Immediate, early and late ventricular fibrillation complicating acute myocardial infarction: insights from a real life registry

Abstract Background Conflicting data exist on the prognosis of ventricular fibrillation (VF) based on its timing in relation to coronary occlusion in acute myocardial infarction (AMI). Purpose To describe the patient profile and prognostic impact of VF during AMI according to its time of occurrence. Methods Data from three nationwide French cohorts carried out 5 years apart from 2005 to 2015 were analysed (FAST-MI program). All consecutive patients with AMI (STEMI and Non-STEMI) over a 1-month period were enrolled and followed 5 years. The present analysis focused on patients who presented with VF according to its time of occurrence defined as: immediate (out-of-hospital cardiac arrest), early (<48h at admission) and late (≥48h after admission). Results Out of 13 125 patients admitted for AMI (66±14 years, 71% men, 52% ST-elevation myocardial infarction-STEMI), 406 (3.1%) presented with VF including 205 immediate (50.5%), 147 early (36.2%) and 54 late (13.3%) VF episodes. Immediate and early VF were most frequent during STEMI (respectively 2.3% and 1.8% vs. 0.8% and 0.4% in non-STEMI, p<0.001), while late VF incidence was as frequent in ST-elevation and non-STEMI (0.5% vs 0.4%, p value). The proportion of immediate, early and late VF in AMI were similar in 2005, 2010 and 2015. Patients with immediate (61.6±13.2 yo) or early (62.4±14.2 yo) VF were younger than those with no VF (66.1±14.1 yo, p<0.001) while patients with late VF patients were the oldest (70.5±13.6 yo). LVEF was lower in patients with early (48±10%) or late (48±13%), compared with patients without VF (52±11%) or those with immediate VF (51±12%) (p<0.001). In-hospital mortality was particularly high in patients who developed immediate (24.9%) and late (37.0%) VF, intermediate in early VF patients (10.9%) and lowest in non-VF patients (3.0%) (p<0.001). Among patients alived at hospital discharge, and compared with patients with no VF (4.2%), the six-month mortality was considerably increased in the late VF subgroup (17.6%, adjusted HR: 6.04 [2.69-13.58], P<0.001), moderately increased in immediate VF (5.2%, HR:2.15 [1.06-4.34, P=0.03) ) and did not differ in early VF (1.5%, HR: 0.52 [0.13-2.11 P=0.36). At five years, mortality was also significantly higher in patients with late VF (38.2%, adjusted HR 1.72 [1.01-2.91]), with no significant difference between immediate, early, and non VF patients (14.2%, 11.5% and 19.3% respectively). Conclusions Patients with immediate, early and late VF during AMI do present with significantly different profiles. Still, only late VF (≥48h after admission) was associated with poorer outcome after the index hospitalisation, especially within the initial 6 months. By contrast with current guidelines based on relatively old data from the 2000s in secondary prevention in ischemic or non-ischemic settings, the extent to which patients with late VF would benefit from rapid cardioverter defibrillator implantation is questionable.3 months survival after VF in AMI5 year survival after VF in AMI