- Email: [email protected]
Accuracy of left ventricular ejection fraction determined by gated myocardial perfusion SPECT with Tl-201 and Tc-99m sestamibi: Comparison with first-pass radionuclide angiography
Accuracy of left ventricular ejection fraction determined by gated myocardial perfusion SPECT with TI-201 and Tc-99m sestamibi: Comparison with first-pass radionuclide angiography Z u o - X i a n g He, MD, E d u a r d o Cwajg, MD, Janice S. Preslar, John J. M a h m a r i a n , MD, a n d M a r i o S. Verani, M D Background. We compared estimates of left ventricular ejection fraction (LVEF) assessed by gated single photon emission computed tomography (SPECT), using both technetium-99m sestamibi and thallium-201, with those obtained by first-pass radionuclide angiography (FPRNA) in patients with a broad spectrum of LVEF and perfusion abnormalities. Methods. Sixty-three patients were r a n d o m l y selected to undergo a dual isotope gated SPECT study (rest TI-201 followed by adenosine Tc-99m sestamibi scintigraphy). Studies were processed by use of the Cedars quantitative gated S P E C T software. FPRNA was acquired during an intravenous bolus injection of Tc-99m sestamibi and processed with a commercially available software. Results. The estimates of LVEF were similar (P = NS) with TI-201 gated S P E C T (54% _+ 15%), Tc-99m gated S P E C T (54% _+ 16%), and FPRNA (54% _+ 12%). There was an excellent correlation between Tc-99m and TI-201 gated S P E C T (Pearson's r = 0.92, P < .0001). There were also good linear correlations between Tc-99m sestamibi gated S P E C T and FPRNA (Pearson's r = 0.85, P < .0001), as well as between T1-201 gated S P E C T and FPRNA (Pearson's r = 0.84, P < .0001). In the 16 patients with LVEF < 50%, Tc-99m sestamibi gated S P E C T and FPRNA (Pearson's r = 0.84, P < .0001) and TI-201 gated SPECT and FPRNA (Pearson's r = 0.92, P < .0001) correlated well. Conclusion. LVEF can be accurately assessed by gated S P E C T with either Tc-99m sestamibi or T1-201 in properly selected patients with normal or depressed left ventricular function. (J Nucl Cardiol 1999;6:412-7.) Key Words: Gated S P E C T • first-pass radionuclide angiography • LVEF • Tc-99m sestamibi
Gated single photon emission computed tomography (SPECT) enables simultaneous assessment of myocardial perfusion and function.l-3 It is well accepted that left ventricular ejection fraction (LVEF) can be accurately measured by gated SPECT using technetium99m-labeled perfusion agents. T M Data are, however, scanty on the accuracy of this measurement by gated SPECT with thallium-201.12,13 Moreover, the accuracy of LVEF measurement by gated SPECT in patients with
depressed left ventricular (LV) function and severe perfusion defects has not been well investigated. Accordingly, this study compared LVEF assessed by gated SPECT by use of both Tc-99m sestamibi and T1201 with that measured by first-pass radionuclide angiography (FPRNA) in patients with a broad spectrum of LV function and peffusion defects.
From the Section of Cardiology, Baylor College of Medicine and The Methodist Hospital,Houston, Texas. Received for publication June 26, 1998; revision accepted Sept 16, 1998. Reprint requests: Mario S. Verani, MD, Baylor College of Medicine, The Methodist Hospital,6550 Fannin, SM 677, Houston,TX 77030;
Study Population
[email protected].
Copyright © 1999 by the AmericanSociety of Nuclear Cardiology. 1071-3581/99/$8.00 + 0 43/1/94581 412
MATERIAL AND METHODS
Patients referred for adenosine myocardial SPECT in our laboratory were randomly selected to undergo a dual isotope study consisting of rest T1-201 gated SPECT followed by adenosine Tc-99m sestamibi gated SPECT. Patients were excluded if they had contraindications to adenosine stress, significant cardiac arrhythmias, or large body habitus (weight > 200 lbs or women with large breasts or breast implants).
Journal of Nuclear Cardiology Volume 6, Number 4;412-7
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
Study Protocol
Table 1. D e m o g r a p h i c d a t a
Rest T1-201 gated SPECT acquisition was initiated 10 minutes after injection of T1-201 (3.1 + 0.2 mCi) with the patient at rest. Adenosine infusion (140 p.g/kg/min for 6 minutes) was begun within the next 30 minutes, Tc-99m sestamibi (30 -+ 2.0 mCi) was injected halfway into the adenosine infusion, and gated SPECT acquisition was initiated 1 hour later.
Total Men Women Age (yrs) Height (inches) Weight History Prior M! CABG PTCA Symptoms Chest pain Other Asymptomatic Risk factors Family history Diabetes mellitus Cigarette smoking Hypercholesterolemia Hypertension
FPRNA FPRNA was acquired immediately after the bolus injection of Tc-99m sestamibi on a SIM-400 multicrystal gamma camera and processed by an operator who was blinded to the gated SPECT results, using a commercially available software, 14 which uses 2 regions of interest for end diastole and end systole, respectively.
Nongated SPECT Nongated projection data were created by summing the gated data set at each projection into a nongated data set. Nongated myocardial tomograms were then reconstructed by use of a standard filtered back-projection algorithm followed by reorientation of the transaxial images into the short and horizontal and vertical long axes. Images were qualitatively and quantitatively interpreted by an expert nuclear cardiologist, as previously reported from our laboratory. 15 Quantification of myocardial perfusion defect size was performed with polar map analysis. The raw data polar maps for each patient were statistically compared with a normal data bank to determine the defect size and the extent of scar and ischemia. 15
Gated SPECT Gated myocardial SPECT was performed on a 90-degree configuration dual-head SPECT system equipped with a low-energy, high-resolution collimator. Sixty-four projections (32 per each detector) were acquired over a 180-degree anterior arc, from the 45-degree left posterior oblique view to the 45-degree right anterior oblique view. Eight frames per cardiac cycle were obtained at each projection. Acquisition time was 30 seconds per projection for T1-201 and 20 seconds per projection for Tc-99m sestamibi. The raw data of T1-201 and Tc-99m sestamibi gated SPECT study were visually inspected for the assessment of image quality. Only images considered to be of good quality were included in the study. T1-201 and Tc-99m gated sestamibi SPECT images were reconstructed by use of a back-projection algorithm with a three-dimensional Butterworth filter (cutoff frequencies of 0.35 Nyquist [pixel size 0.59 cm] and 0.5, respectively, and order 5). The transaxial gated tomographic slices were then reoriented into the short- and horizontal and vertical long-axis views. LVEF by gated SPECT was calculated by an independent operator using the Cedars Quantitative Gated SPECT software.5 This algorithm operates in the 3-dimensional space. It segments the LV, estimates and displays endocardial and epicardial surfaces for all 8 images in the cardiac cycle, calculates the relative LV volumes, and derives LVEF from the end-diastolic and -systolic volumes.5
63 41 22 67 68 176
413
(100%) (65%) (35%) _+ 10 _+4 _+ 30
19 (300/0) 16 (15%) 20 (32%) 30 (48%) 26 (41%) 7 (11%) 18 |6 20 27 39
(28%) (25%) (320/0) (43%) (62%)
Nil, Myocardial infarction; CABG, coronary artery bypast graft; PTCA, percutaneous transluminal coronary angioplasty.
Statistical Analysis Data are reported as mean _+SD or frequency when appropriate. Differences anaong LVEFs measured by gated SPECT with T1-201 and Tc-99m sestamibi and by FPRNA were evaluated by analysis of variance (ANOVA). Agreement among the gated SPECT techniques and FPRNA was evaluated by use of regression analysis (Pearson's r) and Bland-Altman analysis. A P value less than .05 was considered statistically significant.
RESULTS Demographic Data A total of 63 patients were studied. Clinical information on these patients is summarized in Table 1.
Results of Nongated SPECT Myocardial perfusion was normal in 32 patients and abnormal in 31 patients. Perfusion defects were completely reversible in 6 patients, partially reversible in 15 patients, and fixed in 10 patients. Total defect size in patients with an abnormal SPECT measured 19% _+ 15% of the LV.
LVEF by Gated SPECT and by FPRNA L V E F averaged 54% _ 15% b y T1-201 gated SPECT, 54% _+ 16% by Tc-99m sestamibi gated SPECT, and 54% _+ 12% by F P R N A (ANOVA, P = ns).
414
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
Journal of Nuclear Cardiology July/August 1999
90
y=2.84+O.96x r = 0.92, SEE = 6.1 p
80
7o
• ~ "
,
50 ¸
~ ~'~
40
•
t-¢.) 6 0 IM
• .
,~ s0
•
l~:/."
eee
• •e~
•
30 •
.
20
•
"13
10
40
m O1 • - 30
•
•
.=
•
*
6 e°o
0 •
...
O
";.~."".-
~O O
-10
"~' 2 0 I--
10
-20 I
[
r
]
L
I
i
I
10
20
30
40
50
60
70
80
-30 90
Tc-g9m Sestamibi gated-SPECT (%)
Figure 1. Correlation between Tc-99m sestamibi and T1-201 gated SPECT for LVEE
Agreement Between 11-201 and Tc-99m Sestamibi Gated SPECT There was an excellent linear correlation between Tc-99m sestamibi and T1-201 gated SPECT (Pearson's r = 0.92, P <.0001) (Figure 1). Bland-Altman analysis showed no evidence of any systematic bias between these two techniques (Figure 2).
Agreement Between SPECT Techniques and FPRNA There was a good linear correlation between Tc-99m sestamibi gated SPECT and FPRNA (Pearson's r = 0.85, P < .0001), as well as between T1-201 gated SPECT and FPRNA (Pearson's r = 0.84, P < .0001) (Figure 3, A and B). Bland-Altman analysis showed no obvious bias between sestamibi gated SPECT and FPRNA (Figure 4, A) or between T1-201 gated SPECT and FPRNA (Figure 4, B).
LVEF by Gated SPECT in Patients with Depressed LV Function In the subset of 16 patients with an LVEF < 50% by FPRNA, there was a good correlation between Tc-99m sestamibi gated SPECT and T1-201 gated SPECT (Pearson's r = 0.86, P < .0001) (Figure 5). There was also a good correlation between Tc-99m sestamibi gated SPECT and FPRNA (Pearson's r = 0.84, P < .0001), as well as between T1-201 gated SPECT and FPRNA (Pearson's r = 0.92, P < .0001) (Figure 6, A and B).
-40 -50
i
i
I
I
I
[
E
]
I
10
20
30
40
50
60
70
80
90
100
LVEF by T¢-9gm Sestamibi and TI-201 gated-SPECT
Figure 2. Bland-Altman analysis between Tc-99m sestamibi and T1-201 gated SPECT for LVEF. Bias: LVEF by sestamibi gated SPECT minus LVEF by T1-201 gated SPECT. X-axis represents average of LVEF by both Tc-99m sestamibi and T1-201 gated SPECT. (ANOVA, P = NS). The correlation coefficient was 0.88 (P < .0001) between T1-201 gated SPECT and FPRNA, and 0.92 (P < .0001) between Tc-99m sestamibi gated SPECT and FPRNA.
LVEF in Patients with Moderate to Severe Perfusion Defects In the 13 patients who had moderate to severe perfusion defects, the LVEF was 47% _ 12% by T1-201 gated SPECT, 41% _+ 15% by Tc-99m sestamibi gated SPECT, and 44% _ 11% by FPRNA (ANOVA, P = NS). The correlation coefficient was 0.88 (P < .0001) between T1-201 gated SPECT and FPRNA, and 0.92 (P < .0001) between Tc-99m sestamibi gated SPECT and FPRNA.
DISCUSSION To the best of our knowledge, this is the first study comparing LVEF assessed by T1-201 gated SPECT, Tc99m sestamibi gated SPECT, and FPRNA in the same patients. Our study demonstrates that (1) T1-201 and Tc99m gated SPECT correlate very well with respect to determination of LVEF; and (2) both T1-201 and Tc-99m gated SPECT correlate well with FPRNA in properly selected patients, including those with depressed LV function and moderate to severe perfusion abnormalities.
LVEF in Patients with Reversible Perfuslon Defects In the subset of 21 patients who had completely or partially reversible perfusion defects, the LVEF was 48% ___13% by T1-201 gated SPECT, 45% + 15% by Tc-99m sestamibi gated SPECT, and 49% + 13% by FPRNA
Gated SPECT with Tc-99m-Labeled Agents Several algorithms have been recently proposed to calculate LVEF by gated SPECT. German• et al 5 developed a completely automatic algorithm to quantify LVEF
Journal of Nuclear Cardiology Volume 6, Number 4;412-7
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
50 ~
100
.c.
90
I-
8O
415
40 30 -
t.) IM a.
**
y : 1.1107x- 0.84 R:0.85, SEE = 8.3 p < 0.0001
70
•
~
"o
• •
~ 6o
f
20-
•
:
~
• • $
•
10 ~
m= 0•
(4)
50
E ~
40
-30 -
~
30
-40-50
A
117
,, •
i
t
I
[
[
I
E
I
I
10
20
30
40
50
60
70
80
90
100
LVEF by FPRNA and Tc-99rnSestamibi gated-SPECT
5O
0 10
A
• 4`
-20
~ 2o k-
•
-10
e~oo 48e * e • ~
--
20
30
40
50
60
70
80
40
90
First-Pass R a d i o n u c l i d e A n g i o g r a p h y (%)
30 20
100
.:
10
90 y : 1.0868x - 8.600 R : 0.84, SEE = 8.4
8o
•o
el
•
p<0.0001
-~ 70 I0 m 60
¢9
" :
• ~ . im ~e •
o•
. / J "
-10
"
-20
.'~."
•
s•
s
v•
."
,
• *
-30
oe
-50
:
•: . "
B
40
I
I
I
]
I
[
I
[
I
10
20
30
40
50
60
70
80
90
100
LVEF by FPRNAand TI-201 gated-SPECT
__ 3o
Figure 4. Bland-Altman analysis between FPRNA and Tc-99m sestamibi gated SPECT (A) and between FPRNA and T1-201 gated SPECT (B). Bias: LVEF by FPRNA minus LVEF by gated SPECT. X-axis represents average of LVEFs by both FPRNA and gated SPECT.
20 10
B
,
-40
5O
0
..'., • :
•.
o
]
I
~
I
I
I
]
I
• 10
20
30
40
50
60
70
80
90
o First-Pass R a d i o n u c l i d e A n g i o g r a p h y (~)
Figure 3. Correlation between Tc-99m sestamibi gated SPECT and FPRNA (A) and between T1-201 gated SPECT and FPRNA (B) for LVEE
and validated it against F P R N A in 65 patients. Agreement with F P R N A was high (r = 0.9, P < .001, standard error of the estimate = 6.87). Nichols et al 6 developed a different algorithm to calculate LVEF and demonstrated a good correlation between gated SPECT and equilibrium radionuclide angiography (r = 0.87), as well as between gated SPECT and FPRNA (r = 0.87). Other studies have also demonstrated good correlations between gated SPECT and radionuclide angiography,5,7, 9 echocardiography, t0 contrast left ventriculography, 7,8 and magnetic resonance imaging. 8 The results from this study further confirm that T c - 9 9 m sestamibi gated SPECT is a good technique for measuring LVEF. Gated SPECT w i t h TI-201
G e r m a n • et al l• first reported a good correlation between sestamibi and T1-201 gated SPECT with respect to LVEF measurements (r = 0.918, SEE = 6.35). In a
group of 104 patients, Maunoury et a113 found a good correlation between T1-201 gated SPECT and Tc-99m sestamibi gated SPECT (r = 0.93). Our results are consistent with the findings from these two previous studies. LVEF b y Gated SPECT in P a t i e n t s w i t h D e p r e s s e d LV F u n c t i o n a n d M o d e r a t e t o S e v e r e P e r f u s i o n Abnormalities
In this study the correlation between Tc-99m sestamibi gated SPEC, T and T1-201 gated SPECT in patients with depressed LV function was excellent (r = 0.86). Thus our results suggest that even in patients with depressed LV function the LVEF can be accurately measured by gated SPECT with either Tc-99m sestamibi or T1-201. Manoury et a113 also found a good correlation (r = 0.91) between T c - 9 9 m sestamibi and T1-201 gated SPECT in patients with a reduced LVEF (<50%). In the subset of patients with moderate to severe perfusion abnormalities, in whom it may be more difficult to automatically determine the LV borders, the good correlation of gated SPECT with either T1-201 or Tc-99m sestamibi and F P R N A was maintained, with similar values obtained by all 3 techniques.
416
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
Journal of Nuclear Cardiology July/August 1999
60
g 50
I-
/
•
y = 5.0t ÷ 0.88 x r = 0.06, SEE = 6.3
40
•
.
. ~
*•-•50 •
"
~
/
I
y = 1.3317x - t6.418 R : 0.84, .SEE = 6.9
J
* /
40.
