|
|
||||||||
Technical Developments |
1 From the Department of Radiology (H.W.M.K., A.d.R.), Divisions of Nuclear Medicine (C.D.L.B.C., P.D.S., E.K.J.P.) and Cardiology (E.E.v.d.W., D.E.A.), Leiden University Medical Centre, Albinusdreef 2, C4Q-80, 2333 ZA Leiden, the Netherlands; and the Division of Nuclear Medicine, Cedars-Sinai Medical Center, Los Angeles, Calif (G.G.). Received November 10, 1999; revision requested December 15; revision received February 7, 2000; accepted March 30. Address correspondence to C.D.L.B.C. (e-mail: C.Bavelaar-Croon@inter.nl.net).
| ABSTRACT |
|---|
|
|
|---|
Index terms: Heart, function, 524.91 Heart, MR, 524.121412 Heart, SPECT, 524.12162 Heart, ventricles, 524.121412, 524.12162, 524.91 Heart, volume, 524.121412, 524.12162 Magnetic resonance (MR), comparative studies, 524.121412, 524.1262
| INTRODUCTION |
|---|
|
|
|---|
The aim of this study was to validate LVEF, LVEDV, and LVESV measurements with gated SPECT over a wide range of values, with MR imaging as the reference standard.
| Materials and Methods |
|---|
|
|
|---|
Thirteen (62%) of the 21 patients had experienced a myocardial infarction. The infarction was located in the anterior wall in eight patients and in the inferior wall in five. No cardiac events occurred between the two studies. At LV angiography in 12 patients, hypokinesia or akinesia was found in 10 patients and aneurysm in four.
Gated SPECT Acquisition
In our department, 1- and 2-day protocols are used interchangeably (21). Six (29%) of our patients underwent a 1-day protocol; and 15 patients, a 2-day protocol. In the 1-day protocol, 250 MBq of technetium 99m tetrofosmin (Nycomed-Amersham; Oslo, Norway) was administered 4560 minutes prior to acquisition of the rest gated SPECT study. Four hours later, 750 MBq of 99mTc tetrofosmin was injected during peak bicycle exercise or after infusion of adenosine or dobutamine. Thirty minutes later, the gated acquisitions of the stress SPECT study was started.
In the 2-day protocol, 500 MBq of 99mTc-tetrofosmin was administered 4560 minutes prior to rest gated SPECT and 30 minutes prior to stress SPECT. In this protocol, gating was applied during the rest SPECT acquisition because of logistical reasons.
Imaging was performed with a triple-headed gamma camera (model GCA 9300; Toshiba Medical, Tokyo, Japan) with "step-and-shoot" rotation, 90 projections over a 360° arc, 35 seconds per projection, and 64 x 64 matrices. Sixteen frames per cardiac cycle were acquired. Transverse sections were reconstructed with a 9.0 Butterworth filter and a cutoff frequency of 0.32 cycles per pixel for the at-rest MR imaging study or 0.26 cycles per pixel for the after-stress SPECT study (pixel size, 6 mm). Numeric values of LV volumes and LVEF were calculated by using commercially available software (GATED SPECT, version 1.0, revision B; Toshiba Medical) developed by Germano et al (7). This algorithm uses gated SPECT short-axis image volumes. It estimates the location of the LV endocardial surface and the valve plane in the three-dimensional space and then adds the volumes of the voxels bound by these two structures. The technique used to identify the endocardial surface was based on artificial intelligence combined with the Gaussian fitting of count profiles normal to the myocardium. LVEF is derived from the LV volume images without operator interaction (Fig 1).
|
A workstation (Sparc; Sun Microsystems, Mountain View, Calif) was used for further analysis, performed with fixed window width and window level settings. LV volumes were calculated from the gradient-echo MR imaging data by using an analytical software package (MASS, version 4.0; Leiden University Medical Center, Leiden, the Netherlands) (22). The cross-sectional areas of endocardial tracings at end systole and at end diastole were added, and the sum was multiplied by the section thickness. The end-diastolic phase was identified as the first frame of the cine loop and represented the maximum LV volume. The end-systolic phase was identified as the frame that showed the minimal cavity volume. LVEF was calculated on the basis of the LVEDV and LVESV values (Fig 1).
