Resting perfusion MPI-SPECT combined with cardiac 123I-mIBG sympathetic innervation imaging improves prediction of arrhythmic events in non-ischemic cardiomyopathy patients: Sub-study from the ADMIRE-HF trial

Abstract

BackgroundDelayed Iodine-123 meta-iodobenzylguanidine heart/mediastinum (H/M) uptake ratio predicted arrhythmic events in patients with heart failure (HF) and significant left ventricular dysfunction in ADMIRE-HF. We tested the hypothesis that resting perfusion defects on MPI-SPECT, representing scar, would further risk stratify patients beyond H/M ratio in the prediction of ventricular arrhythmic events in both ischemic (ICM) and non-ischemic cardiomyopathy (NICM) patients. MethodsPatients from the ADMIRE-HF database were classified as ICM and NICM and were stratified by delayed H/M ratio (<1.6/≥1.6) and by summed rest score (SRS) (≤8/>8) on MPI-SPECT. The entire cohort was also classified as high risk (H/M <1.6, SRS >8) and low risk (H/M ≥1.6, SRS ≤8). Scores were from visual interpretation of individual and derived consensus (average) reads per ASNC guidelines. ResultsThere were 612 (66%) ICM and 317 (34%) NICM patients. ICM patients had higher mean SRS than NICM patients (25 ± 13 vs 12 ± 10). Sixty-three arrhythmic events occurred over a median follow-up of 17 months. SRS had incremental predictive value among NICM patients with low H/M ratio (<1.6). There was no risk stratification in patients with ICM. Multivariable analysis for NICM with H/M ratio <1.6 demonstrated SRS score >8 as the only independent predictor of arrhythmic events (hazard ratio 3.3, 95% CI 1.1-9.8, P = .03). Patients classified in high-risk subgroup had statistically significant increased risk of arrhythmic events (hazard ratio 2.080, 95% CI 1.112-3.894, P = .022). There was only one event in low-risk subgroup with a trend towards lower risk of arrhythmic events (P = .07). ConclusionResting perfusion defects provide independent risk stratification in addition to Iodine-123 meta-iodobenzylguanidine sympathetic innervation imaging in predicting arrhythmic events in patients with NICM and HF. High and potentially low-risk groups for arrhythmic events can be identified based on the severity of resting perfusion defect score and H/M ratio.