Non–Evidence-Based ICD Implantations in the United States
How often are implantable cardioverter-defibrillator (ICD) implantations for primary prevention of sudden cardiac death (SCD) not in accord with practice guidelines?
This was a retrospective analysis of 111,707 patients (mean age 66 years) enrolled in the National Cardiovascular ICD Data Registry in 2006-2009. All patients had either prior myocardial infarction (MI) and an ejection fraction (EF) ≤30% or congestive heart failure and an EF ≤35%. The ICD implant was considered non–evidence-based if one of the following criterion was present: MI within 40 days of the implant, coronary bypass graft surgery within 3 months of the implant, New York Heart Association class IV heart failure symptoms, or newly diagnosed heart failure.
ICD implantations were not in accord with practice guidelines in 22.5% of patients, most commonly because of newly diagnosed heart failure (62%) or an MI within 40 days (37%). The risks of in-hospital death and complications were significantly lower when the ICD implant was consistent with practice guidelines (0.18% and 2.4%, respectively) than when it was not (0.57% and 3.2%, respectively). The rate of non–evidence-based ICD implants was significantly higher among non-electrophysiologists (25%) than among electrophysiologists (21%).
Approximately one-fourth of ICD implants for primary prevention of SCD in the United States are not in accord with practice guidelines.
The criteria that were used to identify non–evidence-based ICD implants have been shown to not be associated with improved outcomes with ICD therapy. Aside from cost considerations, the higher complication rate among a patient whose ICD implant was not evidence-based provides a compelling reason to comply with practice guidelines when selecting patients for an ICD for primary prevention of SCD.
Keywords: Registries, Myocardial Infarction, Heart Failure, New York, Death, Sudden, Cardiac, Defibrillators, Implantable, Primary Prevention, United States
< Back to Listings