Abstract
Background
Interpreting patient phenotypes is a challenge when screening for hypertrophic cardiomyopathy (HCM). Machine learning (ML) can potentially help with advanced data integration - ...combining information contained in whole-cardiac cycle echo deformation and velocity profiles with standard clinical variables. The aim is to apply an ML approach to integrate whole cardiac cycle echo data with clinical variables to explore HCM phenotypes.
Methods
The cohort consisted of 138 participants from two centres: HCM patients (n=91) and relatives (n=47). Echocardiography was performed, whereas magnetic resonance and genetic testing in 48% and 82%, respectively. Whole cardiac cycle echo data (mitral and aortic velocity profiles, and six regional left ventricular (LV) deformation curves) were combined with clinical variables (age, sex, heart rate, e' medial and e' lateral) and used as the ML input. An unsupervised ML algorithm created a representative space where participants were positioned based on integrated data, blinded to disease status. Clustering was used to determine phenogroups and estimate the average characteristics. Data on family history (FHx), genotype, arrhythmias or syncope, implantable cardioverter-defibrillators (ICD), and late gadolinium enhancement (LGE) were used to interpret the phenogroups. As the LA diameter was not available in the dataset, the HCM risk for sudden cardiac death (SCD) was not calculated, however, the Table shows relevant variables to infer clinical risk.
Results
Clustering divided the participants into 6 phenogroups (P1–6) (Figure). Average echo profiles are shown in the Figure, while the clinical data in the Table. P1/2 was defined by symptomatic patients with a high prevalence of positive genotypes, a positive FHx of SCD, and a burden of comorbidities. Echo findings showed pronounced structural/functional remodeling, and P1 was associated with severe septal hypertrophy and outflow tract obstruction. The high prevalence of ICD devices defined P1/2 as high risk groups. In comparison, patients in P3/4 were younger, with milder LV hypertrophy, but still considerable functional impairment. P3 had a higher burden of FHX and a higher prevalence of pathogenic mutations, whereas P4 a higher incidence of hypertension, high heart rate, mitral inflow fusion and findings of LGE. Finally, P5/6 consisted of younger individuals, predominantly HCM relatives, with a mild phenotype and, thus, low inferred risk. As expected, the majority of patients with the genetic variants of undetermined significance were located in P5.
Conclusion
ML can help derive clinically interpretable phenotypes in HCM based on the automated integration of whole cardiac cycle deformation and velocity data with conventional clinical parameters. The derived phenogroups correspond with established risk profiles in HCM. An expanded dataset is needed to enable further exploration of the phenotype-genotype relations and to define prognostic value.
Funding Acknowledgement
Type of funding sources: Public grant(s) – EU funding. Main funding source(s): This work was supported by the Horizon 2020 European Commission Project H2020-MSCA-ITN
Abstract
Background
Left Ventricular systolic disfunction has already been described in Marfan Syndrome (MS) in patients without valvular dysfunction using 2D and 3D speckle tracking echocardiography ...(STE). This dysfunction has been related to a more severe causal genetic mutation, which suggest the presence of a primary cardiomiopathy in these patients. Right ventricular function has been less studied so far. We sought to evaluate biventricular function in our cohort of MS patients with 2D-STE.
Methods
95 unoperated adult patients with MS and 32 healthy controls were prospectively enrolled. Patients with more than mild mitral or aortic regurgitation were excluded. Using STE we obtanied left ventricular global longitudinal strain (LVGLS) from the average of 16 segments from 4,2 and 3-chamber views and RVGLS values were obtained from the average of 6 segments from the apical 4-chamber view. We also measured classic parameters of systolic biventricular function (LVEF and TAPSE).
Results
Compared to controls, patients with MFS had significantly lower LVGLS and RVGLS (table 1). Values obtanied for LVGLS in MS patients were at the lower limit of normality stablished in the latest cuantification guidelines, while RVGLS and RV free wall LS were slightly above the limit of normality. LVEF and TAPSE were also slightly diminished in MS patients, though the differences found were clinically not relevant.
