Clinical Radiology xxx (xxxx) xxx

Contents lists available at ScienceDirect

Clinical Radiology
journal homepage: www.clinicalradiologyonline.net

Characteristics of coronary artery atherosclerotic

plaques in chronic kidney disease: evaluation
with coronary CT angiography
W. Deng, L. Peng*, J. Yu, T. Shuai, Z. Chen, Z. Li
Department of Radiology, and National Clinical Research Center for Geriatrics, West China Hospital, Sichuan
University, Chengdu, Sichuan Province, China

art icl e i nformat ion AIM: To determine the characteristics of coronary artery atherosclerotic plaques in chronic
kidney disease (CKD) with coronary computed tomography angiography (CTA).
Article history: MATERIALS AND METHODS: Sixty-six patients with CKD who underwent coronary CTA were
Received 20 February 2019 analysed retrospectively. The extent, distribution, and types of plaques and stenosis severity
Accepted 12 June 2019 were evaluated. The imaging features were compared between dialysis and non-dialysis
groups. In the dialysis group, the imaging features were compared between diabetes and
non-diabetes patients.
RESULTS: In total, 152 coronary vessels (2.3
1.3 per patient) and 306 segments (4.6
3.5 per
patient) were found to have plaques. The most common diseased coronary vessel was the left
anterior descending (LAD) artery (53 vessels, 34.9%) followed by the left circumflex (LCX) ar-
tery (39 vessels, 25.7%), and right coronary artery (RCA; 37 vessels, 24.3%) in sequence. The
most commonly involved coronary artery segment was the middle segment of LAD artery
(14.1%). Calcified plaques (65.9%) were detected more frequently than mixed (25.6%) or non-
calcified (8.5%) plaques (p<0.001). Among the degrees of coronary stenosis, minimal steno-
sis (55.8%) was the most common (p<0.001). The majority of calcified plaques were non-
obstructive plaques (n¼134, 78.2%), while about half of non-calcified (n¼14, 63.6%) and
mixed plaques (n¼30, 45.5%) were obstructive plaques (p<0.001).
CONCLUSION: A heavy plaque burden was detected in CKD patients at coronary CTA. Non-
obstructive calcified plaque was the most common imaging feature. CKD patients with type
2 diabetes mellitus had more obstructive mixed plaques.
Ó 2019 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Introduction events when the coronary artery is involved.1 Patients with

CKD have a higher prevalence of cardiovascular disease
Vascular calcification is common in patients with chronic including coronary artery disease, compared with the gen-
kidney disease (CKD) and is a risk factor for cardiovascular eral population.2,3 A causal relationship between vascular
calcification of the coronary artery and cardiovascular dis-
ease seems logical.4
* Guarantor and correspondent: L. Peng, West China Hospital of Sichuan Coronary computed tomography angiography (CTA) plays
University, No. 37 Guoxue Xiang, Chengdu, 610041, China. Tel.: þ86 28 an important role in the non-invasive evaluation of calcifi-
85421065; fax: þ86 28 85422766.
cation of the coronary artery.5e7 Many studies on CKD and
E-mail address: Lqpeng0214@126.com (L. Peng).

https://doi.org/10.1016/j.crad.2019.06.003
0009-9260/Ó 2019 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003
2 W. Deng et al. / Clinical Radiology xxx (xxxx) xxx

coronary artery calcification have used the calcium score at All images were acquired during a single inspiratory
CT to assess the coronary artery involvement8e12; however, breath hold of 10 seconds, while the electrocardiogram was
few focus on the characteristics and distribution of coronary registered simultaneously. Images were reconstructed most
artery calcification and also the type of atherosclerotic plaque often in the diastolic phase, as this is typically the phase
and stenosis severity in patients with CKD. Thus, the features showing the least motion artefacts; however, additional
of coronary artery calcification in patients with CKD, the reconstructions were made throughout the entire cardiac
degree of stenosis, and distribution caused by each type of cycle, when needed.
plaque is still unclear. The purpose of this study was to Depending on the patient’s residual kidney function, the
determine the characteristics of coronary artery atheroscle- following methods were used to prevent further kidney
rotic plaques in CKD patients by using coronary CTA. function deterioration: adequate pre- and post-procedural
hydration, and in haemodialysis patients, the scan was
Materials and methods performed on the day prior to the next dialysis session.
Effective radiation dose (ED) for each CT examination was
Patients calculated as the doseelength product (DLP) multiplied by a
conversion coefficient for the chest (k ¼ 0.014 mSv/
This retrospective study was approved by the institu- [mGy∙cm]).13 The median
quartile of DLP and ED of the
tional review board. Informed patient consent was waived CCTA was 213.8
106.5 and 3
1.5 in 67 patients, respectively.
due to the retrospective nature of this study. Between June
2011 and September 2018, 47,092 patients underwent cor- Image analysis
onary CTA at West China Hospital. The inclusion criteria
were as follows: (1) clinical diagnosis with CKD or sus- Coronary CTA image analysis was performed by two
pected with CKD; and (2) complete clinical data available. A experienced cardiovascular radiologists each with 5 years of
total of 79 patients were found in the database. The exclu- experience. Discrepancies in their interpretations of ste-
sion criteria were as follows: no discharge diagnosis of CKD nosis were resolved in consensus. The diagnostic images
(n¼5), poor image quality inadequate for analysis (n¼2), with optimal quality were transferred to an off-line post-
and a history of stenting or bypass (n¼5). Finally, 67 (67/79, processing workstation (Syngo-Imaging, Siemens Medical
84.8%) patients were enrolled in this study. The reasons for Solution Systems, Forchheim, Germany) for image analysis.
coronary CTA examination were as follows: chest pain Alternative image reconstruction methods for evaluation of
(n¼25), chest tightness (n¼15), palpitations (n¼11), dysp- coronary artery plaques included maximum intensity pro-
noea (n¼7), preoperative evaluation when aged >65 years jection, multiplanar reconstruction, curvature planar
old (n¼4), and other examinations such as electrocardio- reconstruction, and volume rendering.
gram or myocardial enzyme abnormalities indicating cor- The number of diseased coronary vessels and segments,
onary artery disease (n¼5). The clinical data of patients the number and types of plaques, and degree of stenosis
included age, sex, body mass index, smoking history, history were evaluated. In this study, coronary arteries were
of dialysis, diabetes mellitus (DM), and hypertension. divided into four branches: left main (LM), left anterior
descending (LAD), left circumflex (LCX), and right coronary
CT protocols artery (RCA). According to the American Heart Association
standard, the left and right coronary arteries were divided
All coronary CTA examinations were performed on into 15 segments.14 In the present study, all segments were
multidetector CT systems (SOMATOM Definition, Siemens analysed: in the RCA segments 1e4; in the LM segment 5; in
Medical Solutions, Forchheim, Germany; n¼5; SOMATOM the LAD segments 6e10, and in the LCX segments 11e15
Definition FLASH, Siemens Medical Solutions, Forchheim, (Fig 1). Plaques were classified as calcified plaque (plaques
Germany; n¼38; and Revolution CT, GE Healthcare, Wau- with higher CT attenuation than the contrast-enhanced
kesha, WI USA; n¼24). Beta-blockers were not used for to lumen); non-calcified plaque (plaques with lower CT
reduce the heart rate. The scanning range was from the attenuation than the contrast-enhanced lumen without any
tracheal bifurcation to 20 mm below the inferior cardiac calcification) and mixed plaque (non-calcified and calcified
apex. A 50e70 ml (dependent on body mass index) bolus of elements in single plaque).15 The interpretability of coro-
iodinated contrast agent (iopamidol, 370 mg of iodine/ml; nary artery segments was assessed. All segments <1.5 mm
Bracco, Shanghai, China) was injected into the antecubital in diameter were excluded from evaluation. A total of 163
vein at a flow rate of 5 ml/s followed by 30 ml saline at the segments (16.2%) were excluded. The interpretability of
same flow rate. For the SOMATOM Definition and SOMA- each plaque within vessel with a diameter of >1.5 mm was
TOM Definition FLASH systems, scan parameters were analysed before accessing the degree of stenosis. Uninter-
100e120 kV tube voltage (adapted to body mass index), 220 pretable plaque was noted when the severity of stenosis in
mAs tube current, 64/1280.5 mm collimation, and calcified plaque could not be assessed due to blooming ar-
0.33e0.4 seconds rotation time. For the Revolution CT sys- tefacts. The degree of stenosis was assessed based on a
tem, the tube voltage and tube current were set automati- classification system suggested by the Society of Cardio-
cally by kV Assist and Smart-mA based on the scout image vascular Computed Tomography: no visible stenosis (0%),
of the patients, other imagine parameters were 2560.625 minimal stenosis (1e24%), mild stenosis (25e49%), mod-
mm collimation and 0.28 seconds rotation time. erate stenosis (50e69%), severe stenosis (70e99%), and

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003
 W. Deng et al. / Clinical Radiology xxx (xxxx) xxx 3

patients had negative coronary CTA results. The baseline

clinical data of the 67 patients are summarised in Table 1.
There was no significant difference in age, sex, smoking
history, dialysis history, and DM history between the two
groups (all p>0.05) except for the history of hypertension
(p<0.05). The proportion of hypertension patients in the
positive group was higher (72.4% versus 33.3%), but the
duration of the disease was shorter (120 versus 216 months)
compared with the negative group.

