If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author. Department of Medical Ultrasound, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, 1095 Jiefang Road, Wuhan 430030, China.
Affiliations
Department of Medical Ultrasound, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
The study was performed to explore the predictive value of multiple strain parameters for myocardial fibrosis in patients with hypertrophic cardiomyopathy (HCM) by using speckle tracking automated functional imaging (AFI).
Methods
A total of 61 patients diagnosed with HCM were finally enrolled in this study. All patients completed transthoracic echocardiography and cardiac magnetic resonance late gadolinium enhancement (LGE) within 1 month. Twenty age- and sex-matched healthy participants were included as the control group. Multiple parameters, including segmental longitudinal strain (LS), global longitudinal strain (GLS), post-systolic index and peak strain dispersion, were automatically analyzed by AFI.
Results
A total of 1458 myocardial segments were analyzed according to the left ventricular 18-segment model. Among the 1098 segments from HCM patients, segments with LGE had a lower absolute value of segmental LS than those without LGE (p < 0.05). The cutoff values of segmental LS for predicting positive LGE in the basal, intermediate and apical regions were –12.5%, –11.5% and –14.5%, respectively. GLS could predict significant myocardial fibrosis (≥2 positive LGE segments) at a cutoff value of –16.5% with a sensitivity of 80.9% and specificity of 76.5%. As an independent predictor of significant myocardial fibrosis, GLS was substantially associated with the severity of myocardial fibrosis and 5 years sudden cardiac death risk score in HCM patients.
Conclusion
Speckle tracking AFI could efficiently identify left ventricular myocardial fibrosis in patients with HCM by multiple parameters. GLS predicted significant myocardial fibrosis at a cutoff value of –16.5%, which may indicate the adverse clinical outcomes in HCM patients.
Hypertrophic cardiomyopathy (HCM) is a global hereditary cardiovascular disease affecting 1 in 500 people, with heart failure and sudden cardiac death (SCD) being the leading causes of death [
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
]. In addition to heterogeneous myocardial hypertrophy and cardiomyocyte derangement, myocardial fibrosis is also recognized as an important histopathological feature of HCM. Myocardial fibrosis is a progressive irreversible process, which results in myocardial remodeling, electrical inhomogeneity and myocardial systolic and diastolic dysfunction [
]. A growing body of literature reports that diffuse myocardial fibrosis is closely associated with adverse clinical events in patients with HCM, such as malignant arrhythmias, heart failure and SCD [
Currently, myocardial biopsy is the clinical gold standard for detecting myocardial fibrosis. However, because of its invasive nature and limited sampling sites, myocardial biopsy is not routinely used to detect myocardial fibrosis. By contrast, cardiac magnetic resonance late gadolinium enhancement (CMR LGE) is a non-invasive method for detecting myocardial fibrosis and could provide valuable prognostic information for HCM patients [
]. Unfortunately, CMR LGE still cannot be widely used in clinics because of the limits of the patients' physical and mental condition, the high cost and the side effects of contrast agents. Therefore, it is quite necessary to adopt a more convenient, affordable and rapid parametric model for screening myocardial fibrosis.
Transthoracic echocardiography (TTE) is the most common and preferred non-invasive imaging modality for the evaluation and diagnosis of HCM [
]. Although traditional 2-D images provide limited information on myocardial function and prognosis, many previous studies determined that speckle tracking echocardiography (STE) could provide more sensitive parameters for the early identification of subtle myocardial injury, which contribute to improve the patient risk stratification and prognosis assessment [
Association between regional ventricular function and myocardial fibrosis in hypertrophic cardiomyopathy assessed by speckle tracking echocardiography and delayed hyperenhancement magnetic resonance imaging.
Clinical significance of global two-dimensional strain as a surrogate parameter of myocardial fibrosis and cardiac events in patients with hypertrophic cardiomyopathy.
Investigation of global and regional myocardial mechanics with 3-dimensional speckle tracking echocardiography and relations to hypertrophy and fibrosis in hypertrophic cardiomyopathy.
], indicating a close association between the strain parameters and the extent of fibrosis. Different from the previous studies, STE automatic functional imaging (AFI) was performed in the present study to automatically provide multiple strain parameters [
Feasibility and correlation of standard 2D speckle tracking echocardiography and automated function imaging derived parameters of left ventricular function during dobutamine stress test.
Post-systolic shortening is superior to global longitudinal strain in predicting adverse events in patients with stable coronary artery disease and preserved systolic function.
], contributing to the rapid clinical screening and follow-up in HCM patients. It is worth noting that the absolute value of left ventricular (LV) myocardial longitudinal strain normally followed a stepwise increasing trend from the basal to apical segmental levels. Although segmental longitudinal strain was considered a valuable indicator for fibrosis detection in the previous studies [
Quantitative differentiation of LV myocardium with and without layer-specific fibrosis Using MRI in hypertrophic cardiomyopathy and layer-specific strain TTE analysis.
], the differences in segmental values at different levels of the myocardium were not taken into account. In our study, more detailed predictive values of segmental strain parameters according to the myocardial levels were explored. Moreover, we further explored the value of global strain parameters by AFI in risk stratification for HCM patients.
