Prognostic impact of CMR-derived indexed left ventricular end-diastolic volume (LVEDVi) in individuals with preserved and reduced LV ejection fraction - a UK Biobank study

Background Left ventricular ejection fraction (LVEF) is a central prognostic marker in heart failure (HF) and indicates eligibility for neurohumoral treatment strategies when reduced. However, evidence guiding therapeutic decisions in subjects with mildly reduced LVEF (41–49%) remains limited, as most available data originate from subgroup analyses, and these subjects may exhibit structural characteristics typically seen in heart failure with preserved ejection fraction. This highlights an important limitation of relying solely on LVEF for risk stratification and treatment guidance. Purpose The objective of our study was to determine the prognostic value of MRI-derived indexed left ventricular end-diastolic volume (LVEDVi) beyond LVEF in participants from the UK Biobank. Methods MRI-derived LVEDVi was available for N=5527 participants without the diagnosis of HF recruited between 2006 and 2010 in the United Kingdom. Inclusion of participants without HF enabled evaluation of the incremental prognostic value of LVEDVi beyond LVEF within a cohort not confounded by HF-specific therapy Results In participants with a LVEF ≥50%, increased LVEDVi was not associated with adverse outcomes (Log rank p = 0.77). In contrast, among subjects with LVEF < 50%, elevated LVEDVi was associated with a significantly higher incidence of 4-point major adverse cardiovascular events (4P-MACE, i.e. cardiovascular death, myocardial infarction, stroke and hospitalisation for HF; log-rank p < 0.0001) compared with subjects who had normal LVEDVi. Category-free net reclassification showed a statistically significant improvement in prediction of 4P-MACE when LVEDVi was added in participants with an LVEF below 50%, with a net improvement in individual risk prediction of 20% (95%CI 2–40%). Net reclassification was most pronounced for hospitalization for HF, with a significant improvement in prediction of 38% (95%CI 7–71%). In contrast, no improvement in risk prediction was observed in participants with an LVEF above 50%. To further assess the prognostic impact of LVEDVi beyond reduced LVEF, subjects were stratified by LVEF (41–49% or ≤40%) and LVEDVi (normal or increased). After full adjustment for cardiovascular history, biomarkers (hsCRP, HbA1c, LDL, eGFR), sex, age and body mass index, individuals with increased LVEDVi and a LVEF of 41–49% exhibited a 4P-MACE risk comparable to that of subjects with LVEF ≤ 40% and increased LVEDVi (LVEF 41–49%: HR 3.0 [95% CI 2.1–4.0] vs. LVEF < 40% HR 2.8, [95% CI 2.2–3.7]). This heightened risk was predominantly driven by higher rates of myocardial infarction and hospitalization for HF. Conclusion In summary, we demonstrate in a cohort of individuals with pre-clinical HF and reduction in LVEF, that greater CMR-derived LVEDVi identifies people at heightened risk of disease progression. Further studies are needed to determine whether these people may benefit from earlier therapeutic intervention.