Ventricular tachycardia predicts all-cause mortality and non-sudden cardiac death in non-ischaemic cardiomyopathy

Abstract Funding Acknowledgements Type of funding sources: Private company. Main funding source(s): The DCM-VT study was partially supported by an investigator-initiated grant (IIS-310) from Biosense Webster (a Johnson & Johnson company). The DANISH study was supported by unrestricted grants from Medtronic, St. Jude Medical, TrygFonden, and the Danish Heart Foundation. Background Endstage heart failure, ventricular tachycardia (VT) and sudden cardiac death (SCD) are sequels of non-ischaemic cardiomyopathies (NICM). Implantable cardioverter defibrillators (ICDs) can terminate VT and prevent SCD but mortality benefit has been limited in recent trials suggesting that VT may also be associated with non-sudden cardiovascular death (NSCVD). Purpose To assess the effect of VT on all-cause mortality and NSCVD in NICM patients with ICDs. Methods Two multicenter cohorts of NICM patients were merged for analysis: (1) patients randomized to ICD in the DANISH trial, which included patients with NICM, LVEF ≤35% and no prior VT and (2) the DCM-VT study cohort, which included consecutive NICM patients with any LVEF referred for ablation of sustained VT. Predictors for the primary composite outcome of all-cause death and heart transplant (death/HTx) and for the secondary outcome of NSCVD were assessed. In addition, propensity score matching was performed using the parameters age, gender, LVEF and eGFR. Finally, survival following VT occurrence was compared between DANISH and DCM-VT patients. Results A total of 828 patients (556 DANISH, 272 DCM-VT) were followed for a median of 3.9 years (2.1 – 5.8). DANISH patients had a median LVEF of 25% (20 – 30) and DCM-VT patients 35% (27 – 43). A total of 148 patients (DANISH 16.4%, DCM-VT 21.0%) experienced death/HTx and NSCVD occurred in 78 patients (7.7%, 12.9%, respectively). In multivariable analysis history of VT at inclusion was a significant predictor of the primary outcome with an adjusted hazard ratio (HRa) of 3.81 (95% CI 2.33 - 6.24) as were age [HRa 1.02 per year (1.01 - 1.04)], LVEF [HRa 0.96 per 1% increase (0.94 - 0.98)] and reduced eGFR [HRa 1.71 (1.15 – 2.53 )]. History of VT predicted NSCVD with an HRa of 7.54 (3.97 - 14.33). After matching, history of VT still predicted death/HTx [HR 4.12 (2.48 - 6.82)] and NSCVD [HR 6.92 (3.30 - 14.52)]. Risk of death/HTx, as well as NSCVD, for DANISH patients was similar to DCM-VT patients after they experienced their first VT event [HRa 0.90 (0.48 - 1.69) and HRa 1.07 (0.56 - 2.06) resp.]. Conclusions VT occurrence appears to be an important predictor of non-sudden CV death as well as all-cause death in NICM patients independent of other markers of heart failure severity. VT may be a marker of progressive disease and remodelling that could be incorporated into risk-stratification to help predict deterioration and need for HTx.