Perioperative medicine & outcomes
Machine learning-augmented interventions in perioperative care: a systematic review and meta-analysis
Authors' conclusion
HPI decreased the duration of intraoperative hypotension, and NoL decreased postoperative pain scores, but no significant impact on other clinical outcomes was found.
Study context
- Population
- Surgical inpatients13 randomised controlled trials (RCTs)
- Comparison
- Machine learning (ML)-driven interventions vs standard care
- Outcome
- Perioperative outcomes
Methodological detailsrandom-effects model, vote counting, PRISMA
- Random-effects model
- Vote counting
- PRISMA
Pooled result
As reported in the abstract
- Hypotension Prediction Index (HPI) vs Standard care: P = 0.003
- HPI vs Standard care: P < 0.0001
- NoL vs Standard care: P = 0.004
Further pooled results
| Outcome | P value |
|---|---|
| Opioid consumption (NoL)NoL vs Standard care · no significant impact; n=339, I2=92% | P = 0.31 |
| Opioid consumption (NoL)NoL vs Standard care · no significant impact; n=339, I2=0% | P = 0.11 |
| Hospital LOS and PACU stay (HPI and NoL)HPI and NoL vs Standard care · no significant difference; n=361, I2=0% | P = 0.81 |
| Hospital LOS and PACU stay (HPI and NoL)HPI and NoL vs Standard care · no significant difference; n=267, I2=0 | P = 0.44 |
Original publication
Machine learning-augmented interventions in perioperative care: a systematic review and meta-analysis. Br J Anaesth. Published 24 September 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →