Ventricular arrhythmia, heart failure, and death: a triangular relationship in the WICD-MI study

Background/Introduction

In patients with left ventricular (LV) dysfunction, particularly of ischaemic origin, ventricular arrhythmia (VA) and heart failure (HF) frequently coexist but are usually managed separately. Their shared pathophysiological substrate suggests potential bidirectional interactions, yet the nature and magnitude of these associations remain insufficiently defined.

Purpose

To investigate the reciprocal relationship between VA and HF events in patients with persistent LV dysfunction after myocardial infarction (MI), and to assess the relative impact of each event on all-cause mortality.

Methods

We analysed 1,014 patients enrolled in the WICD-MI study across 41 French centres. All received a wearable cardioverter-defibrillator (WCD) after acute MI and subsequently underwent implantable cardioverter-defibrillator (ICD) implantation for primary or secondary prevention with persistent LVEF ≤35%. Follow-up began at the time of ICD implantation. Associations between ICD-treated VA and HF events (HF hospitalisation, heart transplantation, or LV assist device implantation) and their impact on all-cause mortality were assessed using time-dependent Cox models.

Results

The cohort was predominantly male (83.1%), with a mean age of 60.4 ± 10.4 years; 74.8% presented with ST-segment elevation MI. Median WCD wear time was 85 days (IQR 51–120), and mean LVEF at ICD implantation was 29.2 ± 6.2%. Over a median follow-up of 2.6 years post-ICD implantation, VA without HF occurred in 6.7% of patients, HF without VA in 13.7%, and both in 3.0%. Overall mortality was 9.7%, rising to 29.6% among those with an HF event. VA was independently associated with an increased risk of HF events (aHR = 2.97 [1.77–4.99], p < 0.0001), while HF events similarly predicted subsequent VA (aHR = 2.28 [1.28–4.07], p = 0.006) (Table). Both were independently associated with all-cause mortality, with a numerically stronger effect for HF events (aHR = 7.68 [4.91–12.03], p < 0.0001) than for VA (aHR = 4.11 [2.38–7.11], p < 0.0001). No significant interaction was observed between VA and HF events with respect to mortality (aHR = 1.00 [0.99–1.02], p = 0.58). Additional analyses showed similar associations when focusing on HF hospitalisation and ICD shocks.

Conclusions

In ICD-implanted patients with persistent LV dysfunction after MI, VA and HF exhibit a reciprocal relationship and independently predict mortality, with HF exerting the numerically stronger effect. These findings highlight the need for integrated, multidisciplinary management strategies in this high-risk population.

For image description, please refer to the figure legend and surrounding text. Graphical abstract For image description, please refer to the figure legend and surrounding text.