https://orcid.org/0000-0003-4193-0418
Abstract
How the underlying aetiology and pathophysiology of left ventricular (LV) hypertrophy affects left atrial (LA) remodelling and function remains unexplored. The present study aims to investigate the influence of various hypertrophic phenotypes on LA remodelling and function.
Patients with LV hypertrophy who underwent cardiac magnetic resonance (CMR) were compared to a control group. CMR data were analysed retrospectively to assess LA strain, volume, sphericity, and left atrioventricular coupling index (LACI). Independent clinical associates of LA strain were assessed using multivariable linear regression analysis. A total of 375 individuals were included: 148 with hypertrophic cardiomyopathy (HCM), 35 with cardiac amyloidosis (CA), 41 with hypertensive (HTN) heart disease, 97 with severe asymptomatic aortic stenosis (AS), and 54 with normal CMR. Indexed LA end-systolic (iLVmax), diastolic volumes, and LA sphericity were the largest in patients with CA (59.1 ± 16.9 mL/m2, 46.8 ± 16.4 mL/m2, and 83.2 ± 2.1%, respectively). Patients with CA presented a higher LACI when compared with other groups (58 ± 2% vs. 42 ± 2% in HCM, 39 ± 2% in HTN heart disease, 37 ± 2% in AS, and 22 ± 1% in normal), while no differences were observed across others. Patients with CA showed the lowest LA reservoir [9.6% (0.6–18.6%)] and booster strain (9.1 ± 5.4%), whereas no differences were observed across other groups. LACI and iLAVmax were independently correlated with LA reservoir (β = 0.15 and β = −39.33, respectively), LA conduit (β = 0.08 and β = −17.08, respectively), and LA booster strains (β = 0.1 and β = −28.69, respectively). LA sphericity was independently correlated with LA reservoir strain (β = −0.51). Finally, LV global longitudinal strain was independently correlated with LA reservoir (β = −0.43), conduit (β = −0.20), and booster strain (β = −0.24).
LA characteristics differ among LV hypertrophic phenotypes. LACI and iLAVmax are independently correlated with LA function, while LA sphericity correlates independently with LA reservoir strain.
Left atrial (LA) characteristics assessment in various left ventricular (LV) hypertrophic phenotypes. In a cohort of patients with LV hypertrophy of various aetiologies who underwent cardiac magnetic resonance, left atrioventricular coupling index (LACI) and several other parameters of LA function and remodelling were measured. Patients with cardiac amyloidosis showed the worst LACI and the worse LA reservoir strain.
Introduction
Left atrial (LA) dysfunction is a common feature in ventricular dysfunction associated with hypertrophic phenotype. Left ventricular (LV) hypertrophy is characterized by altered LV relaxation and increased LV filling pressures that are transmitted to the LA, leading to LA stretch and remodelling. In addition to this atrioventricular interplay, which is common to all LV hypertrophic phenotypes, the underlying cause of LV hypertrophy may also influence the atrioventricular function. However, how the underlying aetiology and pathophysiology of different cardiomyopathies with LV hypertrophy affects LA remodelling and function remains unexplored.
Currently, a comprehensive assessment of the LA remodelling and function can be performed with non-invasive imaging tools. In addition to chamber size and volumes, many studies support the use of the LA strain to measure the various components of LA function (reservoir, conduit, and booster pump) in the hypertrophic phenotype.1–4 Furthermore, recent studies have suggested to evaluate left atrioventricular coupling in this setting, to simultaneously consider remodelling of both chambers and taking into account their interdependence.5 Finally, LA sphericity is emerging as a promising marker of atrial geometry and remodelling.6,7
Therefore, the objective of the present study is to investigate the influence of various hypertrophic phenotypes on LA remodelling in terms of volumes, geometry, function, and atrioventricular coupling and investigate the independent clinical associates with LA function in these clinical scenarios.
Methods
Study population
Patients with LV hypertrophy phenotype who underwent a cardiac magnetic resonance (CMR) between 2014 and 2022 at the University Hospital Germans Trias i Pujol (Badalona, Spain) were retrospectively evaluated. The aetiology of LV hypertrophy included hypertrophic cardiomyopathy (HCM), cardiac amyloidosis (CA), hypertensive (HTN) heart disease, and severe aortic stenosis (AS).
HCM was defined as abnormal LV thickening of at least 15 mm in one or more LV myocardial segments detected on echocardiography or CMR that was not explained solely by loading conditions.8,9 The patients received a complete clinical and genetic assessment.
CA diagnosis was confirmed based on positive clonal dyscrasia detected with serum-free light-chain assay or with serum and urine protein electrophoresis with immunofixation (for light-chain CA) or with a 99mTc-3,3-diphosphono-1,2-propanodicarboxylic acid scintigraphy showing a Grade 2 or 3 myocardial uptake of radiotracer (for transthyretin CA).10 Cardiac or other clinically affected organ biopsy and histology with amyloid typing were performed when necessary.
HTN heart disease was diagnosed in patients with long-standing systemic arterial hypertension without satisfactory response under medical therapy. In addition, other causes of LV hypertrophy had to be ruled out.11,12
The group of LV hypertrophy due to severe AS consisted of patients with asymptomatic severe AS who underwent CMR in a previous study (PI-14-064 and PI-19-257).13 Severe AS was defined by the presence of an aortic valve area of ≤1.0 cm2, a peak aortic jet velocity of >4 m/s, and a mean transvalvular pressure gradient of ≥40 mmHg. Exercise stress echocardiography was performed in all patients to evaluate the functional capacity and to unmask symptoms.
Additionally, a group of patients with a normal CMR was included as a control group. The normal CMR cohort consisted of patients referred to the Cardiology Unit of the University Hospital Germans Trias i Pujol (Badalona, Spain) for mild cardiovascular symptoms as atypical chest pain or dyspnoea that presented normal cardiac chamber dimensions, function, normal tissue characterization [T1 mapping, T2 mapping, and extracellular volume (ECV)], and absence of late gadolinium enhancement (LGE).
The study was approved by the local ethics committee (PI-19-257), and written informed consent was obtained from all patients.
CMR data acquisition and analysis
CMR was performed on a 1.5-T scanner (Intera dSTREAM 1.5T; Philips, Best, The Netherlands) following a standardized protocol, with the patient in a supine position and a 16-element phased-array coil placed over the chest. Images were acquired during breath-holds with electrocardiographic gating. Cine long-axis (two-, three-, and four-chamber views) and short-axis reconstructions (contiguous slices of 8-mm thickness covering from base to apex) were acquired using segmented k-space steady-state free precession sequences. Delayed enhancement images were acquired with a segmented gradient-echo inversion-recovery sequence at matching cine-image slice locations 10–20 min after intravenous gadolinium-diethylenetriamine penta-acetic acid administration (Gadovist, 0.15 mmol/kg). Inversion time was optimized to achieve myocardial nulling. Pre- and post-contrast myocardial T1 mapping (ms) was acquired using the modified look-locker inversion-recovery sequences. Myocardial ECV was computed considering T1 values and haematocrit. All images were analysed off-line with commercially available post-processing software (QMass-MR, v.8.1; Medis Medical Imaging Systems, Leiden, The Netherlands).
CMR data analysis was performed by a CMR expert. LV and right ventricular volumes were obtained from the short-axis cine images tracing the endocardial borders in the end-diastolic and end-systolic frames excluding the papillary muscles from the tracing. LV mass was calculated by subtracting the endocardial volume from the epicardial volume at end-diastole and then multiplying by the tissue density (1.05 g/mL). LV mass and volumes were indexed to body surface area, according to current recommendations.8 Myocardial fibrosis was assessed visually by signal intensity on LGE sequences, and the distribution, location, and number of segments affected by LGE were reported. Feature-tracking CMR-derived analysis of the LV global longitudinal strain (GLS) was performed utilizing commercially available software (QStrain, version 8.1, Medis, Leiden, The Netherlands) by manually tracing the end-diastolic and end-systolic LV endocardial border in the long-axis cine four-, two-, and three-chamber views. To improve tracking accuracy, the tracking quality was visually assessed, and manual adjustments were made when needed.
Long-axis two- and four-chamber cine images were used to evaluate the LA dimensions. LA volumes were performed by manual tracing of the LA endocardial border excluding the pulmonary veins and the LA appendage. LA volumes were measured in both systole and diastole using commercially available software (QStrain, version 8.1, Medis, Leiden, The Netherlands). The maximum LA volume was assessed at a ventricular end-systolic frame (LAVMax), whereas the minimum LA volume was measured at a late ventricular diastolic frame after atrial contraction (LAVMin), and both volumes were indexed to body surface area, according to current recommendations.10 LA ejection fraction (LAEF) was defined as: LAEF = (LAVMax − LAVMin) × 100%/LAVMax.
