384 Original research

Colocalization of thin-cap fibroatheroma and spotty

calcification is a powerful predictor of procedure-related
myocardial injury after elective coronary stent implantation
Tomoya Ueda, Shiro Uemura, Makoto Watanabe, Yu Sugawara,
Tsunenari Soeda, Satoshi Okayama, Yukiji Takeda, Hiroyuki Kawata,
Rika Kawakami and Yoshihiko Saito

Objectives Percutaneous coronary intervention (PCI)- findings was a powerful predictor of PCI-related cTnT
related cardiac enzyme elevation is an independent risk elevation (odds ratio 8.40, 95% confidence interval
factor for adverse clinical outcomes, but preprocedural 1.65–52.78, P < 0.01). Further, the predictive value of this
predictors of this complication have not been established. colocalization was enhanced when the analysis included
This study evaluated the morphological characteristics of only spotty calcification (odds ratio 21.00, 95% confidence
culprit lesions by frequency-domain optical coherence interval 2.65–454.22, P = 0.003).
tomography (FD-OCT), and examined their predictive value
Conclusion FD-OCT examination showed that
for procedure-related myocardial injury in patients
colocalization of thin-cap fibroatheroma and spotty
undergoing elective PCI.
calcification was a powerful predictor of PCI-related cTnT
Methods Sixty-eight patients treated by FD-OCT-guided elevation. FD-OCT is useful for stratifying risk during PCI to
elective PCI were studied. On the basis of the presence avoid procedure-related complications. Coron Artery Dis
or absence of postprocedural plasma cardiac troponin 25:384–391 c 2014 Wolters Kluwer Health | Lippincott
T (cTnT) elevation, patients were divided into elevation Williams & Wilkins.
(cTnT-E, n = 25) and nonelevation (cTnT-nonE, n = 43) groups. Coronary Artery Disease 2014, 25:384–391
FD-OCT examinations of culprit lesions were performed
before and after stent implantation, and tissue characteristics Keywords: complications, optical coherence tomography, percutaneous
coronary intervention
were evaluated within a 10-mm-long segment of each lesion.
First Department of Internal Medicine, Nara Medical University, Nara, Japan
Results Clinical parameters were similar between the
Correspondence to Shiro Uemura, MD, PhD, First Department of Internal
two groups. Stent length was significantly longer in the Medicine, Nara Medical University, 840 Shijo-cho, Kashihara,
cTnT-E group than in the cTnT-nonE group. On baseline Nara 634-8522, Japan
Tel: + 81 744 22 3051 ¾ 3411; fax: + 81 744 22 9726;
OCT images, thin-cap fibroatheroma and calcium deposition e-mail: suemura@[Link]
were more frequently observed within culprit segments
Received 21 January 2014 Revised 20 February 2014
of the cTnT-E group compared with the cTnT-nonE group Accepted 24 February 2014
(32.0 vs. 11.6%, P = 0.043, and 72.0 vs. 46.5%, P = 0.039,
respectively). In addition, colocalization of these two

Introduction that several characteristic IVUS findings, including large

Percutaneous coronary intervention (PCI) with coronary stent plaque burden, attenuated plaque, and intracoronary
implantation is an established treatment for patients with mobile mass, were risk factors. However, IVUS is limited
coronary artery disease. However, several earlier studies have by its relatively low spatial resolution and the difficulty in
revealed that an increase in plasma levels of cardiac enzymes, differentiating plaque components within culprit lesions.
one of the surrogate markers for myocardial injury, was found
in 10 to almost 50% of patients who underwent elective Intravascular frequency-domain optical coherence tomogra-
PCI [1,2], and that the elevation of these enzymes after PCI phy (FD-OCT) is an intravascular imaging modality with a
was an independent risk for major adverse outcomes during maximal spatial resolution of 10 mm, which is B10-fold
long-term follow-up [3–5]. Therefore, it is important to higher than that of IVUS. In particular, tissue characteriza-
elucidate the factors related to postprocedural myocardial tion by OCT has been validated by pathohistological studies,
injury and to prevent the occurrence of this complication, and OCT is the only clinical modality that can precisely
particularly in patients who undergo elective PCI. determine the thickness of the fibrous cap covering the lipid
pool and the size of calcium deposits, as well as distinguish
Tissue characterization of coronary atherosclerotic lesions macrophage infiltration and intraluminal thrombi [9–11].
with intravascular ultrasonography (IVUS) has been used
to predict the no-reflow phenomenon and PCI-related Accordingly, we hypothesized that FD-OCT may allow us
cardiac enzyme elevation [6–8]. These studies indicated to identify which morphological characteristics of culprit
0954-6928 c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins DOI: 10.1097/MCA.0000000000000114

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 OCT predictor of post-PCI myocardial injury Ueda et al. 385

coronary lesions tightly correlate with the development of lactated Ringer’s solution directly from the guiding
PCI-related myocardial injury. In this study, we evaluated catheter. The first OCT run was conducted before either
the baseline morphological characteristics of culprit balloon predilatation or direct stent implantation, and the
coronary plaques by FD-OCT, and studied the relation- final OCT run was performed just after stent implanta-
ship between these characteristics and procedure-related tion or balloon postdilatation if these were undertaken.
myocardial injury using cardiac troponin T (cTnT) as a
biomarker in patients undergoing elective PCI.
Frequency-domain optical coherence tomography
image analysis
Methods Obtained FD-OCT images were processed and analyzed
Study population offline using proprietary software from LightLab Imaging.
From November 2009 to November 2012, 371 patients
Planimetric parameters of FD-OCT images included
treated with PCI were prospectively registered in the FD-
OCT database of Nara Medical University Hospital. All lumen area at both proximal and distal reference sites,
minimal lumen area (MLA), and minimal lumen diameter.
patients underwent FD-OCTexamination of culprit lesions
In addition, longitudinal length and maximum angle of lipid
before and after index coronary stent implantation. In this
accumulation within culprit segments were measured.
study, we enrolled stable patients who had undergone
elective single coronary stent implantation with FD-OCT Morphological evaluation of FD-OCT images was per-
guidance. Patients with any of the following were excluded formed on each 10-mm-long segment between 5 mm
from the study: (i) acute coronary syndrome, (ii) totally proximal and 5 mm distal to the site of MLA. To evaluate
occluded lesions, (iii) left main and major bifurcation the relation between baseline tissue characteristics of
lesions, (iv) severely calcified lesions requiring rotational culprit lesions and postprocedural cTnT elevation, we
atherectomy, (v) suboptimal angiographic results such as determined the existence of the following 10 OCT-based
coronary dissection or side-branch occlusion, (vi) PCI lesion characteristics within the culprit segment on the
performed for two or more separate lesions during one basis our previous study [12]: (a) eccentric plaque
procedure, (vii) in-stent restenosis, and (viii) poor OCT distribution, (b) concave lumen shape, (c) ruptured
quality. Other exclusion criteria were reduced renal plaque, (d) microchannel, (e) lipid pool, (f) large lipid
function (estimated glomerular filtration rate < 30 ml/ pool covered with thin fibrous cap [thin-cap fibroatheroma
min/1.73 m2), hemodialysis, and declining to participate. (TCFA)], (g) macrophage image, (h) calcium deposition,
We ultimately enrolled 68 patients treated with stent (i) spotty calcification, and (j) luminal thrombus (Fig. 1).
implantation for a single, native, de-novo coronary lesion. Microchannels were defined as nonsignal tubuloluminal
Written informed consent was obtained from all patients. structures without a connection to the vessel lumen,
recognized on more than three consecutive cross-sectional
Diagnostic angiography and therapeutic procedure OCT images. TCFA was defined as a large lipid pool (one
Diagnostic coronary angiography was performed in all or more quadrants) covered by a thin fibrous cap (cap
patients after intracoronary administration of nitroglycer- thickness r 65 mm). Spotty calcification was defined as
in (0.5 mg). All patients received aspirin and clopidogrel the presence of lesions less than 4 mm in length,
before PCI, and PCI procedures were performed with a containing an arc of calcification of less than 901 among
6 or 7 Fr guiding catheter using the radial or femoral all types of calcifications [13,14].
approach after an intravenous bolus injection of 10 000 IU Poststent analysis by FD-OCT was also carried out on the
heparin. Coronary stents were selected on the basis of final OCT run, and existence of edge dissections, intrastent
operator discretion. Successful PCI was defined by the thrombus, and malapposition was examined [15,16]. Edge
angiographic criterion of less than 25% residual stenosis. dissection was defined as the disruption of the luminal
None of the patients showed angiographically evident no- vessel surface at both stent edge segments within 5 mm
reflow. proximal and distal to the stent. Intrastent thrombus was
defined as an irregular mass with dorsal shadowing that
Optical coherence tomography image acquisition protruded into the lumen or a luminal mass with dorsal
After a routine coronary angiography, an intravascular FD- shadowing that was not connected to the vessel wall.
OCT catheter (C7 system, Dragonfly imaging catheter; Malapposition was defined as a clear separation between
LightLab Imaging Inc./St Jude Medical, Westford, at least one stent strut and the vessel wall.
Massachusetts, USA) was inserted into culprit arteries
Cross-sectional FD-OCT images with substantial motion
through a guiding catheter before stent implantation, and
artifact were removed from the assessment.
then advanced to the distal end of the culprit lesion.
OCT images were recorded at 20 mm/s (5 frames/mm)
after intracoronary administration of nitroglycerin (0.5 mg). Cardiac troponin T assay
During image acquisition, coronary blood flow was Blood samples were obtained from all patients between
replaced by continuous flushing with contrast media or 10 and 20 h after the completion of PCI, and cTnT was

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
386 Coronary Artery Disease 2014, Vol 25 No 5

Fig. 1

Ten OCT-based tissue characteristics of culprit lesions. (a) Eccentric plaque distribution, (b) concave lumen shape, (c) ruptured plaque,
(d) microchannel, (e) lipid pool, (f) large lipid pool covered with thin fibrous cap (thin-cap fibroatheroma), (g) macrophage image, (h) calcium
deposition, (i) spotty calcification, and (j) luminal thrombus. OCT, optical coherence tomography.

measured using the electrochemiluminescence immu- and medical history. Baseline laboratory data and medica-
noassay method. In accordance with the third universal tions did not differ between the two groups.
definition of myocardial infarction [17], we used a cutoff
Stent length was significantly longer in the cTnT-E group
value of at least five times the upper reference limit of
than in the cTnT-nonE group, whereas other procedural
normal (0.014 ng/ml) for the diagnosis of cTnT elevation.
parameters were similar. Among 68 culprit vessels, 36
On the basis of the presence or absence of cTnT
lesions were located in the left anterior descending artery,
elevation, patients were divided into the elevation
20 in the right coronary artery, 11 in the left circumflex
(cTnT-E; cTnT Z 0.07 ng/ml, n = 25) and nonelevation
artery, and one in the high lateral branch.
groups (cTnT-nonE; cTnT < 0.07 ng/ml, n = 43).

Optical coherence tomography parameters of culprit

Statistical analysis lesions
All analyses were carried out using JMP 10 (SAS All patients safely completed both baseline and final OCT
Institute, Cary, North Carolina, USA). Continuous examinations, and there were no complications related to
variables were expressed as means±SD, and comparisons the OCT procedure. Table 2 shows baseline and poststent
between groups were made using Student’s t-test or a FD-OCT parameters of 68 culprit lesions. There were no
nonparametric Wilcoxon test, if the normality assumption differences between the two groups in terms of lumen
was violated. Categorical data were evaluated using the area at reference sites or the minimal lumen diameter and
w2-test. Results are reported as odds ratios (ORs), 95% MLA of culprit lesions. In terms of poststent FD-OCT
confidence intervals (CIs), and P-values. All tests were findings, there were no differences between the two
two-sided, and a P-value of less than 0.05 represented groups in the rates of edge dissection, intrastent thrombus
a statistically significant difference. formation, tissue prolapse, or malapposition.

Frequency-domain optical coherence tomography

Results tissue characterization and cardiac troponin T elevation
Patient characteristics and blood chemical parameters Baseline morphological characteristics of culprit lesions in
Baseline clinical and angiographic characteristics are both groups are summarized in Table 3. Among the 10
summarized in Table 1. The mean age of the patients OCT characteristics, TCFA and calcium deposition were
was 68.7 years, and 73.5% were men. Twenty-five patients observed significantly more frequently in the cTnT-E
(36.8%) were assigned to the cTnT-E group (cTnT group than in the cTnT-nonE group (32.0 vs. 11.6%,
Z 0.07 ng/ml) and 43 patients to the cTnT-nonE group P = 0.0428, and 72.0 vs. 46.5%, P = 0.0387, respectively).
(cTnT < 0.07 ng/ml). There was no significant difference In particular, the incidence of spotty calcification was
in baseline clinical parameters, including age, sex, BMI, significantly higher in the cTnT-E group than in the

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 OCT predictor of post-PCI myocardial injury Ueda et al. 387

Table 1 Patient characteristics Table 2 Frequency-domain optical coherence tomography

