Left Atrial Function with MRI Enables Prediction of Cardiovascular Events after Myocardial Infarction: Insights from the AIDA STEMI and TATORT NSTEMI Trials.
Andreas SchusterSören J BackhausThomas StiermaierJenny-Lou NavarraJohannes UhligKarl-Philipp RommelAlexander KoschalkaJohannes T KowallickJoachim LotzMatthias GutberletBoris BigalkeShelby KuttyGerd HasenfussHolger ThieleIngo EitelPublished in: Radiology (2019)
Background The role of left atrial (LA) performance in acute myocardial infarction (AMI) remains controversial. Cardiac MRI myocardial feature tracking (hereafter, MRI-FT) is a method used to quantify myocardial function that enables reliable assessment of atrial function. Purpose To assess the relationship between LA function assessed with MRI-FT and major adverse cardiovascular events (MACE) after AMI. Materials and Methods This secondary analysis of two prospective multicenter cardiac MRI studies (AIDA STEMI [NCT00712101] and TATORT NSTEMI [NCT01612312]) included 1235 study participants with ST-elevation myocardial infarction (n = 795) or non-ST-elevation myocardial infarction (n = 440) between July 2008 and June 2013. All study participants underwent primary percutaneous coronary intervention. MRI-FT analyses were performed in a core laboratory by researchers blinded to clinical status to determine LA performance using LA reservoir function peak systolic strain (εs), LA conduit strain (εe), and LA booster pump function active strain (εa). The relationship of LA performance to a MACE within 12 months after AMI was evaluated by using Cox proportional hazards models and area under the receiver operating characteristic curve (AUC). Results Study participants with MACE had worse LA performance parameters compared with study participants without MACE (εs = 21.2% vs 16.2%, εe = 8.8% vs 6.9%, εa = 11.8% vs 10%; P < .001 for all). All atrial parameters were strongly associated with MACE (hazard ratio [HR], εs = 0.9, εe = 0.88, εa = 0.89; P < .001 for all). For εs, a cutoff of 18.8% was identified as the only independent atrial parameter with which to predict MACE after accounting for confounders and established prognostic markers in adjusted analysis (HR, 0.95; P = .02). The εs yielded incremental prognostic value above left ventricular ejection fraction, global longitudinal strain, microvascular obstruction, and infarct size (AUC comparisons, P < .04 for all). Conclusion Feature tracking of cardiac MRI to derive left atrial peak reservoir strain provided incremental prognostic value for major adverse cardiovascular events prediction versus established cardiac risk factors after acute myocardial infarction. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Almeida in this issue.
Keyphrases
- left ventricular
- left atrial
- acute myocardial infarction
- percutaneous coronary intervention
- cardiovascular events
- st elevation myocardial infarction
- coronary artery disease
- atrial fibrillation
- mitral valve
- magnetic resonance imaging
- hypertrophic cardiomyopathy
- contrast enhanced
- aortic stenosis
- st segment elevation myocardial infarction
- cardiac resynchronization therapy
- heart failure
- acute coronary syndrome
- antiplatelet therapy
- cardiovascular disease
- diffusion weighted imaging
- ejection fraction
- catheter ablation
- risk factors
- machine learning
- blood pressure
- coronary artery bypass
- coronary artery bypass grafting
- type diabetes
- deep learning