Extent and anatomical distribution of coronary artery

plaques

A total of 152 coronary vessels (2.3
1.3 per patient;

Figure 1 Coronary segments: 1 proximal segment of right coronary range 1e4) and 306 segments (4.6
3.5 per patient; range
artery (RCA); 2 middle segment of RCA; 3 distal segment of RCA; 4 1e13) were found to have plaques. The majority of patients
posterior descending artery from RCA; 5 left main coronary artery; 6 had multi-vessel disease (47/58, 81%) rather than single
proximal segment of left anterior descending artery (LAD); 7 middle vessel disease (11/58, 19%). The most common diseased
segment of LAD; 8 distal segment of LAD; 9 first diagonal branch; 10
coronary vessel was the LAD artery (53 vessels, 34.9%) fol-
second diagonal branch; 11 proximal segment of left circumflex
(LCX); 12 first obtuse marginal branch; 13 distal segment of LCX; 14
lowed by LCX artery (39 vessels, 25.7%), RCA artery (37
left posterior-lateral branch; 15 posterior descending artery from LCX. vessels, 24.3%), and LM (23 vessels, 15.1%) in sequence. The
most common involved coronary artery segment was the
occlusion (100%).16 In addition, the lesions were also clas- middle segment of LAD artery (43/306, 14.1%) followed by
sified as obstructive and non-obstructive using a 50% the proximal segment of LAD artery (42/306, 27.8%) and the
threshold.6,9 The length of plaque was assessed for each proximal segment of RCA (33/306, 10.8%). In all patients, no
lesion. For lesions with length 3mm, the length was lesions were detected in segment 14 and segment 15. The
measured on the curvature planar reconstruction image anatomical distribution of plaques in each type is shown in
and mean length of the two measurements were recorded Fig 3. Calcification was also found in the aortic valve (18
(Fig 2). For spot-like lesions (the length <3mm), the cases, 26.9%), mitral valve (14 cases, 20.9%), and left atrial
numbers of lesions were recorded. wall (2 cases, 3%). There was no valvular stenosis or regur-
gitation in all cases with cardiac valvular calcification
Statistical analysis
confirmed at echocardiography or final clinical diagnosis.

Clinical data, number of diseased coronary vessels and Types of coronary artery plaque and degree of coronary
segments, length, types of plaques, and degree of stenosis stenosis
were analysed statistically in each patient. Normally
distributed continuous variables were expressed as mean
A total of 258 lesions (4.17
2.48 per patient; range 1e9)
standard deviation (as assessed by the KolmogoroveSmirnov were analysed. The different types of plaques and degrees of
test) whereas non-normally distributed continuous variables coronary stenosis in each type are shown in Table 2. Calci-
expressed as median (range) and categorical variables as fied plaques (65.9%, Fig 4a) were more frequently detected
number and percentage. Continuous data were compared than mixed (25.6%, Fig 5a) or non-calcified (8.5%, Fig 5b)
using independent sample t-test (for normally distributed plaques (p<0.001). Among the degree of stenosis, minimal
variables) or the ManneWhitney U-test (for non-normally stenosis (55.8%) was the most common (p<0.001). There
distributed variables). Categorical data were compared us- were more uninterpretable lesions in calcified plaques than
ing the chi-square test. Multivariate analysis of variance mixed or non-calcified plaques (20% versus 10.2%;
(MANOVA) was used to determine associations between p<0.001). The majority of calcified plaques were non-
clinical factors and characteristics of plaque. Multivariate obstructive plaques (n¼134, 78.2%), while about half the
logistic regression was used to examine factors associated non-calcified and mixed plaques were obstructive plaques
with characteristics of plaque. All statistical analyses were (n¼14, 63.6% and n¼30, 45.5%; all p<0.001).
performed on SPSS (version 19.0, SPPS. Chicago, IL, USA).
Two-tailed p-value of <0.05 was considered statistically Correlations between clinical history and characteristics
significant. of plaques

Results According to the history of dialysis, the plaques were

divided into the dialysis group and non-dialysis group
Clinical characteristics of patient population (Table 3, Fig 6a). The length of plaque in the dialysis group
was longer than non-dialysis group (2.55
0.22 versus
Among the 67 patients included in this study, a total of 1.64
0.23 cm, p¼0.01). The constituents of the plaque type
58 patients had positive coronary CTA results, and nine were statistically different. The dialysis group had a higher

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003
4 W. Deng et al. / Clinical Radiology xxx (xxxx) xxx

Figure 2 Measurement of calcified plaque length. (a) Multiplanar reconstruction CT image of RCA for drawing the centre line of curvature plane
reconstruction (curved line). (b) Curvature planar reconstruction CT image of RCA generated from the curved centre line in (a). Measurement of
length was performed manually by drawing straight lines on this image using the two-dimensional distance tool in the workstation. The length
of the plaque was the sum of the length of three black lines in this RCA.