Results of statistical analysis
MS (n=93)
Controls (n=32)
p
Age (years)
32.84±12.4
32.41±7.98
0.85
Aortic Root Diameter Valsalva Sinuses (mm)
38.82±5.35
30.91±5.3
<0.001
LVGLS (%)
−18.93±2.62
−21.52±2.26
<0.001
RVGLS (%)
−21.25±3.54
−24.68±3.08
<0.001
RV free wall LS (%)
−22.09±3.92
−25.56±3.63
<0.001
LVEF (%)
59.5±5.34
63.27±4.19
0.001
TAPSE (mm)
23.97±4.57
25.82±3.32
0.03
MS = Marfan Syndrome; LVGLS = Left ventricular global longitudinal strain; RVGLS = right ventricular global longitudinal strain.
Conclusions
Our study suggests that patients with MFS show lower biventricular strain compared with healthy controls. 2D-STE imaging may be useful to detect subclinical changes in cardiac function in patients with MFS and should be added to routine ecocardiographic evaluation in order to improve the follow-up and treatment of these patients.
Abstract
Background
Elastic properties of the thoracic aorta in patients with Marfan Syndrome (MS) have already been evaluated with classic echocardiographic parameters. In the latest years the use ...of Speckle-Tracking (STE) ecocardiography has been widely extended. Our aim is to describe and provide new parameters of aortic deformation measured by STE in patients with MS.
Methods
95 unoperated adult patients with MS and 32 healthy controls were prospectively enrolled. We measured classic parameters of the aortic root using 2D echocardiography. We calculated the posterior aortic wall systolic excursion at the sinuses of Valsalva and ascending aorta using M Mode in TDI colour; with ST 2D ecocardiography we measured the aortic strain at the sinuses of Valsalva (SV) and the anterior and posterior aortic wall displacement at the SV. Aortic distensibility was calculated using the formula: 1000 * (Ds − Dd)/Dd * 1/(Ps − Pd) in mmHg–1 (Ds: systolic and Dd: diastolic diameters, Ps systolic and Pd diastolic blood pressure). Aortic stiffness index was calculated as Ln((Ps/Pd)/(Ds-Dd)/Dd)).
Results
As shown in the table bellow, patients with MS had lower aortic strain, aortic anterior and posterior wall displacement and impaired aortic distensibility and stiffness index compared to healthy controls. We found a strong negative linear correlation between aortic root diameter at the SV and aortic root strain (r=−0.56, figure 1).
Results of statistical analysis
MS (n=95)
Controls (n=32)
p
Age (years)
32.84±12.35
32.41±7.98
0.85
Aortic root diameter at the sinuses of Valsalva (mm)
38.82±5.35
30.92±3.65
<0.001
Aortic root strain (%)
4.66±2.45
9.19±2.49
<0.001
Anterior aortic wall displacement STE (mm)
10.39±3.64
13.10±2.26
<0.001
Posterior aortic wall displacement STE (mm)
9.02±2.87
11.04±1.82
<0.001
Aortic distensibility
0.98±0.46
1.37±0.72
0.01
Aortic stiffness index
3.74±0.43
3.47±0.51
0.01
MS = Marfan Syndrome; STE = Speckle Tracking Ecocardiography.
Figure 1. Dispersion plot
Conclusions
Our results suggest that aortic deformation and displacement obtained by STE echocardiography is impaired in MS, showing a reduced distensibility and an increased stiffness of the aortic wall, with a strong negative correlation between aortic root dilation and aortic strain. All these parameters may be useful as additional tools for the diagnosis and follow-up of Marfan patients, and could be useful to to improve the echocardiographic evaluation of the aortic root.
Abstract
Background/Introduction
Previous studies using conventional echocardiographic measurements have reported subclinical left diastolic dysfunction in patients with Marfan syndrome (MFS). Left ...atrial strain (LAS) has been shown to be an accurate predictor of left ventricular diastolic dysfunction. However, there is no evidence regarding the use of LAS in MFS.
Purpose
To assess feasibility of LAS and compare LAS derived measurements along with traditional diastolic parameters in MFS patients vs healthy controls.
Methods
46 MFS patients (normal LV ejection fraction, no previous cardiovascular surgery, no significant valvular regurgitation) vs. 20 healthy controls (age and sex-matched). We performed LAS analysis using 2D speckle-tracking (QLAB 10, Philips). LA strain was determined as the average value of the longitudinal deformation (7 segments) in the apical 4-chamber view (RR gating).