Statistical Analysis
Regression analysis was performed to determine the correlation between LVEF, LVEDV, and LVESV measured with gated SPECT and MR imaging. A Bland-Altman plot was used to assess systematic trends in the differences between the two methods. With this plot, the mean LVEF value with the two methods was shown on the ordinate, and the difference in LVEF values was shown on the abscissa.
| Results |
|---|
|
|
|---|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Because LVEF and LVESV are major diagnostic and prognostic parameters (1,3,23), the ability to obtain quantitative measurements of LVEF and LVESV represents an important feature of gated SPECT. Before gated SPECT is used routinely, however, it is important to establish the reliability of the functional information, preferably by validating it against an accurate standard of reference. MR imaging has been shown to be an accurate method for measuring LV volumes and LVEF (13). Previous studies found good correlations between wall motion and wall thickening values with obtained gated SPECT and MR imaging. In some patients with severely reduced tracer uptake, however, functional information may be suboptimal (2426). In the patients in our study, a good correlation was found between quantitative parameters with MR imaging and gated SPECT, with both correlation coefficients and scatter around the regression lines comparable to those previously reported for this and other software packages (9,11,1820).
In this study, the endocardial borders on the MR images were drawn manually. The resultant inter- and intraobserver variability was within acceptable limits (27) but would be expected to cause somewhat higher scatter of the LVEF correlations than that of the volume correlations, owing to propagation of error in the former. With quantitative gated SPECT, LVEF may be slightly overestimated in patients with small LVESVs (28). Although it may be difficult to detect the endocardial border in patients with severely reduced tracer uptake (2426), we did not find this to be the case in our study. In none of our patients was LVEF diagnosed as definitely abnormal (<40%) with one method and as definitely normal (>50%) with the other method.
Good correlation between LV function measurements with gated SPECT and MR imaging was found recently by Tadamura et al (20), who also used the software program used in this study. In another study, Vaduganathan et al (19) also found a good correlation in patients with a previous myocardial infarction. Tadamura et al (20) used 99mTc sestamibi and thallium 201, and the gating was performed with eight frames per cycle. In our study, gating was performed with 16 frames per cycle. Although the use of eight frames per cycle may result in an underestimation of LVEF, the use of 16 frames per cycle may reduce count statistics in the individual frames to the point that errors may be introduced in the measurement of LV volumes (11). Vaduganathan et al (19) included only patients with a previous myocardial infarction, but we included patients with and those without myocardial infarction; therefore, the patient population was different.
On the basis of the Bland-Altman plot in our study, no systematic under- or overestimation of LVEF was found with gated SPECT as compared with the MR imaging standard of reference. Although our data demonstrated a slight underestimation of LVEDV and LVESV with gated SPECT as compared with MR imaging, this finding did not reach statistical significance. This finding is most likely due to the inclusion of more outflow tract tissue on MR images than is routinely visible on gated SPECT images. The outflow tract is never part of the LV volume acquired at gated SPECT because the edge of the left ventricle is defined by the mitral valve. The membranous part of the septum is not shown on gated SPECT images, but the calculation of LV volumes on MR images includes this part of the septum. These findings do not appreciably influence the calculation of LVEF, because both the LV volumes are overestimated and, therefore, the ratio of LVEDV to LVESV remains unaffected.
A clear advantage of gated SPECT as compared with MR imaging is the automatic delineation of the myocardial border with gated SPECT in conjunction with the algorithm used in this study and the consequent high reproducibility and lack of operator dependence on the gated SPECT results (11). Although echocardiographically determined LV function may be available for patients seen at a nuclear medicine laboratory, quantitative gated SPECT is also valuable for the following reasons: (a) gated SPECT offers the potential to assess LV function (LV volumes, LVEF, wall motion, wall thickening) and myocardial perfusion simultaneously and on the basis of just one acquisition procedure, (b) the acquisition of functional information adds essentially no extra costs to the standard perfusion study, and (c) functional measurements with gated SPECT are completely automated with the approach used in this study. Moreover, gated SPECT can also be useful in patients with an uninterpretable or unreliable echocardiogram owing to obesity or pulmonary emphysema.
In six (29%) of our 21 patients, the gated SPECT study was performed 30 minutes after stress, whereas the MR imaging study was always performed at rest. In patients with coronary artery disease, LV function 30 minutes after stress may be depressed compared with that at rest, owing to myocardial stunning (2931). However, when data for the six patients were separated from those for the 15 patients, the correlation for LVEF measured with gated SPECT and MR imaging did not change significantly. Therefore, we conclude that myocardial stunning probably did not influence our results.
