Prognostic value of left-ventricular hemodynamic forces by echocardiography in patients with left-ventricular non-compaction

Funding Acknowledgements Type of funding sources: None. Background Left-ventricular (LV) non-compaction (LVNC) is a poorly defined entity characterized by marked LV hypertrabeculation and risk of cardiovascular complications (1). LV ejection fraction (LVEF) by echocardiography (TTE) is the main prognostic factor, but on its own it does not allow for proper risk assessment (2). The evaluation of hemodynamic forces (HDF), a recently-introduced technique for mechanistic description of LV-blood interaction (3), may provide additional information on LVNC mechanics. However, their prognostic value is unknown. Purpose To test whether HDF provide prognostic value in LVNC patients beyond conventional TTE parameters. Methods This is a retrospective longitudinal study including patients diagnosed with LVNC (according to Jenni criteria) (1) who underwent TTE evaluation between 2015 and 2019. Patients were excluded if they had moderate-severe valvular heart disease and if follow-up was <12 months. Major cardiovascular events (MACE) were considered as a composite of heart failure hospitalization, ventricular arrhythmias, systemic embolism, and all-cause death. LV HDF were analyzed on apical 2-, 3-, and 4-chamber view TTE images with a prototype software (Medis Suite Qstrain, The Netherlands). HDF were decomposed into apex-base (HDFab) and lateral-septal (HDFls) components. According to its waveform during the cardiac cycle, HDFab was further decomposed into 3 main components: systolic acceleration, LV suction, and diastolic deceleration (DiastDec) (4). DiastDec was also dichotomized according to its distribution. Results 107 patients were included: 60 (56%) were males, mean age was 45.7±18.5 years, mean LVEF was 50.6±10.0%. During median follow-up of 4.5 [3.4–6.1] years MACE occurred in 19 patients (17.8%). Compared to patients who did not experience MACE, those with MACE had lower DiastDec (2.44±1.62 vs. 3.86±2.24%, p = 0.010), while the other HDF parameters were similar. On Cox univariate analysis age (HR 1.06 [95% CI 1.03–1.09], p<0.001), LVEF (HR 0.93 [95% CI 0.90–0.97], p<0.001), longitudinal strain (GLS) (HR 1.21 [95% CI 1.10–1.33], p<0.001), left atrial volume (LAVi) (HR 1.05 [95% CI 1.02–1.08], p<0.001), and DiaDec (HR 0.69 [95% CI 0.51–0.93], p = 0.015) were significantly associated with MACE. 16 patients (84%) who experienced MACE had low DiastDec (< 3%). Low DiastDec was strongly associated with MACE (HR 7.65 [95% CI 2.23–26.27], p<0.001), and remained independently associated with MACE after adjusting for age, LVEF, GLS, and LAVi on separate multivariate Cox analyses. The best performance for MACE prediction was obtained by the multivariate Cox model including age, GLS, and DiastDec (AUC 71.5%). Conclusions Reduced hemodynamic forces in apex-base direction during diastolic deceleration were significantly associated with MACE in LVNC patients beyond conventional echocardiography parameters. Low DiastDec may be a marker of altered intra-cardiac flow and impaired recoil function of the LV in LVNC.