findings
cTnT-E cTnT-nonE
Total (n = 68) (n = 25) (n = 43) P-value Total cTnT-E cTnT-nonE
(n = 68) (n = 25) (n = 43) P-value
Age (years) 68.7±9.8 71.5±7.1 67.1±10.9 0.1614
Male (%) 73.5 64.0 79.1 0.1790 Baseline OCT
BMI (kg/m2) 25.1±4.2 26.0±5.0 24.6±3.7 0.4193 MLD (mm) 1.11±0.25 1.09±0.21 1.13±0.28 0.8787
Medical history (%) MLA (mm2) 1.34±0.58 1.27±0.44 1.37±0.64 0.9544
Hypertension 86.8 92.0 83.7 0.3154 Distal reference area 5.96±2.21 6.22±2.17 5.51±2.27 0.1192
Diabetes mellitus 45.6 56.0 39.5 0.1886 (mm2)
Dyslipidemia 82.4 84.0 81.4 0.7847 Proximal reference area 7.29±2.96 7.50±3.10 6.94±2.74 0.4922
Smoking 25.0 16.0 30.2 0.1808 (mm2)
Echocardiogram Final OCT
LVEF (%) 67.2±6.8 67.3±7.5 67.2±7.4 0.7212 Edge dissection (%) 8.8 12.0 7.0 0.4885
Laboratory data Protrusion (%) 63.2 52.0 69.8 0.1448
A1c (%) 6.57±1.07 6.73±0.87 6.48±1.18 0.0722 Malapposition (%) 47.1 48.0 46.5 0.9056
eGFR (ml/min/ 63.9±16.4 61.9±11.6 65.1±18.6 0.5844
1.73 m2) Data are shown as percentages, means±SD.
Triglyceride (mg/dl) 135.6±44.0 131.5±36.8 138.1±48.0 0.8238 cTnT-E, plasma cardiac troponin T elevation; cTnT-nonE, plasma cardiac troponin
Total cholesterol 180.1±31.9 180.2±30.9 180.1±32.8 0.9089 T nonelevation; MLA, minimal lumen area; MLD, minimal lumen diameter; OCT,
(mg/dl) optical coherence tomography.
HDL cholesterol 48.9±150 46.0±12.5 50.7±16.2 0.3561
(mg/dl)
LDL cholesterol 104.1±28.7 108.0±29.7 101.8±28.3 0.3800
(mg/dl) Table 3 Frequency of lesion characteristics
CK before PCI 120.6±70.7 123.6±78.2 115.5±56.7 0.9949
Total cTnT-E cTnT-nonE
(mg/dl)
(n = 68) (n = 25) (n = 43) P-value
CK after PCI 105.5±64.6 133.6±78.7 88.7±48.2 0.0074
(mg/dl) Eccentric plaque 30.9 32.0 30.2 0.8793
cTnT after PCI 0.098±0.128 0.210±0.156 0.032±0.019 < 0.0001 distribution (%)
(ng/ml) Concave lumen shape (%) 16.2 16.0 16.3 0.9759
Medication (%) Intimal laceration (%) 13.2 16.0 11.6 0.6116
Aspirin 92.7 92.0 93.0 0.8768 Ruptured plaque (%) 8.8 16.0 4.7 0.1193
Thienopyridine 52.9 56.0 51.2 0.6998 Microchannel (%) 51.5 52.0 51.2 0.9469
Statin 60.3 68.0 55.8 0.3189 Lipid pool (%) 76.5 80.0 74.4 0.5978
Stent size Lipid length (mm) 3.07±2.95 3.10±2.95 3.06±2.98 0.9285
Diameter (mm) 2.99±0.39 3.00±0.37 2.98±0.41 0.7762 Maximal lipid angle 168±130 186±122 159±135 0.4363
Length (mm) 23.1±10.9 27.3±13.0 20.7±8.6 0.0276 (deg.)
Lesion location (%) Thin-cap fibroatheroma (%) 19.1 32.0 11.6 0.0428
LAD 52.9 52.0 53.5 0.1949 Macrophage image (%) 29.4 40.0 23.3 0.1477
RCA 29.4 20.0 34.9 Calcium deposition (%) 55.9 72.0 46.5 0.0387
LCX 16.2 24.0 11.6 Spotty calcification (%) 35.3 56.0 23.3 0.0067
HL 1.5 4.0 0.0 Thrombus formation (%) 17.7 20.0 11.6 0.3544
Angiographic findings (%)
% Stenosis of 77.7±4.9 77.5±5.3 77.8±4.6 0.7313 cTnT-E, plasma cardiac troponin T elevation; cTnT-nonE, plasma cardiac troponin
diameter T nonelevation.
ACC/AHA 50.0/38.2 48.0/48.0 51.2/32.6 0.2183
classification B/
C
Table 4 Univariable and ultivariable analyses for the occurrence of
TIMI flow grade 26.5/73.5 24.0/76.0 27.9/72.1 0.7235
before PCI 2/3
cTnT elevation
Univariable Mutivariable
Data are shown as percentages, means±SD.
ACC/AHA, American College of Cardiology/American Heart Association; CK, OR (95% CI) P-value OR (95% CI) P-value
creatine phosphokinase; cTnT-E, plasma cardiac troponin T elevation; cTnT-nonE,
plasma cardiac troponin T nonelevation; eGFR, estimated glomerular filtration Eccentric plaque 1.09 (0.37–3.12) 0.8793 1.02 (0.27–3.66) 0.9785
rate; HDL, high-density lipoprotein; HL, high lateral; LAD, left anterior distribution
descending; LCX, left circumflex; LDL, low-density lipoprotein; LVEF, left Concave lumen 0.98 (0.23–3.65) 0.9759 0.68 (0.10–3.91) 0.6746
ventricular ejection fraction; PCI, percutaneous coronary intervention; RCA, right shape
coronary artery; TIMI, thrombolysis in myocardial infarction. Intimal laceration 1.45 (0.33–6.05) 0.6116 0.85 (0.14–4.86) 0.8537
Ruptured plaque 3.90 (0.70–29.85) 0.1193 6.27 (0.79–71.75) 0.0828
Microchannel 1.03 (0.38–2.80) 0.9469 0.90 (0.27–2.98) 0.8678
cTnT-nonE group (56.0 vs. 23.3%, P = 0.0067). In Lipid pool 1.38 (0.43–4.90) 0.5978 1.25 (0.29–6.21) 0.7718
addition, the length and maximum angle of lipid Thin-cap 3.58 (1.04–13.40) 0.0428 2.91 (0.68–13.55) 0.1488
accumulation were similar between the groups. fibroatheroma
Macrophage image 2.20 (0.75–6.51) 0.1477 2.02 (0.53–7.83) 0.2965
Calcium deposition 2.96 (1.06–8.97) 0.0387 – –
Predictive values of frequency-domain optical Spotty calcification 4.20 (1.48–12.55) 0.0067 4.82 (1.51–16.85) 0.0076
Thrombus formation 1.90 (0.48–7.60) 0.3544 2.21 (0.42–11.80) 0.3403
coherence tomography parameters for cardiac troponin
T elevation CI, confidence interval; cTnT, plasma cardiac troponin T; OR, odds ratio.
Univariate analysis revealed that TCFA, calcification, and
spotty calcification were correlated with the occurrence Patients were then divided into four groups according to
of cTnT elevation. Particularly, spotty calcification was the presence or absence of TCFA and calcium deposition.
selected as the only independent predictor in multi- As shown in Table 5, the OR value for the occurrence of
variable analysis (Table 4). cTnT elevation was significant in patients who had

