Background: Due to the limitation of spatial resolution, cardiac nuclear medicine images have not been applied to feature-tracking method to automatic extraction of myocardial contours. We have succes
1. Germano G, Erel J, Kiat H, Kavanagh PB, Berman DS. Quantitative LVEF and quantitative regional
function from gated thallium-201 perfusion SPECT. J Nucl Med. 1997; 38: 749-54.
2. Nakao R, Nagao M, Yamamoto A, et al. Papillary muscle ischemia on high-resolution cine imaging of
nitrogen-13 ammonia positron emission tomography: Association with myocardial flow reserve and
prognosis in coronary artery disease. J Nucl Cardiol 2020; doi:10.1007/s12350-020-02231-z.
3. Ishizaki U, Nagao M, Shiina Y, Inai K, Mori H, Takahashi T, et al. Global strain and dyssynchrony of
the single ventricle predict adverse cardiac events after the Fontan procedure: Analysis using featuretracking cine magnetic resonance imaging. J Cardiol 2019;73:163-70.
4. Kawakubo M, Nagao M, Yamamoto A, et al. 13N-ammonia positron emission tomography-derived
endocardial strain for the assessment of ischemia using feature-tracking in high-resolution cine imaging.
J Nucl Cardiol. 2021; doi: 10.1007/s12350-021-02677-9.
5. Cerqueira MD, Weissman, NJ, Dilsizian V, et al. Standardized myocardial segmentation and
nomenclature for tomographic imaging of the heart. J Cardiovasc Magn Reson 2002; 4: 203-10.
6. Yoshinaga K, Chow BJ, Williams K, Chen L, deKemp RA, Garrard L, et al. What is the prognostic
value of myocardial perfusion imaging using rubidium-82 positron emission tomography? J Am Coll
Cardiol 2006; 48: 1029-39.
7. Yamasaki Y, Abe K, Kamitani T, et al. Balloon pulmonary angioplasty improves right atrial reservoir
and conduit functions in chronic thromboembolic pulmonary hypertension. Eur Heart J Cardiovasc
Imaging 2020; 21: 855-62.
8. Kido T, Nagao M, Kido T, et al. Stress/rest circumferential strain in non-ischemia, ischemia, and
infarction: Quantification by 3 tesla tagged magnetic resonance imaging. Circ J 2013; 77: 1235-41.
9. Kawakubo M, Nagao M, Kikuchi N, et al. 13N-ammonia positron emission tomography-derived leftventricular strain in patients after heart transplantation validated using cardiovascular magnetic
resonance feature tracking as reference. Ann Nucl Med 2022; 36: 70-81.
10. Kawakubo M, Nagao M, Yamamoto A, et al. 13N-ammonia PET-derived right ventricular longitudinal
11
strain and myocardial flow reserve in right coronary artery disease. Eur J Nucl Med Mol Imaging 2022;
49: 1870-1880.
12
FIGURE LEGENDS
Fig. 1 Strain analysis using feature-tracking method
A: First, five to seven points are set manually at the end of diastolic frame, including the start and end points
of the region (upper row). The endocardial contour connecting the points is then automatically drawn (lower
row). The left is horizontal long axis image and the right is vertical long axis image. Yellow is right coronary
artery (RCA) territory, blue is left anterior descending artery (LAD) territory, and pink is left circumflex
artery (LCX) territory.
B: Feature-tracking method consists of a local template-matching technique and spline interpolation of
points tracking. The elongation and contraction of the endocardial contours are automatically extracted
throughout the cardiac cycle.
C: The graph shows normalized a time-strain curve for each region. The largest absolute value (red arrow)
on the strain curve was defined as longitudinal strain (LS) and was used as the representative value of the
patient.
Fig. 2 Comparison of longitudinal strain (LS) and longitudinal strain ratio (LSR) between ischemia
and non-ischemia and between stress and rest conditions.
A: RCA territory
In the ischemia, stress LS was significantly smaller than rest LS. In the non-ischemia, there was no
significant difference between stress LS and rest LS. Stress LS was significantly smaller in ischemia than
in non-ischemia (left). LSR was significantly lower for ischemia than non-ischemia (right). *: p<0.05, **:
p<0.01
B: LAD territory
In the ischemia, stress LS was significantly smaller than rest LS. In the non-ischemia, there was no
significant difference between stress LS and rest LS. Stress LS was smaller in ischemia than in nonischemia, but there was no significant difference between them (left). LSR was significantly lower for
ischemia than non-ischemia (right). *: p<0.05, **: p<0.01
13
C: LCX territory
In the ischemia, stress LS was significantly smaller than rest LS. In the non-ischemia, there was no
significant difference between stress LS and rest LS. Stress LS was smaller in ischemia than in nonischemia, but there was no significant difference between them (left). LSR was significantly lower for
ischemia than non-ischemia (right). *: p<0.05, **: p<0.01
Fig. 3 Diagnostic performance of longitudinal strain (LS) and longitudinal strain ratio (LSR) for
myocardial ischemia
In the RCA and LAD territories, the ischemia diagnostic performance of LSR was superior to that of stress
LS and rest LS. The AUC of LSR in the RCA and LAD territories showed good diagnostic performance of
more than 0.8. On the other hand, the AUC of LSR in the LCS territory was low as 0.69 and comparable to
that of stress LS.
