Vaduganathan M, et al. - The rates and predictors of sudden death (SD) or aborted cardiac arrest (ACA) in HFpEF were investigated. Around 20% of deaths in HFpEF are related to SD. With modest discrimination, male sex and insulin-treated diabetes mellitus were associated with greater risk for SD/ACA.

Methods

• From the Americas region of the TOPCAT (Aldosterone Antagonist Therapy for Adults With Heart Failure and Preserved Systolic Function) trial, 1,767 patients with HFpEF (EF ≥ 45%) were studied.
• Using stepwise backward selection with competing risks regression analysis, independent predictors of composite SD/ACA were found that accounted for nonsudden causes of death.

Results

• SD/ACA was reported for 77 patients, and non-SD/ACA was reported for 312 patients during a median 3.0-year (25th to 75th percentile: 1.9 to 4.4 years) follow-up; corresponding incidence rates were 1.4 events/100 patient-years (25th to 75th percentile: 1.1 to 1.8 events/100 patient-years) and 5.8 events/100 patient-years (25th to 75th percentile: 5.1 to 6.4 events/100 patient-years).
• Patients randomized to spironolactone had numerically lower but not statistically reduced SD/ACA: 1.2 events/100 patient-years (25th to 75th percentile: 0.9 to 1.7 events/100 patient-years) vs 1.6 events/100 patient-years (25th to 75th percentile: 1.2 to 2.2 events/100 patient-years); the subdistributional hazard ratio was 0.74 (95% confidence interval: 0.47 to 1.16; p = 0.19).
• They identified male sex and insulin-treated diabetes mellitus as independent predictors of composite SD/ACA after accounting for competing risks of non-SD/ACA (C-statistic = 0.65).
• SD/ACA had no independent association with covariates, including eligibility criteria, age, ejection fraction, coronary artery disease, left bundle branch block, and baseline therapies.
• In sensitivity analyses, sex and diabetes mellitus status continued to be independent predictors, when patients with implantable cardioverter-defibrillators were excluded and when predicting SD alone.