Heart failure & cardiomyopathies
Establishing a Novel Model to Assess Exercise Capacity in Chronic Heart Failure Based on Stress Echocardiography
Authors' conclusion
Compared with the LR, XGBT, CART and RF models, the LR model performed best at predicting the VE/VCO2 slope category of CHF patients.
Study context
- Population
- Patients with chronic HF undergoing cardiopulmonary exercise test (CPET) and exercise stress echocardiography (aged 60 ± 11 years; 78% male)VC1 30, VC2 50
- Participants
- 80 patients
- Comparison
- Logistic regression (LR) vs XGBT vs classification regression tree (CART) vs random forest (RF) classifiers of exercise stress echocardiography (ESE) indicators to distinguish VE/VCO2 slope ≥ 30
- Outcome
- Discrimination of VC1 (VE/VCO2 slope < 30) from VC2 (VE/VCO2 slope ≥ 30)
Methodological detailsK-fold cross-validation, Prospective enrolment; all patients underwent both CPET and ESE
- K-fold cross-validation
- Prospective enrolment; all patients underwent both CPET and ESE
Performance
As reported in the abstract
Discrimination of VC1 vs VC2 (AUC)
| Test or model | AUC |
|---|---|
| LR | 0.82 (95% CI 0.73 to 0.92) |
LR
AUC0.82 (95% CI 0.73 to 0.92)
Further findings
LVEF, VC1 vs VC2
P = 0.617
Independent predictors of VE/VCO2 slope ≥ 30 (multivariable LR)
- ΔE (peak minus rest)
- P = 0.002
- S'peak
- P = 0.005
- Sex
- P = 0.020
- E/e'peak, ΔLVEF, ΔLV global longitudinal strain, Δstroke volume not predictors
- P > 0.05
Original publication
Establishing a Novel Model to Assess Exercise Capacity in Chronic Heart Failure Based on Stress Echocardiography. ESC Heart Fail. Published 1 September 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →