Indipendent predictors of akinetic segment recovery during dobutamine infusion assessed by cardiac magnetic resonance

Abstract Funding Acknowledgements Type of funding sources: None. Background Myocardial viability assessment can guide myocardial revascularization in patients with coronary artery disease. Non-invasive imaging modalities are routinely used for detection of viable myocardium and the highest diagnostic performance is provided by Cardiac Magnetic Resonance (CMR): it combines high sensibility (95%) by delayed enhancement (DE) assessment and high specificity (91%) by evaluation of myocardial systolic thickening recovery at increasing dose of dobutamine (Db-systolic recovery). Purpose Identification of independent predictors for Db-systolic recovery of akinetic segments supplied by a chronic total occlusion (CTO) coronary artery. The second purpose was to develop a diagnostic tool as alternative to dobutamine stress imaging to provide an estimation of myocardial response to inotropic agents. Methods This single center prospective observational study enrolled 47 consecutive patients referred to CMR for evaluation of inducible ischemia and myocardial viability in the territory of one or more CTOs. All enrolled patients presented basal akinetic segments in the CTO territory. Cine-Steady State Free Precession (SSFP) sequences were acquired both in basal and at increasing dose of dobutamine to evaluate Db-systolic recovery. Stress perfusion imaging and DE were then acquired to detect inducible ischemia and necrotic areas. Post-processing CMR-Feature Tracking strain analysis provided both radial and circumferential strain on basal, mid and apical short axis planes. Results Table 1 lists clinical characteristics of the study population. A total of 189 akinetic segments supplied by a CTO were detected and divided into viable and non-viable groups according to their ability to recover or not at increasing dose of dobutamine, respectively. Parameters of each akinetic segment significantly (p value < 0.05) associated with viability at univariate analysis were included in the multivariable binary logistic regression to identify independent predictors of Db-systolic recovery (Table 2). These independent predictors have been then included in a logistic regression model as continuous variable except for Transmural Extent of DE which can be either 1 or 0 according to DE<75% or ≥75%, respectively: L= −5.459 + 0.033 * (Radial Peak Systolic Strain) + 0.067 * (Circumferential Endocardial Peak Systolic Strain) + 1.808 * (Transmural Extent of DE) + 0.358 * (Diastolic Wall Thickness). When L ≥ 0 the model predicts contractile recovery during dobutamine infusion with an accuracy of 83.1%. Conclusion This CMR prospective study identifies independent predictors of contractile recovery at increasing dose of dobutamine and provides a diagnostic tool which might substitute dobutamine assessment with a good accuracy, reducing acquisition time and adverse events in patients with CTOs. An external validation study is needed to assess the consistency of this model for clinical purposes.