Study population
The characteristics of SLE patients and healthy controls were summarized in Table 1. In this study, a total of 35 SLE patients with new-onset disease and 25 healthy controls were enrolled. 51% had stable disease to mild disease activity (n=18), 40% had moderate disease activity (n=14), and 9% had high disease activity (n=3). No cardiac symptoms were referred in SLE patients and controls. Accordingly, all individuals had normal serum markers of cardiac injury including N-terminal pro-B-type natriuretic peptide (NT-proBNP) and Troponin I, indicating absence of clinical evidence of cardiac impairment in SLE patients.
Table 1 Subjects’ characteristics of SLE patients and controls.
Variables
|
SLE(n=35)
|
Healthy controls(n=25)
|
p
|
Female/Male
|
29/6
|
20/5
|
>0.05
|
Age(years)
|
33±6
|
32±6
|
0.863
|
Immune parameters
|
|
|
|
Serum C3, gm/liter
|
0.62±0.18
|
1.13±0.08
|
<0.001
|
Serum C4, gm/liter
|
0.12±0.05
|
0.24±0.03
|
<0.001
|
ANA positive no. (%)
|
35(100)
|
0(0)
|
NA
|
Anti-dsDNA positive, no. (%)
|
25(71)
|
0(0)
|
NA
|
Antiphospholipid antibody positive, no. (%)
|
23(66)
|
0(0)
|
NA
|
High-risk profile aPL, no.(%)
|
15(42.9)
|
0(0)
|
NA
|
Disease Activity
|
|
|
|
SLEDAI-2K≤4
|
7
|
NA
|
NA
|
5≤SLEDAI-2K≤9
|
11
|
NA
|
NA
|
10≤SLEDAI-2K≤14
|
14
|
NA
|
NA
|
SLEDAI-2K≥15
|
3
|
NA
|
NA
|
Traditional cardiac risk factors, no(%)
|
|
|
|
Hypertension
|
0(0)
|
0(0)
|
NA
|
Diabetes
|
0(0)
|
0(0)
|
NA
|
Hyperlipidemia
|
3(9)
|
0(0)
|
NA
|
Smoking
|
0(0)
|
0(0)
|
NA
|
Obesity
|
0(0)
|
0(0)
|
NA
|
Gout
|
0(0)
|
0(0)
|
NA
|
Cardiac symptom, no. (%)
|
0(0)
|
0(0)
|
NA
|
Serum markers of cardiac injury
|
|
|
|
NT-proBNP(pg/L)
|
116±31
|
114±19
|
0.988
|
Troponin I, ng/ml
|
0(0-0.01)
|
0(0-0.01)
|
>0.05
|
Echocardiographic measurements
|
|
|
|
Pulmonary hypertension, no. (%)
|
5(14)
|
0(0)
|
NA
|
LVEF (%)
|
58±3
|
59±3
|
0.132
|
Diastolic dysfunction, no. (%)
|
0(0)
|
0(0)
|
NA
|
hsCRP (mg/L)
|
26.11±12.92
|
6.97±2.12
|
<0.001
|
ESR (mm/H)
|
34±11
|
14±5
|
<0.001
|
ANA= antinuclear antibody; anti-dsDNA= anti–double-stranded DNA; SLEDAI-2K= Systemic Lupus Erythematosus Disease Activity Index-2K; aPL= antiphospholipid antibody; NT-proBNP= n-terminal pro-b-type natriuretic peptide; LVEF= left ventricular ejection fraction; hsCRP= high sensitivity C-reactive protein; ESR= erythrocyte sedimentation rate
Conventional echocardiographic parameters
Conventional echocardiographic results are summarized in Table 2. All participants included had normal LVEF. No significant differences of parameters assessing systolic and diastolic function including LVEDD, LVESD, IVS, PWT, E, A, e’ and E/e’ were found between SLE patients and healthy controls(p>0.05, all). In the follow-up period of seven patients accepting HCQ monotherapy, there were no significant changes of LV diameter and LVEF.
