  • Illinois
    • Peoria, Illinois, United States, 61656
      • University of Illinois
  • Maryland
    • Baltimore, Maryland, United States, 21224
      • Johns Hopkins Hospital - Bayview
  • Massachusetts
    • Boston, Massachusetts, United States, 02114
      • Massachusetts General Hospital
  • Michigan
    • Ann Arbor, Michigan, United States, 48109
      • University of Michigan
  • New York
    • New York, New York, United States, 10029
      • Mt. Sinai Medical Center

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years and older (Adult, Older Adult)
• Accepts Healthy Volunteers: No

Description

Inclusion criteria:

Adult (18 years and older) ED patients with all of the following (all determined pre-randomization):

• VESTIBULAR SYMPTOMS: presenting symptom of "vertigo" OR "dizziness" OR "unsteadiness" (as defined by consensus expert definitions in the International Classification of Vestibular Disorders).
• RELEVANT EXAM SIGNS*: pathologic nystagmus (spontaneous, gaze-evoked, or positional) by bedside VOG testing OR pathologic ataxia (gait, trunk, stance, limbs) by bedside ataxia examination.
• RECENT ONSET: symptoms AND signs* appear to be new or markedly worse in the past month. (*Exam signs are required for randomization, but not for the observational arm)

Exclusion Criteria

1. Excluded from Pre-Randomization Screening

   • Level 1 trauma or critical illness
   • Altered mental status (e.g., delirium, dementia) that would preclude active study participation (this includes patients with abnormal mental state due to alcohol intoxication or illicit substance, which are known, easily-recognized causes of dizziness or vertigo presentations to the ED)
   • Non-English speaking (enrollment of non-English speakers is not feasible given the logistics of identifying a translator and the need for rapid recruitment and randomization in the AVERT study; furthermore, the terms vertigo, dizziness, and unsteadiness may have different meanings in other languages)
   • Known pregnancy (all women of childbearing age who are enrolled will undergo a urine or serum beta human chorionic gonadotropin [hCG] pregnancy test prior to MRI to confirm no pregnancy, per local institutional guidelines)

   • Unable or unsafe to participate in screening, including VOG tests (as deemed by specific pre-enrollment risk assessment questions or ED provider and/or Study Coordinator judgment) including, but not limited to:

     • visual impairment sufficient to prevent visual fixation during the VOG testing
     • clinically-perceived risk to patient of participating in study (ED provider or staff concerns)
     • clinically-perceived risk to research staff (e.g., violence, blood/body fluid/respiratory precautions)
     • unstable cardiac status (given a single reported case of bradycardia with impulse testing)
     • acute cranio-cervical trauma or other condition (e.g., rheumatoid arthritis) that might lead to instability of the cervical spine that would be a contraindication to neck rotation during VOG testing
   • Obvious general medical cause (as judged by treating ED provider) including, but not limited to, acute myocardial infarction, pulmonary embolus, pneumonia, urinary tract infection, drug intoxication, etc.

2. Excluded from Randomization (Eligible for Observational Arm Follow-up)

   • Patient previously randomized in the AVERT Trial (previously screened but not randomized are eligible)

   • Unable to participate fully with study follow-up (particularly MRI) including, but not limited to:

     • unable to return for follow-up testing within 30 days
     • unable to undergo MRI because of contraindications (e.g., pacemaker, metallic foreign body, pregnancy) or other reasons (severe claustrophobia, too large or too heavy for MRI scanner)