proportion of calcified plaque (76.2%), whereas the non- than non-DM patients (1.65
1.67 versus 2.53
3.04 cm,
dialysis group had more non-calcified (17%) and mixed p¼0.02). DM patients also had a shorter history of dialysis
plaque (35.1%). The dialysis group had more non- (31.39
3.19 versus 76.81
6.64 months, p<0.001). The
obstructive plaque than the non-dialysis group (n¼115, proportion of obstructive plaque was higher in DM patients
70.1% and n¼54, 57.4%; p¼0.04). There was no significant (n¼16, 21.1% versus n¼5, 5.7% in non-DM group, p¼0.003);
difference between the dialysis and non-dialysis groups in however, the constituent of the plaque type in DM and non-
uninterpretable plaque. The lesions in the dialysis group DM group had no statistically difference (all p>0.05).
were also divided into three sub-group according to the The associations between clinical factors and character-
tertile of duration of dialysis (Fig 6b). The length of plaque istics of plaque were shown in Table 5. There were no as-
increased gradually with duration of dialysis. sociations between clinical factors and the type of plaque
According to history of DM in the dialysis group, lesions (p>0.05) except for association between dialysis and type of
were divided into the DM group and non-DM group plaque (p¼0.049). No associations were found between all
(Table 4). There were statistically significant differences in factors and obstruction. According to the result of multi-
length of lesion, duration of dialysis, and vessel obstruction variable logistic regression (Table 6), dialysis is a risk factor
(all p<0.05). Lesions in DM patients (Fig 7) were shorter for calcified plaque (OR¼2.778, p¼0.001).

Table 1
Baseline clinical characteristics of patients with chronic kidney disease (CKD) who underwent coronary computed tomography angiography (CTA).

Characteristics Positive group Negative group p-Value

Number of patients 58 9
Age (years) 64
11.7 (20e83) 60 (30e78)a 0.51
Male/Female 36/22 6/3 0.55
Body Mass Index (kg/m2) 23
4.1 (14.2e31.6) 19.8 (17.8e23.5)a 0.20
Current or pervious smoker 20 (34.5) 6 (66.7) 0.17
Dialysis 37 (63.8) 7 (77.8) 0.34
Duration of dialysis (months) 36 (1e228)a 18.6
17.1 (3e48) 0.13
Diabetes mellitus 25 (43.1) 3 (33.3) 0.43
Duration of type 2 diabetes mellitus (months) 178.1
76.0 (48e360) 144 (48e168)a 0.22
Hypertension 42 (72.4) 3 (33.3) 0.03
Duration of hypertension (months) 120 (1e480)a 216 (144e600)a 0.00

Data are expressed as means
SD (range) [normally distributed] or median (range) [non- normally distributed] or numbers (%).
Positive ¼ one or more coronary plaques were detected on CTA. Negative ¼ no coronary plaque was detected on CTA.
a
Non- normally distributed variables.

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003
 W. Deng et al. / Clinical Radiology xxx (xxxx) xxx 5

