J Nucl Med. 2008; 49 (Supplement 1):384P
Instrumentation & Data Analysis: Data Analysis & ManagementData Analysis & Management Posters |
Reproducibility of gated blood pool SPECT regional function normal limits
Kenneth Nichols1,
Andrew Van Tosh2,
Pieter De Bondt3,
Christopher Palestro1 and
Nathaniel Reichek2
1 Nuclear Medicine, North Shore-Long Island Jewish Health System, New Hyde Park, New York;
2 Cardiology, St. Francis Hospital, Roslyn, New York;
3 Nuclear Medicine, OLV Hospital, Aalst, Belgium
1622
Objectives: To establish the reproducibility of left & right ventricular (LV & RV) count-based gated blood pool SPECT (BP) regional functional normal limits.
Methods: Activity-time curves were fit to 3rd-order Fourier series for 17 LV & RV sub-volumes to compute global & regional parameters of ejection fraction (EF), timing, dyssynchrony (z-scores), & phase histogram standard deviation (SD) & bandwidth (BW). Calculations were performed twice (Calc1 and Calc2) independently of one another for 15 pts with CHF & 40 normal subjects (NLs).
Results: There were no significant differences between Calc1 vs. Calc2 for any LV or RV parameter for pts with CHF, nor for NLs (p>0.05), with Pearson correlation coefficients of 0.88-0.96. All parameters categorized as abnormal were the same for Calc1 vs. Calc2 (McNemar’s differences = 0%-7%, p>0.05). The only gender differences were larger volumes for males versus females (p<0.0001). All RV normal limits closely paralleled LV limits. While both ventricles contracted nearly in phase for NLs, dyssynchrony was detected in 93% of pts with CHF for the LV (z-scores > 2.0), & in 40% of pts for the RV.
Conclusions: BP global & regional LV and RV normal limits are reproducible, & application of these normal limits to pts with CHF results in reproducible detection of functional abnormalities.
| |
Reproducibility for NLs & patients with CHF
|
|
