Valvular & structural heart disease
Machine Learning of Cardiac Anatomy and the Risk of New-Onset Atrial Fibrillation After Transcatheter Aortic Valve Replacement (TAVR)
Authors' conclusion
Underlying cardiac structural differences derived from cardiac imaging may be useful in predicting NOAF following transfemoral TAVR, independent of other clinical risk factors.
Study context
- Population
- Patients undergoing elective transfemoral TAVR for severe symptomatic AS935 without pre-existing atrial fibrillation (AF)
- Participants
- 1,385 patients
- Exposure
- Imaging-derived cardiac structure measurements as risk factors for new-onset atrial fibrillation (NOAF) after TAVR; models with vs without imaging
- Outcome
- NOAF
- Follow-up
- Within 30 days following TAVR
Methodological detailsmachine learning-derived cardiac computed tomography (cCT) measurements, random forest variable ranking followed by multivariable logistic regression
- Machine learning-derived cCT measurements
- Random forest variable ranking followed by multivariable logistic regression
Main finding
As reported in the abstract
No pre-existing AF
5.0%
47 events
Odds ratio (95% CI)P value
Left atrial volume index ≥76 cm²
2.538(1.165–5.531)
P = 0.0191BMI <22 kg/m²
4.064(1.500–11.008)
P = 0.0058EATv
1.007(1.000–1.014)
P = 0.043Aortic annulus area ≥659 mm²
6.621(1.849–23.708)
P = 0.004Sinotubular junction diameter ≥35 mm
3.891(1.040–14.552)
P = 0.0435- C-statistic, model with imaging
- 0.737
- C-statistic, model without imaging variables
- 0.646
Observational data: associations, not effects.
Further findings
Left atrial volume index
- Left atrial volume index: P < 0.001
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
Machine Learning of Cardiac Anatomy and the Risk of New-Onset Atrial Fibrillation After TAVR. JACC Clin Electrophysiol. Published 5 June 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →