Trial Review

Technical difficulties have been reported by some users of the search function and is being investigated by technical staff. Thank you for your patience and apologies for any inconvenience caused.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12623000639628p
Ethics application status
Submitted, not yet approved
Date submitted
28/03/2023
Date registered
14/06/2023
Date last updated
10/05/2024
Date data sharing statement initially provided
14/06/2023
Type of registration
Prospectively registered

Titles & IDs
Public title
Electroencephalography in the Neonate: Seizure Detection
Scientific title
Electroencephalography in the Neonate: Seizure Detection
Secondary ID [1] 309326 0
Nil known
Universal Trial Number (UTN)
Trial acronym
ELITE
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Neonatal Seizures 329520 0
Condition category
Condition code
Neurological 326455 326455 0 0
Other neurological disorders
Reproductive Health and Childbirth 326892 326892 0 0
Complications of newborn

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
There are 2 interventions being tested:

1. Seizure detection algorithm software installed into a commercial grade EEG machine
2. Seizure detection algorithm software installed into a custom-built wireless EEG machine

Babies born >35 weeks who are admitted to the intensive care nursery for seizures or who are at-risk of seizures will be eligible to participate. Once consent is obtained, all neonatal participants will undergo a 9 channel EEG using standard scalp electrodes. Electrodes will be placed on the scalp and checked by a neuroscientist who usually performs neonatal clinical EEG requests. The waveform will not be visible to clinicians and the data collected will be analysed offline by Paediatric Neurologists.

The commercial grade EEG machine will be placed onto the baby, after 4 hours of recording, the electrodes will be disconnected from this machine and the wireless EEG machine will be connected using the same scalp electrodes for another 4 hour recording.

The first intervention will test the seizure detection algorithm software and determine its accuracy against the gold standard reading by a Paediatric Neurologist.

The second intervention will test the algorithm installed in a custom-built wireless EEG machine. Accuracy will be determined by comparing its recording with the commercial grade EEG machine. Compared to the commercial EEG machine, the wireless EEG machine is much more compact in size (size of a credit card), is battery operated and small enough to fit inside the baby's cot.

All recordings will be supervised by a neuroscientist or neonatal nurse. The overall duration of recording will be up to a total of 8 hours, depending on cares, feeds, other investigations etc. Apart from switching electrodes between machines, no specific training is required for the neuroscientist. Any training of neonatal staff will be brief and take place prior to the baby being placed on EEG.
Intervention code [1] 325758 0
Diagnosis / Prognosis
Comparator / control treatment
Some of the recruits will inevitably have a normal EEG, which will serve as the control group. Results of the algorithm software will be compared to the gold standard of reading by a Paediatric Neurologist.
Control group
Active

Outcomes
Primary outcome [1] 334298 0
Validation of EEG annotation by a neonatal seizure detection algorithm will be assessed, by comparing these results to the gold standard visual interpretation by the human expert (2 Paediatric Neurologists).
Timepoint [1] 334298 0
Cumulative data will be assessed at the conclusion of the data collection period.
Secondary outcome [1] 420179 0
To determine the accuracy of the seizure detection algorithm across different EEG acquisition machines (is accuracy similar between commercial grade EEG machine and a custom-built EEG machine)
Timepoint [1] 420179 0
Cumulative data will be assessed at the conclusion of the data collection period.
Secondary outcome [2] 422813 0
Determine the applicability of the seizure detection algorithm i.e. visual assessment of the ability of the algorithm to run in real-time at the cotside.
Timepoint [2] 422813 0
Cumulative data will be assessed at the conclusion of the data collection period.

Eligibility
Key inclusion criteria
Infants eligible for inclusion are infants born > 35 weeks gestational age and are at risk of seizures, these include infants who have had a seizure, with hypoxic ischaemic encephalopathy (HIE), or have had abnormal movements, with a clinical concern for seizures.
Minimum age
35 Weeks
Maximum age
44 Weeks
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
Exclusion criteria will be if the neonatal electrodes are unable to be placed adequately for a recording (e.g. other interfering equipment) or if the prognosis is extremely poor and redirection to palliative care is imminent.

Study design
Purpose of the study
Diagnosis
Allocation to intervention
Non-randomised trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Allocation is not concealed
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Not applicable
Masking / blinding
Open (masking not used)
Who is / are masked / blinded?