Feature-tracking CMR-derived analysis of the LA was performed to measure LA strain utilizing commercially available software (QStrain, version 8.1, Medis, Leiden, The Netherlands) by manually tracing the end-diastolic and end-systolic LA endocardial border in the cine longitudinal two-chamber view, excluding LA appendage and pulmonary veins. To improve tracking accuracy, the tracking quality was visually assessed, and manual adjustments were made. LA reservoir strain was estimated from the first peak of the LA strain curve in patients in sinus rhythm, while LA booster strain was evaluated from the second peak of the LA strain curve. Conduit strain was calculated as the difference between the two values obtained. When the patient was in atrial fibrillation, only the LA reservoir peak was assessed.
Left atrioventricular coupling index (LACI) was calculated as the ratio between the LAVMin and the LV end-diastolic volume. The LA and LV volumes were measured in the same end-diastolic phase, defined by the mitral valve closure.5,14–16 Furthermore, LA sphericity, a marker of the roundness or approximation to a spherical shape of the LA that has been previously associated with atrial fibrillation and LA dysfunction,7 was derived using 2D morphology LA measures (four-chamber longitudinal and transversal diameter, four-chamber perimeter, two-chamber longitudinal diameter, three-chamber longitudinal diameter and area), as previously described.6
Statistical analysis
The distribution of continuous variables was estimated using the Shapiro–Wilk test; normally distributed continuous variables are presented as mean ± standard deviation, whereas non-normally distributed data are presented as median and interquartile range (IQR). Categorical variables are presented as numbers and percentages. For continuous variables following a normal distribution, the ANOVA test was used to assess in-between differences across groups, whereas for non-normally distributed continuous variables, the Kruskal–Wallis test with Bonferroni’s correction was used. Categorical variables were compared across the groups with χ2 test.
To investigate the independent clinical associates of LA reservoir strain, LA conduit strain, and LA booster strain, a complete case analysis using multivariate linear regression analyses was performed, introducing sex, age, indexed LAVmax (iLAVmax), LA sphericity, LACI, indexed LV mass, LV GLS, and type of cardiomyopathy as covariates. Furthermore, the interaction between LV hypertrophy phenotype and iLAVmax and between LV hypertrophy phenotype and indexed LV mass was tested. The goodness of fit of the models was evaluated using the Akaike information criterion, the Bayesian information criterion, and the coefficient of determination (R2). The performance of the model was further tested by leave-one-out cross-validation. A two-tailed P value of 0.05 was considered statistically significant. All statistical analyses were performed using Statistical Package for Social Sciences for Windows, version 25 (SPSS, Chicago, IL, USA), and STATA V.13.0 (Stata Corp, College Station, TX, USA).
Results
A total of 375 individuals were analysed (mean age 60.9 ± 17.0 years; 35% female) including 321 patients diagnosed with LV hypertrophy and 54 patients with a normal CMR. Of the 321 patients with LV hypertrophy, 148 were diagnosed with HCM, 35 with CA, 41 with HTN heart disease, and 97 with asymptomatic severe AS.
Baseline demographic and clinical characteristics of the various patient groups are summarized in Table 1. CMR data and strain analysis results are presented in Tables 2 and 3 and Figure 2. In terms of demographic characteristics, patients with CA, AS, and HTN heart disease were significantly older than patients with HCM and normal individuals. In terms of CMR characteristics, patients with CA and HCM exhibited the largest indexed LV mass, the highest myocardial native T1 mapping and ECV values, and the highest percentage of LGE as compared to other cardiomyopathies. Regarding LA volumetric and sphericity measurements, patients with CA had larger iLAVmax, indexed LAVmin (iLAVmin), and larger LA sphericity compared with patients with HCM, HTN heart disease, AS, and normal individuals. Conversely, no significant differences were observed among patients with other LV hypertrophic phenotypes.
Baseline characteristics of the population
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| Clinical characteristics | |||||||
| Age (years) | 60.9 ± 17.0 | 55.2 ± 15.5a | 72.5 ± 10.6 | 65.6 ± 10.3b | 72.9 ± 9. 0 | 43.3 ± 17.1c | <0.001 |
| Female sex (n%) | 131 (35) | 47 (32)d | 14 (40) | 9 22)e | 35 (36) | 26 (48) | 0.081 |
| BSA (mg/m2) | 1.8 ± 0.2 | 1.9 ± 0.2 | 1.8 ± 0.2 | 1.9 ± 0.2b | 1.8 ± 0.2 | 1.8 ± 0.2 | 0.011 |
| Hypertension (n%) | 229 (61) | 55 (37)a | 32 (91) | 41 (100) | 81 (84) | 21 (39)c | <0.001 |
| Dyslipidaemia (n%) | 199 (53) | 66 (45)a | 24 (69) | 28 (69) | 67 (69) | 14 (26)c | <0.001 |
| Diabetes (n%) | 98 (26) | 20 (14)a | 18 (51) | 20 (49) | 39 (40) | 1 (2)c | <0.001 |
| Atrial fibrillation (n%) | 64 (17) | 29 (20) | 15 (43)f | 7 (17) | 8 (8)g | 5 (9) | <0.001 |
| Previous AMI (n%) | 25 (7) | 4 (3) | 8 (23)h | 11 (27)i | 2 (2) | 0 (0) | <0.001 |
| Previous HF (n%) | 40 (11) | 15 (10) | 11 (31)f,j | 12 (29)e | 0 (0) | 2 (4)k | <0.001 |
| Previous HFpEF (n%) | 32 (10) | 13 (9) | 6 (35) | 11 (31) | 0 (0) | 2 (4) | 0.034 |
| Vascular disease (n%) | 11(3) | 3 (2)c | 4 (11)i | 4 (10)j | 0 (0) | 0 (0) | <0.001 |
| Stroke/TIA (n%) | 15 (4) | 7 (5) | 3 (9) | 4 (10) | 0 (0)l | 1 (2) | 0.035 |
| COPD (n%) | 49 (13) | 14 (10) | 7 (20) | 11 (27) | 16 (17) | 1 (2)h | 0.002 |
| Chronic kidney disease (n%) | 48 (13) | 7 (5)m | 9 (26) | 7 (17) | 25 (26) | 0 (0)h | <0.001 |
| Smoking (n%) | 93 (25) | 26 (18)m | 10 (29) | 13 (32) | 34 (35) | 10 (19) | 0.017 |
| NYHA class | |||||||
| I (n%) | 251 (67) | 66 (47) | 10 (29)n | 26 (63) | 97 (100)k | 52 (96)c | <0.001 |
| II (n%) | 109 (29) | 69 (47) | 23 (66)n | 15 (37) | 0 (0)k | 2 (4)c | |
| III (n%) | 15 (4) | 13 (9)d | 2 (6)n | 0 (0) | 0 (0)f | 0 (0) | |
| Creatinine (mg/dL) | 0.9 ± 0.4 | 0.9 ± 0.3 | 1.2 ± 0.8i | 1.0 ± 0.3 j | 0.8 ± 0.2 | 0.9 ± 0.2 | <0.001 |
| NT-proBNP (pg/mL) | 1193 ± 2583 | 1183 ± 2572 | 4146 ± 4778 f | 796 ± 728 | 444 ± 699 | 422 ± 1056 | <0.001 |
| hsTnT (ng/mL) | 50.7 ± 186.3 | 15.4 ± 11.5 | 148.7 ± 222.6d | 132.0 ± 340.4 | 13.7 ± 9.1b | 105.6 ± 368.4 | 0.002 |
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| Clinical characteristics | |||||||
| Age (years) | 60.9 ± 17.0 | 55.2 ± 15.5a | 72.5 ± 10.6 | 65.6 ± 10.3b | 72.9 ± 9. 0 | 43.3 ± 17.1c | <0.001 |
| Female sex (n%) | 131 (35) | 47 (32)d | 14 (40) | 9 22)e | 35 (36) | 26 (48) | 0.081 |
| BSA (mg/m2) | 1.8 ± 0.2 | 1.9 ± 0.2 | 1.8 ± 0.2 | 1.9 ± 0.2b | 1.8 ± 0.2 | 1.8 ± 0.2 | 0.011 |
| Hypertension (n%) | 229 (61) | 55 (37)a | 32 (91) | 41 (100) | 81 (84) | 21 (39)c | <0.001 |
| Dyslipidaemia (n%) | 199 (53) | 66 (45)a | 24 (69) | 28 (69) | 67 (69) | 14 (26)c | <0.001 |
| Diabetes (n%) | 98 (26) | 20 (14)a | 18 (51) | 20 (49) | 39 (40) | 1 (2)c | <0.001 |
| Atrial fibrillation (n%) | 64 (17) | 29 (20) | 15 (43)f | 7 (17) | 8 (8)g | 5 (9) | <0.001 |
| Previous AMI (n%) | 25 (7) | 4 (3) | 8 (23)h | 11 (27)i | 2 (2) | 0 (0) | <0.001 |
| Previous HF (n%) | 40 (11) | 15 (10) | 11 (31)f,j | 12 (29)e | 0 (0) | 2 (4)k | <0.001 |
| Previous HFpEF (n%) | 32 (10) | 13 (9) | 6 (35) | 11 (31) | 0 (0) | 2 (4) | 0.034 |
| Vascular disease (n%) | 11(3) | 3 (2)c | 4 (11)i | 4 (10)j | 0 (0) | 0 (0) | <0.001 |
| Stroke/TIA (n%) | 15 (4) | 7 (5) | 3 (9) | 4 (10) | 0 (0)l | 1 (2) | 0.035 |
| COPD (n%) | 49 (13) | 14 (10) | 7 (20) | 11 (27) | 16 (17) | 1 (2)h | 0.002 |
| Chronic kidney disease (n%) | 48 (13) | 7 (5)m | 9 (26) | 7 (17) | 25 (26) | 0 (0)h | <0.001 |
| Smoking (n%) | 93 (25) | 26 (18)m | 10 (29) | 13 (32) | 34 (35) | 10 (19) | 0.017 |
| NYHA class | |||||||
| I (n%) | 251 (67) | 66 (47) | 10 (29)n | 26 (63) | 97 (100)k | 52 (96)c | <0.001 |
| II (n%) | 109 (29) | 69 (47) | 23 (66)n | 15 (37) | 0 (0)k | 2 (4)c | |
| III (n%) | 15 (4) | 13 (9)d | 2 (6)n | 0 (0) | 0 (0)f | 0 (0) | |
| Creatinine (mg/dL) | 0.9 ± 0.4 | 0.9 ± 0.3 | 1.2 ± 0.8i | 1.0 ± 0.3 j | 0.8 ± 0.2 | 0.9 ± 0.2 | <0.001 |
| NT-proBNP (pg/mL) | 1193 ± 2583 | 1183 ± 2572 | 4146 ± 4778 f | 796 ± 728 | 444 ± 699 | 422 ± 1056 | <0.001 |
| hsTnT (ng/mL) | 50.7 ± 186.3 | 15.4 ± 11.5 | 148.7 ± 222.6d | 132.0 ± 340.4 | 13.7 ± 9.1b | 105.6 ± 368.4 | 0.002 |
Values are expressed as n (%), mean ± SD, or median (IQR).
