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
Heart and Vascular Center, Semmelweis University, Budapest, Városmajor utca 68., Budapest 1122, HungaryDepartment of Sports Medicine, Semmelweis University, Budapest, Városmajor utca, 68., Budapest 1122, Hungary
Heart and Vascular Center, Semmelweis University, Budapest, Városmajor utca 68., Budapest 1122, HungaryDepartment of Sports Medicine, Semmelweis University, Budapest, Városmajor utca, 68., Budapest 1122, Hungary
The 70% and 50% thresholds of the TB method were comparable except TMPi and differed significantly from the CC technique.
•
Diagnostic changes occurred in three-quarters of the population.
•
All the cardiomyopathy groups were affected regarding the decision-making about pharmaco- and device therapy.
•
In LVNC all the diagnostic and therapeutic decisions and risk stratification were influenced using the TB method.
•
In athletes, the TB method can play a role in the interpretation of borderline cases.
Abstract
Background
The quantitative differences of left and right ventricular (LV, RV) parameters of using different cardiac MRI (CMR) post-processing techniques and their clinical impact are less studied. We aimed to assess the differences and their clinical impact between the conventional contouring (CC) and the threshold-based (TB) methods using 70% and 50% thresholds in different hypertrabeculated conditions.
Methods
This retrospective study included 30 dilated cardiomyopathy, 30 left ventricular non-compaction (LVNC), 30 arrhythmogenic cardiomyopathy patients, 30 healthy athletes and 30 healthy volunteers. All participants underwent CMR imaging on 1.5 T. Cine sequences were used to derive measures of the cardiac volumes, function, total muscle mass (TMi) and trabeculae and papillary muscle mass (TPMi) using CC and TB segmentation methods.
Results
Comparing the CC and the 70% and 50% threshold TB methods, the LV and RV volumes were significantly lower, the ejection fraction (EF) and the TMi were significantly higher with the TB methods. Between the two threshold setups, only TPMi was significantly higher with the 70% threshold. Regarding the clinical benefits, the LVNC was the only group in whom all the diagnostic and therapeutic decisions and risk stratification were influenced using the TB method. Diagnostic changes occurred in three-quarters of the population, and all the cardiomyopathy groups were affected regarding the decision-making about pharmaco- and device therapy.
Conclusions
Using the TB method, only TPMi was significantly higher with the 70% threshold than the 50% setup, and both of them differed significantly from the CC technique, with relevant clinical impacts in all patient groups.
Formation and malformation of cardiac trabeculae: biological basis, clinical significance, and special yield of magnetic resonance imaging in assessment.
Left ventricular papillary muscles and trabeculae are significant determinants of cardiac MRI volumetric measurements: effects on clinical standards in patients with advanced systolic dysfunction.
]. Hypertrabeculation can occur in several physiological and pathological conditions in both ventricles, e.g., in athletes, in pregnancy, in dilated (DCM), non-compaction (LVNC) and arrhythmogenic (ACM) cardiomyopathies [
Increased left ventricular trabeculation in highly trained athletes: do we need more stringent criteria for the diagnosis of left ventricular non-compaction in athletes?.
Reversible de novo left ventricular trabeculations in pregnant women: implications for the diagnosis of left ventricular noncompaction in low-risk populations.
] and the assessment of trabeculae in these cases can sometimes be challenging.
Although the trabeculae are part of the ventricles, there are different post-processing cardiac MRI (CMR) analyzing methods in use. Unlike the conventional contouring technique (CC), which counts the intracavitary myocardium to the blood volume, the threshold-based (TB) trabeculae and papillary muscle mass (TPMi) quantification method can distinguish the TPMi from the blood volume using the different signal intensities of blood and myocardium [
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
Left and right ventricular parameters corrected with threshold-based quantification method in a normal cohort analyzed by three independent observers with various training-degree.
Standardized image interpretation and post-processing in cardiovascular magnetic resonance - 2020 update : Society for Cardiovascular Magnetic Resonance (SCMR): Board of Trustees Task Force on standardized post-processing.
], some studies have concluded that TPMi recognized as muscle leads to decreased end-diastolic (EDVi) and end-systolic (ESVi) volumes, increased ejection fraction (EF) and total muscle mass (TMi) [
Left ventricular papillary muscles and trabeculae are significant determinants of cardiac MRI volumetric measurements: effects on clinical standards in patients with advanced systolic dysfunction.
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
Left and right ventricular parameters corrected with threshold-based quantification method in a normal cohort analyzed by three independent observers with various training-degree.
