Propensity score-matched analysis in isolated left ventricular dilation in non-ischaemic dilated cardiomyopathy

Abstract

Introduction

The presence and extent of late gadolinium enhancement (LGE) assessed by cardiac magnetic resonance imaging (CMR) are strong prognosticators of death in patients with non-ischaemic dilated cardiomyopathy (DCM), defined as left ventricular (LV) dilation and left ventricular ejection fraction (LVEF) < 50%. Although the current guidelines defined the concept of “isolated LV dilation” (ILVD) as LV dilation with preserved LVEF ≥ 50%, the prognostic value of the “LGE granularity” is not established in this population.

Objective

To assess the prognostic value of the concept of “LGE granularity” including its extent, location, and pattern for predicting all-cause death above traditional prognosticators in patients with DCM or ILVD, separately.

Method

Between 2008 and 2021, all consecutive patients with DCM and ILVD without implantable cardioverter-defibrillator or history of sustained ventricular arrhythmia referred for CMR were included in two centres. The primary outcome was all-cause death using the French National Registry of Death. A propensity score matching was performed to balance characteristics in patients with DCM vs. those with ILVD. Cox regressions were performed to determine the prognostic value of each LGE findings.

Results

Of 2752 patients analysed (age 52 ± 8 years, 56% male), 15% patients died after a median (inter-quartile range) follow-up of 9 (7–12) years. A total of 737 (27%) patients had LGE. In the propensity-score matched population (n = 1084 in DCM subgroup and n = 1084 in isolated LV dilation), the LGE presence was associated with death (HR = 2.98, 95%CI: 1.97–4.50, p < 0.001). In ILVD patients with LGE (n = 265), the LGE extent (HR = 1.41, 95%CI: 1.09–1.83, p = 0.009), the presence of LGE in multiple areas (HR = 3.86, 95%CI: 1.73–8.61, p < 0.001) and the septal location (HR = 2.97, 95%CI: 1.37–6.46, p = 0.006) were strong prognosticators of death after adjustment for traditional prognosticators (Figure 1). Similarly, in DCM patients with LGE (n = 268), the LGE extent (HR = 1.42, 95%CI: 1.07–1.89, p = 0.014), the LGE presence in multiple areas (HR = 8.41, 95%CI: 3.32–21.3, p < 0.001) and the septal location (HR = 6.65, 95%CI: 3.02–14.6, p < 0.001) were strongly associated with death.

Conclusion

The concept of “LGE granularity” was independently associated with all-cause death after adjustment for all traditional prognosticators in both DCM and ILVD. These results suggest that CMR assessment could improve the risk stratification in this population.