Related Experiment Video
Updated: Jun 12, 2025
05:39
Author Spotlight: A Non-Intubated Video-Assisted Thoracoscopic Surgery with Multimodal Analgesia and Sevoflurane Inhalation Anesthesia
Published on: May 26, 2023
1.4K
Machine learning-augmented interventions in perioperative care: a systematic review and meta-analysis
Divya Mehta1, Xiomara T Gonzalez2, Grace Huang3
1Department of Anesthesiology, Washington University School of Medicine, St. Louis, MO, USA.
British Journal of Anaesthesia
|September 25, 2024
Summary
Machine learning (ML) interventions like the Hypotension Prediction Index (HPI) reduced intraoperative hypotension, and the Nociception Level Index (NoL) decreased postoperative pain. However, ML showed no significant impact on other clinical outcomes, indicating a need for further research.
Area of Science:
- Perioperative Medicine
- Artificial Intelligence in Healthcare
- Clinical Outcomes Research
Background:
- Limited evidence exists on the cumulative effectiveness of machine learning (ML)-driven interventions in perioperative settings.
- A systematic review was conducted to evaluate the impact of ML interventions on perioperative outcomes.
Approach:
- A systematic review and meta-analysis of randomized controlled trials (RCTs) were performed.
- Searches were conducted across multiple databases including Ovid MEDLINE, CINAHL, Embase, Scopus, PubMed, and ClinicalTrials.gov.
- The review adhered to PRISMA guidelines and was registered with PROSPERO (CRD42023433163).
Key Points:
- Thirteen RCTs evaluated ML interventions: Hypotension Prediction Index (HPI), Nociception Level Index (NoL), and a scheduling system.
- HPI significantly decreased absolute and relative intraoperative hypotension.
- NoL significantly reduced postoperative pain scores in the post-anesthesia care unit (PACU).
Conclusions:
- ML interventions demonstrated specific benefits, with HPI reducing hypotension and NoL decreasing postoperative pain.
- No significant impact was observed for other clinical outcomes, including hospital or PACU length of stay and opioid consumption.
- Addressing methodological and clinical practice gaps is crucial for the successful future implementation of ML-driven interventions in perioperative care.