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
388 Coronary Artery Disease 2014, Vol 25 No 5

Table 5 Odds ratios of thin-cap fibroatheroma and calcium colocalization of both TCFA and calcium deposition (OR
deposition 8.40, 95% CI 1.65–52.78, P = 0.0099) in comparison with
TCFA (–) TCFA (+) patients without both characteristics. Further, the pre-
Calcium deposition (–) OR: 1.00 OR: 4.20 dictive value of colocalization was enhanced when only
CI: 0.42–43.02 spotty calcification was considered (OR 21.00, 95% CI
P-value: 0.2067 2.65–454.22, P = 0.0029; Table 6). These correlations
Calcium deposition (+) OR: 2.96 OR: 8.40
CI: 0.91–10.88 CI: 1.65–52.78 remained significant even after adjusting for stent length in
P-value: 0.0724 P-value: 0.0099 colocalization of TCFA both with calcium deposition and
CI, confidence interval; OR, odds ratio; TCFA, thin-cap fibroatheroma.
with spotty calcification (OR 6.12, 95% CI 1.12–40.12,
P = 0.0367, and OR 18.34, 95% CI 2.22–403.61, P = 0.0057,
respectively).
Table 6 Odds ratios of thin-cap fibroatheroma and spotty
calcification
TCFA (–) TCFA (+)
Representative case with postprocedural cardiac
Spotty calcification (–) OR: 1.00 OR: 4.20 troponin T elevation
CI: 0.42–43.02
P-value: 0.2067 Figure 2 shows representative OCT images from a patient
Spotty calcification (+) OR: 4.20 OR: 21.00 who developed elevation of cTnT to 0.228 pg/ml 12 h
CI: 1.14–17.26 CI: 2.65–454.22
P-value: 0.0314 P-value: 0.0029
after stent implantation. The baseline longitudinal image
(Fig. 2a) shows two independent spotty calcium deposi-
CI, confidence interval; OR, odds ratio; TCFA, thin-cap fibroatheroma. tions at both sides of the MLA site. The cross-sectional

Fig. 2

(a)
20 30

(b) (c) (d)

Calc.

Calc.

Lipid
TCFA

Representative case with postprocedural cTnT elevation. (a) Magnified long-axis view of culprit lesion. Note the concomitant existence of TCFA and
two independent spotty calcifications (circles). (b) Distal stenotic region showing spotty calcification (circle). (c) MLA site showing large lipid
accumulation with TCFA (arrow). (d) Proximal stenotic region showing both large lipid accumulation and spotty calcification (circle). Calc., calcification;
cTnT, plasma cardiac troponin T; MLA, minimal lumen area; TCFA, thin-cap fibroatheroma.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 OCT predictor of post-PCI myocardial injury Ueda et al. 389

