Marat Fudim1, Mouhammad Fathallah2, Linda K Shaw2, Peter R Liu2, Olga James3, Zainab Samad4, Jonathan P Piccini5, Paul L Hess6, Salvador Borges-Neto3. 1. Department of Medicine and Division of Cardiology, Duke University, Durham, North Carolina; Duke Clinical Research Institute, Durham, North Carolina. Electronic address: marat.fudim@dm.duke.edu. 2. Department of Medicine and Division of Cardiology, Duke University, Durham, North Carolina. 3. Department of Radiology and Division of Nuclear Medicine, Duke University, Durham, North Carolina. 4. Department of Medicine and Division of Cardiology, Duke University, Durham, North Carolina. Electronic address: https://twitter.com/ZainabASamad. 5. Department of Medicine and Division of Cardiology, Duke University, Durham, North Carolina; Duke Clinical Research Institute, Durham, North Carolina. Electronic address: https://twitter.com/JonPicciniSr. 6. VA Eastern Colorado and Health Care System, Denver, Colorado.
Abstract
OBJECTIVES: The goal of this study was to examine whether diastolic dyssynchrony, measured by using gated single-photon emission computed tomography (GSPECT) myocardial perfusion imaging (MPI) in patients with coronary artery disease (CAD), is independently associated with adverse outcomes. BACKGROUND: Systolic left ventricular dyssynchrony is known to be associated with worse clinical outcome in patients with CAD. METHODS: This study included patients who presented to Duke University for GSPECT MPI between 2003 and 2009. Patients had at least 1 major epicardial obstruction ≥50%. Dyssynchrony was assessed by using Emory Cardiac Toolbox software and compared with a control population without CAD. Abnormal degree of diastolic/systolic dyssynchrony was defined as values above 2 SDs above mean of mechanical dyssynchrony parameters. Using Cox proportional hazards modeling, the adjusted association between dyssynchrony and outcomes, including all-cause and cardiovascular death, was assessed. RESULTS: Among 1,310 patients with a median age of 64 years (interquartile range: 55 to 72 years), 69.7% were male and 2.6% had left bundle branch block. Overall, 241 (18.4%) and 238 (18.2%) patients had significant systolic and diastolic mechanical dyssynchrony, respectively, and 211 (16.1%) had both. After a median follow-up of 7.1 years, 543 deaths occurred. At 5 years, the mortality estimate was 21.2% among patients with a normal degree of diastolic left ventricular mechanical dyssynchrony (LVMD) and 41.7% among those with an abnormal degree of LVMD (p < 0.001). When added to clinical comorbidities, electrical dyssynchrony, and systolic LVMD, diastolic dyssynchrony was incrementally associated with cardiovascular mortality (global chi-square statistic of 211.9 vs. 222.8; 2 degrees of freedom; p = 0.004). In a model that also includes left ventricular ejection fraction, the addition of diastolic dyssynchrony to systolic dyssynchrony maintained an incremental prognostic benefit (global chi-square statistic of 234.8 vs. 241.8; p = 0.030). Adjustment for baseline ischemia and scar burden did not change this relationship. CONCLUSIONS: Systolic and diastolic left ventricular dyssynchrony, as measured by using GSPECT MPI, were associated with adverse outcomes. Moreover, diastolic dyssynchrony appears to provide incremental predictive value to clinical history, electrical dyssynchrony, and left ventricular function.
OBJECTIVES: The goal of this study was to examine whether diastolic dyssynchrony, measured by using gated single-photon emission computed tomography (GSPECT) myocardial perfusion imaging (MPI) in patients with coronary artery disease (CAD), is independently associated with adverse outcomes. BACKGROUND: Systolic left ventricular dyssynchrony is known to be associated with worse clinical outcome in patients with CAD. METHODS: This study included patients who presented to Duke University for GSPECT MPI between 2003 and 2009. Patients had at least 1 major epicardial obstruction ≥50%. Dyssynchrony was assessed by using Emory Cardiac Toolbox software and compared with a control population without CAD. Abnormal degree of diastolic/systolic dyssynchrony was defined as values above 2 SDs above mean of mechanical dyssynchrony parameters. Using Cox proportional hazards modeling, the adjusted association between dyssynchrony and outcomes, including all-cause and cardiovascular death, was assessed. RESULTS: Among 1,310 patients with a median age of 64 years (interquartile range: 55 to 72 years), 69.7% were male and 2.6% had left bundle branch block. Overall, 241 (18.4%) and 238 (18.2%) patients had significant systolic and diastolic mechanical dyssynchrony, respectively, and 211 (16.1%) had both. After a median follow-up of 7.1 years, 543 deaths occurred. At 5 years, the mortality estimate was 21.2% among patients with a normal degree of diastolic left ventricular mechanical dyssynchrony (LVMD) and 41.7% among those with an abnormal degree of LVMD (p < 0.001). When added to clinical comorbidities, electrical dyssynchrony, and systolic LVMD, diastolic dyssynchrony was incrementally associated with cardiovascular mortality (global chi-square statistic of 211.9 vs. 222.8; 2 degrees of freedom; p = 0.004). In a model that also includes left ventricular ejection fraction, the addition of diastolic dyssynchrony to systolic dyssynchrony maintained an incremental prognostic benefit (global chi-square statistic of 234.8 vs. 241.8; p = 0.030). Adjustment for baseline ischemia and scar burden did not change this relationship. CONCLUSIONS: Systolic and diastolic left ventricular dyssynchrony, as measured by using GSPECT MPI, were associated with adverse outcomes. Moreover, diastolic dyssynchrony appears to provide incremental predictive value to clinical history, electrical dyssynchrony, and left ventricular function.