Methods
Study population
From October 2019 to April 2022, 437 patients were clinically diagnosed with HCM in our hospital. According to established diagnostic guidelines, HCM was defined as LV wall thickness ≥15 mm in an adult that could not be explained by abnormal loading conditions [
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
]. Among all the HCM patients, 79 underwent CMR LGE examination within 1 month of TTE examination. Exclusion criteria were (i) age less than 18 years old; (ii) presence of aortic stenosis or myocardial amyloidosis; (iii) history of myocardial infarction disease or chronic kidney disease; (iv) history of cardiac resynchronization therapy (CRT), implantable cardioverter (ICD) or claustrophobia; and (v) poor image quality in speckle tracking AFI analysis or persistent arrhythmia. Finally, a total of 61 patients were included in the present study.
This study was approved by the ethics committee of Tongji Hospital, and informed consent was obtained from all patients. Informed consent for the procedures was signed by patients or family members and was approved by the institutional review board of Tongji Hospital.
Transthoracic echocardiography
Transthoracic echocardiography was performed on patients in the left decubitus position by an experienced doctor using a GE Vivid E95 ultrasound device (GE Vingmed Ultrasound, Horten, Norway). Images were acquired using M5Sc transducers (1.7–3.4 MHz) with high frame rates (>70 fps). After optimization of sector size, gain, depth, compression and time-gain compensation, the LV parasternal long-axis view, short-axis views at the basal, midventricular and apical levels, as well as apical two-, three- and four-chamber views were stored in three-cardiac-cycle cine format. All echocardiograms were analyzed according to the quantitative methods recommended by the American Society of Echocardiography and the European Society for Cardiovascular Imaging [
Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
]. In the parasternal long-axis view, the thickness of the interventricular septum (IVS) was measured at the level of the mitral valve tip at the end of diastole, and left atrial (LA) diameter was the vertical distance from the aortic sinus to the posterior wall of the left atrium at end systole of the left ventricle [
Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
]. In the apical four-chamber view, pulse wave Doppler imaging was used to assess peak velocities of mitral orifice filling in early (E) and late (A) diastole, and tissue wave Doppler imaging was used to assess early diastole at the mitral annulus in the IVS peak speed (e'). The E/A and E/e' ratios were calculated. Left ventricular outflow tract obstruction (LVOTO) was defined as a transient peak pressure gradient in the LV outflow tract at rest or during stress test ≥30 mm Hg [
Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
]. All measurements were analyzed by experienced doctors who were blinded to all clinical data.
Speckle tracking automated functional imaging
The machine built-in speckle tracking AFI was immediately performed after transthoracic echocardiography. The commercial imaging analysis AFI software (version 2.4, GE Vingmed Ultrasound) automatically selected the three best-matched standard apical views and analyzed global and regional myocardial motion [
Post-systolic shortening is superior to global longitudinal strain in predicting adverse events in patients with stable coronary artery disease and preserved systolic function.
Non-invasive global and regional myocardial work predicts high-risk stable coronary artery disease patients with normal segmental wall motion and left ventricular function.
]. Finally, LV 18-segment bull's-eye diagrams related to various strain parameters (Fig. 1), including segmental longitudinal strain (LS), global longitudinal strain (GLS), post-systolic index (PSI), time to peak (TTP) and peak strain dispersion (PSD), were automatically displayed. Segmental LS was maximum systolic longitudinal shortening in each segment [
Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
Post-systolic shortening is superior to global longitudinal strain in predicting adverse events in patients with stable coronary artery disease and preserved systolic function.
Two-dimensional speckle-tracking during dobutamine stress echocardiography in the detection of myocardial ischemia in patients with suspected coronary artery disease.
Figure 1Bull's eye map of speckle tracking AFI in HCM patients without (A, B) and with (C, D) significant myocardial fibrosis assessed by CMR LGE. (A) Mild abnormal GLS with reduced absolute value of segmental LS in localized myocardial segments. (B) No positive LGE on multiple left ventricular CMR views. (C) Obviously abnormal GLS with reduced absolute value of segmental LS in a wide range of myocardial segments (anterior wall, interventricular septum and inferior wall from the basal level to midlevel). (D) Significant positive LGE in the anterior wall, interventricular septum and inferior wall from basal level to midlevel (red arrows) on CMR views. AFI, automated functional imaging; CMR LGE, cardiac magnetic resonance late gadolinium enhancement; GLS, global longitudinal strain; HCM, hypertrophic cardiomyopathy; LS, longitudinal strain; PSD, peak strain dispersion; PSI, post-systolic index.
Post-systolic shortening is superior to global longitudinal strain in predicting adverse events in patients with stable coronary artery disease and preserved systolic function.
