Abstract 2257: The Relationship Between Mortality, Restrictive Mitral Filling Pattern, Ejection Fraction and End-Systolic Volume Index Post Acute Myocardial Infarction: Results from an Individual Patient Meta-Analysis (MeRGE)
The adverse remodelling that occurs post acute myocardial infarction (AMI), including increased LV end-systolic volume (ESV) and reduced ejection fraction (EF), are important determinants of survival. The restrictive filling pattern (RFP) (a diastolic filling pattern associated with high LV filling pressure) also predicts outcome post AMI. However, the relative contribution of RFP, ESV index (ESVi)and EF in predicting survival remains uncertain. The aim of this analysis was to investigate this relationship.
Methods: MeRGE is a global collaboration of 12 prospective echocardiographic outcome studies post AMI, demonstrated an increase in the odds of mortality in the presence of RFP (high E:A ratio, deceleration time < 140 ms). This sub-analysis includes patients in whom LV volumes and EF measurements were also available (7 studies). All of the patient data were merged into a single database. Cox proportional hazards model (including EF, ESVi (ESV/body surface area), age and RFP) was used to determine the independent predictors of outcome. Two models were used: one with EF < or > 35% and the other included EF as a continuous variable. Multivariate models are presented as hazard ratios (HR).
Results: 1482 patients were included of whom 149 died. In univariate analysis deceleration time, RFP, EF and ESVi were all independently associated with outcome. However, in the multivariate model only RFP (HR 1.84 (95%CI:1.3,2.6)), EF (1%) (HR 0.96 (95%CI:0.94,0.98)) and age (1 year) (HR 1.07 (95%CI:1.06,1.09)) were significant independent predictors. When EF < 35% was included in the model the results were similar: RFP (HR 2.05 (95%CI:1.46,2.9)), EF<35% (HR 2.05 (95%CI:1.37,3.07)) and age (HR 1.07 (95%CI:1.06,1.09)) contributed significantly to the model. EF and ESVi are strongly correlated and when EF was excluded from the model ESVi was a significant predictor of outcome.
Conclusion: RFP remained a significant predictor of outcome in this cohort of patients post AMI alongside age and EF. ESVi did not contribute further to the model. Thus, the assessment of EF and the restrictive filling pattern should be an important goal of echocardiography in post AMI patients.