Heart failure & cardiomyopathies
Invasive haemodynamic validation of an artificial intelligence echocardiographic tool for detecting heart failure with preserved ejection fraction
Authors' conclusion
This AI model demonstrated high specificity and positive predictive value (PPV) for detecting HFpEF, reliably identifying patients with more advanced haemodynamic abnormalities. Its performance remained robust across variable intervals between transthoracic echocardiography (TTE) and right heart catheterization (RHC), and in patients with indeterminate clinical scores. Due to limited sensitivity, the tool is best utilized to enrich identification of patients with clearly abnormal haemodynamics rather than to exclude HFpEF, particularly in early or borderline cases. While broader use as a screening tool is promising, prospective validation studies are necessary to confirm its utility in general populations.
Study context
- Population
- Patients with unexplained dyspnoea who underwent transthoracic echocardiography (TTE) and right heart catheterization (RHC)28 with both rest and exercise haemodynamics
- Participants
- 47 patients
- Index test or model
- FDA-approved artificial intelligence (AI) tool on apical 4-chamber echocardiographic video vs invasive haemodynamic criteria (pulmonary capillary wedge pressure (PCWP) ≥15 mmHg at rest or ≥25 mmHg with exercise)
- Target condition
- heart failure with preserved ejection fraction (HFpEF)
Methodological detailsretrospective, real-world referral cohort
- Retrospective
- Real-world referral cohort
Performance
As reported in the abstract
Sensitivity
30%
Specificity
88%
PPV
86%
NPV
33%
Further findings
| Outcome | Estimate | P value |
|---|---|---|
| Haemodynamics, AI-positive vs AI-negativeAI-positive · significantly higher in AI-positive | Resting PCWP 20 vs 15 mmHg | P = 0.029 |
| Haemodynamics, AI-positive vs AI-negativeAI-positive · significantly higher in AI-positive | Mean PA pressure 29 vs 24 mmHg | P = 0.02 |
| Haemodynamics, AI-positive vs AI-negativeAI-positive · significantly higher in AI-positive | PVR 2.1 vs 1.3 WU | P = 0.002 |
Heart failure with preserved ejection fraction (HFpEF) by PCWP/CO slope >2 mmHg/L/min (secondary analysis)
- Specificity
- 100%
- PPV
- 100%
- Sensitivity
- 27%
- NPV
- 16%
- n = 25
Prevalence of haemodynamic HFpEF criteria among patients with exercise RHC (n = 28)
- Met haemodynamic HFpEF criteria
- 71%
Invasively confirmed HFpEF in patients with indeterminate H2FPEF scores (n = 25)
- Correctly identified by AI, of those with invasively confirmed HFpEF
- 80%
Model performance by TTE-RHC interval
- Consistent across TTE-RHC intervals <365 and <90 days
Original publication
Composed by AI from the published abstract and checked against it. How Briefs are made →