Intervention assignment
Crossover
Other design features
N/A
Phase
Not Applicable
Type of endpoint/s
Statistical methods / analysis
There are two aims of data analysis. The first aim is to compare the output of the in situ, automated seizure detection algorithm with the annotation of the human expert. The second aim is show that custom-built wireless EEG records the same activity as the commercial EEG. To evaluate our automated seizure detection algorithm, each recording will be annotated for the presence of artefact by a human expert, blinded to the underlying recording device and the automated seizure detection algorithm. A binary annotation will be performed with a resolution of 1 s (0 will represent 1 s of non-seizure and 1 will represent 1 s of seizure). Differences between the automated and human expert annotations will be assessed using well established metrics such as sensitivity/specificity seizure detection rate and false detections per hour.

We will also compare the human/automated agreement to human/human agreement benchmarks of non-inferiority using protocols we have recently developed. Differences between in agreement between the commercial and wireless EEG machines will be performed by applying the previous analysis on subgroups defined by EEG recording machine. Differences in accuracy per recording such as seizure detection rate, false alarm rate and kappa will be assessed using hypothesis testing. Statistical tests will use a 5% level of significance and be two-sided. Measures of inter-observer agreement will be graded in line with standard practice.

Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
31/05/2024
Actual
Date of last participant enrolment
Anticipated
30/11/2025
Actual
Date of last data collection
Anticipated
30/11/2025
Actual
Sample size
Target
40
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
QLD
Recruitment hospital [1] 24410 0
Royal Brisbane & Womens Hospital - Herston
Recruitment postcode(s) [1] 39990 0
4029 - Herston

Funding & Sponsors
Funding source category [1] 310994 0
Charities/Societies/Foundations
Name [1] 310994 0
Margaret Pemberton Foundation
Country [1] 310994 0
Australia
Primary sponsor type
Other Collaborative groups
Name
Queensland Institute of Medical Research (QIMR) Berghofer
Address
300 Herston Rd, Herston, Qld 4006
Country
Australia
Secondary sponsor category [1] 315300 0
None
Name [1] 315300 0
None
Address [1] 315300 0
Country [1] 315300 0
Other collaborator category [1] 282599 0
Hospital
Name [1] 282599 0
Royal Brisbane and Women's Hospital
Address [1] 282599 0
Butterfield St, Herston, Qld 4029
Country [1] 282599 0
Australia
Other collaborator category [2] 282600 0
Hospital
Name [2] 282600 0
Queensland Children's Hospital
Address [2] 282600 0
501 Stanley St, South Brisbane, Queensland, 4101,
Country [2] 282600 0
Australia

Ethics approval
Ethics application status
Submitted, not yet approved
Ethics committee name [1] 310547 0
Metro North Health Human Research Ethics Committee
Ethics committee address [1] 310547 0
Metro North Office of Research
Level 7, Block 7
Butterfield St, Herston, Qld 4029
Ethics committee country [1] 310547 0
Australia
Date submitted for ethics approval [1] 310547 0
14/02/2023
Approval date [1] 310547 0
Ethics approval number [1] 310547 0
HREC/2022/MNHA/83966

Summary
Brief summary
Continuous Electroencephalography (cEEG) monitoring in neonates provides information on evolving neurological conditions in neonates in the neonatal intensive care unit (NICU). The aim of this study is to clinically validate an automated seizure detection algorithm by comparing EEG interpretations with the gold standard reading by a Paediatric Neurologist. A second aim of the study is to validate the algorithm when housed within a custom built wireless EEG machine and compare that to a commercially available cEEG machine
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 118038 0
Dr Nathan Stevenson
Address 118038 0
QIMR Berghofer
300 Herston Rd, Herston, Qld 4006
Country 118038 0
Australia
Phone 118038 0
+61733620222
Fax 118038 0
Email 118038 0
Nathan.Stevenson@qimrberghofer.edu.au
Contact person for public queries
Name 118039 0
Dr Nathan Stevenson
Address 118039 0
QIMR Berghofer
300 Herston Rd, Herston, Qld 4006
Country 118039 0
Australia
Phone 118039 0
+61733620222
Fax 118039 0
Email 118039 0
Nathan.Stevenson@qimrberghofer.edu.au
Contact person for scientific queries
Name 118040 0
Dr Nathan Stevenson
Address 118040 0
QIMR Berghofer
300 Herston Rd, Herston, Qld 4006
Country 118040 0
Australia
Phone 118040 0
+61733620222
Fax 118040 0
Email 118040 0
Nathan.Stevenson@qimrberghofer.edu.au

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
Yes
What data in particular will be shared?
Deidentified EEGs
When will data be available (start and end dates)?
No end date, available for 5 years after publication.
Available to whom?
The EEGs may be shared anonymously on a public research database.
Available for what types of analyses?
Signal analyses
How or where can data be obtained?
By contacting the principal investigator by email: nathan.stevenson@qimrberghofer.edu.au


What supporting documents are/will be available?

No Supporting Document Provided



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.