Results
LAS analysis was feasible in 40 MFS patients (87%). All participants had normal diastolic function according to current guidelines (ASE/EACVI 2016). MFS patients showed lower TDI e' velocities and higher average E/e' ratio, but still within normal range. Similarly, LVEF was normal but slightly reduced in MFS patients. LA strain and strain rate parameters during reservoir and conduit phase were significantly impaired in MFS patients compared to controls.
MFS vs controls
MFS patients (n=40)
Controls (n=20)
p
MFS patients (n=40)
Controls (n=20)
p
Age
33.8±12.4
34.4±8.3
0.846
Septal e' (cm/s)
9.7±2.5
11.7±2.3
0.006
Male (%)
24 (60%)
12 (60%)
1.000
Average E/e' ratio
6.8±1.5
5.5±1.1
0.002
SBP (mmHg)
120.3±12.4
120.1±9.4
0.969
TR velocity (cm/s)
208.6±21.4
201.6±22.9
0.390
DBP (mmHg)
72.0±10.1
67.1±6.2
0.069
LAVi (ml/m2)
23.5±7.1
25.5±4.8
0.260
Aortic root (mm)
40.3±4.6
31.7±3.7
<0.001
LASr (%)
32.6±8.8
43.0±8.3
<0.001
LVEF (%)
60.9±5.6
64.2±4.2
0.022
LAScd (%)
−20.1±8.0
−29.4±5.5
<0.001
E-wave (cm/s)
74.6±16.5
76.7±16.5
0.651
LASct (%)
−12.8±6.1
−13.6±5.2
0.622
A-wave (cm/s)
55.2±10.9
52.0±12.8
0.327
LASRr
2.02±0.49
2.31±0.43
0.030
E/A ratio
1.4±0.4
1.5±0.4
0.287
LASRcd
−2.22±0.61
−3.07±0.68
<0.001
Lateral e' (cm/s)
13.0±3.6
16.3±3.3
0.002
LASRct
−2.24±0.90
−2.35±0.75
0.600
SBP: Systolic blood pressure. DBP: Diastolic blood pressure. LVEF: Left ventricular ejection fraction. LAVi: Left atrial volume index. LAS: Left atrial strain. LASR: Left atrial strain rate. (r): Reservoir. (cd): Conduit. (ct): Contraction.
Example of LA strain and strain rate
Conclusion
MFS patients showed a subtle impairment in diastolic function compared to controls. Although further evidence is needed, LAS derived parameters could be early markers of diastolic dysfunction in this group of patients.
Acknowledgement/Funding
Programa de Actividades de I+D de la Comunidad de Madrid
Abstract
Background/Introduction
Previous studies using conventional echocardiographic measurements have reported subclinical left diastolic dysfunction in patients with Marfan syndrome (MFS). ...Certain speckle-tracking echocardiography (STE) derived parameters have shown good correlation with invasive measurements of LV relaxation.
Purpose
To evaluate diastolic function in Marfan patients using STE-derived parameters along with traditional diastolic measurements and compare them with healthy controls.
Methods
We consecutively included 127 MFS patients with normal LV ejection fraction and no previous cardiovascular surgery or significant valvular regurgitation. We also studied 38 healthy controls (age and sex-matched). We performed LV and LA strain analysis using 2D STE (QLAB 10, Philips). Echocardiographic parameters recommended to assess diastolic function (according to ASE/EACVI 2016 guidelines) were also considered.
Results
Reported use of BB/ARB was 28/39% in MFS patients. According to current guidelines, all participants had normal diastolic function. MFS patients showed lower TDI e' velocities (septal and lateral) and higher average E/e' ratio compared to controls, but still within normal range. Regarding STE-derived measurements, LVGLS, EDLSR and PALS were significantly reduced in MFS. E/EDLSR ratio was higher among MFS patients, reflecting impaired LV relaxation. Interestingly, E/LVGLS was not different between both groups.
Conclusion
STE-derived parameters showed impaired LV relaxation in MFS patients compared to controls. Our results suggest that STE derived parameters can be markers of early diastolic dysfunction and provide a better insight into Marfan-related cardiomyopathy.
STE derived parameters
Funding Acknowledgement
Type of funding source: Public Institution(s). Main funding source(s): Programa de Actividades de I+D de la Comunidad de Madrid
PDF