Fig. 4 13N-ammonia PET of a man in his 80s
Gray-scale high-resolution cine images (left and center) show a defect in the mid-anterior wall and
hypointensity in the inferior wall during adenosine-stress, with normal accumulation at rest. Perfusion maps
(right) show transient ischemia in the mid-anterior wall and inferior wall. Longitudinal strain (LS) in the
RCA territory is 31% under stress and 28% at rest, LS in the LAD territory is 19% under stress and 22% at
rest, and LS in the LCX territory is 20% under stress and 28% at rest. In the LAD and LCX territories, LS
is decreasing predominantly under adenosine-stress.
14
Fig. 1A
Fig. 1B
Fig. 1C
Fig. 2A
Regional Longitudinal Strain (%)
Ischemia
Non-Ischemia
Rest
Ischemia
Non-Ischemia
Regional LSR (%)
Stress
Ischemia
Non-Ischemia
LSR=longitudinal strain ratio
Fig. 2B
Regional Longitudinal Strain (%)
Ischemia
Non-Ischemia
Rest
Ischemia
Non-Ischemia
Regional LSR (%)
Stress
Ischemia
Non-Ischemia
LSR=longitudinal strain ratio
Fig. 2C
Regional Longitudinal Strain (%)
Ischemia
Non-Ischemia
Rest
Ischemia
Non-Ischemia
Regional LSR (%)
Stress
Non-Ischemia
Ischemia
LSR=longitudinal strain ratio
Fig. 3
LSR
Stress LS
Resting LS
Cut-off
AUC
LSR
0.78
0.82
Stress
-9.2%
Rest
-51.8%
1-specificity
LCX
LSR
Stress LS
Resting LS
Cut-off
AUC
LSR
0.89
0.86
0.65
Stress
-9.1%
0.52
Rest
-31.7%
1-specificity
Sensitivity
LAD
Sensitivity
Sensitivity
RCA
LSR
Stress LS
Resting LS
Cut-off
AUC
LSR
0.82
0.69
0.62
Stress
-7.1%
0.64
0.61
Rest
-10.3%
0.52
1-specificity
LSR=longitudinal strain ratio
LS=longitudinal strain
Fig. 4
Stress
Rest
Table 1. Patient Characteristics
Number of patients
95
Age (years)
68±11
Male/Female
62/33
Cardiovascular risk factors
Hypertension
69 (73%
Dyslipidemia
68 (72%)
Diabetes mellitus
44 (46%)
Past or current smoking
42 (44%)
Familiry history of CAD
22 (23%)
Clinical history of CAD
Myocardial infaction
14 (15%)
Percutaneous coronary intervension
21 (22%)
Coronary artery bypass grafting
CAD = coronary artery disease
8 (8%)
Table 2
Intra-observer reproducibility
Parameter
Inter-observer reproducibility
Bias
(LOA)
SDD
ICC
(95%CI)
Bias
(LOA)
SDD
ICC
(95%CI)
0.5
(-3.0 to 4.1)
1.8
0.98
(0.98 to 0.99)
2.4
(-1.9 to 6.7)
2.2
0.97
(0.97 to 0.98)
0.7
1.00
1.0
0.99
(0.99 to 1.00)
1.3
(-2.2 to 4.7)
1.6
(-2.0 to 5.2)
Stress LS
RCA
LAD
LCX
0.1
(-1.2 to 1.4)
-0.1
(-2.0 to 1.8)
1.8
1.8
0.98
(0.97 to 0.98)
0.98
(0.97 to 0.98)
Rest LS
RCA
-0.1
(-2.0 to 1.8)
1.0
0.99
(0.99 to 1.00)
1.6
(-2.0 to 5.2)
1.8
0.98
(0.97 to 0.98)
LAD
0.0
(-1.7 to 1.7)
0.9
0.99
(0.99 to 1.00)
0.9
(-2.5 to 4.3)
1.7
0.98
(0.97 to 0.98)
LCX
-0.1
(-2.0 to 1.8)
1.0
0.99
(0.99 to 1.00)
1.6
(-2.0 to 5.2)
1.8
0.98
(0.97 to 0.98)
LS=longitudinal strain, LOA=limit of agreement, SDD=standard deviation of difference,
ICC=intraclass correlation coefficient, CI=coefficient interval
...
論文の公開元へ