30 •
30-
E ~20 ¢/)
20-
=s
1.-
10
10-
k0
I
I
I
I
I
10
20
30
40
50
Tc-99m
Sestamibi gated-SPECT
0 60
(%)
A
Figure 5. Correlation between sestamibi and T1-201 gated SPECT for LVEF in patients with depressed LV function.
i p 10
LVEF in Patients with Reversible Defects
Study Limitations First, our first-pass studies were not performed in a true basal condition but during adenosine infusion. Although LV regional wall motion deterioration resulting from myocardial ischemia may occasionally develop during adenosine infusion, the prevalence of true myocardial ischemia in patients with coronary artery disease undergoing adenosine stress is very low. 16 Ogilby et al, 17 reporting on 15 patients with documented coronary artery disease, showed no significant change in contrast LVEF during adenosine infusion (73% at baseline versus 75% during adenosine infusion). Thus adenosine should not have had a major impact on the global LVEF by FPRNA nor on the comparison between FPRNA and gated SPECT. Second, we excluded patients who were very large or heavy. In our laboratory, patients weighing more than 200 lbs or women with large breasts or breast implants
30
40
50 (%)
y = 1.3902x - t7.S04 R = 0.02, SEE = 4.7 .
/ . .~" **
60
60-
50
In patients with perfusion defects, two issues are pertinent. First, it is conceivable that if myocardial ischemia occurs during stress and is followed by myocardial stunning, it might reduce the post-stress LVEF. This, of course, would be more likely to occur during exercise stress, when true ischemia is more likely to be triggered; moreover, the postexercise images are often initiated earlier after stress (15 to 30 minutes) than after pharmacologic stress (typically 1 hour). Nevertheless, in our subset of patients with reversible defects, there was no systematic difference between the LVEF assessed in the poststress Tc-99m sestamibi gated SPECT images and in the rest T1-201 gated SPECT images, supporting the notion that no substantial stunning occurred after adenosine stress.
20
First-Pass Radionuclide Angiography
I- 40
3
30.
~
20-
110-
0
B
r
10
20
30
40
Flint-Pass Radionuclide Angiography
50
60
(%)
Figure 6. Correlation between Tc-99m sestamibi gated SPECT and FPRNA (A) and between T1-201 gated SPECT and FPRNA (B) for LVEF in patients with depressed LV function:
are routinely studied with Tc-99m agents. These patients account for approximately 40% of patients referred to our laboratory for myocardial SPECT imaging. Although both Tc-99m agents and T1-201 have adequate accuracy for routine determination of LVEF in patients with average body build, we would still prefer to use Tc99m-based agents for patients with large body habitus, because of the better count statistics with the latter agents in comparison with T1-201. Third, most of our patients had LVEF > 50% by FPRNA. Although LVEF by gated SPECT with T1-201 and Tc-99m sestamibi were comparable, further studies in large numbers of patients with depressed LVEF should be performed. Finally, in our study we did not compare wall motion by the 2 different techniques. Because the thallium gated SPECT images generally have lower count statistics, the assessment of LV wall motion and right ventricular function might be potentially difficult. However, Germano et al t2 recently demonstrated a good correlation between T1-201 and sestamibi gated SPECT for wall motion and thickening assessment (kappa values of 0.619 and 0.586, respectively).
Journal of Nuclear Cardiology Volume 6, N u m b e r 4;412-7
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
References 1. Grucker D, Florentz P, Oswald T, Chambron J. Myocardial gated tomoscintigrapby with 99Tcm-methoxy-isobutyl-isonitrile (MIBI): regional and temporal activity curve analysis. Nucl Med Commun 1989;10:723-32. 2. Mannting F, Morgan-Mannting MG. Gated SPECT with technetium99m-sestamibi for assessment of myocardial perfusion abnormalities. J Nucl Med 1993;34:601-8. 3. Berman DS, Germano G. Evaluation of ventricular ejection fraction, wall motion, wall thickening, and other parameters with gated myocardial perfusion single-photon emission computed tomography. J Nucl Cardiol 1997;4:S169-71. 4. DePuey EG, Nichols K, Dobrinsky C. Left ventricular ejection fraction assessed from gated technetium-99m-sestamibi SPECT. J Nucl Med 1993;34:1871-6. 5. Germano G, Kiat H, Kavanagh PB, Moriel M, Mazzanti M, Su HT, et al. Automatic quantification of ejection fraction from gated myocardial perfusion SPECT. J Nucl Med 1995;36:2138-47. 6. Nichols K, DePuey EG, Rozanski A. Automation of gated tomographic left ventricular ejection fraction. J Nucl Cardiol 1996;3:475-82. 7. Williams KA, TaiUon LA. Left ventricular function in patients with coronary artery disease assessed by gated tomographic myocardial perfusion images. Comparison with assessment by contrast ventriculography and first-pass radionuclide angiography. J Am Coil Cardiol 1996;27:173-81. 8. Mochizuki T, Murase K, Tanaka H, Kondoh T, Hamamoto K, Tauxe WN. Assessment of left ventricular volume using ECG-gated SPECT with technetium-99m-MIBI and technetium-99m-tetrofosmin. J Nucl Med 1997;38:53-7. 9. Hambye AS, Van Den Branden F, Vandevivere J. Diagnostic value of
10.
11.
12.
13.
14,
15.
16.
17.