aCM vs. all other groups P < 0.05.
bHTN heart disease vs. AS P < 0.05.
cNormal vs. all other groups P < 0.05.
dHCM vs. normal P < 0.05.
eHTN heart disease vs. normal P < 0.05.
fCA vs. all other groups P < 0.05.
gAS vs. HCM P < 0.05.
hCA vs. HCM, AS, and normal P < 0.05.
iHTN heart disease vs. HCM, AS, and normal P < 0.05.
jCA vs. HCM and AS P < 0.05.
kNormal vs. CA, HTN heart disease, and AS P < 0.05.
lAS vs. HCM, CA, and HTN heart disease P < 0.05.
mHCM vs. CA, AS, and HTN heart disease P < 0.05.
nCA vs. AS P < 0.05.
AMI, acute myocardial infarction; BSA, body surface area; COPD, chronic obstructive pulmonary disease; HF, heart failure; HFpEF, heart failure with preserved left ventricular ejection fraction; hsTnT, high-sensitivity troponin T; NYHA, New York Heart Association; NT-proBNP, N-terminal prohormone of brain natriuretic peptide TIA, transitory ischaemic attack.
Baseline characteristics of the population
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| Clinical characteristics | |||||||
| Age (years) | 60.9 ± 17.0 | 55.2 ± 15.5a | 72.5 ± 10.6 | 65.6 ± 10.3b | 72.9 ± 9. 0 | 43.3 ± 17.1c | <0.001 |
| Female sex (n%) | 131 (35) | 47 (32)d | 14 (40) | 9 22)e | 35 (36) | 26 (48) | 0.081 |
| BSA (mg/m2) | 1.8 ± 0.2 | 1.9 ± 0.2 | 1.8 ± 0.2 | 1.9 ± 0.2b | 1.8 ± 0.2 | 1.8 ± 0.2 | 0.011 |
| Hypertension (n%) | 229 (61) | 55 (37)a | 32 (91) | 41 (100) | 81 (84) | 21 (39)c | <0.001 |
| Dyslipidaemia (n%) | 199 (53) | 66 (45)a | 24 (69) | 28 (69) | 67 (69) | 14 (26)c | <0.001 |
| Diabetes (n%) | 98 (26) | 20 (14)a | 18 (51) | 20 (49) | 39 (40) | 1 (2)c | <0.001 |
| Atrial fibrillation (n%) | 64 (17) | 29 (20) | 15 (43)f | 7 (17) | 8 (8)g | 5 (9) | <0.001 |
| Previous AMI (n%) | 25 (7) | 4 (3) | 8 (23)h | 11 (27)i | 2 (2) | 0 (0) | <0.001 |
| Previous HF (n%) | 40 (11) | 15 (10) | 11 (31)f,j | 12 (29)e | 0 (0) | 2 (4)k | <0.001 |
| Previous HFpEF (n%) | 32 (10) | 13 (9) | 6 (35) | 11 (31) | 0 (0) | 2 (4) | 0.034 |
| Vascular disease (n%) | 11(3) | 3 (2)c | 4 (11)i | 4 (10)j | 0 (0) | 0 (0) | <0.001 |
| Stroke/TIA (n%) | 15 (4) | 7 (5) | 3 (9) | 4 (10) | 0 (0)l | 1 (2) | 0.035 |
| COPD (n%) | 49 (13) | 14 (10) | 7 (20) | 11 (27) | 16 (17) | 1 (2)h | 0.002 |
| Chronic kidney disease (n%) | 48 (13) | 7 (5)m | 9 (26) | 7 (17) | 25 (26) | 0 (0)h | <0.001 |
| Smoking (n%) | 93 (25) | 26 (18)m | 10 (29) | 13 (32) | 34 (35) | 10 (19) | 0.017 |
| NYHA class | |||||||
| I (n%) | 251 (67) | 66 (47) | 10 (29)n | 26 (63) | 97 (100)k | 52 (96)c | <0.001 |
| II (n%) | 109 (29) | 69 (47) | 23 (66)n | 15 (37) | 0 (0)k | 2 (4)c | |
| III (n%) | 15 (4) | 13 (9)d | 2 (6)n | 0 (0) | 0 (0)f | 0 (0) | |
| Creatinine (mg/dL) | 0.9 ± 0.4 | 0.9 ± 0.3 | 1.2 ± 0.8i | 1.0 ± 0.3 j | 0.8 ± 0.2 | 0.9 ± 0.2 | <0.001 |
| NT-proBNP (pg/mL) | 1193 ± 2583 | 1183 ± 2572 | 4146 ± 4778 f | 796 ± 728 | 444 ± 699 | 422 ± 1056 | <0.001 |
| hsTnT (ng/mL) | 50.7 ± 186.3 | 15.4 ± 11.5 | 148.7 ± 222.6d | 132.0 ± 340.4 | 13.7 ± 9.1b | 105.6 ± 368.4 | 0.002 |
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| Clinical characteristics | |||||||
| Age (years) | 60.9 ± 17.0 | 55.2 ± 15.5a | 72.5 ± 10.6 | 65.6 ± 10.3b | 72.9 ± 9. 0 | 43.3 ± 17.1c | <0.001 |
| Female sex (n%) | 131 (35) | 47 (32)d | 14 (40) | 9 22)e | 35 (36) | 26 (48) | 0.081 |
| BSA (mg/m2) | 1.8 ± 0.2 | 1.9 ± 0.2 | 1.8 ± 0.2 | 1.9 ± 0.2b | 1.8 ± 0.2 | 1.8 ± 0.2 | 0.011 |
| Hypertension (n%) | 229 (61) | 55 (37)a | 32 (91) | 41 (100) | 81 (84) | 21 (39)c | <0.001 |
| Dyslipidaemia (n%) | 199 (53) | 66 (45)a | 24 (69) | 28 (69) | 67 (69) | 14 (26)c | <0.001 |
| Diabetes (n%) | 98 (26) | 20 (14)a | 18 (51) | 20 (49) | 39 (40) | 1 (2)c | <0.001 |
| Atrial fibrillation (n%) | 64 (17) | 29 (20) | 15 (43)f | 7 (17) | 8 (8)g | 5 (9) | <0.001 |
| Previous AMI (n%) | 25 (7) | 4 (3) | 8 (23)h | 11 (27)i | 2 (2) | 0 (0) | <0.001 |
| Previous HF (n%) | 40 (11) | 15 (10) | 11 (31)f,j | 12 (29)e | 0 (0) | 2 (4)k | <0.001 |
| Previous HFpEF (n%) | 32 (10) | 13 (9) | 6 (35) | 11 (31) | 0 (0) | 2 (4) | 0.034 |
| Vascular disease (n%) | 11(3) | 3 (2)c | 4 (11)i | 4 (10)j | 0 (0) | 0 (0) | <0.001 |
| Stroke/TIA (n%) | 15 (4) | 7 (5) | 3 (9) | 4 (10) | 0 (0)l | 1 (2) | 0.035 |
| COPD (n%) | 49 (13) | 14 (10) | 7 (20) | 11 (27) | 16 (17) | 1 (2)h | 0.002 |
| Chronic kidney disease (n%) | 48 (13) | 7 (5)m | 9 (26) | 7 (17) | 25 (26) | 0 (0)h | <0.001 |
| Smoking (n%) | 93 (25) | 26 (18)m | 10 (29) | 13 (32) | 34 (35) | 10 (19) | 0.017 |
| NYHA class | |||||||
| I (n%) | 251 (67) | 66 (47) | 10 (29)n | 26 (63) | 97 (100)k | 52 (96)c | <0.001 |
| II (n%) | 109 (29) | 69 (47) | 23 (66)n | 15 (37) | 0 (0)k | 2 (4)c | |
| III (n%) | 15 (4) | 13 (9)d | 2 (6)n | 0 (0) | 0 (0)f | 0 (0) | |
| Creatinine (mg/dL) | 0.9 ± 0.4 | 0.9 ± 0.3 | 1.2 ± 0.8i | 1.0 ± 0.3 j | 0.8 ± 0.2 | 0.9 ± 0.2 | <0.001 |
| NT-proBNP (pg/mL) | 1193 ± 2583 | 1183 ± 2572 | 4146 ± 4778 f | 796 ± 728 | 444 ± 699 | 422 ± 1056 | <0.001 |
| hsTnT (ng/mL) | 50.7 ± 186.3 | 15.4 ± 11.5 | 148.7 ± 222.6d | 132.0 ± 340.4 | 13.7 ± 9.1b | 105.6 ± 368.4 | 0.002 |
Values are expressed as n (%), mean ± SD, or median (IQR).
