ClinicalTrials.Veeva

Menu

AI-Assisted Blood Pressure Control During Anesthesia (RL-PRAIS PoC)

B

Beijing Tsinghua Chang Gung Hospital

Status

Not yet enrolling

Conditions

General Anesthesia

Treatments

Device: Device: AI Clinical Decision Support System
Procedure: Procedure: Standard of Care Anesthesia

Study type

Interventional

Funder types

Other

Identifiers

NCT07123675
24444-4-03

Details and patient eligibility

About

This study is being done to see if a new computer-based tool can help anesthesiologists keep patients' anesthesia levels more stable during surgery.

Background: During general anesthesia, doctors carefully adjust medication doses to keep patients safely and comfortably unconscious. This requires constant attention. Researchers want to find out if a smart computer assistant can help with this task.

Purpose of the Study: The main goal is to compare two ways of managing anesthesia. One way is the standard method used by doctors today. The other way is having the doctor use suggestions from a new computer program designed to recommend medication doses in real-time. We want to see if using the computer tool can lead to more stable anesthesia, potentially using less medication and helping patients recover more smoothly.

Who Can Participate: The study will include adult patients who are scheduled to have surgery that requires general anesthesia.

What Will Happen in the Study: If you choose to join, you will be randomly assigned (like flipping a coin) to one of two groups:

Standard Care Group: Your anesthesiologist will manage your anesthesia medications based on their expert judgment and normal practice, as is done every day.

Computer-Assisted Group: Your anesthesiologist will also manage your anesthesia, but they will see suggestions from a computer system on a screen. The doctor will always have the final say and can choose to follow or ignore the computer's advice to ensure your safety.

In both groups, you will receive safe and complete anesthetic care. Your participation in the study will last for the duration of your surgery and the immediate recovery period. Researchers will look at information like the amount of medication used and how stable your anesthesia level was.

Full description

Rationale: Precise control of anesthetic depth during general anesthesia is crucial for patient safety and optimal outcomes. Suboptimal dosing can lead to intraoperative awareness, hemodynamic instability, or excessive drug administration, which may delay recovery. This study will evaluate a novel reinforcement learning (RL) based clinical decision support system designed to provide real-time, personalized dosing recommendations for anesthetic agents.

Hypothesis: The use of an RL-based decision support system for guiding anesthetic drug administration results in a greater percentage of time within a target range of anesthetic depth compared to standard manual practice by anesthesiologists.

Study Design: This is a prospective, dual-center, parallel-group, randomized controlled superiority trial.

Objectives:

Primary Objective: To compare the percentage of case time that the Bispectral Index (BIS) is maintained within the target range of 40 to 60 between the RL-guided group and the standard care group.

Secondary Objectives: To compare total consumption of anesthetic agents (e.g., propofol and remifentanil), incidence and duration of intraoperative hypotension, time to tracheal extubation, and length of stay in the post-anesthesia care unit (PACU).

Interventions:

Intervention Group (RL-Guided): Anesthesiologists will manage general anesthesia with the aid of a real-time decision support system. The system, powered by a reinforcement learning algorithm, will display continuous dosing recommendations for propofol and remifentanil. The attending anesthesiologist retains full clinical autonomy and is responsible for all final dosing decisions.

Control Group (Standard Care): Anesthesiologists will manage general anesthesia according to their clinical experience and institutional standard of care, without input from the decision support system.

Study Population: Adult patients (aged 18-65 years), ASA physical status I-III, scheduled for elective surgery expected to last at least two hours under general anesthesia. Key exclusion criteria include known allergies to anesthetic agents, severe cardiovascular or respiratory disease, and emergency surgery.

Enrollment

40 estimated patients

Sex

All

Ages

18 to 65 years old

Volunteers

No Healthy Volunteers

Inclusion criteria

    1. Age range: 18 to 85 years. 2. Patients undergoing non-cardiac surgery. 3. Receiving intravenous-inhalation combined general anesthesia with tracheal intubation during surgery.

    2. Patients undergoing elective surgery. 5. Anesthesia maintenance plan includes:

    3. Propofol for continuous sedation during the maintenance phase;

    4. Remifentanil for continuous analgesia during the maintenance phase;

    5. Sevoflurane or desflurane for inhalation anesthesia. 6. ASA physical status I-IV. 7. Continuous monitoring of blood pressure, heart rate, and Bispectral Index (BIS) during surgery.

    6. Continuous invasive arterial blood pressure monitoring during surgery.

Exclusion criteria

    1. Emergency surgery. 2. Continuous infusion of remifentanil or propofol for less than 30 minutes during surgery.

    2. Receiving continuous intravenous sedatives other than propofol, or continuous infusion of intravenous sedatives other than propofol.

    3. Receiving continuous intravenous analgesics other than remifentanil, or continuous infusion of intravenous analgesics other than remifentanil.

    4. Inhaled anesthetic maintenance concentration is not equal to 0.5 Minimum Alveolar Concentration (MAC).

    5. Allergy to propofol or allergic reaction to remifentanil. 7. Severe obesity (BMI ≥ 35 kg/m²).

Trial design

Primary purpose

Treatment

Allocation

Randomized

Interventional model

Parallel Assignment

Masking

Double Blind

40 participants in 2 patient groups

Experimental: AI-Assisted Anesthesia Management
Experimental group
Description:
Participants in this arm will have their general anesthesia managed by an anesthesiologist with real-time guidance from an AI-based clinical decision support system. The system provides recommendations for drug administration and physiological parameter adjustments.
Treatment:
Device: Device: AI Clinical Decision Support System
Active Comparator: Standard Anesthesia Care
Active Comparator group
Description:
Participants in this arm will receive standard-of-care general anesthesia management. The anesthesiologist will make all clinical decisions based on their professional judgment and standard institutional practices, without the aid of the investigational AI system.
Treatment:
Procedure: Procedure: Standard of Care Anesthesia

Trial contacts and locations

2

Loading...

Central trial contact

Gao Zhifeng, MD; Zheng Zhang, MD

Data sourced from clinicaltrials.gov

Clinical trials

Find clinical trialsTrials by location
© Copyright 2026 Veeva Systems