especially in dialysis patients. A long duration of dialysis,

hyperphosphataemia, hypercalcaemia, hyperparathyroid-
ism, use of high-dose vitamin D, and hypoalbuminaemia
represent known risk factors for vascular calcification pro-
gression in dialysis patients.4,18 According to the previous
research, cardiovascular mortality is strongly associated
with vascular calcification in end-stage renal disease
(ESRD).19e21 In the present study, the CKD patients had a
heavy plaque burden that was mainly distributed in the LAD
artery and proximal segment of each coronary vessel.
Analysis of plaque composition revealed a relatively high
proportion of calcified plaques. Regarding stenosis severity,
about half of the detected lesions were minimal stenosis.
Thus, the most common imaging features of coronary artery
lesions in patients with CKD were non-obstructive calcified
lesions. For most patients, the severity of coronary artery
stenosis was not consistent with symptoms such as chest
pain, chest tightness, and palpitations.
There were also nine patients with negative coronary
CTA results, although each patient had clinical symptoms.
The demographic features and clinical information in
negative patients were similar to positive patients except
for hypertension; however, the bias of the small sample size
of negative patients should be considered. The CTA results
indicate that clinical symptoms in these negative patients
Figure 3 Anatomical distribution of plaques. (a) Distribution of pla- may not be caused by coronary artery disease. Thus, in pa-
ques in four branches: RCA, LM, LAD, and LCX. The LAD was the most tients with positive symptoms and non-obstructive lesions
common location for each type of plaque with calcified plaque the or normal coronary arteries in negative cases, non-
most frequent type. (b) Distribution of plaques in each segment of the atherosclerotic causes of the clinical manifestations of
coronary artery. The proximal segment of the LAD (segment 6) was these CKD patients should be considered. Whether diffuse
the most common location followed by the middle segment of the myocardial fibrosis in patients with CKD and ESRD might
LAD (segment 7). No lesions were detected in segments 14 and 15.
lead to these manifestations is still unclear.22,23
Much of the research on coronary CTA in CKD and ESRD
Discussion patients has shown that CT calcium scores in these patients
were high, especially in dialysis patients.9,10,24,25 Few
Cardiovascular disease is an important cause of death studies have focused on the specific morphology of
among patients with CKD. At present, the pathophysiolog- atherosclerotic plaque. In the present study, lesion length
ical mechanism of cardiovascular injury in patients with was longer in dialysis patients than non-dialysis patients.
CKD is still unclear; however, many studies have shown that Moreover, the length of plaque increased gradually with
there are a variety of CKD-specific cardiovascular disease duration of dialysis, indicating that these patients have a
risk factors, such as inflammation, anaemia, volume over- heavy plaque burden and the morphology of calcification is
load, oxidative stress, renineangiotensin system, sympa- different from the general population. Coronary artery
thetic nerve system, uraemic toxins, and CKD mineral bone calcification in CKD patients with dialysis often involves
disorder.17 Vascular calcification, which is associated with along a continuous segment of the vessel, which is rare in
vascular stiffness, is much more frequent in CKD patients, other populations.6,26
There are two forms of vascular calcification in CKD:
atherosclerotic calcification within neo-intimal plaques
Table 2
and medial calcification within the smooth muscle layer
Different types of plaque and degree of stenosis. that can occur in the absence of atherosclerosis.2,4,27
Arterial medial calcification might be result from disrup-
Types of plaque Total
tion in bone and mineral metabolism in patients with
Calcified Non-calcified Mixed CKD.28,29 The prevalence and progression of this form of
Uninterpretable 34 1 8 43 vascular calcification increases rapidly once patients are
Degree of stenosis
on dialysis.30,31 According to the imaging morphology of
Minimal stenosis (1e24%) 128 4 12 144
Mild stenosis (25e49%) 6 3 16 25
vascular calcification in CKD patients, especially in the
Moderate stenosis (50e69%) 1 8 11 20 dialysis group in the present study, the calcified elements
Severe stenosis (70e99%) 1 6 19 26 of plaque might mainly result from the calcification of the
Total 170 22 66 medial arterial wall or both medial and intimal arterial
Data are expressed in numbers. wall. In either case, calcification of the medial arterial wall

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003
6 W. Deng et al. / Clinical Radiology xxx (xxxx) xxx

Figure 4 Calcified plaques in patients with and without CKD. (a) A 46-year-old woman with a history of 8 years of haemodialysis for CKD. The
patient had coronary CTA for angina. Curvature planar reconstruction CT image of the LCX artery shows diffused continuous calcification with
minimal stenosis of the vessel. (b) A 79-year-old woman without history of CKD underwent coronary CTA due palpitations. Curvature planar
reconstruction CT image of the RCA shows several calcified plaques in the proximal segment of the RCA with minimal stenosis.

is an important factor in the imaging morphology of frequent as calcified plaque in CKD patients, especially
vascular lesions in CKD patients, especially in dialysis pa- non-calcified plaque. To the authors’ knowledge, few im-
tients, which is different from that in other populations. aging studies have focused on the morphology and dis-
Non-calcified plaques and mixed plaques are not as tribution of these types of plaques. The mechanism and

Figure 5 Mixed and non-calcified plaques in a 62-year-old woman with CKD for 9 years without a history of dialysis. The patient underwent
coronary CTA for palpitations. (a) Curvature planar reconstruction CT image of the LAD artery shows several mixed plaques (white arrows) in the
proximal segment of the LAD with severe stenosis. (b) Curvature planar reconstruction CT image of the RCA shows several non-calcified plaques
(white arrows) in the proximal and middle segments of the RCA with mild to severe stenosis.

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003
 W. Deng et al. / Clinical Radiology xxx (xxxx) xxx 7

Table 3
Correlations between dialysis and characteristics of plaque.

Dialysis (n¼164) Non-dialysis (n¼94) p-Value

Length of lesion (cm)a 2.55
0.22 (n¼135) 1.64
0.23 (n¼64) 0.01
Type of plaques
Calcified 125 45 <0.001
Non-calcified 6 16 <0.001
Mixed 33 33 0.008
Uninterpretable 28 15 0.82
Obstructive 21 25 0.005
Non-obstructive 115 54 0.040

Data are expressed as means
SD or numbers.

a
Spot-like lesions were excluded.