Table 2 Conventional and TDI echocardiographic parameters in SLE patients and controls
Variables
|
SLE(n=35)
|
Healthy controls(n=25)
|
p
|
LVEDD (mm)
|
45.8±2.5
|
45.6±3.3
|
>0.05
|
LVESD (mm)
|
31.6±1.9
|
31.4±2,8
|
>0.05
|
IVS (mm)
|
9.4±0.8
|
9.6±0.7
|
>0.05
|
PWT (mm)
|
8.9±0.7
|
9.1±0.9
|
>0.05
|
E(cm/s)
|
88.0±11.3
|
90.2±11.6
|
>0.05
|
A(cm/s)
|
59.5±11.3
|
62.4±9.1
|
>0.05
|
e’(cm/s)
|
12.3±1.7
|
12.2±1.8
|
>0.05
|
E/e’
|
7.2±0.7
|
7.4±0.7
|
>0.05
|
LVEDD= left ventricular end‐diastolic diameter; LVESD= left ventricular end‐systolic diameter; IVS= interventricular septum; PWT= posterior wall thickness; E= transmitral peak velocity of early diastole; A= transmitral peak velocity of late diastole; e′ = velocity of early diastole of the mitral annulus
Layer-specific STE characteristics, left ventricular dyssynchrony index (PSD) and disease parameters in SLE patients
Representative images of layer-specific STE of one SLE patient and one health control individual are shown in Figure 1 and 2. The results of layer-specific STE parameters are displayed in Table 3. Although no clinical evidence of cardiac involvement was detected in SLE patients through routine assessment, impaired left ventricular GLS at three layers (whole layer GLS, endocardial GLS and epicardial GLS) were observed in drug-naïve patients with new-onset SLE, compared with controls (p=0.02, 0.01, 0.04, respectively). Elevation of PSD, which reflects left ventricular dyssynchrony, was found in SLE patients (p<0.01). In contrast, there’s also no significant difference of the GCS of the three layers (midmyocardium GCS, endocardium GCS and epicardium GCS) at the basal level, papillary level and apical level between SLE patients and controls (p>0.05, all).
In addition, more severe impairment of GLS was observed in SLE patients with moderate to severe active disease, high-risk aPL profile or lupus nephritis, compared with those with inactive to mild active disease, negative aPL to low-risk aPL profile or without evidence of renal involvement, respectively (Table 4). In contrast, PSD significantly increased in patients with higher disease activity or high-risk aPL, but not in those with lupus nephritis (Table 4).
Correlation analysis showed that endocardial GLS, whole layer GLS, epicardial GLS and PSD correlated with high-sensitivity CRP (hsCRP) levels. PSD also correlated with the three layer-specific GLS parameters above (Figure 3). Partial correlation analysis showed that PSD correlated with epicardial GLS, but not whole layer GLS or endocardial GLS, when treating hsCRP level, renal involvement, profile of aPL and disease activity as control variable (Table 5). Multivariate regression analysis showed that hsCRP level and epicardial GLS are the predictors of layer-specific GLS impairment and PSD change, respectively (Table 6).
Table 3 Lay-specific strain data and left ventricular dyssynchrony index parameters in SLE patients and controls
Variables
|
SLE
|
Healthy controls
|
p
|
Whole layer GLS
|
-17.6±3.0
|
-19.3±2.6
|
0.02
|
Endocardial GLS
|
-20.0±3.2
|
-22.1±3.0
|
0.01
|
Epicardial GLS
|
-15.6±2.7
|
-16.8±2.4
|
0.04
|
GCS at basal level
|
|
|
|
Endocardium
|
-25.6±6.3
|
-24.4±6.0
|
0.458
|
Mid‐myocardium
|
-18.0±4.4
|
-16.9±4.5
|
0.376
|
Epicardium
|
-12.8±4.0
|
-12.2±3.9
|
0.675
|
GCS at papillary muscle level
|
|
|
|
Endocardium
|
-29.0±5.5
|
-27.1±7.0
|
0.239
|
Mid‐myocardium
|
-20.3±3.9
|
-18.3±5.3
|
0.124
|
Epicardium
|
-14.2±4.5
|
-12.5±4.3
|
0.136
|
GCS at apical level
|
|
|
|
Endocardium
|
-33.8±5.7
|
-32.7±8.4
|
0.605
|
Mid‐myocardium
|
-23.4±4.9
|
-22.5±6.3
|
0.686
|
Epicardium
|
-16.0±5.6
|
-15.5±5.9
|
0.708
|
PSD
|
41.0±18.9
|
28.8±10.1
|
0.007
|
STE= speckle tracking echocardiography; GCS= global circumferential strain; GLS= global longitudinal strain; PSD= peak systolic dispersion.
