Perioperative medicine & outcomes
Mortality associated with the perioperative management of patients with proximal femur fractures: A machine learning prediction model
Authors' conclusion
Our results provide interesting elements to optimize the perioperative management of these patients.
Study context
- Population
- Patients operated for osteoporotic proximal femur fractures (PFF), Marseille, November 2018 to June 2023
- Participants
- 2,442 patients
- Exposure
- 90-day non-survivors vs survivors
- Non-survivorsn = 279
- Outcome
- 90-day mortality (factors associated; predictive model)
- Follow-up
- 90 days
Methodological detailsbicentric, observational, neural network prediction model
- Bicentric
- Observational
- Neural network prediction model
Main finding
As reported in the abstract
Factors associated with 90-day mortality
Factors associated with 90-day mortality Non-survivors 11.4%
Age, years Non-survivors 87.2 (0.8) · Survivors 79.8 (0.3)
Male
Charlson score Non-survivors 1[0-3] · Survivors 1[0-2]
ASA score Non-survivors 3[3-3] · Survivors 3[2-3]
- Non-survivors vs Survivors: P < 0.0001
- Non-survivors vs Survivors: P = 0.003
- Non-survivors vs Survivors: P < 0.0001
- Non-survivors vs Survivors: P < 0.0001
Observational data: associations, not effects.
Further findings
Neural network model for 90-day mortality (age, gender, ASA score, perioperative confusion, haemoglobin, creatinine, albumin), test set
- AUC
- 0.91 [0.91-0.92]
- Sensitivity
- 50.5%
- Specificity
- 75.0%
- PPV
- 20.6%
- NPV
- 92.1%
- Test set
Original publication
Mortality associated with the perioperative management of patients with proximal femur fractures: A machine learning prediction model. Anaesth Crit Care Pain Med. Published 27 November 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →