Trial Review

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
ACTRN12623000040662
Ethics application status
Approved
Date submitted
28/03/2022
Date registered
16/01/2023
Date last updated
16/01/2023
Date data sharing statement initially provided
16/01/2023
Type of registration
Prospectively registered

Titles & IDs
Public title
Using facial patterns from artificial intelligence facial recognition to guide and optimise treatment with opioid medications following painful surgery
Scientific title
Prospective Randomized Controlled Trial- Artificial Intelligence facial recognition to guide opioid de-escalation following painful surgery
Secondary ID [1] 306782 0
Nil known
Universal Trial Number (UTN)
Trial acronym
Linked study record
This study is a follow up study to ACTRN12620000865910

Health condition
Health condition(s) or problem(s) studied:
opioid use disorder 325822 0
post operative surgical pain 325823 0
Condition category
Condition code
Anaesthesiology 323148 323148 0 0
Pain management
Public Health 323150 323150 0 0
Epidemiology

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
The intervention will utilise artificial intelligence facial pattern recognition using specialised software to interpret films of facial expressions of patients after they have received premedication with opioid prior to their surgery. The filming will involve asking the patient to look ahead with a normal facial expression at which time the camera will record their face for 2 minutes. The patient will not be asked to produce any specific facial expressions, for example smiling or frowning. The film will be interpreted by experts in facial recognition who will identify special features of facial expression that predispose previously opioid naïve patients to prolonged opioid requirement following surgery. The patients who are considered at risk of prolonged opioid requirement will then receive opioid de-escalation intervention delivered by trained pharmacists. The opioid de-escalation will involve repeated visits with education about opioids and review of postoperative pain from trained pharmacists. The education will involve casual verbal conversation which will focus on pain severity ratings, the role of opioids to control and not necessarily take away pain following surgery and the need to reduce opioid dosing with time following surgery provided pain severity is controlled. Each meeting with pharmacists will take approximately 20 minutes. Reduction of opioid will be approximately 25-50% every (1-3) days provided pain intensity keeps falling or remains the same and there is no surgical complication. Pharmacists will visit participants daily while the patient is in hospital and communicate via phone after the patient is discharged from hospital twice weekly. Pharmacist intervention will continue until the patient has ceased their opioid. Treatment provided by pharmacists will be personalized in accordance with participant recorded pain intensity. Adherence to opioid de-escalation will be monitored though recording of opioid use. Family doctors will be informed as to patient treatment and may follow up as per usual practice.
Intervention code [1] 323233 0
Early detection / Screening
Intervention code [2] 323234 0
Prevention
Intervention code [3] 323235 0
Treatment: Other
Comparator / control treatment
Facial recognition data will be collected and interpreted for patient blinding but will not be used to guide therapy Surgical residents will manage opioids during the surgical recovery period using their existing knowledge with involvement of pharmacists who help with dispensing and provide basic education. Following discharge from hospital communication of the patients surgery and therapy will be made with the family physician continues to provide follow-up.
Control group
Active

Outcomes
Primary outcome [1] 330904 0
Duration of opioid therapy following surgery. This will be collected by data linkage to medical records while the patients remain in hospital and via telephone follow-up after the patient leaves hospital if relevant i.e. if they continue to take opioids.
Timepoint [1] 330904 0
The data collection for the primary timepoint will cease at 5 days after the patient has remained opioid free and be measured in total days of opioid requirement. this may range from days to weeks.
Secondary outcome [1] 408009 0
Early postoperative NRS pain intensity score
Timepoint [1] 408009 0
pain intensity for 48 hrs after surgery
Secondary outcome [2] 408010 0
Mean NRS pain intensity during opioid therapy duration
Timepoint [2] 408010 0
Pain intensity measured daily for the period during which the patient requires opioid therapy. This may range from days to weeks.

Eligibility
Key inclusion criteria
- Painful surgery, all types allowed, with stratification for surgeries perceived as more painful incorporated in the randomisation.
- Age 18-80 yrs
- Able to understand concept of Artificial Intelligence
- Where significant perioperative opioid therapy is planned.
- General anaesthetic planned
- Where first dose of intravenous opioid administered pre-induction is allowed by anaesthetist (to allow AI facial recognition filming).
Minimum age
18 Years
Maximum age
80 Years
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
- Those receiving opioids or atypical opioids (tapentadol, tramadol) in the two weeks preceding surgery
- Not willing to have AI facial recognition evaluation
- Unable to understand procedure
- Current opioid use disorder and current illicit drug use
- Contraindications to use of significant opioid
- Parkinsons Disease, Cerebrovascular accident impairing facial expression, Use of anti-psychotic medication likely to affect facial expression, Surgery involving head and neck.