aCM vs. all other groups P < 0.05.
bHTN heart disease vs. AS P < 0.05.
cNormal vs. all other groups P < 0.05.
dHCM vs. normal P < 0.05.
eHTN heart disease vs. normal P < 0.05.
fCA vs. all other groups P < 0.05.
gAS vs. HCM P < 0.05.
hCA vs. HCM, AS, and normal P < 0.05.
iHTN heart disease vs. HCM, AS, and normal P < 0.05.
jCA vs. HCM and AS P < 0.05.
kNormal vs. CA, HTN heart disease, and AS P < 0.05.
lAS vs. HCM, CA, and HTN heart disease P < 0.05.
mHCM vs. CA, AS, and HTN heart disease P < 0.05.
nCA vs. AS P < 0.05.
AMI, acute myocardial infarction; BSA, body surface area; COPD, chronic obstructive pulmonary disease; HF, heart failure; HFpEF, heart failure with preserved left ventricular ejection fraction; hsTnT, high-sensitivity troponin T; NYHA, New York Heart Association; NT-proBNP, N-terminal prohormone of brain natriuretic peptide TIA, transitory ischaemic attack.
CMR characteristics of the population
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| CMR characteristic | |||||||
| LVEF (%) | 63.4 ± 10.7 | 66.6 ± 9.0a | 51.1 ± 16.5b | 61. ± 12.3 | 64.3 ± 8.9 | 62.9 ± 5.3 | <0.001 |
| Indexed LV EDV (mL/m2) | 74.4 ± 16.6 | 73.9 ± 16.1 | 79. ± 25.8 | 78.2 ± 20.2 | 70.9 ± 15.0 | 77.1 ± 11.6 | 0.047 |
| Indexed LV ESV (mL/m2) | 27.5 ± 12.6 | 25.2 ± 9.9a | 38.8 ± 27.1c | 31.7 ± 17.4 | 25.9 ± 10.3 | 28.7 ± 6.3 | <0.001 |
| Indexed mass (g/m2) | 83.7 ± 27.3 | 93.6 ± 31.8 | 95.1 ± 30.6 | 87.4 ± 24.2 | 75.6 ± 17.3d | 65. 0 ± 11.5e | <0.001 |
| RVEF (%) | 64.4 (58.2–71.4) | 68.1 (55.5–80.7)f | 58.3 (42.9–58.4)b | 63 (50.2–75.8) | 65 (51–79) | 62.8 (53.1–72.5) | <0.001 |
| Indexed RV EDV (mL/m2) | 63.1 ± 15.6 | 61.0 ± 15.6 | 60.8 ± 16.0 | 65.3 ± 16.3 | 59.8 ± 13.9 | 73.3 ± 13.5g | <0.001 |
| Indexed RV ESV (mL/m2) | 22.4 ± 9.4 | 19.7 ± 8.1a | 25.6 ± 13.3 | 25.1 ± 10.6 | 21.4 ± 8.5 | 28.1 ± 8.6h | <0.001 |
| T1 mapping (ms) | 1038.8 ± 59.0 | 1052.4 ± 56.3 | 1129.1 ± 69.9b | 1034.5 ± 46.5 | 1014.8 ± 44.9d | 1003.7 ± 28.0g | <0.001 |
| ECV | 30.3 ± 9.2 | 31.2 ± 9.1a | 47.1 ± 15.1b | 29.4 ± 5.0 | 26.5 ± 4.1 | 26.3 ± 3.0i | <0.001 |
| LV LGE | 212 (59) | 111 (76)a | 29 (83) | 24 (59) | 48 (50)d | 0 (0)j | <0.001 |
| LV LGE localization | |||||||
| Diffuse (n%) | 25 (7) | 6 (4) | 18 (51)b | 0 (0) | 1 (1) | 0 (0) | <0.001 |
| Intramyocardic (n%) | 163 (44) | 103 (71)f | 7 (20) | 17 (42) | 36 (37) | 0 (0)j | <0.001 |
| Epicardic (n%) | 5 (1) | 1 (1) | 1 (3) | 0 (0) | 3 (3) | 0 (0) | 0.328 |
| Endocardic (n%) | 46 (12) | 15 (10)k | 8 (23) | 7 (17) | 16 (17) | 0 (0)j | 0.007 |
| LV GLS (%) | 22.5 (8–36) | 23.8 (15.5–32.1)a | 16.7 (8.41–25)b | 20.5 (13.9–27.1) | 22.2 (16.6–27.9) | 22.4 (17.1–27.7) | <0.001 |
| LAEF (%) | 44.5 ± 16.1 | 44.6 ± 13.7 | 21.7 ± 15.6b | 40.5 ± 16.9 | 46.2 ± 12.9 | 58.0 ± 9.4j | <0.001 |
| Indexed LA Vmin (mL/m2) | 28.6 ± 16.1 | 30.6 ± 16.8 | 46.8 ± 16.4b | 29.9 ± 15.7 | 25.4 ± 12.4 | 17. ± 5.8j | <0.001 |
| Indexed LA Vmax (mL/m2) | 49.2 ± 16.7 | 53.0 ± 18.1 | 59.1 ± 16.9b | 48.6 ± 14.8 | 45.5 ± 14.8 | 40.4 ± 10.4j | <0.001 |
| LACI (%) | 39 ± 2 | 42 ± 2 | 58 ± 2b | 39 ± 2 | 37 ± 2 | 22 ± 1j | <0.001 |
| LA sphericity | 81.6 ± 2.4 | 81.8 ± 2.1 | 83.2 ± 2.1b | 82.1 ± 2.2 | 81.3 ± 2.6 | 80.7 ± 2.4j | <0.001 |
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| CMR characteristic | |||||||
| LVEF (%) | 63.4 ± 10.7 | 66.6 ± 9.0a | 51.1 ± 16.5b | 61. ± 12.3 | 64.3 ± 8.9 | 62.9 ± 5.3 | <0.001 |
| Indexed LV EDV (mL/m2) | 74.4 ± 16.6 | 73.9 ± 16.1 | 79. ± 25.8 | 78.2 ± 20.2 | 70.9 ± 15.0 | 77.1 ± 11.6 | 0.047 |
| Indexed LV ESV (mL/m2) | 27.5 ± 12.6 | 25.2 ± 9.9a | 38.8 ± 27.1c | 31.7 ± 17.4 | 25.9 ± 10.3 | 28.7 ± 6.3 | <0.001 |
| Indexed mass (g/m2) | 83.7 ± 27.3 | 93.6 ± 31.8 | 95.1 ± 30.6 | 87.4 ± 24.2 | 75.6 ± 17.3d | 65. 0 ± 11.5e | <0.001 |
| RVEF (%) | 64.4 (58.2–71.4) | 68.1 (55.5–80.7)f | 58.3 (42.9–58.4)b | 63 (50.2–75.8) | 65 (51–79) | 62.8 (53.1–72.5) | <0.001 |
| Indexed RV EDV (mL/m2) | 63.1 ± 15.6 | 61.0 ± 15.6 | 60.8 ± 16.0 | 65.3 ± 16.3 | 59.8 ± 13.9 | 73.3 ± 13.5g | <0.001 |
| Indexed RV ESV (mL/m2) | 22.4 ± 9.4 | 19.7 ± 8.1a | 25.6 ± 13.3 | 25.1 ± 10.6 | 21.4 ± 8.5 | 28.1 ± 8.6h | <0.001 |
| T1 mapping (ms) | 1038.8 ± 59.0 | 1052.4 ± 56.3 | 1129.1 ± 69.9b | 1034.5 ± 46.5 | 1014.8 ± 44.9d | 1003.7 ± 28.0g | <0.001 |
| ECV | 30.3 ± 9.2 | 31.2 ± 9.1a | 47.1 ± 15.1b | 29.4 ± 5.0 | 26.5 ± 4.1 | 26.3 ± 3.0i | <0.001 |
| LV LGE | 212 (59) | 111 (76)a | 29 (83) | 24 (59) | 48 (50)d | 0 (0)j | <0.001 |
| LV LGE localization | |||||||
| Diffuse (n%) | 25 (7) | 6 (4) | 18 (51)b | 0 (0) | 1 (1) | 0 (0) | <0.001 |
| Intramyocardic (n%) | 163 (44) | 103 (71)f | 7 (20) | 17 (42) | 36 (37) | 0 (0)j | <0.001 |
| Epicardic (n%) | 5 (1) | 1 (1) | 1 (3) | 0 (0) | 3 (3) | 0 (0) | 0.328 |
| Endocardic (n%) | 46 (12) | 15 (10)k | 8 (23) | 7 (17) | 16 (17) | 0 (0)j | 0.007 |
| LV GLS (%) | 22.5 (8–36) | 23.8 (15.5–32.1)a | 16.7 (8.41–25)b | 20.5 (13.9–27.1) | 22.2 (16.6–27.9) | 22.4 (17.1–27.7) | <0.001 |