417
Tc-99m sestamibi gated SPECT to assess viability in a patient after acute myocardial infarction. Clin Nucl Med 1996;21:19-23. Miron SD, Finkelhor R, Penuel JH, Bahler R, Bellon EM. A geometric method of measuring the left ventricular ejection fraction on gated Tc99m sestamibi myocardial imaging. Clin Nucl Med 1996;21:439-44. Everaert H, Franken PR, Flamen P, Goris M, Momen A, Bossuyt A. Left ventricular ejection fraction from gated SPECT myocardial peffusion studies: a method based on the radial distribution of count rate density across the myocardial wall. Eur J Nucl Med 1996;23:1628-33. Germano G, Erel J, Kiat H, Kavanagh PB, Berman DS. Quantitative LVEF and qualitative regional function from gated thallium-201 perfusion SPECT. J Nucl Med 1997;38:749-54. Maunoury C, Chen CC, Chua KB, Thompson CJ. Quantification of left ventricular function with thallium-201 and technetium-99m-sestamibi myocardial gated SPECT. J Nucl Med 1997;38:958-61. Gallik DM, Obermueller SD, Swarna US, Guidry GW, Mahmarian JJ, Verani MS. Simultaneous assessment of myocardial perfusion and left ventricular function during transiet coronary occlusion. J Am Coil Cardiol 1995;25:1529-38. Mahmarian JJ, Boyce TM, Goldberg RK, Cocanougher MK, Roberts R, Verani, MS. Quantative exercise thallium-201 single photon emission computed tomography for the enhanced diagnosis of ischemic heart disease. J Am Coll Cardiol 1990;15:318-29. Fenster MS, Feldman MD, Camarano G, Johnson WH, Ellis M, Linden J, et al. Correlation of adenosine thallium-201 perfusion patterns with markers for inducible iscbemia. Am Heart J 1997;133:406-12. Ogilby JD, Iskandrian AS, Untereker WJ, Heo J, Nguyen TN, Mercuro J. Effect of intravenous adenosine infusion on myocardial perfusion and function: hemodynamic/angiographic and scintigraphic study. Circulation 1992;86:887-95.
Gated single photon emission computed tomography (SPECT) enables simultaneous assessment of myocardial perfusion and function.l-3 It is well accepted that left ventricular ejection fraction (LVEF) can be accurately measured by gated SPECT using technetium99m-labeled perfusion agents. T M Data are, however, scanty on the accuracy of this measurement by gated SPECT with thallium-201.12,13 Moreover, the accuracy of LVEF measurement by gated SPECT in patients with
depressed left ventricular (LV) function and severe perfusion defects has not been well investigated. Accordingly, this study compared LVEF assessed by gated SPECT by use of both Tc-99m sestamibi and T1201 with that measured by first-pass radionuclide angiography (FPRNA) in patients with a broad spectrum of LV function and peffusion defects.
From the Section of Cardiology, Baylor College of Medicine and The Methodist Hospital,Houston, Texas. Received for publication June 26, 1998; revision accepted Sept 16, 1998. Reprint requests: Mario S. Verani, MD, Baylor College of Medicine, The Methodist Hospital,6550 Fannin, SM 677, Houston,TX 77030;
Study Population
[email protected].
Copyright © 1999 by the AmericanSociety of Nuclear Cardiology. 1071-3581/99/$8.00 + 0 43/1/94581 412
MATERIAL AND METHODS
Patients referred for adenosine myocardial SPECT in our laboratory were randomly selected to undergo a dual isotope study consisting of rest T1-201 gated SPECT followed by adenosine Tc-99m sestamibi gated SPECT. Patients were excluded if they had contraindications to adenosine stress, significant cardiac arrhythmias, or large body habitus (weight > 200 lbs or women with large breasts or breast implants).
Journal of Nuclear Cardiology Volume 6, Number 4;412-7
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
Study Protocol
Table 1. D e m o g r a p h i c d a t a
Rest T1-201 gated SPECT acquisition was initiated 10 minutes after injection of T1-201 (3.1 + 0.2 mCi) with the patient at rest. Adenosine infusion (140 p.g/kg/min for 6 minutes) was begun within the next 30 minutes, Tc-99m sestamibi (30 -+ 2.0 mCi) was injected halfway into the adenosine infusion, and gated SPECT acquisition was initiated 1 hour later.
Total Men Women Age (yrs) Height (inches) Weight History Prior M! CABG PTCA Symptoms Chest pain Other Asymptomatic Risk factors Family history Diabetes mellitus Cigarette smoking Hypercholesterolemia Hypertension
FPRNA FPRNA was acquired immediately after the bolus injection of Tc-99m sestamibi on a SIM-400 multicrystal gamma camera and processed by an operator who was blinded to the gated SPECT results, using a commercially available software, 14 which uses 2 regions of interest for end diastole and end systole, respectively.
Nongated SPECT Nongated projection data were created by summing the gated data set at each projection into a nongated data set. Nongated myocardial tomograms were then reconstructed by use of a standard filtered back-projection algorithm followed by reorientation of the transaxial images into the short and horizontal and vertical long axes. Images were qualitatively and quantitatively interpreted by an expert nuclear cardiologist, as previously reported from our laboratory. 15 Quantification of myocardial perfusion defect size was performed with polar map analysis. The raw data polar maps for each patient were statistically compared with a normal data bank to determine the defect size and the extent of scar and ischemia. 15
Gated SPECT Gated myocardial SPECT was performed on a 90-degree configuration dual-head SPECT system equipped with a low-energy, high-resolution collimator. Sixty-four projections (32 per each detector) were acquired over a 180-degree anterior arc, from the 45-degree left posterior oblique view to the 45-degree right anterior oblique view. Eight frames per cardiac cycle were obtained at each projection. Acquisition time was 30 seconds per projection for T1-201 and 20 seconds per projection for Tc-99m sestamibi. The raw data of T1-201 and Tc-99m sestamibi gated SPECT study were visually inspected for the assessment of image quality. Only images considered to be of good quality were included in the study. T1-201 and Tc-99m gated sestamibi SPECT images were reconstructed by use of a back-projection algorithm with a three-dimensional Butterworth filter (cutoff frequencies of 0.35 Nyquist [pixel size 0.59 cm] and 0.5, respectively, and order 5). The transaxial gated tomographic slices were then reoriented into the short- and horizontal and vertical long-axis views. LVEF by gated SPECT was calculated by an independent operator using the Cedars Quantitative Gated SPECT software.5 This algorithm operates in the 3-dimensional space. It segments the LV, estimates and displays endocardial and epicardial surfaces for all 8 images in the cardiac cycle, calculates the relative LV volumes, and derives LVEF from the end-diastolic and -systolic volumes.5
63 41 22 67 68 176
413
(100%) (65%) (35%) _+ 10 _+4 _+ 30
19 (300/0) 16 (15%) 20 (32%) 30 (48%) 26 (41%) 7 (11%) 18 |6 20 27 39
(28%) (25%) (320/0) (43%) (62%)
Nil, Myocardial infarction; CABG, coronary artery bypast graft; PTCA, percutaneous transluminal coronary angioplasty.