image at the MLA site (Fig. 2c) shows a TCFA with a believe that two factors, the thickness of the fibrous cap
large lipid pool. and the volume of lipid content, play a role in this
complication, and that the fragility of the fibrous cap
Discussion covering the lipid pool is a particularly important factor in
The salient findings of the present FD-OCT study are as determining leakage of the plaque component. In this
follows: (i) the incidence of PCI-related myocardial injury regard, previous IVUS studies have reported that plaque
was common when we applied a sensitive biomarker, length and volume are positively correlated with the
cTnT; (ii) the presence of TCFA and calcium deposition occurrence of postprocedural myocardial injury or the no-
in culprit lesions was correlated with this complication; reflow phenomenon [7,8,21]. However, in our study, the
and (iii) colocalization of TCFA and calcium deposition, quantitative OCT parameters of length and maximum
especially spotty calcification, was a powerful predictor angle of lipid accumulation did not correlate with cTnT
for the development of PCI-related cTnT elevation. elevation. The reason is unclear, but this may have
resulted from our patient selection. Previous IVUS
In this study on stable angina pectoris, 37.6% of patients studies enrolled patients with acute coronary syndrome
fulfilled the criteria of significant cTnT elevation, even (ACS) [7,8], and all patients were stable in our study. It
though all PCI procedures were completed with angio- has been shown that lipid length and arc are significantly
graphically successful findings. However, this high greater in ACS culprit plaques compared with plaques in
incidence of postprocedural myocardial injury is compar- stable angina pectoris [21,22].
able to that in earlier studies [3–5,18]. Creatine
phosphokinase myocardial band (CK-MB) has previously The finding that lipid-rich plaques and TCFA are the
been used as a biomarker for myocardial injury after predictors for cardiac enzyme elevation is consistent with
elective PCI, and the association between CK-MB those of previous studies using OCT. However, those
elevations and adverse prognosis after PCI has been well studies have also reported that intraluminal thrombus is a
documented [4,18]. We adopted a cTnT value of at least predictor of cardiac enzyme elevation [20,23]. In this
five times the upper reference limit on the basis of the study on stable angina pectoris, the prevalence of
third universal definition of myocardial infarction [17]. intraluminal thrombus was relatively low and intraluminal
Although this new criterion using the isolated elevation of thrombus was not selected as a predictor for this
cTnT after PCI has still not been validated in terms complication. This discrepancy might be derived from
of association with an adverse post-PCI prognosis [19], patient selection because previous studies included
patients with PCI-related myocardial damage should be patients with both unstable and stable angina.
properly diagnosed and then undergo careful long-term
follow-up.
Calcification
Our new finding is that calcium deposition, in particular
Thin-cap fibroatheroma
spotty calcification, is independently correlated with the
This study showed that the presence of both TCFA and
occurrence of PCI-related cTnT elevation. In contrast to
calcium deposition within the culprit segment was a risk
TCFA, calcification itself has not been reported as a risk
factor for postprocedural cTnT elevation in patients with
factor for this complication, and thus the role of calcium
elective stent implantation. Because this complication is
deposition in enzyme elevation is not clear. It is unlikely
believed to result from the distal embolization of plaque
that large calcium deposits detach from the vascular wall
material into the peripheral coronary circulation [6–8,20],
and enter the coronary circulation as emboli during the
it is highly likely that mechanical laceration of the TCFA
mechanical compression of plaques on coronary stent
luminal surface on stent implantation causes the leakage
implantation. However, one of the important findings of
of plaque contents. Earlier studies using IVUS have
this study was that 77.8% of calcium depositions had
shown that low-echoic or echo-attenuated plaques at the
spotty characteristics in the cTnT-E group.
culprit segments are independent risk factors for cardiac
enzyme elevation, and researchers concluded that such Recent studies using multislice computed tomography
IVUS features indicate a large amount of lipid or necrotic and IVUS have shown that small calcifications observed
tissue within the culprit plaque [6–8]. However, conven- in coronary atherosclerotic plaques are one of the
tional IVUS cannot clearly image the fine structures of vulnerable characteristics that contribute to plaque
plaque components or arterial walls partly because of its instability and subsequent progression [24,25]. The
limited spatial resolution. Lee et al. [20] recently mechanism underlying the development of spotty calci-
reported the characteristics of IVUS-derived, echo- fications in atherosclerotic lesions has not yet been
attenuated plaques examined by OCT. They showed elucidated, but continuing inflammation with the ex-
that the mean cap thickness of attenuated plaques was pression of osteogenic cytokines is assumed to play an
significantly less than that of nonattenuated plaques, and essential role [26]. It is therefore possible that the
that the mean thickness of fibrous caps was 60 mm, which colocalization of spotty calcifications and TCFA might be
fulfills the definition of TCFA. It is thus reasonable to common within plaques with ongoing inflammation.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
390 Coronary Artery Disease 2014, Vol 25 No 5

