Dutch Outcome in ICD Therapy (Do It)

Summary: Objective: The objective of this study is to identify patients that will not benefit from ICD-implantation for primary prevention of sudden cardiac death. The specific purpose is to develop two prediction rules. The first rule identifies patients that die within years of the ICD implantation despite ICD implantation. The second rule identifies patient that do not need ICD implantation, i.e. patients that survive long time without receiving ICD shocks. Also health economic effects, cost minimization and budget impact of targeted supply of ICD implantation in patients with reduced left ventricular function in a setting of structural heart disease will be assessed. Study design: A national, multicenter cohort registry with a two-year clinical follow up to evaluate the practice of ICD implantation in The Netherlands. This registry involves the collection of demographic, clinical, and ICD data (see next paragraph) in all patients in whom an ICD was implanted for primary prevention of sudden death. The following baseline information is entered into the registry: Gender and age; cardiac risk factors (such as body mass index, smoking habits, etc); cardiac history (atrial fibrillation, previous myocardial infarction, heart failure hospitalization, non-sustained ventricular tachycardia, etc); history of previous stroke, TIAS or peripheral vascular disease; indication for ICD implantation; functional status (NYHA classification); left-ventricular ejection fraction; aetiology of LV-dysfunction (ischemic versus non-ischemic); diabetic and renal status; medication usage; height, weight, blood pressure, and heart rate; laboratory parameters such as creatinine, eGFR, ureum, NT-proBNP, etc.; electrocardiographic parameters (LBBB, QTc and QRS duration) Dataset: Data of approximately 1500 consecutive patients are entered into the registry in participating high-volume ICD-implanting centers in The Netherlands. The following information is entered into the registry at 6 months visits: ICD therapy; electronic ICD readouts; recent history and functional status; clinical events. Intervention: not applicable. Outcome measures: All patients are assigned to one of the following endpoint categories: 1. Alive after successful ICD-treatment, i.e. the patient is alive and has received at least one critical ICD shock. 2. Alive but no ICD-treatment received, i.e. the patient is alive and has never received an ICD shock. 3. Dead after unsuccessful ICD-treatment, i.e. the patient has died after unsuccessful ICD-shocks. 4. Dead but having received prior successful ICD-treatment, i.e. the patient has died but has received prior successful ICD-shocks. 5. Dead & no ICD-treatment given, i.e. the patient has died of another cause without ever having received an ICD shock. The secondary endpoints are: death of any cause, appropriate shocks, inappropriate shocks, hospitalization for ICD-related complications, hospitalizations for heart failure. Health economic analysis will be performed by means of cost-minimization analysis Sample size calculation / data analysis: The registry includes 1500 patients with a mean follow-up of two years. Based on the latest RCT and literature, it is expected that 150 (10%) patients experience at least one appropriate shock. We estimate that 150 patients with appropriate shocks (in combination with 150 fatalities) are sufficient to develop effective prediction rules. We will use binomial and multinomial logistic regression modelling to develop rules that predict the patient’s outcome status in terms of the primary endpoint as defined above. The prediction factors include parameters from the patient’s history, the ECG, functional status, LV-ejection fraction, renal function, etc. Cost-effectiveness analysis / budget impact analysis: cost minimization analysis will be performed. Scenario-analyses will be performed based on alternative cut-off values of the multivariate prediction rules of (i) death within years of the ICD implantation despite ICD implantation and (ii) survival for a long period without receiving appropriate ICD shocks. Time schedule: Entry into registry is expected to start by January 2014 and enrollment is completed by July 2016. All patients entered into the registry are followed until 18 months after inclusion of the last patient. Early termination of follow-up only occurs in case a patient explicitly demands that follow-up be discontinued. The average duration of follow-up is two years.