Prediction of sudden cardiac death in Japanese heart failure patients: international validation of the Seattle Proportional Risk Model

Abstract

AIMS: Heart failure (HF) is associated with an increased risk of sudden cardiac death (SCD). This study sought to demonstrate the incidence of SCD within a multicentre Japanese registry of HF patients hospitalized for acute decompensation, and externally validate the Seattle Proportional Risk Model (SPRM). METHODS AND RESULTS: We consecutively registered 2240 acute HF patients from academic institutions in Tokyo, Japan. The discrimination and calibration of the SPRM were assessed by the c-statistic, Hosmer-Lemeshow statistic, and visual plotting among non-survivors. Patient-level SPRM predictions and implantable cardioverter-defibrillator (ICD) benefit [ICD estimated hazard ratio (HR), derived from the Cox proportional hazards model in the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT)] was calculated. During the 2-year follow-up, 356 deaths (15.9%) occurred, which included 76 adjudicated SCDs (3.4%) and 280 non-SCDs (12.5%). The SPRM showed acceptable discrimination [c-index = 0.63; 95% confidence interval (CI) 0.56-0.70], similar to that of original SPRM-derivation cohort. The calibration plot showed reasonable conformance. Among HF patients with reduced ejection fraction (EF; < 40%), SPRM showed improved discrimination compared with the ICD eligibility criteria (e.g. New York Heart Association functional Class II-III with EF ≤ 35%): c-index = 0.53 (95% CI 0.42-0.63) vs. 0.65 (95% CI 0.55-0.75) for SPRM. Finally, in the subgroup of 246 patients with both EF ≤ 35% and SPRM-predicted risk of ≥ 42.0% (SCD-HeFT defined ICD benefit threshold), mean ICD estimated HR was 0.70 (30% reduction of all-cause mortality by ICD). CONCLUSION: The cumulative incidence of SCD was 3.4% in Japanese HF registry. The SPRM performed reasonably well in Japanese patients and may aid in improving SCD prediction.