Frequency and Determinants of Implantable Cardioverter Defibrillator Deployment Among Primary Prevention Candidates With Subsequent Sudden Cardiac Arrest in the Community

BACKGROUND: The prevalence rates and influencing factors for deployment of primary prevention implantable cardioverter defibrillators (ICDs) among subjects who eventually experience sudden cardiac arrest in the general population have not been evaluated. METHODS AND RESULTS: Cases of adult sudden cardiac arrest with echocardiographic evaluation before the event were identified from the ongoing Oregon Sudden Unexpected Death Study (population approximately 1 million). Eligibility for primary ICD implantation was determined from medical records based on established guidelines. The frequency of prior primary ICD implantation in eligible subjects was evaluated, and ICD nonrecipients were characterized. Of 2093 cases (2003-2012), 448 had appropriate pre- sudden cardiac arrest left ventricular ejection fraction information available. Of these, 92 (20.5%) were eligible for primary ICD implantation, 304 (67.9%) were ineligible because of left ventricular ejection fraction >35%, and the remainder (52, 11.6%) had left ventricular ejection fraction ≤35% but were ineligible on the basis of clinical guideline criteria. Among eligible subjects, only 12 (13.0%; 95% confidence interval, 6.1%-19.9%) received a primary ICD. Compared with recipients, primary ICD nonrecipients were older (age at ejection fraction assessment, 67.1±13.6 versus 58.5±14.8 years, P=0.05), with 20% aged ≥80 years (versus 0% among recipients, P=0.11). Additionally, a subgroup (26%) had either a clinical history of dementia or were undergoing chronic dialysis. CONCLUSIONS: Only one fifth of the sudden cardiac arrest cases in the community were eligible for a primary prevention ICD before the event, but among these, a small proportion (13%) were actually implanted. Although older age and comorbidity may explain nondeployment in a subgroup of these cases, other determinants such as socioeconomic factors, health insurance, patient preference, and clinical practice patterns warrant further detailed investigation.