Conversely, OCT imaging indicating the concomitant Finally, the blood sampling of cTnT was not serially
existence of these two characteristics may indicate the followed after PCI; hence, some of the patients who were
presence of active inflammation within the culprit lesion. assigned to the cTnT-nonE group might have had late
Kataoka et al. [14] reported that the total volume of increases in cTnT.
atheromas, which are thought to be a possible source of
coronary emboli, was higher in lesions with spotty Conclusion
calcifications than in those without. In addition, it is This FD-OCT study showed that colocalization of
known that matrix metalloproteinase expression is TCFA and spotty calcification is a powerful predictor of
enhanced in plaques with active inflammation [27]. PCI-related cTnT elevation after elective PCI. FD-OCT
Overexpression of matrix metalloproteinase will degrade is feasible and useful for risk stratification during PCI to
the matrix collagen of fibrous caps, resulting in increased avoid procedure-related complications.
fragility and thinning and enhanced vulnerability to
mechanical stress. Acknowledgements
The other new finding of this study is that colocalization This study was partly supported by research grant from
of TCFA and calcification is a powerful predictor of PCI- the Japanese Ministry of Education, Culture, Sports
related myocardial injury. According to the above discus- Science and Technology, and Takeda Science Foundation.
sion, colocalization may not only be a marker for cTnT
elevation, but may also be the mechanism that enhances Conflicts of interest
leakage of plaque content into coronary circulation. There are no conflicts of interest.
Although this mechanism should be further elucidated,
it is possible that during the compression of plaques by
References
stent expansion, the concomitant existence of hard 1 Nallamothu BK, Bates ER. Periprocedural myocardial infarction and
calcified tissue around lipid-rich culprit lesions might mortality: causality versus association. J Am Coll Cardiol 2003; 42:
1412–1414.
disturb the outward or circumferential movement of 2 Lansky AJ, Stone GW. Periprocedural myocardial infarction: prevalence,
relatively soft lipid pools, thus enhancing the leakage of prognosis, and prevention. Circ Cardiovasc Interv 2010; 3:602–610.
plaque components into coronary circulation through the 3 Fujii K, Carlier SG, Mintz GS, Kobayashi Y, Jacoboff D, Nierenberg H, et al.
Creatine kinase-MB enzyme elevation and long-term clinical events after
disruption of fragile fibrous caps [21,22]. successful coronary stenting in lesions with ruptured plaque. Am J Cardiol
2005; 95:355–359.
4 Feldman DN, Kim L, Rene AG, Minutello RM, Bergman G, Wong SC.
Clinical implications
Prognostic value of cardiac troponin-I or troponin-T elevation following
The findings of this study indicate that preinterventional nonemergent percutaneous coronary intervention: a meta-analysis. Catheter
FD-OCT evaluation of culprit lesion allows us to modulate Cardiovasc Interv 2011; 77:1020–1030.
5 Kizer JR, Muttrej MR, Matthai WH, McConnell J, Nardone H, Sonel AF, et al.
the strategy of subsequent PCI procedures. If culprit Role of cardiac troponin T in the long-term risk stratification of patients
lesions show both TCFA and spotty calcification, high-risk undergoing percutaneous coronary intervention. Eur Heart J 2003;
characteristics, additional proactive strategies should be 24:1314–1322.
6 Hong YJ, Mintz GS, Kim SW, Lee SY, Okabe T, Pichard AD, et al. Impact
considered to prevent distal embolization of plaque of plaque composition on cardiac troponin elevation after percutaneous
contents. Embolic protection devices, such as filters, distal coronary intervention: an ultrasound analysis. JACC Cardiovasc Imaging
occlusion systems, and newly developed mesh-covered 2009; 2:458–468.
7 Lee SY, Mintz GS, Kim SY, Hong YJ, Kim SW, Okabe T, et al. Attenuated
coronary stents [28], are thought to be good candidates in plaque detected by intravascular ultrasound: clinical, angiographic, and
this regard. As a conservative strategy, it might be morphologic features and post-percutaneous coronary intervention
preferable to defer PCI treatment until the optimal complications in patients with acute coronary syndromes. JACC Cardiovasc
Interv 2009; 2:65–72.
medical therapy stabilizes the culprit lesion(s). Previous 8 Kimura S, Kakuta T, Yonetsu T, Suzuki A, Iesaka Y, Fujiwara H, et al.
studies have shown that high-dose atorvastatin treatment Clinical significance of echo signal attenuation on intravascular ultrasound
significantly reduces the incidence of periprocedural in patients with coronary artery disease. Circ Cardiovasc Interv 2009;
2:444–454.
myocardial injury in patients undergoing elective PCI [29]. 9 Brezinski ME, Tearney GJ, Bouma BE, Izatt JA, Hee MR, Swanson EA, et al.
Optical coherence tomography for optical biopsy. Properties and
demonstration of vascular pathology. Circulation 1996; 93:1206–1213.
Study limitations 10 Yabushita H, Bouma BE, Houser SL, Aretz HT, Jang IK, Schlendorf KH, et al.
This study has several limitations that should be noted. Characterization of human atherosclerosis by optical coherence
First, culprit lesions were obtained in a small study tomography. Circulation 2002; 106:1640–1645.
11 Jang IK, Bouma BE, Kang DH, Park SJ, Park SW, Seung KB, et al.
population. The present results from a single-center Visualization of coronary atherosclerotic plaques in patients using optical
cohort should be confirmed by larger multicenter clinical coherence tomography: comparison with intravascular ultrasound. J Am Coll
investigations. Second, we excluded patients with ACS Cardiol 2002; 39:604–609.
12 Uemura S, Ishigami K, Soeda T, Okayama S, Sung JH, Nakagawa H, et al.
or complex lesions; hence, our findings cannot be applied Thin-cap fibroatheroma and microchannel findings in optical coherence
to these patient populations. Third, because the near- tomography correlate with subsequent progression of coronary
infrared light of FD-OCT is attenuated at the surface of atheromatous plaques. Eur Heart J 2012; 33:78–85.
13 Ehara S, Kobayashi Y, Yoshiyama M, Shimada K, Shimada Y, Fukuda D, et al.
lipid pools, some calcium depositions might be missed Spotty calcification typifies the culprit plaque in patients with acute myocardial
when they are located within or behind large lipid pools. infarction: an intravascular ultrasound study. Circulation 2004; 110:3424–3429.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
 OCT predictor of post-PCI myocardial injury Ueda et al. 391