| LAEF (%) | 44.5 ± 16.1 | 44.6 ± 13.7 | 21.7 ± 15.6b | 40.5 ± 16.9 | 46.2 ± 12.9 | 58.0 ± 9.4j | <0.001 |
| Indexed LA Vmin (mL/m2) | 28.6 ± 16.1 | 30.6 ± 16.8 | 46.8 ± 16.4b | 29.9 ± 15.7 | 25.4 ± 12.4 | 17. ± 5.8j | <0.001 |
| Indexed LA Vmax (mL/m2) | 49.2 ± 16.7 | 53.0 ± 18.1 | 59.1 ± 16.9b | 48.6 ± 14.8 | 45.5 ± 14.8 | 40.4 ± 10.4j | <0.001 |
| LACI (%) | 39 ± 2 | 42 ± 2 | 58 ± 2b | 39 ± 2 | 37 ± 2 | 22 ± 1j | <0.001 |
| LA sphericity | 81.6 ± 2.4 | 81.8 ± 2.1 | 83.2 ± 2.1b | 82.1 ± 2.2 | 81.3 ± 2.6 | 80.7 ± 2.4j | <0.001 |
Values are expressed as n (%), mean ± SD, or median (IQR).
aHCM vs. HTN heart disease P < 0.05.
bCA vs. all other groups P < 0.05.
cCA vs. HCM, AS, and normal P < 0.05.
dAS vs. HTN heart disease P < 0.05.
eNormal vs. CA, HCM, and HTN heart disease P < 0.05.
fHCM vs. all other groups P < 0.05.
gNormal vs. HCM, CA, and AS P < 0.05
hNormal vs. HCM and AS P < 0.05.
iNormal vs. HCM P < 0.05.
jNormal vs. all other groups P < 0.05.
kHCM vs. CA P < 0.05.
EDV, end-diastolic volume; ESV, end-systolic volume; LVEF, left ventricular ejection fraction; RVEF, right ventricular ejection fraction.
CMR characteristics of the population
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| CMR characteristic | |||||||
| LVEF (%) | 63.4 ± 10.7 | 66.6 ± 9.0a | 51.1 ± 16.5b | 61. ± 12.3 | 64.3 ± 8.9 | 62.9 ± 5.3 | <0.001 |
| Indexed LV EDV (mL/m2) | 74.4 ± 16.6 | 73.9 ± 16.1 | 79. ± 25.8 | 78.2 ± 20.2 | 70.9 ± 15.0 | 77.1 ± 11.6 | 0.047 |
| Indexed LV ESV (mL/m2) | 27.5 ± 12.6 | 25.2 ± 9.9a | 38.8 ± 27.1c | 31.7 ± 17.4 | 25.9 ± 10.3 | 28.7 ± 6.3 | <0.001 |
| Indexed mass (g/m2) | 83.7 ± 27.3 | 93.6 ± 31.8 | 95.1 ± 30.6 | 87.4 ± 24.2 | 75.6 ± 17.3d | 65. 0 ± 11.5e | <0.001 |
| RVEF (%) | 64.4 (58.2–71.4) | 68.1 (55.5–80.7)f | 58.3 (42.9–58.4)b | 63 (50.2–75.8) | 65 (51–79) | 62.8 (53.1–72.5) | <0.001 |
| Indexed RV EDV (mL/m2) | 63.1 ± 15.6 | 61.0 ± 15.6 | 60.8 ± 16.0 | 65.3 ± 16.3 | 59.8 ± 13.9 | 73.3 ± 13.5g | <0.001 |
| Indexed RV ESV (mL/m2) | 22.4 ± 9.4 | 19.7 ± 8.1a | 25.6 ± 13.3 | 25.1 ± 10.6 | 21.4 ± 8.5 | 28.1 ± 8.6h | <0.001 |
| T1 mapping (ms) | 1038.8 ± 59.0 | 1052.4 ± 56.3 | 1129.1 ± 69.9b | 1034.5 ± 46.5 | 1014.8 ± 44.9d | 1003.7 ± 28.0g | <0.001 |
| ECV | 30.3 ± 9.2 | 31.2 ± 9.1a | 47.1 ± 15.1b | 29.4 ± 5.0 | 26.5 ± 4.1 | 26.3 ± 3.0i | <0.001 |
| LV LGE | 212 (59) | 111 (76)a | 29 (83) | 24 (59) | 48 (50)d | 0 (0)j | <0.001 |
| LV LGE localization | |||||||
| Diffuse (n%) | 25 (7) | 6 (4) | 18 (51)b | 0 (0) | 1 (1) | 0 (0) | <0.001 |
| Intramyocardic (n%) | 163 (44) | 103 (71)f | 7 (20) | 17 (42) | 36 (37) | 0 (0)j | <0.001 |
| Epicardic (n%) | 5 (1) | 1 (1) | 1 (3) | 0 (0) | 3 (3) | 0 (0) | 0.328 |
| Endocardic (n%) | 46 (12) | 15 (10)k | 8 (23) | 7 (17) | 16 (17) | 0 (0)j | 0.007 |
| LV GLS (%) | 22.5 (8–36) | 23.8 (15.5–32.1)a | 16.7 (8.41–25)b | 20.5 (13.9–27.1) | 22.2 (16.6–27.9) | 22.4 (17.1–27.7) | <0.001 |
| LAEF (%) | 44.5 ± 16.1 | 44.6 ± 13.7 | 21.7 ± 15.6b | 40.5 ± 16.9 | 46.2 ± 12.9 | 58.0 ± 9.4j | <0.001 |
| Indexed LA Vmin (mL/m2) | 28.6 ± 16.1 | 30.6 ± 16.8 | 46.8 ± 16.4b | 29.9 ± 15.7 | 25.4 ± 12.4 | 17. ± 5.8j | <0.001 |
| Indexed LA Vmax (mL/m2) | 49.2 ± 16.7 | 53.0 ± 18.1 | 59.1 ± 16.9b | 48.6 ± 14.8 | 45.5 ± 14.8 | 40.4 ± 10.4j | <0.001 |
| LACI (%) | 39 ± 2 | 42 ± 2 | 58 ± 2b | 39 ± 2 | 37 ± 2 | 22 ± 1j | <0.001 |
| LA sphericity | 81.6 ± 2.4 | 81.8 ± 2.1 | 83.2 ± 2.1b | 82.1 ± 2.2 | 81.3 ± 2.6 | 80.7 ± 2.4j | <0.001 |
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| CMR characteristic | |||||||
| LVEF (%) | 63.4 ± 10.7 | 66.6 ± 9.0a | 51.1 ± 16.5b | 61. ± 12.3 | 64.3 ± 8.9 | 62.9 ± 5.3 | <0.001 |
| Indexed LV EDV (mL/m2) | 74.4 ± 16.6 | 73.9 ± 16.1 | 79. ± 25.8 | 78.2 ± 20.2 | 70.9 ± 15.0 | 77.1 ± 11.6 | 0.047 |
| Indexed LV ESV (mL/m2) | 27.5 ± 12.6 | 25.2 ± 9.9a | 38.8 ± 27.1c | 31.7 ± 17.4 | 25.9 ± 10.3 | 28.7 ± 6.3 | <0.001 |
| Indexed mass (g/m2) | 83.7 ± 27.3 | 93.6 ± 31.8 | 95.1 ± 30.6 | 87.4 ± 24.2 | 75.6 ± 17.3d | 65. 0 ± 11.5e | <0.001 |
| RVEF (%) | 64.4 (58.2–71.4) | 68.1 (55.5–80.7)f | 58.3 (42.9–58.4)b | 63 (50.2–75.8) | 65 (51–79) | 62.8 (53.1–72.5) | <0.001 |
| Indexed RV EDV (mL/m2) | 63.1 ± 15.6 | 61.0 ± 15.6 | 60.8 ± 16.0 | 65.3 ± 16.3 | 59.8 ± 13.9 | 73.3 ± 13.5g | <0.001 |
| Indexed RV ESV (mL/m2) | 22.4 ± 9.4 | 19.7 ± 8.1a | 25.6 ± 13.3 | 25.1 ± 10.6 | 21.4 ± 8.5 | 28.1 ± 8.6h | <0.001 |
| T1 mapping (ms) | 1038.8 ± 59.0 | 1052.4 ± 56.3 | 1129.1 ± 69.9b | 1034.5 ± 46.5 | 1014.8 ± 44.9d | 1003.7 ± 28.0g | <0.001 |
| ECV | 30.3 ± 9.2 | 31.2 ± 9.1a | 47.1 ± 15.1b | 29.4 ± 5.0 | 26.5 ± 4.1 | 26.3 ± 3.0i | <0.001 |
| LV LGE | 212 (59) | 111 (76)a | 29 (83) | 24 (59) | 48 (50)d | 0 (0)j | <0.001 |
| LV LGE localization | |||||||
| Diffuse (n%) | 25 (7) | 6 (4) | 18 (51)b | 0 (0) | 1 (1) | 0 (0) | <0.001 |
| Intramyocardic (n%) | 163 (44) | 103 (71)f | 7 (20) | 17 (42) | 36 (37) | 0 (0)j | <0.001 |