clinical significance of these plaques is still undetermined DM. Atherosclerosis in DM could result from endothelial
and further research is needed. dysfunction via various mechanisms, such as hyper-
It is well known that DM patients had a higher incidence glycaemia, insulin resistance, hyperamylinaemia, inflam-
of coronary artery calcification than the general pop- mation, changes in coagulation and fibrinolysis processes,
ulation.32e34 Other risk factors for coronary artery calcifi- dyslipidaemia, and hypertension.35 In the present study,
cation include age and CKD, which are also related to DM. nearly half of the CKD patients had a history of DM. The
Atherosclerosis, which leads to narrowing of the arterial result showed that DM patients undergoing dialysis had
walls throughout the body, is the most important patho- more mixed plaques and obstructive plaques than non-DM
logical mechanism underlying the vascular complications of patients undergoing dialysis. Moreover, lesion length in DM
patients was shorter. This result is different from what was
anticipated: DM patients on dialysis should have a higher
plaque burden and thus lesion length should be longer than
non-DM patients on dialysis. This might be caused by
differing durations of dialysis and the types of plaque. The
duration of dialysis in non-DM patients was statistically
longer than DM patients. Second, obstructive mixed plaques
were the most common type in DM patients whereas non-
obstructive calcified plaques were seen in non-DM patients.
Calcified plaque is more likely to involve the long segment
of vessels rather than lumen stenosis, whereas lesions with
non-calcified elements favoured lumen stenosis. Therefore,
DM patients had a higher total plaque volume and higher
degree of coronary stenosis, but shorter in the lesion length.
According to the MANOVA and multivariate logistic
regression results, there was an association between dial-
ysis and type of plaque. This is consistent with previous
studies that dialysis patients are inclined to develop calci-
fied plaques; however, no associations were found between
other single or multiple factors and characteristics of pla-
que. There was no association between DM and character-
istics of plaque according to the present study. This result is
inconsistent with previous studies. Previous studies on DM
and coronary atherosclerotic plaque have mainly focused on
DM patients, whereas the present study focused on CKD
patients. The total sample size included in the present study
was small; moreover, there were fewer cases of DM.
Therefore, no association between DM and plaque charac-
teristics were found in the present study. A correlation
might be found if the sample size was large enough.
A major limitation of the present study is that coronary
Figure 6 Correlation between dialysis and length of plaque. (a) The
CTA in not a routine examination for CKD. The inclusion
length of plaque in the dialysis group was longer than the non- population of the present study does not represent the
dialysis group. (b) The length of plaque was different between the majority of patients with CKD. In West China Hospital,
subgroups with differing durations of dialysis. Plaque length was coronary CTA is only scheduled for CKD patients with car-
positively correlated with duration of dialysis. diovascular symptoms or for preoperative evaluation of

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003
8 W. Deng et al. / Clinical Radiology xxx (xxxx) xxx

Table 4
Difference of lesion feature between DM and non-DM patients in dialysis population.

DM group (n¼76) Non-DM group (n¼88) p-Value

Length of lesion (cm)a 1.65
1.67 2.53
3.04 0.02
Duration of dialysis (month) 31.39
3.19 76.81
6.64 <0.001
Type of plaques
Calcified 54 71 0.15
Non-calcified 2 4 0.52
Mixed 20 13 0.07
Uninterpretable 15 13 0.40
Obstructive 16 5 0.003
Non-obstructive 45 70 0.005

Data are expressed as means
SD or numbers.

DM, diabetes mellitus.
a
Spot-like lesions were excluded.

comprehensive features of the plaques in these patients.

The present study has clinical significance in increasing the
understanding of imaging characteristics of atherosclerotic
plaques in CKD. Another limitation is that traditional coro-
nary angiography results, which is the reference standard in
terms of degree of coronary stenosis, are lacking. The most
frequent detected plaque type in the present study is
calcified plaque, which might impact the accuracy in
accessing the degree of stenosis due to blooming artefacts;
however, most calcified plaques in the present study
resulted in minimal stenosis, thus, assessment of the degree
of stenosis in such lesions is reliable.
In conclusion, coronary CTA detected a heavy plaque
burden in CKD patients. Non-obstructive calcified plaque
was the most common imaging feature. Dialysis was a risk
factor in the progression of coronary calcification with the
duration positively related to lesion length. CKD patients
with DM had more obstructive mixed plaques. In the pre-
sent series, the coronary plaques of most symptomatic pa-
tients with CKD were non-obstructive. Whether symptoms
Figure 7 Curvature planar reconstruction of RCA in a 74-year-old
are caused by coronary lesions or CKD-related cardiomy-
man with CKD and DM for 20 years and 24 years, respectively. The opathy should be further studied.
examination was scheduled for angina. There are lots of mixed pla-
ques (white arrows) in the proximal, middle, and distal segments of Table 6
the RCA with moderate to severe stenosis. Multivariate logistic regression of dialysis and type of plaques.