A comprehensive CMR examination was performed by operators blinded to the patient's echocardiographic data using a 3.0-T MR scanner (MAGNETOM Skyra, Siemens Healthcare, Erlangen, Germany). Image acquisition was performed with electrocardiogram gating during breath hold, resulting in four-chamber and parallel contiguous short-axis cines from the mitral plane to the apex. After 8–10 min of intravenous injection of gadolinium-based contrast agent, visual analysis was performed to determine the presence of LGE on LV short-axis sections (Fig. 1). The positive LGE segments were counted, and positive LGE ≥2 segments were defined as a significant fibrosis [
] in patients with HCM. In addition, segmental wall thickness (SWT) and maximum wall thickness (MWT) was measured on end-diastolic LV short-axis views [
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
where prognostic index = [0.15939858 × MWT (mm)][0.00294271× MWT2 (mm2)] + [0.0259082 × LA diameter (mm)] + [0.00446131× maximal (rest/Valsalva) LVOT (mm Hg)] + [0.4583082 × family history SCD] + [0.82639195 × non-sustained ventricular tachycardia] +[0.71650361 × unexplained syncope][0.01799934 × age at clinical evaluation (y)].
Reproducibility
To determine intra- and inter-observer variability, 20 patients were randomly selected and remeasured by two observers blinded to patient data and each other's results. One of the researchers did the analysis once again after 2 week. Intra- and inter-observer reproducibility and variability were calculated using the intra-class correlation coefficient (ICC).
Statistical analysis
All continuous variables were assessed for normality using the Shapiro–Wilk test. Normally distributed data are expressed as the mean ± standard deviation (SD) and compared using Student's t-test. The median and interquartile range (IQR) were used for non-normally distributed data, and the Mann–Whitney U-test was performed for comparison. Categorical data are expressed as frequencies and percentages, which were compared with the χ2-test or Fisher's exact test. To explore the predictive value of each parameter in detecting myocardial fibrosis, a receiver operating characteristic (ROC) curve was used, and the area under the curve (AUC) was calculated. In addition, the optimal cutoff value for specificity or sensitivity was calculated by maximizing the Youden index. The DeLong method in MedCalc was used to compare AUCs. To determine the relationship between measurements and significant myocardial fibrosis, binary logistic regression was used for analysis and expressed as odds ratios (ORs). To detect independent risk factors for predicting significant myocardial fibrosis, variables with significant p values in univariate analysis were further subjected to multivariate binary logistic regression analysis. Pearson's correlation coefficient (r) was used to examine the association between GLS and the number of positive LGE segments and 5 years SCD risk score in HCM patients. Inter- and intra-observer variability of strain parameters was assessed using Bland–Altman analysis and expressed using the ICC. A two-tailed p < 0.05 was considered significant in all tests. All statistical analyses were performed using SPSS, version 25.0 (IBM Corp., Armonk, NY, USA), and MedCalc version 18.11.3 (MedCalc Software, Ostend, Belgium).
Results
Clinical characteristics of the study population
A total of 61 HCM patients (mean age: 50 ± 13 y, 72% men) and 20 age- and sex-matched healthy participants (mean age: 50 ± 4 y, 75% men) were enrolled in the present study. Among them, 46 patients were tested for the disease-causing gene, of whom 14 were determined to be positive. According to the number of positive LGE segments, HCM patients were divided into two groups, LGE <2 segments (n = 14) and LGE ≥2 segments (n = 47). Detailed patient characteristics according to study group are summarized in Table 1. No significant difference was found in age, sex, body mass index (BMI), heart rate and blood pressures between the control and HCM groups (p > 0.05). Of all the HCM patients, no significant difference was detected in symptoms, New York Heart Association (NYHA) class and histories between two groups with or without significant myocardial fibrosis.
Table 1Clinical characteristics of the study population
Controls (n = 20)
All HCM (n = 61)
p Value
All HCM (n = 61)
p Value
LGE <2 segments (n = 14)
LGE ≥2 segments (n = 47)
Age, y
50 ± 4
50 ± 13
0.649
47 ± 14
52 ± 13
0.251
Male, n (%)
15 (75.0)
44 (72.1)
0.499
11 (78.6)
33 (70.2)
0.540
Body mass index, kg/m2
23.9 ± 3.2
24.8 ± 4.3
0.415
23.9 ± 4.3
25.1 ± 4.4
0.403
Systolic blood pressure, mm Hg
126 ± 14
134 ± 20
0.082
136 ± 16
134 ± 22
0.781
Diastolic blood pressure, mm Hg
79 ± 8
80 ± 12
0.972
82 ± 10
80 ± 12
0.380
Heart rate, beats/min
72 ± 10
76 ± 13
0.242
76 ± 10
76 ± 14
0.997
QRS duration, ms
79 (71, 83)
98 (90, 108)
<0.001
96 ± 14
100 ± 17
0.385
BBB, n (%)
0 (0)
8 (13.6)
0.192
2 (15.4)
6 (13.0)
0.830
NSVT, n (%)
0 (0)
1 (1.6)
0.565
0 (0)
1 (2.1)
0.582
NT-proBNP, pg/mL
37 (22, 72)
–542 (323, 974)
<0.001
314 (98, 468)
762 (398, 1392)
0.002
hs-cTnI, pg/mL
2.7 (1.6, 4.1)
32 (13, 111)
<0.001
38 (4, 1740)
32 (14, 82)
0.831
Chest pain, n (%)
0 (0)
23 (41.1)
<0.001
4 (30.8)
19 (44.2)
0.389
Dyspnea, n (%)
0 (0)
12 (21.4)
0.029
4 (30.8)
8 (18.6)
0.349
Hypertension, n (%)
0 (0)
27 (48.2)
<0.001
6 (46.2)
21 (48.8)
0.865
Diabetes, n (%)
0 (0)
5 (8.9)
0.317
1 (7.7)
4 (9.3)
0.858
Coronary heart disease, n (%)
0 (0)
12 (21.8)
0.029
3 (25.0)
9 (20.9)
0.763
NYHA class >2, n (%)
0 (0)
5 (9.1)
0.316
1 (7.7)
4 (9.5)
0.841
History of atrial fibrillation, n (%)
0 (0)
5 (8.9)
0.317
1 (7.7)
4 (9.3)
0.858
Family history of HCM, n (%)
0 (0)
10 (21.7)
0.026
1 (9.1)
9 (25.7)
0.341
Disease-causing mutation, n (%)
0 (0)
15 (32.0)
0.003
2 (16.7)
14 (36.6)
0.410
5-y SCD risk score, %
NA
2.0 ± 1.2
NA
1.6 ± 0.6
2.1 ± 1.3
0.208
Data are expressed as the mean ± standard deviation or median (interquartile range) for continuous variables and count (%) for categorical variables. Values of p for statistically significant comparisons (p < 0.05) are in boldface.
