Outcomes Following Implantable Cardioverter-Defibrillator Insertion in Patients 80 Years of Age or Older: A Statewide Population Cohort Study

Background: Patients >= 80 years of age are underrepresented in major implantable cardioverter-defibrillator (ICD) trials, and real -world data are lacking. In this study, we sought to assess ICD utilisation, outcomes, and their predictors, in an unselected statewide population including patients >= 80 years old. Methods: We extracted details of ICDs implanted from 2009 to 2018 in New South Wales (NSW), Australia from the Centre for Health Record Linkage administrative data sets. Analysis was stratified into age groups of < 60 years, 60-79 years, and >= 80 years. Results: A total of 9304 patients (mean age 66.1 +/- 13.1 years; 12.1% >= 80 years) had de novo ICD placement at an average rate of 1163 +/- 122 patients per annum, with more implants in men in all age groups. After adjusting for NSW population size by sex, age group, and calendar year, mean implantation rates were 5.5 +/- 0.6, 63.2 +/- 8.6, and 52.7 +/- 10.8 per 100,000 persons per annum in patients aged < 60 years, 60-79 years, and >= 80 years, respectively. In -hospital mortality was 0.4% and did not differ among age groups. However, 1 -year mortality was 2.1%, 5.9%, and 10.7%, in those < 60 years, 60-79 years, and >= 80 years of age, respectively (P < 0.001), with hazard ratios for those aged >= 80 years of 4.3 (95% confidence interval [CI] 3.1-6.0) and those aged 60-79 years of 2.6 (95% CI 1.9-3.5) relative to those aged < 60 years (both P < 0.001) after adjusting for ICD indications, sex, implantation year, referral source, and comorbidities. In those aged >= 80 years, age > 83 years, congestive cardiac failure, chronic renal failure, neurodegenerative disease, and a higher Charlson comorbidity index score were each independent predictors of 1year mortality. Conclusions: ICD use in patients aged >= 80 years and 60-79 years was 10 -fold that in patients aged < 60 years, and perioperative outcomes were good in all ages, but there was substantially increased 1year mortality in those aged >= 80 years. Careful selection based on age and comorbidity may further reduce 1 -year mortality in patients >= 80 years old receiving ICDs.