A random sample of 3,809 patients (13.4%) who received one or more shock therapies from an ICD or CRT-D device were selected from the LATITUDE patient database. All shock EGMs were adjudicated by a panel of seven board-certified electrophysiologists. Survival (by cross-reference to the United States Social Security Death Index) was compared by 1) time from first shock to death by adjudicated rhythm and 2) matched pair analysis of patients with and without a shock.
Patients were followed for an average of about three years post device implant and two years following their initial shock. The mean age of the patient was 64 years with a 78% male predominance. Devices included single chamber ICD (24.9%), dual chamber ICD (34.6%), CRT-D (40.5%). 29% of patients had unsuccessful ATP that preceded their initial shock.
Of all first shock episodes- 58.7 % were appropriate (ventricular arrhythmias) and 41.3% were inappropriate. Of all first shocks, 36% were noted to be monomorphic ventricular tachycardia (MVT) with 16% representing ventricular fibrillation (VF)/polymorphic VT (PMVT). Monomorphic/polymorphic VT represented 7% of first shocks. Atrial fibrillation (18% of all first shocks) was the most common rhythm precipitating inappropriate shock with sinus tachycardia comprising 17% of all first shocks. Nonsustained ventricular tachycardia and nonarrhythmic shocks (noise,artifacts, oversensing) comprised 1.4% and 5% respectively of this group.
Patients receiving a first shock for ventricular arrhythmias had an increased risk of death during follow up as compared to those shocked for non-ventricular arrhythmias. Among those with appropriate shocks, those with polymorphic VT/VF demonstrated an increased risk of death compared to monomorphic VT within the ICD population, although not within the CRT-D population.
As compared to survival in the monomorphic VT population, survival in the sinus tachycardia and non-arrhythmic shock population was better while in the atrial fibrillation/atrial flutter group, survival was similar.
When compared to the no-shock matched comparison, patients receiving any shock (appropriate or inappropriate) had an increased risk of death (hazard ratios: 2.82- appropriate; 1.81 inappropriate). However, when adjudicated by rhythm on presentation, patients with shocks for ventricular arrhythmias or atrial arrhythmias had an increased risk of death (hazard rations: 2.10- ventricular; 1.61- atrial). Shocks for sinus tachycardia, SVT, nonsustained arrhythmias or noise/oversensing demonstrated no change in overall survival as compared to a matched cohort with no shocks. (hazard ratios: sinus tach/SVT- 0.97; noise- 0.91)
Cardiac rhythm (ventricular arrhythmias and atrial fibrillation) rather than the ICD shock itself was the major predictor of reduced survival in both ICD and CRT-D patients.
To shock or not to shock- since the emergence of the implantable defibrillator, this hotly debated topic has driven therapy development (anti-tachycardia pacing) and programming (single, dual or multi-zone programming).1,2 The prevailing scientific thought suggesting mortality detriment with ICD shocks has led to evolution of ICD programming from "defaulting to shocks" to "shock avoidance" with disparate data fueling both sides of the debate.3-7
In this article, Dr. Powell and his colleagues are to be commended for shedding light on the debate with an outcomes-driven analysis of the effects of ICD shocks and their impact on survival in a real world population. Utilizing LATITUDE patient remote monitoring data, patients with high-risk status (prior shocks) were compared to matched controls. This observation of substrate versus "shock-effect" driving overall risk-with a specific focus on mortality- will evolve our view of disease progression.8,9 With this data, the dilemma of programming towards shock-avoidance should put the focus squarely on quality of life instead of mortality. Coupled with data from MADIT-RIT suggesting poorer outcome in certain patients programmed with anti-tachycardia pacing (ATP)- this data should force us to more dynamically tailor our therapies and scrutinize whether there are times when ATP therapy may be more harmful than good.10
Despite the limitations of an observational analysis, this article advances our knowledge of arrhythmias within our heart failure and sudden death populations and the implications of the therapies we provide. These data should substantively advance our ability to tailor therapy for our patients. As we advance our understanding of population-based care, device therapy and stored diagnostic information will play a vital role in how we care for large groups of patients. We must now develop a better understanding of these device-based risk "signposts" within our patients and how they drive a "value-based" approach to more advanced therapies such as cardiac ablation, advanced device therapy, mechanical cardiac support and cardiac transplantation and in some instances where appropriate a more patient-centered palliative care approach.
< Back to Listings