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Diagnostic
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: VRT Care; Patients randomized to VRT (VOG-guided Rapid Triage) care will have an algorithm-determined patient-specific diagnosis and treatment pathway in the emergency department. Participants will complete a 1-week in-person follow-up and a 1-month and 6-month phone follow-up.	Device: VRT Care; The VOG report includes direct device output (physiologic traces, quantitative measures) plus most likely diagnosis, category, and clinical trial care pathway (peripheral, equivocal, central) instructions. The VOG report becomes part of the patient's emergency department clinical record.; Other Names: GN Otometrics (Instrumentation & Control Systems, Inc)
No Intervention: Standard of Care (SOC); Patients randomized to Standard of Care will undergo usual emergency department care without revealing results of VOG testing. Participants will complete a 1-week in-person follow-up and a 1-month and 6-month phone follow-up.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
VRT vs. ED SOC Six-Category Diagnosis Accuracy (Primary Analysis-eligible Participants, Two-arm Comparison)	Total diagnosis accuracy VRT vs. ED SOC using 30-day adjudicated final diagnoses categorized in one of six diagnosis categories (3 peripheral, 1 central, 1 medical/other, 1 non-diagnosis). VRT diagnoses were based on automated interpretation of ED index VOG tests in the context of structured medical history information and examination findings from the ED index visit (clinically supervised for safety), while ED SOC diagnoses were based on all clinical information from the ED index visit, including neuroimaging and consultations. Final diagnoses were based on ED index visit, 1-week, and 30-day follow-up clinical assessments. The population was limited to those with complete 1-week follow-up testing including in-person vestibular specialist exam, repeat VOG assessment, and MRI with diffusion-weighted images for ischemic stroke detection.	30-day follow-up time point
VRT vs. ED SOC Total Diagnostic Utilization Costs at the ED Index Visit (Primary Analysis-eligible Participants, Two-arm Comparison)	Total US dollar costs VRT vs. ED SOC for diagnostic tests and consultations obtained during the ED index visit and associated hospital admission (for those admitted at the index visit). For the VRT arm, this does not include costs of safety MRIs required by the institutional review board (IRB)-approved protocol or any tests ordered "off protocol" by ED physicians (i.e., it represents VRT-recommended utilization-based costs); however, it does include tests, consultations, or admissions ordered "on protocol" by consultants or ED physicians in the VRT "equivocal" diagnosis pathway. For the SOC arm, this includes all utilization-related costs from the ED index visit (tests, consultations, or admissions). Total costs are calculated by multiplying fixed cost estimates (2025 national average Medicare reimbursement in US dollars) by utilization rates for each ED index visit service tracked.	30-day follow-up time point
Participants With Short-Term Prespecified Medical Event(s) of Interest (PMEIs) After a Correct vs. Incorrect Diagnosis (Primary Analysis-eligible Participants, One-arm Comparison [SOC Arm Only])	PMEIs included ED revisits, falls, vascular events, and test or treatment complications. PMEIs occurring between the ED index visit disposition and 1-week follow-up visit (after which the two arms joined the same diagnostic pathway) were considered. Events diagnosed at ED index visit were not counted. Events newly diagnosed at 1-week follow-up or in the interval prior to 1-week follow-up were counted, regardless of relatedness to ED index dizziness symptoms, with the exception of test or treatment complications, which were required to be related directly or indirectly to the dizziness symptoms. To avoid "double counting" misdiagnoses as follow-on PMEIs pursuant to an initial misdiagnosis, 1-week stroke diagnoses not rendered at the ED index visit were not counted unless neurologic or vestibular symptoms/signs worsened after the ED index visit. Six-category accuracy was used to determine "correct" vs. "incorrect" index ED SOC diagnoses relative to 30-day adjudicated final diagnoses.	1-week follow-up time point