Statistical Analysis Data are reported as mean _+SD or frequency when appropriate. Differences anaong LVEFs measured by gated SPECT with T1-201 and Tc-99m sestamibi and by FPRNA were evaluated by analysis of variance (ANOVA). Agreement among the gated SPECT techniques and FPRNA was evaluated by use of regression analysis (Pearson's r) and Bland-Altman analysis. A P value less than .05 was considered statistically significant.
RESULTS Demographic Data A total of 63 patients were studied. Clinical information on these patients is summarized in Table 1.
Results of Nongated SPECT Myocardial perfusion was normal in 32 patients and abnormal in 31 patients. Perfusion defects were completely reversible in 6 patients, partially reversible in 15 patients, and fixed in 10 patients. Total defect size in patients with an abnormal SPECT measured 19% _+ 15% of the LV.
LVEF by Gated SPECT and by FPRNA L V E F averaged 54% _ 15% b y T1-201 gated SPECT, 54% _+ 16% by Tc-99m sestamibi gated SPECT, and 54% _+ 12% by F P R N A (ANOVA, P = ns).
414
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
Journal of Nuclear Cardiology July/August 1999
90
y=2.84+O.96x r = 0.92, SEE = 6.1 p
80
7o
• ~ "
,
50 ¸
~ ~'~
40
•
t-¢.) 6 0 IM
• .
,~ s0
•
l~:/."
eee
• •e~
•
30 •
.
20
•
"13
10
40
m O1 • - 30
•
•
.=
•
*
6 e°o
0 •
...
O
";.~."".-
~O O
-10
"~' 2 0 I--
10
-20 I
[
r
]
L
I
i
I
10
20
30
40
50
60
70
80
-30 90
Tc-g9m Sestamibi gated-SPECT (%)
Figure 1. Correlation between Tc-99m sestamibi and T1-201 gated SPECT for LVEE
Agreement Between 11-201 and Tc-99m Sestamibi Gated SPECT There was an excellent linear correlation between Tc-99m sestamibi and T1-201 gated SPECT (Pearson's r = 0.92, P <.0001) (Figure 1). Bland-Altman analysis showed no evidence of any systematic bias between these two techniques (Figure 2).
Agreement Between SPECT Techniques and FPRNA There was a good linear correlation between Tc-99m sestamibi gated SPECT and FPRNA (Pearson's r = 0.85, P < .0001), as well as between T1-201 gated SPECT and FPRNA (Pearson's r = 0.84, P < .0001) (Figure 3, A and B). Bland-Altman analysis showed no obvious bias between sestamibi gated SPECT and FPRNA (Figure 4, A) or between T1-201 gated SPECT and FPRNA (Figure 4, B).
LVEF by Gated SPECT in Patients with Depressed LV Function In the subset of 16 patients with an LVEF < 50% by FPRNA, there was a good correlation between Tc-99m sestamibi gated SPECT and T1-201 gated SPECT (Pearson's r = 0.86, P < .0001) (Figure 5). There was also a good correlation between Tc-99m sestamibi gated SPECT and FPRNA (Pearson's r = 0.84, P < .0001), as well as between T1-201 gated SPECT and FPRNA (Pearson's r = 0.92, P < .0001) (Figure 6, A and B).
-40 -50
i
i
I
I
I
[
E
]
I
10
20
30
40
50
60
70
80
90
100
LVEF by T¢-9gm Sestamibi and TI-201 gated-SPECT
Figure 2. Bland-Altman analysis between Tc-99m sestamibi and T1-201 gated SPECT for LVEF. Bias: LVEF by sestamibi gated SPECT minus LVEF by T1-201 gated SPECT. X-axis represents average of LVEF by both Tc-99m sestamibi and T1-201 gated SPECT. (ANOVA, P = NS). The correlation coefficient was 0.88 (P < .0001) between T1-201 gated SPECT and FPRNA, and 0.92 (P < .0001) between Tc-99m sestamibi gated SPECT and FPRNA.
LVEF in Patients with Moderate to Severe Perfusion Defects In the 13 patients who had moderate to severe perfusion defects, the LVEF was 47% _ 12% by T1-201 gated SPECT, 41% _+ 15% by Tc-99m sestamibi gated SPECT, and 44% _ 11% by FPRNA (ANOVA, P = NS). The correlation coefficient was 0.88 (P < .0001) between T1-201 gated SPECT and FPRNA, and 0.92 (P < .0001) between Tc-99m sestamibi gated SPECT and FPRNA.
DISCUSSION To the best of our knowledge, this is the first study comparing LVEF assessed by T1-201 gated SPECT, Tc99m sestamibi gated SPECT, and FPRNA in the same patients. Our study demonstrates that (1) T1-201 and Tc99m gated SPECT correlate very well with respect to determination of LVEF; and (2) both T1-201 and Tc-99m gated SPECT correlate well with FPRNA in properly selected patients, including those with depressed LV function and moderate to severe perfusion abnormalities.
LVEF in Patients with Reversible Perfuslon Defects In the subset of 21 patients who had completely or partially reversible perfusion defects, the LVEF was 48% ___13% by T1-201 gated SPECT, 45% + 15% by Tc-99m sestamibi gated SPECT, and 49% + 13% by FPRNA
Gated SPECT with Tc-99m-Labeled Agents Several algorithms have been recently proposed to calculate LVEF by gated SPECT. German• et al 5 developed a completely automatic algorithm to quantify LVEF
Journal of Nuclear Cardiology Volume 6, Number 4;412-7
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
50 ~
100
.c.
90
I-
8O
415
40 30 -
t.) IM a.
