Pulmonary vascular disease & thromboembolismCohort study

The added value of echocardiography in pulmonary arterial hypertension risk assessment an artificial intelligence machine learning–derived analysis of the ULTRA RIGHT VALUE registry

Eur Heart J Cardiovasc ImagingPublished

  • multicentre European prospective cohort
  • machine learning

Authors' conclusion

Incorporation of echocardiographic parameters improves risk scores in PAH.

Study context

Population
Patients with prevalent pulmonary arterial hypertension (PAH) enrolled in the ULTRA RIGHT VALUE study between January 2022 and December 2023patients with prevalent PAH
Participants
401 patients
Exposure
European Society of Cardiology (ESC)/European Respiratory Society (ERS) and REVEAL 2.0 risk scores with vs without echocardiographic right ventricle (RV) variables

Main finding

As reported in the abstract

Odds of achieving/maintaining a low-risk status per one-unit decrease in any RV morphological parameter

ESC/ERS score: odds increased up to

69%

REVEAL score: odds increased up to

92%

Observational data: associations, not effects.

Further findings

Outcomec-index (95% CI)P value
c-index for morbi-mortality prediction — ESC/ERS score with echocardiographic variables vs ESC/ERS score · improved from 0.73 (95% CI 0.69-0.76)0.78 (0.75–0.81)P < 0.01
c-index for morbi-mortality prediction — REVEAL 2.0 score with echocardiographic variables vs REVEAL 2.0 score · improved from 0.74 (95% CI 0.67-0.83)0.79 (0.76–0.82)P < 0.01

Odds of achieving/maintaining a low-risk status per one-unit increase in any RV functional parameter

ESC/ERS score: odds increased up to
1.28
REVEAL score: odds increased up to
1.78
Echocardiographic assessments of RV-PA coupling were the most performant

c-index for morbi-mortality prediction

Machine learning: indexes of RV-PA coupling had the highest influence on the c-index

Original publication

The added value of echocardiography in pulmonary arterial hypertension risk assessment an artificial intelligence machine learning–derived analysis of the ULTRA RIGHT VALUE registry. Eur Heart J Cardiovasc Imaging. Published 17 September 2026. doi:10.1093/ehjci/jeag232

Summarised from the published abstract; every number is checked against it