J Nucl Med. 2007; 48 (Supplement 2):223P
Cardiovascular: Basic Science
Basic Science Posters |
SPECT myocardial perfusion image assessment of cardiac geometry changes: Validation by CMR
Shoji Suzuki1,
Piotr Slomka1,
Fieno David1,
Ramesh Amit1,
Yasuyuki Suzuki1,
Muneo Oba1,
Dey Damini1,
Louise Thomson1,
Sean Hayes1,
John Friedman1,
Aiden Abidov1,
Guido Germano1 and
Daniel Berman1
1 Departments of Medicine and Imaging, Cedars-Sinai Medical Center, Los Angeles, California
1063
Objectives: Changes in cardiac geometry result from worsening left ventricular (LV) performance and are predictive of heart disease. We previously reported that LV shape index (LVSI), the ratio of the maximum 3D short- and long-axis dimensions of LV, measured from gated Tc-99m sestamibi myocardial perfusion SPECT (MPS) predicted hospitalization of patients with congestive heart failure. The aim of this study was to validate LVSI at end-systole (ES) and end-diastole (ED) quantified from gated rest Tl MPS by comparison with cardiac magnetic resonance (CMR). Methods: We studied 98 patients (pts) (range 21-92 years, 69 male) who underwent dual isotope gated MPS (rest Tl-201/stress Tc-99m) and CMR within a month. There were 50 small heart cases with LV end-systolic volume (LVESV) < 40ml by MPS and 48 non-small heart cases, with LVESV > 40 ml by MPS. LVSI-ED, LVSI-ES and LV ejection fraction (LVEF) were calculated from MPS and CMR, by the same LV computational model. Results: LVSI-ED did not differ between MPS and CMR (0.64±0.06 vs. 0.64±0.06, p=0.7) but LVSI-ES was lower for MPS than for CMR (0.47±0.09 vs. 0.52±0.08, p<0.01). LVSI-ED (r=0.81, p<0.01) and LVSI-ES (r=0.78, p<0.01) correlated strongly between MPS and CMR. In small heart cases, LVSI-ED was not different between MPS and CMR (0.62±0.06 vs. 0.63±0.06, p=0.6), but LVSI-ES by MPS was significantly smaller (0.42±0.06 vs. 0.48±0.05, p<0.01) and correlated weakly (r=0.38, p<0.01). In non-small heart cases (n=48), LVSI-ED and LVSI-ES were not significantly different between MPS and CMR, and correlated strongly (LVSI-ED: r=0.77, p<0.01; LVSI-ES: r=0.82, p<0.01). There was significant negative correlation between LVEF and LVSI-ES in MPS (r=-0.8, p<0.01), and in CMR (r=-0.69, p<0.01). There was significant correlation between LVESV and LVSI-ES in MPS (r=0.72, p<0.01) and CMR (r=0.74, p<0.01). Conclusions: LVSI-ED by MPS and CMR and LVSI-ES in non-small hearts were similar and were strongly correlated. In small hearts, LVSI-ES was smaller by MPS than by CMR and the correlation was weak; however, in these pts, LVSI-ES values by CMR were relatively low and not likely to be clinically relevant.