General imaging characteristics of the study population
General echocardiographic and cardiovascular magnetic resonance parameters are listed in Table 2. HCM patients had increased MWT, E/e', LAVI, LVMI and LVEF compared with the control individuals (p < 0.05 for all). LVOT obstruction was found in 20 (32.8%) patients, and in these, the mean resting peak gradient was 47 mm Hg. In the HCM group, patients with significant myocardial fibrosis (LGE ≥2 segments) had higher NT-proBNP, larger MWT, more SWT ≥15 mm and more LGE-positive segments in comparison with the patients without or with mild myocardial fibrosis (p < 0.05 for all). However, there were no significant differences found in the number with LVOTO, LV volume indexes, LA volume index and LVEF between the aforementioned two groups.
Table 2General imaging characteristics of the study population
Controls (n = 20)
All HCM (n = 61)
p Value
All HCM (n = 61)
p Value
LGE <2 segments (n = 14)
LGE ≥2 segments (n = 47)
MWT, mm
9.7 ± 0.9
20.5 ± 6.1
<0.001
16.4 ± 2.2
21.5 ± 6.3
0.021
Number with SWT ≥15 mm
0 (0,0)
3 (1, 5)
<0.001
1 (0, 2)
4 (2, 5.8)
0.003
Number of positive LGE segments
—
5.3 ± 4.2
—
0.6 ± 0.9
6.7 ± 3.8
<0.001
IVS, mm
8.5 ± 1.6
17.7 ± 5.1
<0.001
15.4 ± 4.5
18.3 ± 5.1
0.062
LA diameter, mm
27.7 ± 9.1
38.6 ± 6.0
<0.001
37.0 ± 7.2
39.1 ± 5.5
0.258
Apical HCM, n (%)
0 (0)
16 (26.2)
0.009
7 (50.0)
9 (19.1)
0.021
Asymmetrical (RWT >1.3), n (%)
0 (0)
50 (82.0)
<0.001
10 (71.4)
40 (85.1)
0.243
Apical aneurysm, n (%)
0 (0)
2 (3.3)
0.412
0 (0)
2 (4.3)
0.433
Obstructive HCM, n (%)
0 (0)
20 (32.8)
0.002
3 (21.4)
17 (36.2)
0.302
Mitral E/A
1.2 ± 0.3
1.0 ± 0.4
0.133
1.02 ± 0.3
0.99 ± 0.4
1.000
Septal E/e′
9.1 ± 2.3
15.2 ± 7.6
0.001
13.1 ± 6.2
15.8 ± 7.9
0.245
LVESVI, mL/m2
30.6 ± 7
27.8 ± 13
0.358
26 ± 15
28 ± 12
0.459
LVEDVI, mL/m2
65.8 ± 12
62.1 ± 23
0.372
63 ± 28
62 ± 22
0.947
LAVI, mL/m2
27.7 ± 9.1
44.6 ± 18
0.001
39.7 ± 20.1
44.9 ± 18.4
0.439
LVMI, g/m2
70.3 ± 6.4
84.6 ± 8.1
<0.001
81.9 ± 9.7
85.4 ± 7.6
0.181
LVEF, %
61.5 ± 4.2
67.1 ± 5.9
0.001
68.7 ± 4.8
67.9 ± 5.5
0.352
Data are expressed as the mean ± standard deviation or median (interquartile range) for continuous variables and count (%) for categorical variables. Values of p for statistically significant comparisons (p < 0.05) are in boldface.
