Prognostic Value of Cardiac MRI-derived Diastolic Function for Adverse Cardiovascular Outcomes: Insights from UK Biobank.
Background Although diastolic function markers can help predict adverse outcomes in cardiac disease, the prognostic value of cardiac MRI-derived metrics at a population level remains unclear. Purpose To assess the independent and incremental prognostic value of cardiac MRI-derived diastolic function metrics for adverse cardiovascular outcomes in the UK Biobank. Materials and Methods This secondary analysis of the UK Biobank used data from participants enrolled between March 2006 and August 2010. Univariable and multivariable Cox models were used to evaluate the prognostic influence of each diastolic function marker (peak early to peak late mitral filling rate ratio [E/A], peak early diastolic longitudinal strain rate [e'SR], peak early mitral filling rate to peak e'SR ratio [E/e'SR], and maximum left atrial volume [LAVmax]) on adverse outcomes, including heart failure (HF), arrhythmia, and all-cause death. Multivariable models were adjusted for clinical cardiovascular risk factors (age, sex, smoking, body mass index, hypertension, diabetes, and hyperlipidemia) and known prognostic imaging markers (left ventricular [LV] ejection fraction [LVEF], global longitudinal strain [GLS], LV end-diastolic volume index, and LV mass index). Results The cohort included 59 416 individuals (mean age, 65.4 years ± 7.7 [SD]; 30 887 women) with a median follow-up of 6 years. Peak e'SR, E/A, and LAVmax were independently associated with incident HF and arrhythmia, with peak e'SR showing the strongest association (hazard ratio [HR], 0.55 [95% CI: 0.50, 0.61] for HF, 1.37 [95% CI: 1.26, 1.49] for E/e'SR, and 1.28 [95% CI: 1.17, 1.40] for LAVmax; all P < .001). E/e'SR provided incremental value beyond LVEF and GLS. There was no evidence that diastolic markers were associated with incident all-cause death in fully adjusted models. Conclusion Cardiac MRI-derived diastolic function metrics independently predicted HF and arrhythmia in the UK Biobank with incremental value beyond established imaging markers. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Ishida in this issue.
Authors
Mahmood Mahmood, Chadalavada Chadalavada, Hesse Hesse, Fung Fung, Woodbridge Woodbridge, Khanji Khanji, Jensen Jensen, Garg Garg, Munroe Munroe, Petersen Petersen, Aung Aung
View on Pubmed