Dependent variables Odds ratio p-Value

patients mainly due to safety assessments before anaes- Calcified plaque 2.778 (1.538e5.025) 0.001
thesia. Therefore, a selection bias is inevitable in this study; Non-calcified plaque 0.375 (0.131e1.078) 0.068
however, previous studies using CT calcium score for Mixed plaque - -

accessing coronary lesions could not reflect the Data in parenthesis are 95% confidence intervals.

Table 5
Multivariate analysis of variance associations between clinical factors and characteristics of plaques.

Clinical factors Characteristics of plaques

Type of plaques Obstruction

Dialysis 0.049 0.902
Diabetes mellitus 0.611 0.105
Hypertension 0.619 0.353
Dialysis þ diabetes mellitus 0.861 0.114
Dialysis þ hypertension 0.947 0.421
Diabetes mellitus þ hypertension 0.252 0.102
Dialysis þ diabetes mellitus þ Hypertension - -

Data are p-values.

Obstruction¼degrees of stenosis 50%.

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003
 W. Deng et al. / Clinical Radiology xxx (xxxx) xxx 9

Conflict of interest 15. van Velzen JE, Schuijf JD, de Graaf FR, et al. Plaque type and composition
as evaluated non-invasively by MSCT angiography and invasively by VH
IVUS in relation to the degree of stenosis. Heart 2009;95(24):1990e6.
The authors declare no conflict of interest. 16. Cury RC, Abbara S, Achenbach S, et al. Coronary Artery Disease -
Reporting and Data System (CAD-RADS): an expert consensus document
of SCCT, ACR and NASCI: endorsed by the ACC. JACC Cardiovasc Imaging
2016;9(9):1099e113.
Acknowledgements
17. Fujii H, Kono K, Nishi S. Characteristics of coronary artery disease in
chronic kidney disease. Clin Exp Nephrol 2019;23(6):725e32.
The work was supported by the Sichuan Province Science 18. Gungor O, Kocyigit I, Yilmaz MI, et al. Role of vascular calcification in-
and Technology Support plan (no. 2016FZ0107), National hibitors in preventing vascular dysfunction and mortality in hemodial-
Natural Science Foundation of China (no. 81601462), ysis patients. Semin Dial 2018;31(1):72e81.
19. Go AS, Chertow GM, Fan D, et al. Chronic kidney disease and the risks of
Sichuan Provincial Health and Family Planning Commission
death, cardiovascular events, and hospitalization. N Engl J Med
Scientific Research Project (Key Research Projects; no. 2004;351(13):1296e305.
16ZD013) and 1$3$5 project for disciplines of excellence, 20. Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk
West China Hospital, Sichuan University (no. ZYGD18013). factor for development of cardiovascular disease: a statement from the
American heart association councils on kidney in cardiovascular disease,
high blood pressure research, clinical cardiology, and epidemiology and
prevention. Circulation 2003;108(17):2154e69.
References 21. Wheeler DC, London GM, Parfrey PS, et al. Effects of cinacalcet on
atherosclerotic and nonatherosclerotic cardiovascular events in patients
1. Peeters MJ, van den Brand JA, van Zuilen AD, et al. Abdominal aortic receiving hemodialysis: the EValuation of Cinacalcet HCl Therapy to
calcification in patients with CKD. J Nephrol 2017;30(1):109e18. Lower CardioVascular Events (EVOLVE) trial. J Am Heart Assoc
2. London GM, Gue rin AP, Marchais SJ, et al. Arterial media calcification in
2014;3(6):e001363.
end-stage renal disease: impact on all-cause and cardiovascular mor- 22. Graham-Brown MP, Rutherford E, Levelt E, et al. Native T1 mapping:
tality. Nephrol Dial Transplant 2003;18(9):1731e40. inter-study, inter-observer and inter-center reproducibility in hemodi-
3. Detrano R, Guerci AD, Carr JJ, et al. Coronary calcium as a predictor of alysis patients. J Cardiovasc Magn Reson 2017;19(1):21.
coronary events in four racial or ethnic groups. N Engl J Med 23. Graham-Brown MP, Patel AS, Stensel DJ, et al. Imaging of myocardial
2008;358(13):1336e45. fibrosis in patients with end-stage renal disease: current limitations and
4. Vervloet M, Cozzolino M. Vascular calcification in chronic kidney dis- future possibilities. BioMed Res Int 2017;2017:5453606.
ease: different bricks in the wall? Kidney Int 2017;91(4):808e17. 24. Charytan D, Kuntz RE, Mauri L, et al. Distribution of coronary artery
5. den Dekker MA, de Smet K, de Bock GH, et al. Diagnostic performance of disease and relation to mortality in asymptomatic hemodialysis pa-
coronary CT angiography for stenosis detection according to calcium tients. Am J Kidney Dis 2007;49(3):409e16.
