P2876Repeating non-invasive risk stratification tests improves the prediction of outcomes of ICD patients in the EUTrigTreat study

Abstract

Abstract Background Non-invasive risk stratification of SCD aims to predict the risk by assessing measures of substrate (LVEF), of triggers (PVCs; T-wave alternans, TWA) and of autonomic function (heart rate turbulence, HRT). However, the value of repeating these tests during follow-up is unclear. Purpose To study the predictive value of repeated non-invasive risk assessment. Methods EUTrigTreat is a prospective trial aimed to improve non-invasive risk stratification in ICD patients. The study protocol included non-invasive testing at baseline and a repeat after 6 to 12 months. The population included ischemic and non-ischemic cardiomyopathies and arrhythmogenic heart disease. Test results were categorized as pathologic (1) versus non-pathologic (0) for LVEF ≤40%, PVCs >400 in 24h, abnormal exercise TWA (Cambridge Heart) and abnormal HRT (TO >0.1% and/or TS ≤2.0ms/RRI). Time dependent Cox regression modelling was performed for mortality, and a Fine-and- Gray competing risk analysis for shocks, including adjustment for independent predictors in the overall study population (mortality: age, LVEF, history of AF, NT-proBNP, NYHA class, eGFR; shocks: LVEF, secondary prevention). Results A total of 635 patients were included with a follow-up of 4.3±1.5 years, 96 (15%) received an ICD shock and 108 (17%) died. The table shows the results at baseline and with repeating the tests after 8±1 months. Worsening of LVEF compared to a stable LVEF >40% and persistent abnormal HRT were independent predictors of mortality. Improvement in HRT was associated with a lower mortality. Worsened results upon TWA testing was associated with a 3 times higher risk of shocks. A persistent low LVEF was an independent predictor of both mortality and ICD shocks. Baseline 1 (vs. 0) Worsening 0–1 (vs. 0–0) Improvement 1–0 (vs. 1–1) Stable 1–1 (vs. 0–0) HR (CI) HR (CI) HR (CI) HR (CI) Mortality LVEF (n=315) 1.85 (1.06–3.24) 3.47 (1.13–10.68) 0.88 (0.40–1.94) 2.22 (1.19–4.15) TWA (n=204) 0.62 (0.28–1.37) 0.80 (0.25–2.59) 0.67 (0.18–2.47) 0.59 (0.20–1.77) PVC (n=329) 1.26 (0.73–2.16) 0.99 (0.36–2.72) 0.63 (0.26–1.52) 1.38 (0.75–2.55) HRT (n=163) 2.57 (0.85–7.77) 4.01 (0.39–41.17) 0.10 (0.01–0.81) 8.71 (1.11–68.24) Shocks LVEF (n=338) 1.73 (0.93–3.22) 0.92 (0.13–6.67) 0.26 (0.06–1.08) 2.02 (1.09–3.76) TWA (n=256) 0.82 (0.39–1.70) 2.91 (1.04–8.13) 1.31 (0.48–3.59) 1.54 (0.54–4.43) PVC (n=366) 1.28 (0.70–2.34) 1.48 (0.54–4.09) 0.54 (0.16–1.81) 1.70 (0.86–3.35) HRT (n=188) 1.12 (0.50–2.50) 0.57 (0.13–2.52) 0.73 (0.20–2.64) 1.06 (0.43–2.61) Conclusion Repeating LVEF, TWA and HRT have the potential to improve risk stratification for mortality and shocks in ICD patients. Acknowledgement/Funding This research has received funding from European Community's Seventh Framework Program FP7: EUTrigTreat (grant agreement no. HEALTH-F2-2009-241526).