E/A, mitral inflow peak early velocity/mitral inflow peak late velocity; E/e′, mitral inflow peak early velocity/mitral annular peak early velocity; HCM, hypertrophic cardiomyopathy; IVS, interventricular septum; LA, left atrial; LAVI, left atrial volume index; LGE, late gadolinium enhancement; LVEDVI, left ventricular end-diastolic volume index; LVEF, left ventricular ejection fraction; LVESVI, left ventricular end-systolic volume index; LVMI, left ventricular mass index; MWT, maximum wall thickness; RWT, relative wall thickness; SWT, segmental wall thickness.
Regional myocardial characteristics and predictive value for segmental myocardial fibrosis
A total of 1458 LV myocardial segments were analyzed, including 360 (24.7%) segments in controls and 1098 (75.3%) segments in HCM patients. Regional AFI analysis results in the present study are outlined in Table 3. Of all the segments in HCM patients, LGE was found in 79 (7.2%), 90 (8.2%) and 84 (7.7%) segments in the basal, intermediate and apical regions, respectively. Comparison with the segments in controls, segmental LS, PSI and TTP in HCM patients were significantly abnormal (p < 0.05 for all). Among all the segments in patients with HCM, segments with LGE had lower absolute LS values and larger SWT values than those without LGE from the basal to apical region (p < 0.05). However, significantly elevated PSI and TTP levels in LGE segments were only found in basal regions compared with those without LGE (p < 0.05 for all).
Table 3Regional myocardial characteristics in controls and HCM patients
Segments of controls
Segments of all HCM
p Value
Segments of all HCM (n = 366)
p Value
Segments without LGE
Segments with LGE
Basal
n = 120
n = 366
n = 287
n = 79
SLS, %
–15.4 ± 4.7
–12.7 ± 7.9
<0.001
–14.1 ± 7.4
–7.2 ± 7.4
<0.001
PSI, %
7.8 ± 11
22.0 ± 82.7
0.001
12.9 ± 34.8
41.8 ± 109.5
<0.001
TTP, ms
381 ± 66
401 ± 108
0.016
391 ± 97
435 ± 135
0.002
SWT, mm
8.7 ± 1.4
12.0 ± 5.7
<0.001
10.6 ± 5.1
16.7 ± 5.4
<0.001
Intermediate
n = 120
n = 366
n = 276
n = 90
SLS, %
–19.2 ± 3.4
–12.5 ± 6.3
<0.001
–13.4 ± 6.2
–10.7 ± 5.9
<0.001
PSI, %
1.0 ± 2.8
9.6 ± 20.2
<0.001
10.0 ± 22.3
8.6 ± 14.6
0.526
TTP, ms
342 ± 48
386 ± 98
<0.001
384 ± 87
392 ± 117
0.479
SWT, mm
8.6 ± 1.5
12.1 ± 6.2
<0.001
10.3 ± 5.4
15.5 ± 6.0
<0.001
Apical
n = 120
n = 366
n = 282
n = 84
SLS, %
–24.8 ± 4.3
–15.1 ± 9.3
<0.001
–16.4 ± 9.3
–12.6 ± 8.8
<0.001
PSI, %
1.2 ± 2.7
13.9 ± 39.2
0.001
11.6 ± 28.1
18.6 ± 55.2
0.115
TTP, ms
350 ± 52
382 ± 128
0.014
385 ± 124
374 ± 137
0.437
SWT, mm
8.8 ± 1.3
10.4 ± 5.4
<0.001
8.6 ± 3.3
13.9 ± 6.7
<0.001
Data are expressed as the mean ± standard deviation or median (interquartile range) for continuous variables and count (%) for categorical variables. Values of p for statistically significant comparisons (p < 0.05) are in boldface.
HCM, hypertrophic cardiomyopathy; LGE, late gadolinium enhancement; PSI, post-systolic strain index; SLS, segmental longitudinal strain; SWT, segmental wall thickness; TTP, time to peak.
The ROC analysis of regional myocardial parameters used to detect segmental myocardial fibrosis in different LV myocardium regions is outlined in Table 4. Among various regional AFI parameters, segmental LS was superior to PSI and TTP in identifying fibrosis segments in HCM patients (p < 0.05 for all). The cutoff values of segmental LS in identifying positive LGE in the basal, intermediate and apical regions were –12.5% (AUC = 785, p < 0.001), –11.5% (AUC = 0.709, p < 0.001), and –14.5% (AUC = 0.700, p < 0.001), respectively.
Table 4ROC curve for predicting regional fibrosis in HCM patients
AUC
SE
AUC (95% CI)
p Value
p Value (vs. SLS)
Cutoff value
Sensitivity (%)
Specificity (%)
Basal
SLS, %
0.785
0.027
0.738–0.838
<0.001
—
–12.5
79.7
64.8
PSI, %
0.594
0.041
0.514–0.675
0.010
<0.001
13.5
44.0
77.8
TTP, ms
0.655
0.040
0.577–0.732
<0.001
0.004
457
49.3
83.3
SWT, mm
0.809
0.028
0.753–0.864
<0.001
0.431
12.4
82.9
69.6
Intermediate
SLS, %
0.709
0.027
0.656–0.763
<0.001
—
–11.5
61.3
72.5
PSI, %
0.545
0.032
0.483–0.608
0.149
0.035
5.5
39.3
70.8
TTP, ms
0.564
0.033
0.500–0.628
0.041
0.011
467
25.0
89.3
SWT, mm
0.768
0.027
0.715–0.822
<0.001
0.001
11.3
73.7
74.0
Apical
SLS, %
0.700
0.027
0.647–0.754
<0.001
—
–14.5
66.9
69.5
PSI, %
0.596
0.032
0.534–0.658
0.002
0.006
9.5
38.5
80.6
TTP, ms
0.539
0.034
0.473–0.605
0.206
0.001
407
48.3
68.8
SWT, mm
0.792
0.026
0.741–0.842
<0.001
<0.001
11.0
63.2
81.1
Data are expressed as the mean ± standard deviation or median (interquartile range) for continuous variables, and count (%) for categorical variables. Values of p for statistically significant comparisons (p < 0.05) are in boldface.
