Correlation of Ejection Fractions Obtained from PostStress Gated Myocardial Perfusion Studies and Pr

1
Correlation of Ejection Fractions Obtained from
Post-Stress Gated Myocardial Perfusion Studies
and Pre-and Post-Stress Planar Gated Blood Pool
Studies

• JE Koss, DL Kirch, TP Bublitz,
• and PP Steele
• Western Cardiology Associates
• Englewood, CO

2
Perfusion Current Methods

• Rotational SPECT - parallel hole collimators
• But, resolution doesnt permit resolving wall
  edges.
• Assume Inner Wall Surface is Smooth
• Rely on geometric fit.

Short Axis Slices at End Systole From Hurrell et
al., JNM, 1997
3
Current Methods II

• Usually validated by comparing RGBP EF and a
  Resting RSPECT EF.
• Problem 1 Cant resolve wall edges.
• Problem 2 Most common perfusion protocols
  call for gating after stress portion of test.
  SO WHAT EF ARE WE ACTUALLY GETTING?

4
NR SPECT

• Disadvantage is a limited number of views
• Basal planes resolution is degraded
• Improve by adding more detectors and views
• Advantages to multi-pinhole Non-Rotational SPECT
• Works well on small organs
• Good count statistics
• Simultaneous rather than sequential acquisition

5
Perfusion Protocol

• We use a simultaneous dual isotope (Tc-99m at
  rest Tl-201 at peak stress) technique.
• Gated imaging is acquired in list-mode format and
  completed within 25 minutes of exercise
  cessation.
• List-mode data processed a posteriori to
  produce 16 segment gated data.
• Can tag or reject beats individually

6
Start EF Calculation From A Known
Apex
Base
One known is the Center of Wall (CofW) location.
7
Determining EF from CofW

• Find areas (using the short axis) of center of
  wall at end diastole and at end systole.
• Gives a center of wall EFCofW.
• Correct EFCofW to Inner Wall EFIW using a 2nd
  order empirically derived correction.
• EFiw (-.571)EFCofW 2 (1.514)EFCofW (.03)
• Correction depends on Wall Thickness and Chamber
  Size.

8
Reproducibility

• n88 patients
• two observers

Mean0.5 2 sd4.2
27 (27/88) are exact, 76 (67/88) are within
1, 93 (82/88) are within 2
EF1 .99 EF2 .2 r.97
9
Example of Gated Perfusion Data

• 16 segment gated perfusion data
• EF 67

10
Stress Gated Blood Pool Imaging (SGBP)

• Bicycle ergonometry
• 3 minute stages (both exercise and recovery)
• Usually do 3-4 exercise followed by 3-4 recovery
  stages.
• End recovery is 9-12 minutes post exercise - or
  about 1/2 way through a GSPECT acquisition.

11
Sample SGBP EFs
Rest Stress 1 Stress 2 Stress 3 Stress
4 Recov 1 Recov 2 Recov 3 EF 67
72 74 75 80
77 74 67
Rest Str 1 Str 2 Str 3 Str 4
Str 5 Rec 1 Rec 2 Rec 3 Rec 4
EF 46 52 61 69
74 76 71 66
60 58
12
Sample SGBP EFs
77
74
75
74
80
72
67
67
76
74
71
70
66
61
60
58
52
47
13
EF Recovery Time
n85
Even after 3 Recovery Stages, half the patients
have a recovery EF over 5 higher than resting
EF. A quarter of them have gt10 elevated EF.
14
Comparison Between Perfusion EF and Resting GBP EF

• Bland Altman Regression Plots
• n90 patients

Mean1.5 2 sd14.5
y1.02 x - 1.2 r.85
15
Comparison Between Perfusion EF and Recovery GBP
EF

• Bland Altman Regression Plots
• n85 patients

Mean -1.7 2 sd11.0
y.96 x 1.0 r.91
16
Conclusions

• Perfusion EFs acquired post-stress are not well
  correlated with resting EF.
• The heart undergoes significant volumetric
  changes as it returns to a resting condition.
• Perfusion EFs correlate better with a late
  recovery stage rather than a resting GBP.
• Center of Wall tracking is a reliable method for
  determining Perfusion EFs.
• The correlation between perfusion and recovery
  EFs is comparable to those published comparing
  EFs from other SPECT perfusion techniques.