OtherClinical Investigations

Quantitative Evaluation of Myocardial Blood Flow and Ejection Fraction with a Single Dose of 13NH3 and Gated PET

Hidehiko Okazawa, Masaaki Takahashi, Tatsuhiko Hata, Kanji Sugimoto, Yoshihiko Kishibe and Takafumi Tsuji
Journal of Nuclear Medicine August 2002, 43 (8) 999-1005;

Article Figures & Data

Figures

  • FIGURE 1.
    FIGURE 1.

    Graphic plotting method (Patlak plot) was applied for calculation of MBF. Representative Patlak plot shows excellent linear regression for total frame time of 5 min. Insert shows time-activity curves for LV cavity and myocardial tissue obtained from same dynamic 13NH3 PET data. θ = normalized time; DV = volume of distribution; Sep = septal; Lat = lateral.

  • FIGURE 2.
    FIGURE 2.

    Representative images of MBF of healthy volunteer (NV) and of patient with OMI (with defect) in anterior wall. Images of MBF were calculated pixel by pixel on basis of graphic plotting method. Slopes of linear fit using time frames of 30–150 s in plots (Fig. 1) were converted into MBF values. Images were resliced into LV short-axis and long-axis planes.

  • FIGURE 3.
    FIGURE 3.

    Images of gated 13NH3 PET in 3D mode of pFAST (A) and of GBP PET in 2D mode (B). Ratio of volumes in end-systolic phase (red part) and end-diastolic phase (meshed frame) provides LVEF. In GBP imaging, right and left ventricles (RV and LV) are clearly separated by septal myocardial wall. Bitmap images (pink transparent areas) were generated on each short-axis slice in end-diastolic and end-systolic phases of GBP PET. ANT = anterior; LAT = lateral; INF = inferior.

  • FIGURE 4.
    FIGURE 4.

    (A) Correlation of LVEF obtained from LVG and GBP PET (C15O) in 20 patients who underwent both studies. LVG and GBP PET show excellent linear correlation regardless of whether defect was (□) or was not (•) on perfusion image. Dashed line is line of identity. (B) Bland-Altman plot shows no significant degree of systematic measurement bias between 2 methods. Lines indicate mean and mean ± 2 SD.

  • FIGURE 5.
    FIGURE 5.

    (A) Correlation of LVEF obtained from GBP PET (C15O) and pFAST (13NH3 PET) in all subjects (n = 40). LVEF values from GBP and pFAST correlate well, although there was slight underestimation of LVEF with pFAST, and stronger tendency for underestimation was observed in patients with defect (□) compared with patients without defect (•). Dashed line is line of identity. (B) Bland-Altman plot also shows underestimation of LVEF by pFAST compared with GBP by −4.58% ± 7.49%. Lines indicate mean and mean ± 2 SD.

Tables

  • TABLE 1

    Comparison Between LVEF Obtained Using LVG and Gated PET

    PatientsnLVGGBPpFAST
    With CVD2054 ± 1653 ± 1446 ± 16*
     No defect867 ± 1366 ± 959 ± 16
     With defect1246 ± 1244 ± 1038 ± 8*†

    • ↵* P < 0.001 in comparison with LVG.

    • ↵† P < 0.005 in comparison with GBP (post hoc Fisher’s PLSD).

    • GBP = PET with C15O; pFAST = perfusion PET with 13NH3; CVD = cardiovascular disease.

    • Data are mean ± SD.

  • TABLE 2

    LVEF, ESV, and EDV Obtained Using Different Methods

    SubjectsnGBPpFAST
    LVEF (%)ESV (mL)EDV (mL)LVEF (%)ESV (mL)EDV (mL)
    Healthy volunteers665 ± 533 ± 593 ± 1063 ± 655 ± 9*149 ± 17*
    Patients with CVD3451 ± 1461 ± 31113 ± 3346 ± 15†96 ± 58†168 ± 62†
     No defect1560 ± 1243 ± 2196 ± 2556 ± 1357 ± 21‡130 ± 32†
     With defect1945 ± 1175 ± 32127 ± 3239 ± 11†127 ± 59†198 ± 63†

    • ↵* P < 0.005 comparing 2 methods of gated PET (paired t test).

    • ↵† P < 0.001 comparing 2 methods of gated PET (paired t test).

    • ↵‡ P < 0.05 comparing 2 methods of gated PET (paired t test).

    • GBP = PET with C15O; pFAST = perfusion PET with 13NH3; CVD = cardiovascular disease.

    • Data are mean ± SD.

Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Bookmark this article

Jump to section