Is inappropriate implantable defibrillator shock therapy predictable?

OBJECTIVE: To identify implantable cardioverter defibrillator (ICD) patients who are at risk of receiving inappropriate shock.
BACKGROUND: Inappropriate ICD shock, usually from atrial fibrillation (AF) or sinus tachycardia (ST), is a common problem. We hypothesized that clinical variables would predict which patients with single chamber ICDs would be more likely to receive inappropriate therapy and be candidates for more accurate discriminators such as those available in dual chamber ICDs.
METHODS: The ICD registry at St. Michael's Hospital has it's clinical information and demographic data updated after each clinic visit. Inappropriate shock was considered as the outcome variable. Possible predictors considered were age, gender, ejection fraction, NYHA class, prior CABG and prior history of AF. Univariate predictors were identified using t-test for continuous variables and Chi-square test for categorical variables. Multivariate predictors were identified using stepwise logistic regression analyses.
RESULTS: Of 299 patients, 261 had complete data for analysis. In this population 78% were male, mean age was 60 +/- 13 years, mean ejection fraction was 37 +/- 15% and mean follow up was 53 +/- 36 months. One hundred and sixteen of the 261 patients (44%) received one or more inappropriate therapies (73% within 2 years of receiving their device), and 140 (51%) received one or more appropriate therapies. Significant predictors of inappropriate therapy by multivariate model were prior AF (OR 2.6, 95% CI 1.5-4.5) and NYHA class 1 vs. classes 2-4 (OR 2.2, 95% CI 1.2-3.7).
CONCLUSION: Clinical characteristics of ICD patients can predict those at risk for inappropriate shock and should be considered for interventions to decrease such shocks.