AUC, area under receiver operating characteristic curve; CI, confidence interval; HCM, hypertrophic cardiomyopathy; PSI, post-systolic strain index; ROC, receiver operating characteristic; SE, standard error; SLS, segmental longitudinal strain; SWT, segmental wall thickness; TTP, time to peak.
Global AFI parameters and predictive value for significant myocardial fibrosis in HCM patients
GLS, PSI and PSD were significantly abnormal in HCM patients compared with controls (p < 0.05 for all). Among the HCM patients, however, no significant difference was found in global PSI and PSD (p > 0.05 for all). Only the absolute value of GLS in the LGE ≥2 segment group was significantly lower than that in the LGE <2 segment group (p < 0.05). Global AFI parameters of the study population are illustrated in Fig. 2.
Figure 2Global speckle tracking AFI parameters of the study population. Column plots of GLS (A), GPSI (B) and PSD (C) between controls, HCM patients with LGE <2 segments and HCM patients with LGE ≥2 segments. *p < 0.05, **p < 0.001. AFI, automated functional imaging; GLS, global longitudinal strain; GPSI, global post-systolic index; HCM, hypertrophic cardiomyopathy; LGE, late gadolinium enhancement; PSD, peak strain dispersion.
Receiver operating characteristic curve analysis of the global parameters of predicting significant myocardial fibrosis is outlined in Fig. 3. In controls and HCM patients, the AUCs of GLS, GPSI and PSD for predicting significant myocardial fibrosis were 0.825 (p < 0.001, cutoff value, –16.5%; sensitivity, 80.9%; specificity, 76.5%), 0.786 (p < 0.001, cutoff value, 8.6%; sensitivity, 77.8%; specificity, 82.4%) and 0.810 (p < 0.001, cutoff value, 72.5%; sensitivity, 78.7%; specificity, 76.5%), respectively, whereas LVEF had no statistical significance. In HCM patients, only GLS had a statistically significant value predicting significant myocardial fibrosis (AUC = 0.682; p = 0.04; cutoff value, –16.5%; sensitivity, 80.9%; specificity, 57.1%).
Figure 3Receiver operating characteristic curves of GLS, GPSI, PSD and LVEF in predicting significant myocardial fibrosis in controls and HCM patients (A) and in HCM patients (B). GLS, global longitudinal strain; GPSI, global post-systolic index; LVEF, left ventricular ejection fraction; PSD, peak strain dispersion.
Univariate and multivariate logistic regression analysis for predictors of significant myocardial fibrosis in HCM patients
The logistic regression analysis for predicting significant myocardial fibrosis in HCM patients is outlined in Table 5. The univariate logistic analysis found that NT-proBNP, MWT, SWT ≥15 mm, apical HCM and GLS were significantly related to significant myocardial fibrosis (p < 0.05 for all). Further multivariate logistic analysis indicated that GLS was an independent predictor of significant myocardial fibrosis (odds ratio [OR] = 1.541; 95% confidence interval [CI]: 1.005–2.346; p < 0.05).
Table 5Univariate and multivariate logistic regression analysis for predicting significant myocardial fibrosis in HCM patients
Univariable analysis
Multivariate analysis
OR
95% CI
p Value
OR
95% CI
p Value
Age, y
1.028
0.981–1.076
0.249
Body mass index, kg/m2
1.065
0.921–1.232
0.398
NT-proBNP, pg/mL
1.003
1.000–1.005
0.019
1.001
0.998–1.005
0.535
MWT, mm
1.318
1.025–1.694
0.031
1.241
0.771–1.998
0.373
Number with SWT ≥15 mm
1.738
1.116–2.707
0.014
1.402
0.802–2.448
0.235
Apical HCM
0.237
0.066–0.848
0.027
0.066
0.003–1.291
0.073
LAVI, mL/m2
1.017
0.976–1.059
0.431
LVEF, %
0.952
0.859–1.055
0.348
GLS, %
1.212
1.008–1.458
0.041
1.541
1.005–2.364
0.047
GPSI, %
1.029
0.987–1.073
0.175
PSD, ms
1.016
0.955–1.037
0.143
Values of p for statistically significant comparisons (p < 0.05) are in boldface.
