Who is shocked and who survives? A multi-state analysis of the NORDIC ICD trial

Abstract

Abstract Background/Purpose The interaction between the risk of arrhythmic death and a competing non-arrhythmic risk of death in patients suitable for implantable cardioverter defibrillator (ICD) implantation is not well understood. Commonly, identification of subpopulations with the largest benefit of ICD implantation has been performed by separate risk models for the outcomes death and appropriate shock therapy. The interrelation between the outcomes was not sufficiently studied. Methods Data were derived from the safety population of the multinational, prospectively randomized NORDIC ICD trial (N=1067) with real-word patients implanted with a single, dual or triple chamber ICD for primary or secondary prevention. Since all outcome adjudication was performed by an independent Clinical Event Committee supported by full telemonitoring data transmission, a high validity of ICD interventions could be achieved. To investigate the impact of baseline characteristics on time to first appropriate shock, death without prior appropriate shock therapy and death after appropriate shock therapy, a multi-state Cox model was computed. Missing data have been multiply imputed before analysis. Results At 36 months follow-up, 86.4% of the patients were alive (7.8% after appropriate shock). 11.0% and 2.6% patients died without or after a foregoing appropriate shock, respectively. The primary randomization allocation showed no significant effect on the 3 outcome types. Higher age (per 5 years) and NYHA functional class (≥III vs. ≤II) were associated with an increased risk of death without appropriate shock (HR 1.31, 95% CI 1.14–1.50, p<0.001, and HR 2.17, 95% CI 1.26–3.74, p=0.005, fig.1, accordingly). The presence of diabetes mellitus at baseline was associated with the reduced risk of the occurrence of an appropriate shock (HR 0.57, 95% CI 0.35–0.92, p=0.022). Patients with secondary prevention indication for an ICD had very high risk for an appropriate shock after ICD implantation (HR 3.21, 95% CI 2.02–5.11, p<0.001), but not for death without or with previous appropriate shock (HR 1.42, 95% CI 0.72–2.79, p=0.306, or HR 0.73, 95% CI 0.23–2.34, p=0.594 after ICD shock). Renal insufficiency and ischemic vs. nonischemic disease showed a significantly increased global effect on all three transitions (HR 1.63, 95% CI 1.18–2.24, p=0.003 and HR 1.53, 95% CI 1.06–2.20, p=0.025, respectively). Conclusion The new multi-state model shows the interrelation between appropriate shocks and death, as well a remarkable variation of risk factors for the transitions. Specifically, the presence of higher age and NYHA functional class ≥III at baseline were strong prognostic factors for all-cause mortality without a foregoing shock therapy, but were not predictive for an appropriate shock therapy. In this all-comer study, a significant discriminator predictive for appropriate shock therapy, but not for death was an indication for secondary prevention of sudden cardiac death. Multi-state graph for NYHA class Funding Acknowledgement Type of funding source: Private company. Main funding source(s): This work was supported by Biotronik SE & Co. KG (Berlin, Germany)