| Epicardic (n%) | 5 (1) | 1 (1) | 1 (3) | 0 (0) | 3 (3) | 0 (0) | 0.328 |
| Endocardic (n%) | 46 (12) | 15 (10)k | 8 (23) | 7 (17) | 16 (17) | 0 (0)j | 0.007 |
| LV GLS (%) | 22.5 (8–36) | 23.8 (15.5–32.1)a | 16.7 (8.41–25)b | 20.5 (13.9–27.1) | 22.2 (16.6–27.9) | 22.4 (17.1–27.7) | <0.001 |
| LAEF (%) | 44.5 ± 16.1 | 44.6 ± 13.7 | 21.7 ± 15.6b | 40.5 ± 16.9 | 46.2 ± 12.9 | 58.0 ± 9.4j | <0.001 |
| Indexed LA Vmin (mL/m2) | 28.6 ± 16.1 | 30.6 ± 16.8 | 46.8 ± 16.4b | 29.9 ± 15.7 | 25.4 ± 12.4 | 17. ± 5.8j | <0.001 |
| Indexed LA Vmax (mL/m2) | 49.2 ± 16.7 | 53.0 ± 18.1 | 59.1 ± 16.9b | 48.6 ± 14.8 | 45.5 ± 14.8 | 40.4 ± 10.4j | <0.001 |
| LACI (%) | 39 ± 2 | 42 ± 2 | 58 ± 2b | 39 ± 2 | 37 ± 2 | 22 ± 1j | <0.001 |
| LA sphericity | 81.6 ± 2.4 | 81.8 ± 2.1 | 83.2 ± 2.1b | 82.1 ± 2.2 | 81.3 ± 2.6 | 80.7 ± 2.4j | <0.001 |
Values are expressed as n (%), mean ± SD, or median (IQR).
aHCM vs. HTN heart disease P < 0.05.
bCA vs. all other groups P < 0.05.
cCA vs. HCM, AS, and normal P < 0.05.
dAS vs. HTN heart disease P < 0.05.
eNormal vs. CA, HCM, and HTN heart disease P < 0.05.
fHCM vs. all other groups P < 0.05.
gNormal vs. HCM, CA, and AS P < 0.05
hNormal vs. HCM and AS P < 0.05.
iNormal vs. HCM P < 0.05.
jNormal vs. all other groups P < 0.05.
kHCM vs. CA P < 0.05.
EDV, end-diastolic volume; ESV, end-systolic volume; LVEF, left ventricular ejection fraction; RVEF, right ventricular ejection fraction.
LA strain across the LV hypertrophy phenotypes
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| LA reservoir strain (%) | 24 (5–60) | 24.5 (6.8–42.2) | 9.6 (0.6–18.6)a | 23.0 (10.4–35.6) | 24.7 (13.7–35.7) | 38.6 (27.3–49.9)b | <0.001 |
| LA booster strain (%) | 16.1 ± 7.7 | 15. ± 6.4 | 9.1 ± 5.4a | 16.5 ± 8.1 | 17.2 ± 7.4 | 19.5 ± 9.3c | <0.001 |
| LA conduit strain (%) | 10.9 ± 8.2 | 11.1 ± 7.3b | 4.5 ± 4.1 | 9.6 ± 6.4 | 7.9 ± 6.3 | 19.4 ± 9.4b | <0.001 |
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| LA reservoir strain (%) | 24 (5–60) | 24.5 (6.8–42.2) | 9.6 (0.6–18.6)a | 23.0 (10.4–35.6) | 24.7 (13.7–35.7) | 38.6 (27.3–49.9)b | <0.001 |
| LA booster strain (%) | 16.1 ± 7.7 | 15. ± 6.4 | 9.1 ± 5.4a | 16.5 ± 8.1 | 17.2 ± 7.4 | 19.5 ± 9.3c | <0.001 |
| LA conduit strain (%) | 10.9 ± 8.2 | 11.1 ± 7.3b | 4.5 ± 4.1 | 9.6 ± 6.4 | 7.9 ± 6.3 | 19.4 ± 9.4b | <0.001 |
Values are expressed as n (%), mean ± SD, or median (IQR).
aCA vs. all other groups P < 0.05.
bNormal vs. all other groups P < 0.05.
cHCM vs. CA and AS P < 0.05.
LA strain across the LV hypertrophy phenotypes
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| LA reservoir strain (%) | 24 (5–60) | 24.5 (6.8–42.2) | 9.6 (0.6–18.6)a | 23.0 (10.4–35.6) | 24.7 (13.7–35.7) | 38.6 (27.3–49.9)b | <0.001 |
| LA booster strain (%) | 16.1 ± 7.7 | 15. ± 6.4 | 9.1 ± 5.4a | 16.5 ± 8.1 | 17.2 ± 7.4 | 19.5 ± 9.3c | <0.001 |
| LA conduit strain (%) | 10.9 ± 8.2 | 11.1 ± 7.3b | 4.5 ± 4.1 | 9.6 ± 6.4 | 7.9 ± 6.3 | 19.4 ± 9.4b | <0.001 |
| Total (n = 375) | HCM (n = 148) | CA (n = 35) | HTN heart disease (n = 41) | AS (n = 97) | Normal (n = 54) | P | |
|---|---|---|---|---|---|---|---|
| LA reservoir strain (%) | 24 (5–60) | 24.5 (6.8–42.2) | 9.6 (0.6–18.6)a | 23.0 (10.4–35.6) | 24.7 (13.7–35.7) | 38.6 (27.3–49.9)b | <0.001 |
| LA booster strain (%) | 16.1 ± 7.7 | 15. ± 6.4 | 9.1 ± 5.4a | 16.5 ± 8.1 | 17.2 ± 7.4 | 19.5 ± 9.3c | <0.001 |
| LA conduit strain (%) | 10.9 ± 8.2 | 11.1 ± 7.3b | 4.5 ± 4.1 | 9.6 ± 6.4 | 7.9 ± 6.3 | 19.4 ± 9.4b | <0.001 |
Values are expressed as n (%), mean ± SD, or median (IQR).
aCA vs. all other groups P < 0.05.
bNormal vs. all other groups P < 0.05.
cHCM vs. CA and AS P < 0.05.
Patients with CA showed a higher LACI compared with the other groups (Table 2, Figures 1 and 2). In contrast, no significant differences on LACI were observed across patients with HCM, HTN heart disease and AS.
Finally, patients with CA exhibited more impaired LA reservoir and booster strains compared with the other groups (Table 3, Figures 1–3), whereas no significant differences were observed across other hypertrophic phenotypes (Figures 1 and 2).
CMR-derived LA remodelling and functional parameters. 3CH views of each type of hypertrophic phenotype and of a normal CMR are shown. Patients with CA exhibit more pronounced atrial involvement, whereas those with HCM, asymptomatic AS, and HTN heart disease display similar patterns of atrial dysfunction, LACI, and LA sphericity. 3CH, three-chamber view.
LA characteristics in different groups. The box plot shows differences in LA functional, geometrical, and volumetric parameters according to each cohort. LA iVmax, indexed left atrial end-systolic volume.
LA strain curves in different groups. The graph shows LA strain curve of the different cohorts traced using LA reservoir and conduit strain results (given in Table 3) adjusted with the same cardiac cycle duration.