14 Kataoka Y, Wolski K, Uno K, Puri R, Tuzcu EM, Nissen SE, et al. Spotty 22 Prati F, Pawlowski T, Gil R, Labellarte A, Gziut A, Caradonna E, et al.
calcification as a marker of accelerated progression of coronary Stenting of culprit lesions in unstable angina leads to a marked reduction in
atherosclerosis: insights from serial intravascular ultrasound. J Am Coll plaque burden: a major role of plaque embolization? A serial intravascular
Cardiol 2012; 59:1592–1597. ultrasound study. Circulation 2003; 107:2320–2325.
15 Gonzalo N, Serruys PW, Okamura T, Shen ZJ, Onuma Y, Garcia-Garcia HM, 23 Porto I, Di VL, Burzotta F, Niccoli G, Trani C, Leone AM, et al. Predictors of
et al. Optical coherence tomography assessment of the acute effects of periprocedural (type IVa) myocardial infarction, as assessed by frequency-
stent implantation on the vessel wall: a systematic quantitative approach. domain optical coherence tomography. Circ Cardiovasc Interv 2012; 5:
Heart 2009; 95:1913–1919. 89–96.
16 Bouma BE, Tearney GJ, Yabushita H, Shishkov M, Kauffman CR, DeJoseph 24 Fujii K, Carlier SG, Mintz GS, Kobayashi Y, Jacoboff D, Nierenberg H, et al.
Gauthier D, et al. Evaluation of intracoronary stenting by intravascular optical Intravascular ultrasound study of patterns of calcium in ruptured coronary
coherence tomography. Heart 2003; 89:317–320. plaques. Am J Cardiol 2005; 96:352–357.
17 Thygesen K, Alpert JS, White HD. Joint ESC/ACCF/AHA/WHF Task Force 25 Motoyama S, Kondo T, Sarai M, Sugiura A, Harigaya H, Sato T, et al.
for the Redefinition of Myocardial Infarction. Third universal definition of Multislice computed tomographic characteristics of coronary lesions in
myocardial infarction. J Am Coll Cardiol 2012; 60:1581–1598. acute coronary syndromes. J Am Coll Cardiol 2007; 50:319–326.
18 Javaid A, Buch AN, Steinberg DH, Pinto Slottow T, Roy P, Pichard AD, et al. 26 Abedin M, Tintut Y, Demer LL. Vascular calcification: mechanisms
Does creatine kinase-MB (CK-MB) isoenzyme elevation following and clinical ramifications. Arterioscler Thromb Vasc Biol 2004; 24:
percutaneous coronary intervention with drug-eluting stents impact late 1161–1170.
clinical outcome? Catheter Cardiovasc Interv 2007; 70:826–831. 27 Galis ZS, Sukhova GK, Lark MW, Libby P. Increased expression of
19 Zimarino M, Affinito V. The prognosis of periprocedural myocardial infarction matrix metalloproteinases and matrix degrading activity in vulnerable
after percutaneous coronary interventions. Cardiovasc Revasc Med 2013; regions of human atherosclerotic plaques. J Clin Invest 1994; 94:
14:32–36. 2493–2503.
20 Lee T, Kakuta T, Yonetsu T, Takahashi K, Yamamoto G, Iesaka Y, et al. 28 Stone GW, Abizaid A, Silber S, Dizon JM, Merkely B, Costa RA, et al.
Assessment of echo-attenuated plaque by optical coherence tomography Prospective, randomized, multicenter evaluation of a polyethylene
and its impact on post-procedural creatine kinase-myocardial band terephthalate micronet mesh-covered stent (MGuard) in ST-segment
elevation in elective stent implantation. JACC Cardiovasc Interv 2011; elevation myocardial infarction: the MASTER trial. J Am Coll Cardiol 2012;
4:483–491. 60:1975–1984.
21 Ando H, Amano T, Matsubara T, Uetani T, Nanki M, Marui N, et al. 29 Briguori C, Visconti G, Focaccio A, Golia B, Chieffo A, Castelli A, et al. Novel
Comparison of tissue characteristics between acute coronary syndrome and approaches for preventing or limiting events (Naples) II trial: impact of a
stable angina pectoris. An integrated backscatter intravascular ultrasound single high loading dose of atorvastatin on periprocedural myocardial
analysis of culprit and non-culprit lesions. Circ J 2011; 75:383–390. infarction. J Am Coll Cardiol 2009; 54:2157–2163.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.