CI, confidence interval; GLS, global longitudinal strain; GPSI, global post-systolic index; HCM, hypertrophic cardiomyopathy; MWT, maximum wall thickness; LAVI, left atrial volume index; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro-brain natriuretic peptide; PSD, peak strain dispersion; OR, odds ratio; SWT, segmental wall thickness.
Pearson's correlation analysis between GLS and positive LGE segments and 5 years SCD risk score in HCM patients
The mean 5 years SCD risk score in enrolled HCM patients was 2.0 ± 1.2%, of which 40 patients (66%) had low risk scores (<2%), 19 patients (31%) had moderate risk scores (4%–6%) and 2 patients (3%) had high risk scores (>6%). The average number of positive LGE segments was 5.3 ± 4.2. GLS was significantly associated with the number of positive LGE segments (r = 0.337, p = 0.008) and 5 years SCD risk score (r = 0.287, p = 0.025), as illustrated in Fig. 4.
Figure 4Plots of correlation between GLS and number of positive LGE segments and 5-y SCD risk score. r, Pearson's correlation coefficient; GLS, global longitudinal strain; LGE, late gadolinium enhancement; SCD, sudden cardiac death.
Inter- and intra-observer variability analysis of AFI parameters
All strain parameters provided by speckle tracking AFI exhibited excellent intra- and inter-observer reproducibility, as reflected in the Bland–Altman analysis. The ICCs for intra- and inter-observer variation of AFI parameters are provided in Table S1 and Figure S1 (online only).
Discussion
The present study explored for the first time the feasibility of using the strain parameters provided by AFI in the diagnosis and assessment of myocardial fibrosis in HCM patients. The primary findings are as follows: First, among the HCM patients, segments with LGE had lower absolute values of LS than those without LGE. The cutoff values of segmental LS for predicting positive LGE segments in the basal, intermediate and apical regions were –12.5%, –11.5% and –14.5%, respectively. Second, GLS could predict significant myocardial fibrosis at a cutoff value of –16.5%, sensitivity of 80.9% and specificity of 76.5% in the general population. In HCM patients, the cutoff value of GLS was –16.5% with a sensitivity of 80.9% and specificity of 57.1%. Third, as an independent predictor of significant myocardial fibrosis, GLS was statistically associated with the severity of myocardial fibrosis (r = 0.337, p = 0.008) and 5 y ears SCD risk score (r = 0.287, p = 0.025), providing valuable prognostic information for HCM patients.
In recent years, STE has been recommended for early diagnosis and risk stratification in various cardiovascular diseases [
]. In our recent studies, speckle tracking AFI was applied in the routine workup for risk stratification in patients with stable coronary artery disease and was not only a faster and more convenient method for non-invasive quantitative assessment of LV myocardial function but also effectively improved inter-observer and intra-observer consistency [
Post-systolic shortening is superior to global longitudinal strain in predicting adverse events in patients with stable coronary artery disease and preserved systolic function.
Rapidly and accurately detecting significant coronary artery stenosis in patients with suspected stable coronary artery disease and normal segmental wall motion by speckle tracking automated functional imaging.
Myocardial fibrosis tended to alter the organizational structure and mechanical properties of the myocardium, facilitating the alterations in electrical activity and cardiac function, which contributes to the development of heart failure and subsequent poor outcomes [
]. Over the past decade, STE has been performed by several investigators to explore the association of strain parameters with myocardial fibrosis in HCM patients [
Association between regional ventricular function and myocardial fibrosis in hypertrophic cardiomyopathy assessed by speckle tracking echocardiography and delayed hyperenhancement magnetic resonance imaging.
Clinical significance of global two-dimensional strain as a surrogate parameter of myocardial fibrosis and cardiac events in patients with hypertrophic cardiomyopathy.
Investigation of global and regional myocardial mechanics with 3-dimensional speckle tracking echocardiography and relations to hypertrophy and fibrosis in hypertrophic cardiomyopathy.
]. These studies found that the mean absolute value of segmental LS in segments with LGE provided by STE was significantly reduced compared with that of segments without LGE [
Distinguishing focal fibrotic lesions and non-fibrotic lesions in hypertrophic cardiomyopathy by assessment of regional myocardial strain using two-dimensional speckle tracking echocardiography: comparison with multislice CT.
]. As a result of myocardial fibrosis, an increase in the tensile strength and elasticity of the regional myocardium caused a decrease in the local contractility of the heart, which could be reflected by strain [
]. Consistent with previous findings, our study confirmed that segmental LS provided by AFI was significantly lower in segments with LGE than in segments without LGE. Differently, we explored in detail the predictive value of segmental strain parameters for myocardial fibrosis according to the myocardial level, taking into account that the absolute value of the LV myocardial longitudinal strain increases gradually from the basal to the apical level in healthy individuals. In Pagourelias’ study, 2-D segmental LS was the best strain parameter for detecting fibrotic segments in HCM patients, with a cutoff value of –15.83% [
]. However, our results indicated that the optimal cutoff values of segmental LS in predicting LGE segments in the basal, intermediate and apical regions were –12.5%, –11.5% and –14.5%, respectively. The different cutoff values of segmental LS according to the myocardial level could help us to identify fibrosis segments more accurately. Consistent with the previous study, the present study also found that myocardial fibrosis is more likely to occur in hypertrophic segments, where segmental LS was more impaired. These findings further indicate the close association between myocardial function and the histopathological features, including myocardial fibrosis and hypertrophy [
]. It is worth noting that segmental LS was superior to both PSI and TTP in detecting segmental myocardial fibrosis in the present study. The underlying reason could be that the regional myocardial deformation rather than the time of peak contraction is more sensitive to myocardial fibrosis. A more complete study design on a large research sample should be performed to explore this issue in the future.