Correlates of LA strains in patients with LV hypertrophic phenotype
Independent clinical correlates of LA reservoir, LA conduit, and LA booster strains are given in Table 4. Based on linear regression models, there was no statistically significant interaction between LV hypertrophy phenotype and iLAVmax or indexed LV mass (for LA reservoir analysis: P = 0.531 and P = 0.696, respectively; for LA conduit analysis: P = 0.818 and P = 0.956, respectively; for LA booster analysis: P = 0.485 and P = 0.600, respectively). Age, LV GLS, iLAVMax, and LACI were independently correlated with all three LA strain components: for LA reservoir analysis: β=−0.15, P < 0.001, β = −0.43, P < 0.001, β = 0.15, P < 0.001, and β = −37.77, P < 0.001, respectively; for LA conduit analysis: β = −0.22, P < 0.001, β = −0.20, P = 0.002, β = 0.07, P = 0.028, and β = −15.83, P < 0.001, respectively; and for LA booster strain: β = 0.09, P = 0.002, β = −0.24, P < 0.001, β = 0.10, P = 0.004, and β = −28.25, P < 0.001, respectively. Additionally, indexed LV mass was also an independent correlate of LA reservoir (β = −0.06, P = 0.001) and LA conduit strain (β = −0.05, P < 0.001). Finally, LA sphericity was independently correlated only with LA reservoir strain (β = −0.49, P = 0.002); male sex was independently correlated only with LA conduit strain (β = 1.69, P = 0.022). Supplementary data online, Tables S1–S3 present the goodness-of-fit indices for the different regression models. The regression models explain 65.0% of the variance in the LA reservoir, 53.0% of the variance in the LA conduit, and 37.0% of the variance in the LA booster with a root mean squared error of 7.28, 5.72, and 6.13, respectively. The variance of the three dependent variables explained by the equations derived by the cross-validation samples and the corresponding root mean squared error were nearly the same as those obtained in the derivation sample (62% and 7.43 for LA reservoir, 50% and 5.82 for LA conduit, and finally, 32% and 6.27 for LA booster, respectively).
Independent correlates of LA reservoir, LA conduit, and LA booster functions
| LA reservoir | LA conduit | LA booster | ||||
|---|---|---|---|---|---|---|
| β (95% CI) | P | β (95% CI) | P | β (95% CI) | P | |
| Male sex | 0.72 (−1.03 to 2.43) | 0.423 | 1.69 (0.25–3.14) | 0.022 | −0.81 (−2.36 to 0.74) | 0.305 |
| Age | −0.15 (−0.21 to −0.09) | <0.001 | −0.22 (−0.27 to −0.17) | <0.001 | 0.09 (0.03–0.14) | 0.002 |
| Indexed LA VMax | 0.15 (0.07–0.22) | <0.001 | 0.07 (0.01–0.14) | 0.028 | 0.10 (0.03–0.17) | 0.004 |
| LA sphericity | −0.49 (−0.81 to −0.18) | 0.002 | −0.25 (−0.50 to 0.01) | 0.059 | −0.22 (−0.49 to 0.06) | 0.124 |
| LACI | −37.77 (−43.98 to −31.55) | <0.001 | −15.83 (−21.68 to −9.98) | <0.001 | −28.25 (−34.55 to −21.95) | <0.001 |
| Indexed LV mass | −0.06 (−0.09 to −0.02) | 0.001 | −0.05 (−0.08 to −0.02) | <0.001 | −0.01 (−0.04 to 0.03) | 0.740 |
| LV GLS | −0.43 (−0.58 to −0.28) | <0.001 | −0.20 (−0.32 to −0.07) | 0.002 | −0.24 (−0.38 to −0.11) | <0.001 |
| LA reservoir | LA conduit | LA booster | ||||
|---|---|---|---|---|---|---|
| β (95% CI) | P | β (95% CI) | P | β (95% CI) | P | |
| Male sex | 0.72 (−1.03 to 2.43) | 0.423 | 1.69 (0.25–3.14) | 0.022 | −0.81 (−2.36 to 0.74) | 0.305 |
| Age | −0.15 (−0.21 to −0.09) | <0.001 | −0.22 (−0.27 to −0.17) | <0.001 | 0.09 (0.03–0.14) | 0.002 |
| Indexed LA VMax | 0.15 (0.07–0.22) | <0.001 | 0.07 (0.01–0.14) | 0.028 | 0.10 (0.03–0.17) | 0.004 |
| LA sphericity | −0.49 (−0.81 to −0.18) | 0.002 | −0.25 (−0.50 to 0.01) | 0.059 | −0.22 (−0.49 to 0.06) | 0.124 |
| LACI | −37.77 (−43.98 to −31.55) | <0.001 | −15.83 (−21.68 to −9.98) | <0.001 | −28.25 (−34.55 to −21.95) | <0.001 |
| Indexed LV mass | −0.06 (−0.09 to −0.02) | 0.001 | −0.05 (−0.08 to −0.02) | <0.001 | −0.01 (−0.04 to 0.03) | 0.740 |
| LV GLS | −0.43 (−0.58 to −0.28) | <0.001 | −0.20 (−0.32 to −0.07) | 0.002 | −0.24 (−0.38 to −0.11) | <0.001 |
CI, confidence interval.
Independent correlates of LA reservoir, LA conduit, and LA booster functions
| LA reservoir | LA conduit | LA booster | ||||
|---|---|---|---|---|---|---|
| β (95% CI) | P | β (95% CI) | P | β (95% CI) | P | |
| Male sex | 0.72 (−1.03 to 2.43) | 0.423 | 1.69 (0.25–3.14) | 0.022 | −0.81 (−2.36 to 0.74) | 0.305 |
| Age | −0.15 (−0.21 to −0.09) | <0.001 | −0.22 (−0.27 to −0.17) | <0.001 | 0.09 (0.03–0.14) | 0.002 |
| Indexed LA VMax | 0.15 (0.07–0.22) | <0.001 | 0.07 (0.01–0.14) | 0.028 | 0.10 (0.03–0.17) | 0.004 |
| LA sphericity | −0.49 (−0.81 to −0.18) | 0.002 | −0.25 (−0.50 to 0.01) | 0.059 | −0.22 (−0.49 to 0.06) | 0.124 |
| LACI | −37.77 (−43.98 to −31.55) | <0.001 | −15.83 (−21.68 to −9.98) | <0.001 | −28.25 (−34.55 to −21.95) | <0.001 |
| Indexed LV mass | −0.06 (−0.09 to −0.02) | 0.001 | −0.05 (−0.08 to −0.02) | <0.001 | −0.01 (−0.04 to 0.03) | 0.740 |
| LV GLS | −0.43 (−0.58 to −0.28) | <0.001 | −0.20 (−0.32 to −0.07) | 0.002 | −0.24 (−0.38 to −0.11) | <0.001 |
| LA reservoir | LA conduit | LA booster | ||||
|---|---|---|---|---|---|---|
| β (95% CI) | P | β (95% CI) | P | β (95% CI) | P | |
| Male sex | 0.72 (−1.03 to 2.43) | 0.423 | 1.69 (0.25–3.14) | 0.022 | −0.81 (−2.36 to 0.74) | 0.305 |
| Age | −0.15 (−0.21 to −0.09) | <0.001 | −0.22 (−0.27 to −0.17) | <0.001 | 0.09 (0.03–0.14) | 0.002 |
| Indexed LA VMax | 0.15 (0.07–0.22) | <0.001 | 0.07 (0.01–0.14) | 0.028 | 0.10 (0.03–0.17) | 0.004 |
| LA sphericity | −0.49 (−0.81 to −0.18) | 0.002 | −0.25 (−0.50 to 0.01) | 0.059 | −0.22 (−0.49 to 0.06) | 0.124 |
| LACI | −37.77 (−43.98 to −31.55) | <0.001 | −15.83 (−21.68 to −9.98) | <0.001 | −28.25 (−34.55 to −21.95) | <0.001 |
| Indexed LV mass | −0.06 (−0.09 to −0.02) | 0.001 | −0.05 (−0.08 to −0.02) | <0.001 | −0.01 (−0.04 to 0.03) | 0.740 |
| LV GLS | −0.43 (−0.58 to −0.28) | <0.001 | −0.20 (−0.32 to −0.07) | 0.002 | −0.24 (−0.38 to −0.11) | <0.001 |
CI, confidence interval.
Discussion
Among patients with LV hypertrophic phenotype, LA remodelling and dysfunction are more pronounced in CA as compared to patients with HCM, asymptomatic AS, and HTN heart disease. Patients with HCM, asymptomatic AS, and HTN heart disease show similar patterns of LA dysfunction and remodelling (Graphical Abstract). Moreover, CMR-derived iLAVmax and LACI measurements are independently correlated with LA reservoir, conduit, and booster strains across all different cardiomyopathies whereas LA sphericity is independently correlated only with LA reservoir strain. Additionally, CMR-derived LV GLS is independently correlated with LA reservoir, conduit, and booster strain.