score: systematic review and meta-analysis. Eur Radiol 25. Kim CD, Cho JH, Choi HJ, et al. Coronary-artery calcium scores using
2012;22(12):2688e98. electron beam CT in patients with chronic renal failure. J Korea Med Sci
6. Chu ZG, Yang ZG, Dong ZH, et al. Characteristics of coronary artery 2005;20(6):994e9.
disease in symptomatic type 2 diabetic patients: evaluation with CT 26. Halon DA, Lavi I, Barnett-Griness O, et al. Plaque morphology as pre-
angiography. Cardiovasc Diabetol 2010;9:74. dictor of late plaque events in patients with asymptomatic type 2 dia-
7. Budoff MJ, Achenbach S, Blumenthal RS, et al. Assessment of coronary betes: a long-term observational study. JACC Cardiovasc Imaging 2018.
artery disease by cardiac computed tomography: a scientific statement [Epub ahead of print].
from the American heart association committee on cardiovascular im- 27. Gross ML, Meyer HP, Ziebart H, et al. Calcification of coronary intima and
aging and intervention, council on cardiovascular radiology and inter- media: immunohistochemistry, backscatter imaging, and x-ray analysis
vention, and committee on cardiac imaging, council on clinical in renal and nonrenal patients. Clin J Am Soc Nephrol 2007;2(1):121e34.
cardiology. Circulation 2006;114(16):1761e91. 28. Isakova T, Wahl P, Vargas GS, et al. Fibroblast growth factor 23 is
8. Hokanson JE, MacKenzie T, Kinney G, et al. Evaluating changes in coro- elevated before parathyroid hormone and phosphate in chronic kidney
nary artery calcium: an analytic method that accounts for interscan disease. Kidney Int 2011;79(12):1370e8.
variability. AJR Am J Roentgenol 2004;182(5):1327e32. 29. Hill Gallant KM, Spiegel DM. Calcium balance in chronic kidney disease.
9. de Bie MK, Buiten MS, Gaasbeek A, et al. CT coronary angiography is Curr Osteoporos Rep 2017;15(3):214e21.
feasible for the assessment of coronary artery disease in chronic dialysis 30. Noordzij M, Korevaar JC, Bos WJ, et al. Mineral metabolism and car-
patients, despite high average calcium scores. PloS One diovascular morbidity and mortality risk: peritoneal dialysis patients
2013;8(7):e67936. compared with haemodialysis patients. Nephrol Dial Transplant
10. Jansz TT, van Reekum FE, Ozyilmaz A, et al. Coronary artery calcification 2006;21(9):2513e20.
in hemodialysis and peritoneal dialysis. Am J Nephrol 31. Shroff RC, McNair R, Figg N, et al. Dialysis accelerates medial vascular
2018;48(5):369e77. calcification in part by triggering smooth muscle cell apoptosis. Circu-
11. Sabour S, Atsma F, Rutten A, et al. Multi detector-row computed to- lation 2008;118(17):1748e57.
mography (MDCT) had excellent reproducibility of coronary calcium 32. Blanke P, Naoum C, Ahmadi A, et al. Long-term prognostic utility of
measurements. J Clin Epidemiol 2008;61(6):572e9. coronary CT angiography in stable patients with diabetes mellitus. JACC
12. Lee SM, Lee HW, Son YK, et al. Abdominal aortic calcification score Cardiovasc Imaging 2016;9(11):1280e8.
among several vascular calcification scores of plain radiograph is the 33. Deseive S, Straub R, Kupke M, et al. Impact of diabetes on coronary ar-
most reliable predictor of severe coronary artery calcification in dialysis tery plaque volume by coronary CT angiography and subsequent
patients. Ren Fail 2017;39(1):729e35. adverse cardiac events. J Cardiovasc Comput Tomogr 2019;13(1):31e7.
13. Deak PD, Smal Y, Kalender WA. Multisection CT protocols: sex- and age- 34. Nadjiri J, Hausleiter J, Deseive S, et al. Prognostic value of coronary CT
specific conversion factors used to determine effective dose from angiography in diabetic patients: a 5-year follow up study. Int J Car-
doseelength product. Radiology 2010;257(1):158e66. diovasc Imaging 2016;32(3):483e91.
14. Austen WG, Edwards JE, Frye RL, et al. A reporting system on patients 35. Calles-Escandon J, Garcia-Rubi E, Mirza S, et al. Type 2 diabetes: one
evaluated for coronary artery disease. Report of the ad hoc committee disease, multiple cardiovascular risk factors. Coron Artery Dis
for grading of coronary artery disease, council on cardiovascular surgery, 1999;10(1):23e30.
American heart association. Circulation 1975;51(4 Suppl):5e40.

Please cite this article as: Deng W et al., Characteristics of coronary artery atherosclerotic plaques in chronic kidney disease: evaluation with
coronary CT angiography, Clinical Radiology, https://doi.org/10.1016/j.crad.2019.06.003