Haemodynamics, fluids & transfusion
Identification of intraoperative hypotension endotypes and revolution with a temporal deep learning algorithm
Authors' conclusion
Five endotypes of IOH were identified
Study context
- Population
- Surgical patients with continuous intraoperative monitoring of systemic vascular resistance index (SVRI), stroke volume index (SVI), stroke volume variation (SVV), CI and HR who had at least one episode of intraoperative hypotension (IOH) (mean arterial pressure (MAP) < 65 mmHg for at least 1 min)
- Participants
- 184 patients · 1,304 hypotensive episodes
- Outcome
- Distinct hemodynamic endotypes of IOH
Methodological detailsdeep learning (long short-term memory (LSTM)-based autoencoder), unsupervised k-means clustering
- Deep learning (LSTM-based autoencoder)
- Unsupervised k-means clustering
Main finding
As reported in the abstract
IOH endotypes
- Five distinct IOH endotypes identified by k-means
- 5
Observational data: associations, not effects.
Further findings
Timing of endotypes within the procedure
- Mild vasodilation with preserved CI mainly in the first and second quartiles of the procedure
- Bradycardia throughout the procedure, more frequently in the fourth quartile
- Myocardial depression primarily in the second quartile
- Hypovolemia frequently throughout the procedure
Intra-event transitions between endotypes within the same hypotensive episode
- Transitions between endotypes within the same episode (events)
- 5
Original publication
Identification of intraoperative hypotension endotypes and revolution with a temporal deep learning algorithm. Perioper Med (Lond). Published 14 October 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →