Whether the risk factors for cardiovascular (CV) outcomes are different in primary versus secondary prevention implantable cardioverter-defibrillator (ICD) patients is unclear. We sought to identify predictors of CV outcomes in ICD recipients for primary (G1) versus secondary prevention (G2). Consecutive patients who had ICD implanted during August 2005 to December 2009 were included. The primary outcome was a composite of appropriate shock, acute coronary syndrome, ischemic stroke, coronary revascularization, heart failure exacerbation, CV hospitalization, or all-cause death. We used Cox proportional hazards model and a stepwise selection method to fit the most parsimonious model to predict the primary outcome in all patients and separately for G1 and G2 patients. We followed 223 (184 G1 and 39 G2, mean age 61 years) patients through December 31, 2012; 141 (63.2%) developed the primary outcome. In all patients, atrial fibrillation (AF; hazard ratio 6.72, 95% CI 4.20 to 10.75; p <0.001), use of antiarrhythmic drug (1.55, 1.02 to 2.36; p = 0.04), and lower estimated glomerular filtration rate (0.99, 0.98 to 0.997; p = 0.01) were associated with increased risk of the primary outcome; the attributable risks were 21.6%, 16.0%, and 15.9%, respectively. In G1, AF, hypertension, and lower estimated glomerular filtration rate were associated with increased risk, whereas in G2, AF, use of antiarrhythmic drug, and nonischemic cardiomyopathy were associated with increased risk. In conclusion, although risk factors are different for primary and secondary prevention patients, AF is a strong and consistent risk factor for adverse outcomes in both populations.
The benefit of implantable cardioverter-defibrillators (ICDs) for both primary and secondary sudden cardiac death (SCD) prevention has been well demonstrated in numerous clinical trials, and as a result, the use of ICD has become prevalent in clinical practice. However, the magnitude of ICD survival benefit in real-world settings may be altered by multiple co-morbidities prevalent in the ICD population. In this regard, for example, chronic kidney disease and atrial fibrillation (AF) have been found to attenuate the benefits from ICD. Several studies have reported the co-morbidity predictors for adverse cardiovascular (CV) outcomes after ICD implantation, but whether the risk factors are different in primary versus secondary prevention patients is not entirely clear. Knowing the differences in predictors may help the decision-making process before ICD implantation and facilitate prevention strategies targeting these risk factors after implantation. In this retrospective hospital-based cohort study, we aimed to identify and compare the risk factors for adverse CV outcomes in patients who received ICD therapy for primary versus secondary prevention indication.
Methods
This retrospective study was conducted at the University of Minnesota Medical Center in Minneapolis, Minnesota. From August 2005 to December 2009, consecutive patients who underwent an ICD insertion procedure at the University of Minnesota Medical Center were included. The follow-up period was from the date of implantation through December 31, 2012. The Institutional Review Board at the University of Minnesota approved the research protocol.
Data on baseline clinical characteristics, CV risk factors, and CV outcomes during follow-up were manually abstracted from the electronic medical records, including cardiology clinic notes, pacemaker interrogation notes, history and physical notes on admission, and discharge summaries. Information considered at baseline included age, gender, type of defibrillator (ICD or biventricular ICD), indication (primary or secondary prevention), a history of hypertension, hyperlipidemia, coronary artery disease, baseline AF, congestive heart failure (HF), valvular heart disease, ischemic stroke or transient ischemic attack (TIA), peripheral artery disease, diabetes mellitus, estimated glomerular filtration rate (eGFR), end-stage renal disease requiring chronic dialysis, cigarette use, family history of early acute coronary syndrome or sudden cardiac death, percutaneous coronary intervention, coronary artery bypass graft surgery, left ventricular ejection fraction (EF), baseline use of β blocker, angiotensin-converting enzyme inhibitor, angiotensin II receptor blocker, antiarrhythmic drug, aspirin, anticoagulant, and statin. Incident AF during follow-up was ascertained by electrophysiology clinic notes and electrocardiogram and was confirmed by ICD interrogation whenever possible.
Primary and secondary prevention for SCD used in this study was defined in accordance with the published guidelines for device-based therapy and randomized trials on primary or secondary prevention. Patients who received an ICD placement for secondary prevention had a near-fatal ventricular fibrillation (VF), syncope from sustained ventricular tachycardia (VT), sustained, symptomatic VT with impaired ventricular function, or syncope with inducible sustained VT in an electrophysiology study. Patients who received an ICD placement with primary prevention indication for SCD were at risk for SCD but had not developed an episode of cardiac arrest, sustained VT, or VF.
The primary outcome was a composite of appropriate shock, acute coronary syndrome (unstable angina, non–ST-segment elevation myocardial infarction, or ST-segment elevation myocardial infarction), ischemic stroke or TIA, coronary revascularization (coronary artery bypass graft surgery or percutaneous coronary intervention), HF exacerbation, CV hospitalization, or death from all causes. Appropriate shock was defined as a shock for either VT or VF.
Person-years at risk were calculated from the date of ICD implantation until the date of primary outcome, loss to follow-up, or end of follow-up (December 31, 2012), whichever occurred first. For AF, baseline or incident AF was considered as one variable for analysis. We used multivariable Cox proportional hazards regression to construct multivariable models in relation to the primary outcome in all patients and separately for primary and secondary prevention patients. Next, using a backward stepwise approach, we constructed parsimonious models in relation to the primary outcome. The parsimonious model is the simplest model with the fewest variables that are all significantly associated with the composite outcome at the 5% level. This model is obtained by removing variables in an iterative manner from the initial full model that are not associated with the composite outcome. The parsimonious model may be clinically useful as a risk prediction model for the composite outcome. In all patients, we constructed full and parsimonious models for each individual component of the composite primary outcome. In addition to hazard ratios (HRs), we computed attributable risks for factors that were significantly associated with the primary outcome in all patients. Finally, we excluded hospitalization from the composite primary outcome to determine if the significant associations were primarily driven by hospitalization. Statistical analysis was performed using SAS, version 9.2 (SAS Institute Inc., Cary, North Carolina). All p values reported were 2-sided, and statistical significance was evaluated at the 5% level.
Results
From August 2005 to December 2009, 310 consecutive patients underwent an ICD insertion or generator replacement procedure. Eighty-seven patients undergoing only generator replacement were excluded, leaving 223 patients for analyses. Baseline characteristics of the study subjects, including the primary (n = 184) and secondary (n = 39) prevention groups, are summarized in Table 1 . Compared with the secondary prevention group, the proportion of patients using β blocker and biventricular ICD was higher in the primary prevention group. The average left ventricular EF was higher in the secondary group than in the primary prevention group. There was no statistical difference in the proportion of patients with AF between the 2 groups.
