Guidelines for implantation of primary prevention defibrillators: Who is listening?

Abstract

i r a p i m m m a b r w m S t c c Pivotal trials over the past decade have established that implantable cardioverter-defibrillators (ICDs) prevent sudden death and improve overall survival in high-risk groups, identified broadly on the basis of low left ventricular ejection fraction (EF) and heart failure. In 2005, the Center or Medicare & Medicaid Services extended the US national overage determination for ICD implantation to include rimary prevention indications based on the results of these ivotal trials. The Center for Medicare & Medicaid Services estricted coverage for patients with recent myocardial inarctions and newly diagnosed heart failure, since these ere exclusions from the pivotal trials and other studies ailed to demonstrate benefit in these situations. Guidelines rom ACC/AHA/HRS identify specific class I indications or ICD implantation in patients with left ventricular dysunction (ejection fraction 35% and class II/III heart failre; ejection fraction 30% and class I heart failure in schemic cardiomyopathy). Why then are primary prevenion ICDs implanted in only a minority of eligible recipints? Barriers to ICD implantation occur at several levels. The treating physician may not understand the indications for ICD implantation or may believe that implantation is not warranted. The patient may decline to have the implantation, particularly if he or she cannot expect symptomatic improvement from a purely prophylactic device. In some localities, access to an implanting physician may be difficult or cumbersome. Highly publicized generator and lead recalls have probably eroded the enthusiasm of some physicians and patients for prophylactic ICDs. It also has been shown that wide variations exist in ICD utilization, depending on geographical region, practice setting, availability of electrophysiologists, race, and insurance status. Just as puzzling as the failures to refer high-risk patients are the implantations that clearly fall outside of current guidelines. This issue was highlighted by an analysis of the US ICD Registry that found that 23% of the ICD implantations were “non–evidence-based.” Arguably, many of