Clinical Evaluation of EEG Device for the Triage of Stroke Patients in the Ambulance

Not Applicable

Not yet recruiting

• Conditions: Ischemic Stroke; Stroke; Cardiovascular Diseases; Vascular Diseases; Central Nervous System Diseases; Brain Diseases; Nervous System Diseases; Cerebrovascular Disease

• Registration Number: NCT06871969
• Lead Sponsor: Academisch Medisch Centrum - Universiteit van Amsterdam (AMC-UvA)

Brief Summary

Endovascular thrombectomy (EVT) is the standard treatment for large vessel occlusion (LVO) strokes, but it can only be performed in specialized hospitals. Since ambulance personnel cannot determine if a patient is eligible for EVT, 54% of LVO stroke patients are initially taken to non-EVT-capable hospitals, resulting in an average delay of 1 hour in time-to-EVT in the Netherlands. To reduce this delay, it is crucial for ambulance personnel to identify potential LVO stroke patients and directly transport them to EVT-capable hospitals. Dry electrode electroencephalography (EEG) has shown high diagnostic accuracy for detecting LVO strokes, but in 32% of patients, the EEG signal quality was too poor to analyze.

To address this issue, TrianecT developed StrokePointer, a portable EEG-based triage device designed to collect and analyze EEG data in patients with suspected acute stroke. The objective of this study is to validate the effectiveness and safety of StrokePointer in detecting LVO stroke among patients with a suspected stroke in the pre-hospital setting.

Detailed Description

RATIONALE Endovascular thrombectomy (EVT) is the standard treatment for large vessel occlusion (LVO) stroke. However, EVT can only be performed in specialized hospitals and its effect on functional outcome rapidly decreases with passing time (time = brain). Since ambulance personnel cannot determine whether a patient has a stroke that is eligible for EVT, 54% of patients with an LVO stroke are primarily presented at a non-EVT capable hospital. These patients then require interhospital transfer, resulting in average delay in time-to-EVT of 1 hour in the Netherlands. Therefore, providing ambulance personnel with tools to identify patients with a possible LVO stroke in the ambulance, allowing direct transport to an EVT capable hospital, is much needed. Dry electrode electroencephalography (EEG) has shown to have a high diagnostic accuracy for LVO stroke detection among patients with a suspected stroke (area under the receiving operating curve \[AUC\]: 0.91). However, in 32% of patients EEG signal quality was too poor to analyse. A new portable EEG-based triage device (StrokePointer) has been developed by TrianecT with the aim to collect and analyse EEG data in patients suspected of acute stroke. In this study, we intend to validate the safety and effectiveness of the device.

HYPOTHESIS:

1. StrokePointer device can measure EEG data of sufficient quality in \>85% of patients, and has a good diagnostic accuracy (AUC\>0.8) for LVO stroke detection in the pre-hospital setting.

2. Usability of StrokePointer device is rated as "good" on average by ambulance personnel.

3. StrokePointer is safe to use in an acute care setting.

OBJECTIVE Primary objective is to validate the data quality and diagnostic accuracy of StrokePointer to detect LVO stroke among patients with a suspected stroke in the pre-hospital setting.

STUDY DESIGN CROSSROADS-EEG is an investigator-initiated, prospective, multi-centre cohort study.

STUDY POPULATION Adult patients with a suspected stroke, onset of symptoms (or last seen well) \<24 hours in the pre-hospital setting.

INTERVENTION A single measurement with a dry electrode headset EEG (approximately 2 minutes recording duration) will be performed in each patient. Clinical and radiological data will be collected. EEG data will be acquired with the improved TrianecT EEG device, StrokePointer.

MAIN STUDY END POINTS

* Proportion of patients with a technically successful EEG dataset: at least 20 seconds of usable EEG (at least 3 electrodes with good skin-electrode quality on either side, no movement artifacts, no muscle artifacts) within a measurement time of 3 minutes.

* Diagnostic accuracy of StrokePointer for LVO stroke among patients with a suspected stroke, as measured with AUC as well as sensitivity and specificity.

Study Timeline

• Study First Submitted: 2025-02-27
• Status Verified: 2025-03
• Study First Submitted QC: 2025-03-06
• Last Update Submitted: 2025-03-06
• Study First Posted: 2025-03-12
• Last Update Posted: 2025-03-12
• Study Start: 2025-06
• Primary Completion: 2026-12
• Study Completion: 2026-12

Recruitment & Eligibility

• Status: NOT_YET_RECRUITING
• Sex: All
• Target Recruitment: 275

Inclusion Criteria

• Suspected acute stroke as per judgement of the ambulance personnel.
• Age 18 years or older.
• Onset of symptoms (or last seen well) <24 hours.
• Written informed consent by patient or legal representative (deferred).

Exclusion Criteria

• Injuries or infections of the scalp in the area of the electrode headset placement.

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: SINGLE_GROUP

Primary Outcome Measures

Name	Time	Method
Data quality of StrokePointer	EEG-data for analysis will be recorded within 24 hours after onset of symptoms or last seen well.	Proportion of patients with a technically successful EEG dataset: at least 20 seconds of usable EEG (at least 3 electrodes with good skin-electrode quality on either side, no movement artifacts, no muscle artifacts) within a measurement time of 3 minutes.
Diagnostic accuracy of StrokePointer for LVO stroke	EEG-data for analysis will be recorded within 24 hours after onset of symptoms or last seen well.	Diagnostic accuracy of StrokePointer for LVO stroke among patients with a suspected stroke, as measured with Area under the curve (AUC) as well as sensitivity and specificity

Secondary Outcome Measures

Name	Time	Method
Predictive value of StrokePointer in identifying LVO stroke	EEG-data for analysis will be recorded within 24 hours after onset of symptoms or last seen well.	Positive and negative predictive value of StrokePointer in identifying LVO stroke
User friendliness	EEG-data for analysis will be recorded within 24 hours after onset of symptoms or last seen well.	User-friendliness rating of StrokePointer by ambulance personnel and researchers: (1) after the training 80% of the users should be able to start StrokePointer and start measuring EEG-data within 120s. (2) Interviewed users score (a) usability of StrokePointer hardware on average as "makkelijk" (easy) or better, (b) StrokePointer software on average as "duidelijk" (clear) or better.
Incidence of serious adverse device-related events (Safety of StrokePointer)	EEG-data for analysis will be recorded within 24 hours after onset of symptoms or last seen well.	Provided that the device is being used in line with the intended use, there are no occurrences of serious adverse device-related events in the study.
Diagnostic accuracy for identifying LVO stroke subgroups	EEG-data for analysis will be recorded within 24 hours after onset of symptoms or last seen well.	Diagnostic accuracy for identifying LVO stroke within the following subgroups: sex (men vs. women) and age (above vs. below 60), as measured with sensitivity and specificity.
Incidence of skin reactions (Safety of StrokePointer)	EEG-data for analysis will be recorded within 24 hours after onset of symptoms or last seen well.	Number of patients with an (allergic) skin reaction observed at the site of the electrode.
Discriminative power of StrokePointer	EEG-data for analysis will be recorded within 24 hours after onset of symptoms or last seen well.	Discriminative power of StrokePointer for ischemic stroke vs stroke mimic as measured with Area under the curve (AUC) as well as sensitivity and specificity.

Trial Locations

Locations (1)

Amsterdam University Medical Centers, location AMC; 🇳🇱 Amsterdam, Noord-Holland, Netherlands