Definition of left ventricular dilation in heart failure using cardiovascular magnetic resonance: prognostic implications of end-diastolic diameter and volume assessment

Abstract

Abstract Background Patients with heart failure (HF) and reduced or mildly reduced ejection fraction (HFrEF/HFmrEF) display variable grades of left ventricular (LV) dilation. LV end-diastolic diameter (LVEDD) and volume (LVEDV) may not be concomitantly increased. The respective prognostic implications of LV dilation have not been explored. Purpose To assess the prognostic implications of LVEDD and LVEDV index (LVEDVi) in HFrEF/HFmrEF. Methods Data from consecutive HF outpatients undergoing cardiovascular magnetic resonance (CMR) and with LV ejection fraction <50% were analyzed. LVEDD was measured in a basal short axis slice. Established age- and gender-specific cut-offs for LVEDVi and a cut-off of 58 mm for LVEDD were used to define LV dilation. Patients were followed for the occurrence of the combined endpoint of cardiovascular death or HF hospitalization. Results A total of 564 patients (79% men, median age 65 years [56–73], ischaemic aetiology in 32%) were included. HFrEF was present in 85% of patients, and HFmrEF in 15%. Median values for LVEDD and LVEDVi were 63 mm (58–69) and 124 mL/m2 (104–149) respectively. LVEDD was significantly correlated with LVEDVi (r=0.804, p<0.001). Sixty-nine percent of patients had concomitant dilation of LV based on LVEDD and LVEDVi. In contrast, 18% of patients did not have concordant definition of LV dilation based on LVEDVi and LVEDD, including 14% with small LVEDD and large LVEDVi. LV dilation based on LVEDVi was associated with a shorter survival free from CV death or HF hospitalization (log-rank 4.6, p=0.032), while LVEDD >58 mm was not (p=0.278). There was a trend toward an independent association between LV dilation based on LVEDVi and outcome adjusting for HFrEF/HFmrEF status and ischaemic aetiology (hazard ratio 2.10, 95% confidence interval 1.00–4.41, p=0.050). An independent association was not found for LV dilation based on LVEDD. An LVEDD >66 mm was associated with increased risk of CV death or HF hospitalization on spline curve analysis (Figure 11). All patients with an LVEDD >66 mm had also LV dilation based on LVEDVi. Conclusions In patients with HFrEF and HFmrEF, LV dilation based on LVEDVi does not always coincide with a dilated LVEDD. The cut-off value of LVEDD associated with an excess of the combined endpoint is much larger than the value used in clinical practice. However, in patients with dilated LV based on LVEDVi, LVEDD does not add to the risk stratification. Funding Acknowledgement Type of funding sources: None.