Study design
Purpose of the study
Prevention
Allocation to intervention
Randomised controlled trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Central randomisation by phone/fax/computer
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Simple randomisation using a randomisation table created by computer software (i.e. computerised sequence generation)
Masking / blinding
Blinded (masking used)
Who is / are masked / blinded?
The people receiving the treatment/s

The people assessing the outcomes
The people analysing the results/data
Intervention assignment
Parallel
Other design features
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
This is a feasibility study that will focus on whether methodology is practical and realistically and easily applicable . It will assess 40 patients. It will be conducted with a view to planning a larger study that will require approximately 1500 participants. Data from the 40 patients will be used to calculate opioid duration data that will be used for power analysis to confirm numbers needed for the larger study ( looking for a 20% reduction in duration of postoperative opioid therapy duration with intervention)

Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
1/02/2023
Actual
Date of last participant enrolment
Anticipated
1/06/2024
Actual
Date of last data collection
Anticipated
1/07/2024
Actual
Sample size
Target
40
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 22053 0
The Alfred - Melbourne
Recruitment postcode(s) [1] 37176 0
3004 - Melbourne

Funding & Sponsors
Funding source category [1] 311119 0
Commercial sector/Industry
Name [1] 311119 0
Strong Room technologies
Country [1] 311119 0
Australia
Primary sponsor type
Individual
Name
Dr Alex Konstantatos
Address
Alfred Hospital Department of Anesthesiology and Perioperative Medicine
Alfred Hospital
Commercial Rd
Melbourne
Victoria
3004
Country
Australia
Secondary sponsor category [1] 312455 0
Commercial sector/Industry
Name [1] 312455 0
Strong Room Technologies
Address [1] 312455 0
Strong Room Technologies
1/58 Lorimer St
Docklands
Victoria
3008
Country [1] 312455 0
Australia

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 310649 0
Alfred Hospital Research and Ethics Committee
Ethics committee address [1] 310649 0
C/o Alfred Hospital
Commercial Rd
Melbourne
VIC 3004
Ethics committee country [1] 310649 0
Australia
Date submitted for ethics approval [1] 310649 0
04/05/2022
Approval date [1] 310649 0
09/11/2022
Ethics approval number [1] 310649 0

Summary
Brief summary
This is a study that aims to identify patients that might be susceptible to prolonged requirement for opioids after painful surgery. It will do this by using facial recognition technology before surgery to see whether participants respond to a specialised pharmacologic intervention to minimise unnecessary opioid use. It will do so by comparing this intervention with usual care to see whether there is an overall reduction of days where opioid therapy is needed following surgery between two groups. One group will use information from Artificial Intelligence prior to surgery to identify facial characteristics predicting prolonged opioid requirement following surgery, Those with predicted prolonged opioid requirement will have postoperative pain managed in close conjunction with a pharmacist. They will be involved in a detailed program where they are reviewed regularly and, if circumstances are appropriate, have their opioid dose reduced proactively. This is known as pharmacy-led opioid de-escalation. It comprises management by an accredited clinical pharmacist providing an individualised plan aiming for detailed patient education together with opioid reduction of approximately 25-50% every (1-3) days. The plan will take pain intensity scores into consideration. Increasing pain intensity scores will result in no de-escalation, whilst unchanged and decreasing pain scores will result in reduction of opioid of approximately 25-50% every (1-3) days. The clinical pharmacist will also continue to closely follow progress of patients assigned to opioid de-escalation after discharge from hospital to continue the de-escalation pathway until cessation of all opioid analgesics, often working in conjunction with a family doctor. The other group will be randomly allocated to receive usual care for their postoperative pain. This comprises filming the participant but not acting on the results of the AI facial recognition analysis. All participants in this group will be subject to current practice whereby surgical residents (young doctors working with the surgical team) manage opioids after the acute recovery period using their existing knowledge with involvement of pharmacists who help with dispensing and provide basic education. Following discharge from hospital the family physician continues to provide follow-up. The main measure of the study will be the time in days of opioid requirement compared between the two groups.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 118390 0
Dr Alex Konstantatos
Address 118390 0
Alfred Hospital
Department of Anesthesiology and Perioperative Medicine
Alfred hospital
commercial Rd
Melbourne VIC 3004
Country 118390 0
Australia
Phone 118390 0
+61425702049
Fax 118390 0
Email 118390 0
[email protected]
Contact person for public queries
Name 118391 0
Dr Alex Konstantatos
Address 118391 0
Alfred Hospital
Department of Anesthesiology and Perioperative Medicine
Alfred hospital
commercial Rd
Melbourne VIC 3004
Country 118391 0
Australia
Phone 118391 0
+61 32763176
Fax 118391 0
Email 118391 0
[email protected]
Contact person for scientific queries
Name 118392 0
Dr Alex Konstantatos
Address 118392 0
Alfred Hospital
Department of Anesthesiology and Perioperative Medicine
Alfred hospital
commercial Rd
Melbourne VIC 3004
Country 118392 0
Australia
Phone 118392 0
+61 32763176
Fax 118392 0
Email 118392 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment
IPD will not be made available as the data is sensitive


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.