Introduction: The aim of the NORDIC ICD sub-study was to compare the incidence of implantable cardioverter-defibrillator (ICD) shocks and clinical outcome of a mixed population implanted with a Cardiac Resynchronization Therapy Defibrillator (CRT-D) or ICD.
Methods: The NORDIC ICD study included 1077 patients randomized to ICD implantation with or without defibrillation testing. Based on current guidelines, 1067 were eventually implanted with an ICD for primary or secondary prevention. For the comparison of ICD shock incidence a mixed logistic regression model with center as random effect was used. For the comparison of mortality rates a shared frailty Cox-regression model with center as frailty term was applied. All models were adjusted for random group.
Results: The sub-study included 356 patients with a CRT-D and 711 patients with an ICD. Patients with a CRT-D were significantly older (p<0.001), suffered more often from non-ischemic cardiomyopathy (p<0.001), had higher NYHA function class (p<0.001), had more often a reduced left ventricular ejection fraction (p<0.001). The incidence of appropriate ICD shocks was lower in the CRT-D group (6.2% vs. 9.8%; adjusted for random group and center effect, OR: 0.6 95%-CI [0.4-1.0]; p=0.046). The frequency of inappropriate ICD shocks showed a non-significant trend for less inappropriate shocks for CRT-D recipients (2.5% in CRT-D group vs. 4.2% in ICD group; adjusted for random group and center effect, OR: 0.6 95%-CI [0.3-1.3]; p=0.169). All-cause mortality did not differ between groups (HR: 0.9 95%-CI [0.6-1.4]; p=0.701).
Conclusions: CRT-D systems were implanted in older and sicker patients with advanced heart failure. Compared with ICD recipients, those implanted with a CRT-D had a lower risk of appropriate ICD shocks and non-significant trend towards lower occurrence of inappropriate ICD shocks. There was no statistical significant difference between the mortality rates of ICD and CRT-D patients.