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Expert VOG vs. ED SOC Six-Category Diagnosis Accuracy (Participants With a Known Final Diagnosis, One-arm Comparison [SOC Arm Only])	Total diagnosis accuracy Expert VOG vs. ED SOC using 30-day adjudicated final diagnoses categorized in one of six diagnosis categories (3 peripheral, 1 central, 1 medical/other, 1 non-diagnosis). Expert VOG diagnoses were based on masked interpretation of ED index VOG tests in the context of basic demographic and medical history information from the ED index visit, while ED SOC diagnoses were based on all clinical information from the ED index visit, including neuroimaging and consultations. Final diagnoses were based on ED index visit, 1-week, and 30-day follow-up clinical assessments. The SOC arm population was limited to those with known final diagnoses to avoid counting as "incorrect" cases with unknown final diagnoses after 30-day follow-up. This within-subject comparison reflects current potential accuracy of expert VOG-based tele-diagnosis and the targeted maximum diagnostic accuracy (i.e., expert level performance) for future automated algorithms, relative to current care.	30-day follow-up time point
VRT vs. ED SOC Stroke-No Stroke Diagnosis Accuracy (Primary Analysis-eligible Participants, Two-arm Comparison)	Total diagnosis accuracy VRT vs. ED SOC using 30-day adjudicated final diagnoses categorized as stroke (any cerebrovascular event) versus no stroke (including peripheral vestibular, medical, psychiatric, or other central neurologic causes such as multiple sclerosis, traumatic brain injury, epilepsy, or anticonvulsant toxicity). "Index VRT Diagnosis" and "ED SOC Diagnosis" were compared to the "Adjudicated Final Diagnosis" based on ED index visit and 30-day follow-up clinical assessments.	30-day follow-up time point

Collaborators and Investigators

• Sponsor: Johns Hopkins University
• Collaborators: National Institute on Deafness and Other Communication Disorders (NIDCD); GN Otometrics
• Investigators: Principal Investigator: David Newman-Toker, MD, PhD, Johns Hopkins University

Publications and helpful links

General Publications

• Kotwal S, Fanai M, Fu W, Wang Z, Bery AK, Omron R, Tevzadze N, Gold D, Garibaldi BT, Wright SM, Newman-Toker DE. Real-world virtual patient simulation to improve diagnostic performance through deliberate practice: a prospective quasi-experimental study. Diagnosis (Berl). 2021 Mar 8;8(4):489-496. doi: 10.1515/dx-2020-0127. Print 2021 Nov 25.
• Edlow JA, Carpenter C, Akhter M, Khoujah D, Marcolini E, Meurer WJ, Morrill D, Naples JG, Ohle R, Omron R, Sharif S, Siket M, Upadhye S, E Silva LOJ, Sundberg E, Tartt K, Vanni S, Newman-Toker DE, Bellolio F. Guidelines for reasonable and appropriate care in the emergency department 3 (GRACE-3): Acute dizziness and vertigo in the emergency department. Acad Emerg Med. 2023 May;30(5):442-486. doi: 10.1111/acem.14728.

Helpful Links

• Brain Injury OutcomeS (BIOS) website

Study record dates

Study Major Dates

• Study Start (Actual): December 4, 2017
• Primary Completion (Actual): March 17, 2023
• Study Completion (Actual): March 17, 2023

Study Registration Dates

• First Submitted: March 3, 2015
• First Submitted That Met QC Criteria: June 24, 2015
• First Posted (Estimated): June 29, 2015

Study Record Updates

• Last Update Posted (Actual): December 26, 2025
• Last Update Submitted That Met QC Criteria: December 8, 2025
• Last Verified: December 1, 2025

More Information

Terms related to this study

• Keywords: Stroke; Vertigo; Benign paroxysmal positional vertigo; Dizziness; Emergency department; Video-oculography; Vestibular neuritis; Unsteadiness; Vestibulo-ocular reflex
• Additional Relevant MeSH Terms: Neurologic Manifestations; Cerebrovascular Disorders; Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Vascular Diseases; Cardiovascular Diseases; Pathologic Processes; Disease Attributes; Otorhinolaryngologic Diseases; Sensation Disorders; Ear Diseases; Cranial Nerve Diseases; Vestibulocochlear Nerve Diseases; Retrocochlear Diseases; Labyrinth Diseases; Vestibular Diseases; Pathological Conditions, Signs and Symptoms; Signs and Symptoms; Stroke; Emergencies; Vertigo; Benign Paroxysmal Positional Vertigo; Dizziness; Vestibular Neuronitis
• Other Study ID Numbers: IRB00044228; 1U01DC013778-01A1 (U.S. NIH Grant/Contract)

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: Yes