Title : Risk prediction of major adverse cardiac events in post-infarction patients based on combination of autonomic reactivity and ejection fraction parameters
Abstract:
Background: A compromised cardiac autonomic reactivity with an impaired left ventricular ejection fraction (LVEF) leads to increased mortality and morbidity in post myocardial infarction (MI) patients. We examined the combination of autonomic profile parameters with LVEF for prediction of Major Adverse Cardiac Events (MACE) in post-MI patients.
Hypothesis: The combination of compromised autonomic reactivity and low LVEF predicts cardiovascular events in post-MI patients.
Methods: A retrospective cross-sectional observational study in 203 diagnosed post-MI patients was conducted with MACE as the primary endpoint, including all-cause-mortality, non-fatal myocardial infarction, non-fatal stroke, unstable angina, and late revascularization. The combination of autonomic reactivity (Ewing's battery of tests) and low (LVEF 50%) or preserved (LVEF>50%) EF as a predictor-model for MACE were assessed. Patients were classified into three categories based on autonomic reactivity: no autonomic neuropathy (CAN=0-1), early autonomic neuropathy (CAN= 2-3), and definite autonomic neuropathy (CAN=4-6) using the Cardiac Autonomic Neuropathy (CAN) score system. Patients with low CAN (both early and definite CAN) were further divided into two groups of low EF and preserved EF. Multivariable logistic regression and area under Receiver Operating Characteristic (ROC) curves (AUC) of low autonomic reactivity measures and LVEF were used to create the final MACE predictive model.
Results: 76 (37.4%) of 203 post-MI patients had MACE throughout the 9-year follow-up period (n=14, all-cause death; n=22, non-fatal MI; n=10, non-fatal stroke; n=18, unstable angina; and n=12, late revascularization). Compromised autonomic parameters were found in 122 (60.09%) of the 203 post-MI patients (n=76, CAN=2-3; n=46, CAN=4-6). Out of 122 post-MI patients with poor autonomic profile 51 (41.8%) had low EF (LVEF <50%) and 71 (58.18%) had preserved EF (LVEF >50%). Compromised autonomic parameters with low EF or preserved EF predictive models had AUCs of 0.8873 (95% CI 0.798-0.976), 83.87% positive predictive value (PPV) and 80.00% negative predictive value (NPV) and 0.790 (95% CI 0.687-0.893), PPV 68.89% and NPV 57.69%, respectively (p<0.0001).
Conclusion: MACE in post-MI patients is well predicted by a combination of poor autonomic reactivity measures and low ejection fraction. An integrated prediction model may be developed to prevent MACE in post-MI patients.