**
y : 1.1107x- 0.84 R:0.85, SEE = 8.3 p < 0.0001
70
•
~
"o
• •
~ 6o
f
20-
•
:
~
• • $
•
10 ~
m= 0•
(4)
50
E ~
40
-30 -
~
30
-40-50
A
117
,, •
i
t
I
[
[
I
E
I
I
10
20
30
40
50
60
70
80
90
100
LVEF by FPRNA and Tc-99rnSestamibi gated-SPECT
5O
0 10
A
• 4`
-20
~ 2o k-
•
-10
e~oo 48e * e • ~
--
20
30
40
50
60
70
80
40
90
First-Pass R a d i o n u c l i d e A n g i o g r a p h y (%)
30 20
100
.:
10
90 y : 1.0868x - 8.600 R : 0.84, SEE = 8.4
8o
•o
el
•
p<0.0001
-~ 70 I0 m 60
¢9
" :
• ~ . im ~e •
o•
. / J "
-10
"
-20
.'~."
•
s•
s
v•
."
,
• *
-30
oe
-50
:
•: . "
B
40
I
I
I
]
I
[
I
[
I
10
20
30
40
50
60
70
80
90
100
LVEF by FPRNAand TI-201 gated-SPECT
__ 3o
Figure 4. Bland-Altman analysis between FPRNA and Tc-99m sestamibi gated SPECT (A) and between FPRNA and T1-201 gated SPECT (B). Bias: LVEF by FPRNA minus LVEF by gated SPECT. X-axis represents average of LVEFs by both FPRNA and gated SPECT.
20 10
B
,
-40
5O
0
..'., • :
•.
o
]
I
~
I
I
I
]
I
• 10
20
30
40
50
60
70
80
90
o First-Pass R a d i o n u c l i d e A n g i o g r a p h y (~)
Figure 3. Correlation between Tc-99m sestamibi gated SPECT and FPRNA (A) and between T1-201 gated SPECT and FPRNA (B) for LVEE
and validated it against F P R N A in 65 patients. Agreement with F P R N A was high (r = 0.9, P < .001, standard error of the estimate = 6.87). Nichols et al 6 developed a different algorithm to calculate LVEF and demonstrated a good correlation between gated SPECT and equilibrium radionuclide angiography (r = 0.87), as well as between gated SPECT and FPRNA (r = 0.87). Other studies have also demonstrated good correlations between gated SPECT and radionuclide angiography,5,7, 9 echocardiography, t0 contrast left ventriculography, 7,8 and magnetic resonance imaging. 8 The results from this study further confirm that T c - 9 9 m sestamibi gated SPECT is a good technique for measuring LVEF. Gated SPECT w i t h TI-201
G e r m a n • et al l• first reported a good correlation between sestamibi and T1-201 gated SPECT with respect to LVEF measurements (r = 0.918, SEE = 6.35). In a
group of 104 patients, Maunoury et a113 found a good correlation between T1-201 gated SPECT and Tc-99m sestamibi gated SPECT (r = 0.93). Our results are consistent with the findings from these two previous studies. LVEF b y Gated SPECT in P a t i e n t s w i t h D e p r e s s e d LV F u n c t i o n a n d M o d e r a t e t o S e v e r e P e r f u s i o n Abnormalities
In this study the correlation between Tc-99m sestamibi gated SPEC, T and T1-201 gated SPECT in patients with depressed LV function was excellent (r = 0.86). Thus our results suggest that even in patients with depressed LV function the LVEF can be accurately measured by gated SPECT with either Tc-99m sestamibi or T1-201. Manoury et a113 also found a good correlation (r = 0.91) between T c - 9 9 m sestamibi and T1-201 gated SPECT in patients with a reduced LVEF (<50%). In the subset of patients with moderate to severe perfusion abnormalities, in whom it may be more difficult to automatically determine the LV borders, the good correlation of gated SPECT with either T1-201 or Tc-99m sestamibi and F P R N A was maintained, with similar values obtained by all 3 techniques.
416
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
Journal of Nuclear Cardiology July/August 1999
60
g 50
I-
/
•
y = 5.0t ÷ 0.88 x r = 0.06, SEE = 6.3
40
•
.
. ~
*•-•50 •
"
~
/
I
y = 1.3317x - t6.418 R : 0.84, .SEE = 6.9
J
* /
40.
30 •
30-
E ~20 ¢/)
20-
=s
1.-
10
10-
k0
I
I
I
I
I
10
20
30
40
50
Tc-99m
Sestamibi gated-SPECT
0 60
(%)
A
Figure 5. Correlation between sestamibi and T1-201 gated SPECT for LVEF in patients with depressed LV function.
i p 10
LVEF in Patients with Reversible Defects
Study Limitations First, our first-pass studies were not performed in a true basal condition but during adenosine infusion. Although LV regional wall motion deterioration resulting from myocardial ischemia may occasionally develop during adenosine infusion, the prevalence of true myocardial ischemia in patients with coronary artery disease undergoing adenosine stress is very low. 16 Ogilby et al, 17 reporting on 15 patients with documented coronary artery disease, showed no significant change in contrast LVEF during adenosine infusion (73% at baseline versus 75% during adenosine infusion). Thus adenosine should not have had a major impact on the global LVEF by FPRNA nor on the comparison between FPRNA and gated SPECT. Second, we excluded patients who were very large or heavy. In our laboratory, patients weighing more than 200 lbs or women with large breasts or breast implants
30
40
50 (%)
y = 1.3902x - t7.S04 R = 0.02, SEE = 4.7 .
/ . .~" **
60
60-
50
In patients with perfusion defects, two issues are pertinent. First, it is conceivable that if myocardial ischemia occurs during stress and is followed by myocardial stunning, it might reduce the post-stress LVEF. This, of course, would be more likely to occur during exercise stress, when true ischemia is more likely to be triggered; moreover, the postexercise images are often initiated earlier after stress (15 to 30 minutes) than after pharmacologic stress (typically 1 hour). Nevertheless, in our subset of patients with reversible defects, there was no systematic difference between the LVEF assessed in the poststress Tc-99m sestamibi gated SPECT images and in the rest T1-201 gated SPECT images, supporting the notion that no substantial stunning occurred after adenosine stress.