The LA plays a crucial role in modulating LV filling and cardiac output.10 LA reservoir, conduit, and booster strains are important parameters reflecting LA function during different phases of the cardiac cycle, and they are influenced by LA compliance, LA pre-load, LA contractile function, LV end-diastolic pressure, and LV systolic reserve. The complexity of this interplay between LA and LV is pivotal to understand how each cardiomyopathy with LV hypertrophic phenotype affects LA remodelling and function.
The pathophysiology of each LV hypertrophic phenotype is an important determinant of the alteration of LA strains and left atrioventricular coupling. LV hypertrophy is characterized by altered LV relaxation and increased LV filling pressures that are transmitted to the LA leading to LA stretch and remodelling. In addition to this mechanism of atrioventricular interplay common to all LV hypertrophic phenotypes, the underlying cause of LV hypertrophy may also influence the atrioventricular function. While in CA there is infiltration of anomalous amyloid fibrils and expansion of the extracellular matrix,3,17 in HCM there is unexplained LV hypertrophy with myofibrillar disarray and myocardial fibrosis caused by mutations in genes encoding various components of the sarcomere apparatus.18–20 Those sarcomere mutations lead to abnormal sarcomere function and increased energy consumption across the cardiac cycle and sensitivity of the myofilaments to calcium.21 Therefore, it could be hypothesized that the LA remodelling and dysfunction could differ across the various phenotypes of LV hypertrophy.
In the present study, patients with CA showed a more pronounced LA remodelling and dysfunction, including increased LA dimensions and sphericity and impaired strains compared with individuals with other cardiomyopathies or normal CMR. Similar findings were reported by Rausch et al.2 demonstrating that patients with CA exhibited lower LA function compared with patients with HTN heart disease despite similar degree of LV wall thickness. Likewise, recent studies demonstrated worse LA function in patients with CA compared with those with HCM, regardless of LV mass and LV function. Nochioka et al.3 suggested a direct injury of the LA walls by amyloid, resulting in impaired function across all phases of LA function. This impairment was observed independently of LA size and was more severe in patients with CA compared with age- and gender-matched healthy controls after adjustment for LA volume and LV function.3 Moreover, a cardiotoxic damage of LA cardiomyocytes by amyloid precursors has been recently hypotized.17,22
Furthermore, the present results suggest that patients with HCM, asymptomatic severe AS, and HTN heart disease may show comparable patterns of LA dysfunction. The lack of differences across those LV hypertrophic phenotypes may be due to the imaging method that may not be sensitive enough to distinguish underlying differences.
In patients with HCM, LA remodelling and dysfunction stem primarily from LV diastolic dysfunction and increased filling pressure due to myocardial hypertrophy and fibrosis. Studies have shown that as LV stiffness increases and LV GLS decreases, LA stiffness progresses and reservoir and conduit strain decrease.23 Initial impairment often occurs in LA reservoir strain before any dilatation of the LA.24 Conversely, with the progression of the disease, LA conduit and booster strains decrease alongside increased LA volumes.25 Factors such as mitral regurgitation and LV outflow tract obstruction can exacerbate LA dysfunction.4,15 Recent results of studies evaluating the effect of mavacamten on LV hypertrophy have shown that the LA has a favourable response with a reduction in LA volumes and an improvement in LA strains, particularly in those patients who do not have atrial fibrillation.26 Whether these changes in LA remodelling and function are independent of the magnitude of LV hypertrophy regression remains elusive.
Similarly, in patients with severe AS, LA dilatation, partly driven by increased LV filling pressures, correlates with reduced LA strain across all phases.27,28
In patients with HTN heart disease, LA dysfunction primarily results from haemodynamic mechanisms, with increased LV afterload leading to elevated LV filling pressure transmitted to the LA.29 LA reservoir function is typically the first to be affected, and CMR studies suggest that LA reservoir strain may also be influenced by LA myocardial fibrosis.30,31 In this clinical scenario, impairment of LA conduit and booster strain has been described to progress with the progression of the pathology, independent of LV mass.28,32
Determinants of abnormal LA strains in LV hypertrophy
In the present study, iLAVmax, LACI, and LV GLS were independently correlated with LA functional parameters. LA volumes have been extensively described as markers of disease progression in HCM, AS, CA, and HTN heart disease.1–4 LACI has been recently described as a promising marker of left atrioventricular coupling, calculated as the ratio between LA and LV volumes measured in the ventricular end-diastole. LACI was initially described as a good predictor for cardiovascular events in large healthy population cohort.14,15 Moreover, the role of this ratio has recently been explored in patients with HCM. Meucci et al.15 reported that an increased LACI value (≥ 40%) was associated with new-onset atrial fibrillation in patients with HCM and demonstrated its incremental value over conventional LA volumetric parameters to predict events in this specific clinical scenario. The potential strength of this ratio when compared with LV and LA volumetric parameters in isolation relies on the fact that LACI reflects both chambers remodelling simultaneously, offering a more comprehensive assessment of overall left-sided cardiac remodelling. Furthermore, an increase of iLAVmin has been identified as the most closely correlated LA parameter with elevated LV filling pressure, directly reflecting the impairment of LV compliance.5,33 Hence, evaluating LACI may enable the detection of early stages of LA remodelling when compared with other LA parameters.
In the present study, LA sphericity was independently correlated with LA reservoir strain. LA sphericity is an important geometry-based marker that quantifies the difference between the shape of the LA and a perfect sphere.6 This parameter is a comprehensive index of LA geometrical remodelling, and it was recently associated with clinical events as atrial fibrillation and stroke.34,35 The interrelation between LA geometry and function in different LV hypertrophic phenotypes has not been yet extensively explored, and the present results suggest that an increased sphericity mainly reflects changes in reservoir strain. Previous studies suggested that LA non-uniform spherical remodelling is associated with increased atrial stretch and that it might lead to a more rigid and less compliant chamber.36 Similar findings about the relationship between atrial shape and function were reported by Hopman et al.37 demonstrating that patients with atrial fibrillation with an increased LA sphericity had an impaired passive LA function when compared with patients with a non-spherical LA. Despite these findings, data on LA sphericity are still scarce, and this parameter is not yet extensively evaluated in everyday clinical practice. Conventionally, LA sphericity is evaluated as a three-dimensional parameter and its assessment might be laborious, requiring dedicated CMR software.
LV GLS was independently correlated with all the three LA strain components. The interdependence of LA strain parameters with LV function has been extensively analysed in different clinical scenarios.38 These results are consistent with the study of Mălăescu et al.39 that showed that LA strain in all cardiac phases is highly dependent on LV strain.
Study limitations
Several limitations should be acknowledged, including the retrospective nature of the study, the lack of age- and sex-matched normal cohorts, and the relatively small sample sizes within each cardiomyopathy group that could affect the power of the study to detect meaningful differences between phenotypes. Nevertheless, the performance of the models has been tested by leave-one-out cross-validation leading to results that were similar to those obtained in the derivation sample, confirming that our findings are sufficiently robust despite the aforementioned limitation. Additionally, due to the relatively small number of cases in each cohort and the wide range of aetiologies of LV hypertrophy, it was not possible to assess the clinical impact of the CMR variables explored. There are no parameters for LA fibrosis available in the present study, which could affect both functional parameters and the volumetric or geometrical adaptation of the LA. LV filling pressures assessed with echocardiography were not available at the same time of the CMR data acquisition, and therefore, we have not included this information in the analysis. Finally, strain analysis was performed using single-vendor software, potentially limiting generalizability due to intervendor variability. Despite these limitations, the present study includes a large group of patients with various forms of LV hypertrophy and various aetiologies that may lead to varying forms of LA remodelling and impact differently on LA functions. In addition, the analysis is performed with CMR which is considered the reference standard imaging modality to assess cardiac chambers. Certainly, larger and prospective studies will be needed to further validate the present findings and to assess the effects of novel therapies on specific aetiologies such as CA and HCM.
Conclusion
LA remodelling and function vary across hypertrophic phenotypes, and patients with CA exhibit more severe LA involvement compared with other groups. CMR-derived parameters such as iLAVmax, LACI, and LV GLS correlate independently with LA function across different hypertrophy patterns, suggesting potential utility in patient risk stratification. Additionally, LA spherical morphology independently correlates only with an impaired LA reservoir strain.
Supplementary data
Supplementary data are available at European Heart Journal - Cardiovascular Imaging online.
Funding
This research received no external funding.
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
References
Author notes
Martina De Raffele and Albert Teis contributed equally to this work.
Conflict of interest: V.D. received speaker fees from Abbott Structural, Edwards Lifesciences, GE Healthcare, JenaValve, Medtronic, Philips, Siemens and Products & Features. A.B.-G. reports personal fees and/or advisory board from AstraZeneca, Novartis, Boehringer Ingelheim, Abbott, Roche Diagnostics, and Vifor Pharma. The remaining authors have nothing to disclose.