Abnormal GLS assessed by STE has been reported to be associated with adverse cardiac outcomes in HCM patients, such as ventricular arrhythmia and heart failure [
Incremental prognostic utility of left ventricular global longitudinal strain in hypertrophic obstructive cardiomyopathy patients and preserved left ventricular ejection fraction.
]. In our study, multivariate regression analysis revealed that GLS provided by AFI was an independent predictor of significant myocardial fibrosis (OR = 1.541; 95% CI: 1.005–2.346; p < 0.05) in HCM patients. Moreover, GLS was significantly associated with the number of positive LGE segments (r = 0.337, p = 0.008) and 5 years SCD risk score (r = 0.287, p = 0.025). Notably, the 2020 American College of Cardiology/American Heart Association (ACC/AHA) guidelines have classified the extensive LGE detected on CMR imaging as a clinical risk factor for stratifying the risk of SCD in HCM [
]. According to our findings, GLS is expected to be a more convenient indicator of SCD risk stratification in HCM patients. On the whole, the segmental and global LS parameters provided by AFI not only provide a basis for non-invasive and cost-effective auxiliary screening for myocardial fibrosis, but also benefit the long-term monitoring and individualized management of HCM patients.
Limitations
The present study has limitations. First, the results are based on a single-center study of a small subset of HCM patients. Second, myocardial fibrosis was assessed on the basis of visual analysis of CMR LGE images in the present study, which should be improved to accurately quantify the extent of the myocardial fibrosis. Third, as LV mechanics and deformation are spatially in 3-D motion, 3-D speckle tracking imaging could provide ideal parameters reflecting the various histopathological features in HCM patients [
]. Unfortunately, the present speckle tracking AFI imaging could only be performed on three 2-D apical views.
Conclusion
Speckle tracking AFI could automatically provide multiple non-invasive parameters for identifying LV myocardial fibrosis, contributing to the rapid clinical screening and follow-up of HCM patients. Segmental LS exhibited excellent predictive performance in the localized diagnosis of myocardial fibrosis with cutoff values of –12.5%, –11.5% and –14.5% in the basal, intermediate and apical regions, respectively. As an independent predictor of significant myocardial fibrosis, GLS was associated with severe myocardial fibrosis and 5 years SCD risk score, which may indicate adverse clinical outcomes in HCM patients.
Conflict of interest
The authors declare no competing interests.
Acknowledgments
Y.L. is supported by the Natural Science Foundation of Hubei Province in China (Grant No. 2021 CFB336, 2021).
Data availability statement
The data sets used or analyzed during the present study are available from the corresponding author on reasonable request.
2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC).
Association between regional ventricular function and myocardial fibrosis in hypertrophic cardiomyopathy assessed by speckle tracking echocardiography and delayed hyperenhancement magnetic resonance imaging.
Clinical significance of global two-dimensional strain as a surrogate parameter of myocardial fibrosis and cardiac events in patients with hypertrophic cardiomyopathy.
Investigation of global and regional myocardial mechanics with 3-dimensional speckle tracking echocardiography and relations to hypertrophy and fibrosis in hypertrophic cardiomyopathy.
Feasibility and correlation of standard 2D speckle tracking echocardiography and automated function imaging derived parameters of left ventricular function during dobutamine stress test.
Post-systolic shortening is superior to global longitudinal strain in predicting adverse events in patients with stable coronary artery disease and preserved systolic function.
Quantitative differentiation of LV myocardium with and without layer-specific fibrosis Using MRI in hypertrophic cardiomyopathy and layer-specific strain TTE analysis.
Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging.
Non-invasive global and regional myocardial work predicts high-risk stable coronary artery disease patients with normal segmental wall motion and left ventricular function.
Two-dimensional speckle-tracking during dobutamine stress echocardiography in the detection of myocardial ischemia in patients with suspected coronary artery disease.
Rapidly and accurately detecting significant coronary artery stenosis in patients with suspected stable coronary artery disease and normal segmental wall motion by speckle tracking automated functional imaging.
Distinguishing focal fibrotic lesions and non-fibrotic lesions in hypertrophic cardiomyopathy by assessment of regional myocardial strain using two-dimensional speckle tracking echocardiography: comparison with multislice CT.
Incremental prognostic utility of left ventricular global longitudinal strain in hypertrophic obstructive cardiomyopathy patients and preserved left ventricular ejection fraction.
00033-9&id=gr1.jpg){kind=link}
00033-9&id=gr2.jpg){kind=link}
00033-9&id=gr3.jpg){kind=link}
00033-9&id=gr4.jpg){kind=link}