20
First-Pass Radionuclide Angiography
I- 40
3
30.
~
20-
110-
0
B
r
10
20
30
40
Flint-Pass Radionuclide Angiography
50
60
(%)
Figure 6. Correlation between Tc-99m sestamibi gated SPECT and FPRNA (A) and between T1-201 gated SPECT and FPRNA (B) for LVEF in patients with depressed LV function:
are routinely studied with Tc-99m agents. These patients account for approximately 40% of patients referred to our laboratory for myocardial SPECT imaging. Although both Tc-99m agents and T1-201 have adequate accuracy for routine determination of LVEF in patients with average body build, we would still prefer to use Tc99m-based agents for patients with large body habitus, because of the better count statistics with the latter agents in comparison with T1-201. Third, most of our patients had LVEF > 50% by FPRNA. Although LVEF by gated SPECT with T1-201 and Tc-99m sestamibi were comparable, further studies in large numbers of patients with depressed LVEF should be performed. Finally, in our study we did not compare wall motion by the 2 different techniques. Because the thallium gated SPECT images generally have lower count statistics, the assessment of LV wall motion and right ventricular function might be potentially difficult. However, Germano et al t2 recently demonstrated a good correlation between T1-201 and sestamibi gated SPECT for wall motion and thickening assessment (kappa values of 0.619 and 0.586, respectively).
Journal of Nuclear Cardiology Volume 6, N u m b e r 4;412-7
He et al LVEF determined by SPECT with T1-201 and Tc-99m sestamibi
References 1. Grucker D, Florentz P, Oswald T, Chambron J. Myocardial gated tomoscintigrapby with 99Tcm-methoxy-isobutyl-isonitrile (MIBI): regional and temporal activity curve analysis. Nucl Med Commun 1989;10:723-32. 2. Mannting F, Morgan-Mannting MG. Gated SPECT with technetium99m-sestamibi for assessment of myocardial perfusion abnormalities. J Nucl Med 1993;34:601-8. 3. Berman DS, Germano G. Evaluation of ventricular ejection fraction, wall motion, wall thickening, and other parameters with gated myocardial perfusion single-photon emission computed tomography. J Nucl Cardiol 1997;4:S169-71. 4. DePuey EG, Nichols K, Dobrinsky C. Left ventricular ejection fraction assessed from gated technetium-99m-sestamibi SPECT. J Nucl Med 1993;34:1871-6. 5. Germano G, Kiat H, Kavanagh PB, Moriel M, Mazzanti M, Su HT, et al. Automatic quantification of ejection fraction from gated myocardial perfusion SPECT. J Nucl Med 1995;36:2138-47. 6. Nichols K, DePuey EG, Rozanski A. Automation of gated tomographic left ventricular ejection fraction. J Nucl Cardiol 1996;3:475-82. 7. Williams KA, TaiUon LA. Left ventricular function in patients with coronary artery disease assessed by gated tomographic myocardial perfusion images. Comparison with assessment by contrast ventriculography and first-pass radionuclide angiography. J Am Coil Cardiol 1996;27:173-81. 8. Mochizuki T, Murase K, Tanaka H, Kondoh T, Hamamoto K, Tauxe WN. Assessment of left ventricular volume using ECG-gated SPECT with technetium-99m-MIBI and technetium-99m-tetrofosmin. J Nucl Med 1997;38:53-7. 9. Hambye AS, Van Den Branden F, Vandevivere J. Diagnostic value of
10.
11.
12.
13.
14,
15.
16.
17.
417
Tc-99m sestamibi gated SPECT to assess viability in a patient after acute myocardial infarction. Clin Nucl Med 1996;21:19-23. Miron SD, Finkelhor R, Penuel JH, Bahler R, Bellon EM. A geometric method of measuring the left ventricular ejection fraction on gated Tc99m sestamibi myocardial imaging. Clin Nucl Med 1996;21:439-44. Everaert H, Franken PR, Flamen P, Goris M, Momen A, Bossuyt A. Left ventricular ejection fraction from gated SPECT myocardial peffusion studies: a method based on the radial distribution of count rate density across the myocardial wall. Eur J Nucl Med 1996;23:1628-33. Germano G, Erel J, Kiat H, Kavanagh PB, Berman DS. Quantitative LVEF and qualitative regional function from gated thallium-201 perfusion SPECT. J Nucl Med 1997;38:749-54. Maunoury C, Chen CC, Chua KB, Thompson CJ. Quantification of left ventricular function with thallium-201 and technetium-99m-sestamibi myocardial gated SPECT. J Nucl Med 1997;38:958-61. Gallik DM, Obermueller SD, Swarna US, Guidry GW, Mahmarian JJ, Verani MS. Simultaneous assessment of myocardial perfusion and left ventricular function during transiet coronary occlusion. J Am Coil Cardiol 1995;25:1529-38. Mahmarian JJ, Boyce TM, Goldberg RK, Cocanougher MK, Roberts R, Verani, MS. Quantative exercise thallium-201 single photon emission computed tomography for the enhanced diagnosis of ischemic heart disease. J Am Coll Cardiol 1990;15:318-29. Fenster MS, Feldman MD, Camarano G, Johnson WH, Ellis M, Linden J, et al. Correlation of adenosine thallium-201 perfusion patterns with markers for inducible iscbemia. Am Heart J 1997;133:406-12. Ogilby JD, Iskandrian AS, Untereker WJ, Heo J, Nguyen TN, Mercuro J. Effect of intravenous adenosine infusion on myocardial perfusion and function: hemodynamic/angiographic and scintigraphic study. Circulation 1992;86:887-95.