Table 4 Layer-specific longitudinal parameters, PSD and certain disease-related contributors of cardiovascular risk in SLE patients
|
Whole layer GLS
|
Endocardial GLS
|
Epicardial GLS
|
PSD
|
Disease activity
|
|
|
|
|
Inactive to mild active disease
|
-19.3±2.3
|
-21.8±2.6
|
-17.1±2.2
|
34.1±16.1
|
Moderate to severe active disease
|
-15.8±2.5
|
-18.1±2.8
|
-13.9±2.3
|
48.2±19.3
|
p
|
<0.001
|
<0.001
|
<0.001
|
0.014
|
Renal involvement
|
|
|
|
|
Without lupus nephritis
|
-19.5±2.1
|
-22.0±2.4
|
-17.4±2.0
|
35.1±17.4
|
With lupus nephritis
|
-16.6±2.9
|
-19.0±3.2
|
-14.6±2.6
|
44.0±19.3
|
p
|
<0.001
|
0.001
|
0.002
|
0.062
|
aPL profile
|
|
|
|
|
Negative and low-risk aPL profile
|
-18.9±2.5
|
-21.3±2.7
|
-16.8±2.3
|
36.4±18.4
|
High-risk aPL profile
|
-15.9±2.8
|
-18.3±3.2
|
-14.0±2.5
|
47.1±18.4
|
p
|
0.002
|
0.002
|
0.004
|
0.037
|
GCS= global circumferential strain; GLS= global longitudinal strain; PSD = peak systolic dispersion; aPL= antiphospholipid antibody
Table 5 Partial correlation analysis between PSD and layer-specific GLS parameter
Control Variable
|
Variable
|
r
|
p
|
HsCRP level
Renal involvement
Profile of aPL
Disease activity
|
PSD
|
Whole layer GLS
Endocardial GLS
Epicardial GLS
|
0.296
0.212
0.360
|
0.106
0.252
0.047
|
|
|
|
|
|
PSD= peak systolic dispersion; GLS= global longitudinal strain; hsCRP= high sensitivity C-reactive protein; aPL= antiphospholipid antibody
Table 6 Multivariate linear regression to analyze predictor of layer-specific GLS parameter and PSD
|
Model
|
Unstandardized coefficients
|
Standardized coefficients
|
p
|
B
|
Std. error
|
Beta
|
t
|
Whole layer GLS
Endocardial GLS
Epicardial GLS
PSD
|
Constant
HsCRP level
Disease activity
Renal involvement
Profile of aPL
Constant
HsCRP level
Disease activity
Renal involvement
Profile of aPL
Constant
HsCRP level
Disease activity
Renal involvement
Profile of aPL
Constant
HsCRP level
Disease activity
Profile of aPL
Epicardial GLS
|
-21.725
0.121
1.728
-0.164
0.545
-24.454
0.139
2.028
-0.433
0.295
-19.294
0.104
1.385
0.055
0.730
95.394
-0.123
5.306
-0.995
3.429
|
0.815
0.032
1.211
1.064
0.952
0.909
0.036
1.351
1.187
1.063
0.767
0.031
1.140
1.001
0.896
32.226
0.315
8.505
7.972
1.635
|
0.528
0.296
-0.027
0.092
0.556
0.317
-0.064
0.046
0.492
0.258
0.010
0.134
-0.084
0.142
-0.026
0.494
|
-26.656
3.717
1.426
-0.154
0.572
-26.893
3.849
1.500
-0.365
0.278
-25.161
3.400
1.215
0.055
0.815
2.960
-0.390
0.624
-0.125
2.097
|
<0.001
0.001
0.164
0.878
0.572
<0.001
0.001
0.144
0.718
0.783
<0.001
0.002
0.234
0.957
0.422
0.006
0.699
0.537
0.901
0.045
|
|
|
|
|
|
|
|
|
|
|
|
|
|
GLS= global longitudinal strain; PSD= peak systolic dispersion; hsCRP= high sensitivity C-reactive protein; aPL= antiphospholipid antibody
Effect of HCQ monotherapy on layer-specific GLS parameters and PSD in SLE patients with stable disease
The effect of HCQ monotherapy for 6 months and 12 months on GLS parameters and PSD was assessed in 7 SLE patients with stable disease. The results were summarized in Table 7. Figure 4 showed change of Bull’s eye plot during HCQ monotherapy in one patient at different points in time. Cardiac enzyme, Troponin I, NT-proBNP and LVEF were within normal arrange during follow-up.
Endocardial GLS and whole layer GLS showed no change after HCQ treatment for 6 months, compared with baseline. During the second 6 months of HCQ treatment, significant increase of the two strain parameters was detected. Similarly, significant decrease of PSD was observed in the second 6 months of HCQ treatment. No change of epicardial GLS was observed during follow-up.
Table 7 Effect on GLS parameters and PSD by HCQ monotherapy in SLE patients with stable disease
Variables
|
Baseline
|
6 months
|
12 months
|
p
|
Whole layer GLS
|
-19.0±1.8
|
-19.3±2.1
|
-21.1±2.6**
|
<0.05
|
Endocardial GLS
|
-21.5±2.0
|
-21.8±2.4
|
-23.8±2.8**
|
<0.05
|
Epicardial GLS
|
-16.8±1.8
|
-17.1±2.0
|
-18.7±2.5
|
>0.05
|
PSD
|
49.9±15.7
|
39.9±7.8
|
29.7±10.1*,**
|
<0.01
|
|
|
|
|
|
|
|
|
|
GLS= global longitudinal strain; PSD= peak systolic dispersion; HCQ= Hydroxychloroquine
*p<0.05 compared with baseline.
**p<0.05 compared with 6 months.