On the basis of findings in this study, we believe gated SPECT is a reliable, accurate method for measuring LV function in patients over a wide range of LV volume and LVEF values.
| FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, C.D.L.B.C., D.E.A.; study concepts, D.E.A., C.D.L.B.C., G.G.; study design, D.E.A., C.D.L.B.C., G.G.; definition of intellectual content, D.E.A.; literature research, C.D.L.B.C., D.E.A.; clinical studies, D.E.A., E.E.v.d.W.; data acquisition, P.D.S.; data analysis, H.W.M.K., D.E.A.; statistical analysis, D.E.A.; manuscript preparation, C.D.L.B.C.; manuscript editing, C.D.L.B.C., E.E.v.d.W., D.E.A.; manuscript review, E.E.v.d.W., A.d.R., E.K.J.P.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. Knollmann, O. H. Winz, P. T. Meyer, M. Raptis, T. Krohn, K.-C. Koch, and W. M. Schaefer Gated Myocardial Perfusion SPECT: Algorithm-Specific Influence of Reorientation on Calculation of Left Ventricular Volumes and Ejection Fraction J. Nucl. Med., October 1, 2008; 49(10): 1636 - 1642. [Abstract] [Full Text] [PDF] |
||||
![]() |
M S KRISHNAM, A TOMASIAN, M. Iv, S G RUEHM, R SALEH, C PANKNIN, and J G GOLDIN Left ventricular ejection fraction using 64-slice CT coronary angiography and new evaluation software: initial experience Br. J. Radiol., June 1, 2008; 81(966): 450 - 455. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Schepis, O. Gaemperli, P. Koepfli, I. Valenta, K. Strobel, A. Brunner, S. Leschka, L. Desbiolles, L. Husmann, H. Alkadhi, et al. Comparison of 64-Slice CT with Gated SPECT for Evaluation of Left Ventricular Function J. Nucl. Med., August 1, 2006; 47(8): 1288 - 1294. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. M. Schaefer, C. S.A. Lipke, D. Standke, H. P. Kuhl, B. Nowak, H.-J. Kaiser, K.-C. Koch, and U. Buell Quantification of Left Ventricular Volumes and Ejection Fraction from Gated 99mTc-MIBI SPECT: MRI Validation and Comparison of the Emory Cardiac Tool Box with QGS and 4D-MSPECT J. Nucl. Med., August 1, 2005; 46(8): 1256 - 1263. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Yamamuro, E. Tadamura, S. Kubo, H. Toyoda, T. Nishina, M. Ohba, R. Hosokawa, T. Kimura, N. Tamaki, M. Komeda, et al. Cardiac Functional Analysis with Multi-Detector Row CT and Segmental Reconstruction Algorithm: Comparison with Echocardiography, SPECT, and MR Imaging Radiology, February 1, 2005; 234(2): 381 - 390. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. K. Paul and H. A. Nabi Gated Myocardial Perfusion SPECT: Basic Principles, Technical Aspects, and Clinical Applications J. Nucl. Med. Technol., December 1, 2004; 32(4): 179 - 187. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. M. Schaefer, C. S.A. Lipke, H. P. Kuhl, K.-C. Koch, H.-J. Kaiser, P. Reinartz, B. Nowak, and U. Buell Prone Versus Supine Patient Positioning During Gated 99mTc-Sestamibi SPECT: Effect on Left Ventricular Volumes, Ejection Fraction, and Heart Rate J. Nucl. Med., December 1, 2004; 45(12): 2016 - 2020. [Abstract] [Full Text] [PDF] |
||||
![]() |
S.R. Underwood, J. J Bax, J. v. Dahl, M. Y Henein, A. C van Rossum, E. R Schwarz, J.-L. Vanoverschelde, E. E.v. d. Wall, and W. Wijns Imaging techniques for the assessment of myocardial hibernation: Report of a Study Group of the European Society of Cardiology Eur. Heart J., May 2, 2004; 25(10): 815 - 836. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Go, M. R. Bhatt, and R. C. Hendel The Diagnostic and Prognostic Value of ECG-Gated SPECT Myocardial Perfusion Imaging J. Nucl. Med., May 1, 2004; 45(5): 912 - 921. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. A. Ioannidis, T. A. Trikalinos, and P. G. Danias Electrocardiogram-gated single-photonemission computed tomography versus cardiacmagnetic resonance imaging for the assessmentof left ventricular volumes and ejection fraction: A meta-analysis J. Am. Coll. Cardiol., June 19, 2002; 39(12): 2059 - 2068. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |