82Rb PET measurement of diastolic dyssynchrony during physiologic stress

• Published in: The Journal of nuclear medicine (1978) Vol. 59; p. 1523
• Main Authors: Van Tosh, Andrew, Mathew, Jaison, Cooke, Charles, Palestro, Christopher, Nichols, Kenneth
• Format: Journal Article
• Language: English
• Published: New York Society of Nuclear Medicine 01.05.2018
• Subjects: Abnormalities; Cardiovascular disease; Chi-square test; Contraction; Coronary artery; Coronary artery disease; Correlation; Curve fitting; Data acquisition; Diagnostic systems; Diastole; Evaluation; Heart diseases; Heart function; Histograms; Medical imaging; Normality; Parameter identification; Rest; Single photon emission computed tomography; Stenosis; Subgroups; Systole; Territory
• ISSN: 0161-5505; 1535-5667

Objectives: Systolic dyssynchrony (SDys) has been shown to predict prognosis in pts with heart failure & coronary artery disease (CAD). Diastolic dyssynchrony (DDys) is less well recognized, & may also be prognostic. Routinely DDys is assessed at rest by SPECT. In this investigation we computed DDys parameters from 82Rb PET data acquired at rest & during pharmacologic stress, to determine if they aid in diagnosing CAD. Methods: Data were reviewed retrospectively for 135 pts who had undergone rest/regadenoson-stress 82Rb gated PET for evaluation of potential CAD, including 46 with bundle branch block (BBB). Rest & stress PET data were rebinned as gated data sets during the myocardial equilibrium portion of the acquisition. Multiple harmonic Fourier curve fitting using Emory Cardiac Toolbox v4 was performed to identify onset of myocardial contraction in systole, & onset of myocardial relaxation in diastole. Phase histograms were constructed from regional relative myocardial count changes versus time to measure peak phase, phase standard deviation & phase bandwidth (BW) for SDys & DDys. Polar maps were constructed to display SDys computations showing phase systolic onset, & DDys computations showing phase diastolic onset, which were examined & scored by a medical imaging physicist based on deviations from normality on a 5-point scale (0 = "normal" to 4 = "markedly dyssynchronous extensive territory") for each of the major arterial territories. In a subgroup of 103 pts, regional stenoses were graded from coronary angiograms at a core lab (Boston Cardiac Research Institute), using quantitative coronary angiographic methods. Results: DDys phase histogram bandwidth (BW), measuring range of the R-R interval including 95% of the LV relaxation, was larger than SDys BW of LV contraction at rest (137±80º versus 126±80 º, p < 0.0001) & at stress (133±88º versus 125±85º, p < 0.0001), but rest-stress changes were similar (-4±63º versus -1±60º, p = 0.34). DDys BW correlated strongly with SDys BW for rest (r = 0.95, p < 0.0001) & stress (r = 0.97, p < 0.0001) & for rest-stress changes (r = 0.85, p < 0.0001). Correlations were similar for rest-stress changes in DDys & SDys BW for pts without BBB (r = 0.88, p < 0.0001; N = 72), & with BBB (r = 0.97, p < 0.0001; N = 18). DDys & SDys phase shifts from rest to stress changed in the same direction (r = 0.40, p < 0.0001). DDys & SDys visual scores of regional phase abnormalities judged from polar phase maps also correlated strongly for rest (χ2 = 101.6, p < 0.0001), stress (χ2 = 161.8, p < 0.0001), & rest-stress changes (χ2 = 165.6, p < 0.0001). The diagnostic value of DDys & SDys phase parameters for identifying individual coronary stenoses appeared to be additive: overall accuracy was 78±4% by SDys parameters, 77±4% by DDys parameters, & 82±3% by combining SDys with DDys phase parameters. Conclusion: In a referral population being evaluated for CAD, DDys measurements closely paralleled SDys parameters, & had additive diagnostic value for identifying coronary stenosis. Research Support: Grant from Astellas, Inc.