]. Despite these studies, which were conducted mainly on healthy volunteers, the differences in the hypertrabeculated populations are less well studied, and the precise evaluation of the TPMi using the TB method might influence the diagnostic, differential diagnostic, prognostic and therapeutic decisions in these conditions [
Left ventricular papillary muscles and trabeculae are significant determinants of cardiac MRI volumetric measurements: effects on clinical standards in patients with advanced systolic dysfunction.
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
]. Namely, it could affect the interpretation of cardiac adaptation in athletes; the evaluation of the diagnostic criteria in LVNC and ACM; the risk stratification in LVNC; the progression in DCM, LVNC and ACM; the anticoagulation therapy in LVNC; and the need for implantable devices in DCM and LVNC [
Recommendations for participation in competitive and leisure time sport in athletes with cardiomyopathies, myocarditis, and pericarditis: position statement of the sport cardiology section of the European Association of Preventive Cardiology (EAPC).
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
Left and right ventricular parameters corrected with threshold-based quantification method in a normal cohort analyzed by three independent observers with various training-degree.
]. To the best of our knowledge, there is no evidence of determining the optimal threshold using the TB method.
In this study, we aimed to compare the CC and TB post-processing techniques to assess the differences in left ventricular (LV) and right ventricular (RV) volumetric, functional and muscle mass values, which might have importance in clinical decision-making for patients with different etiologies of ventricular hypertrabeculation. Our further purpose was to evaluate the different threshold setups using the TB method.
2. Methods
2.1 Study population
Overall, 30 patients with DCM, 30 patients with LVNC, 30 patients with ACM, 30 healthy athletes, and 30 healthy volunteers from a Caucasian population were included in our retrospective study. The baseline characteristics are reported in Table 1.
Table 1Baseline characteristics of the study population.
Total population
LVNC
ACM
DCM
Athletes
Healthy volunteers
Population (n/men)
150/80
30/15
30/18
30/17
30/15
30/15
Age (years)
39.8 ± 15.8
43.4 ± 13.5
44.0 ± 17.1
50.7 ± 14.7
23.0 ± 3.7
37.6 ± 11.0
BSA (m2)
1.9 ± 0.2
1.9 ± 0.3
1.9 ± 0.2
2.0 ± 0.2
1.9 ± 0.2
1.9 ± 0.2
BMI (kg/m2)
25.3 ± 4.5
26.2 ± 5.2
25.0 ± 4.2
28.2 ± 5.0
23.7 ± 2.3
23.3 ± 3.8
Contrast agent administred (n)
76
24
25
27
0
0
LVNC = left ventricular non-compaction, ACM = arrhythmogenic cardiomyopathy, DCM = dilated cardiomyopathy, BSA = body surface area, BMI = body mass index, contrast agents: gadobutrol n = 62 (0.1 mL/kg), gadobenate n = 11 (0.2 mL/kg), gadoterate n = 3 (0.2 mL/kg).
According to the “Classification of cardiomyopathies…” by Elliott et al., patients with dilated LV chamber, increased LV volume, impaired LV function (EF <50%) and without a known cause of LV dysfunction were enrolled in the DCM group [
Classification of the cardiomyopathies: a position statement from the European Society of Cardiology working group on myocardial and pericardial diseases.
In the LVNC group, we included patients who fulfilled the Petersen criteria (non-compacted to compacted myocardial layer ratio > 2.3) and had at least one clinical manifestation [
Patients with definitive diagnoses of ACM based on the revised Task-Force criteria including imaging modalities, pathological and clinical data, ECG recording, genetic testing and family history were enrolled in the ACM group. Notably, in the case of two major or one major and two minor or four minor criteria, the diagnoses can be established [
] and healthy volunteers without known cardiovascular or other systemic or relevant nonsystemic diseases were also included.
From the studied populations, patients with congenital, ischemic or valvular heart diseases, other cardiomyopathies, relevant comorbidities, e.g., hypertension or diabetes mellitus, and individuals with a training activity of >6 h/week (except in athletes) were excluded. Other exclusion criteria included technical reasons, e.g., artifacts, implanted devices or contrast agent administration before segmentation [
All procedures performed in this study were in accordance with 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Ethical approval was obtained from the Central Ethics Committee of Hungary, and all participants provided informed consent.
2.2 Image acquisition
The CMR examinations were performed using 1.5 T MRI scanners (Magnetom Aera, Siemens Healthineers, Erlangen, Germany and Achieva, Philips Medical System, Eindhoven, the Netherlands). Retrospectively gated, balanced steady-state free precession (bSSFP) cine sequences were performed with short-axis (SA) and two-, three-, and four-chamber long-axis views from base to apex, covering the whole LV and RV. The slice thickness was 8 mm with no interslice gap, and the field of view was 350 mm on average adapted to body size.
Contrast agent was not given to athletes and healthy volunteers; in the other groups, it was administered to 76 patients, and their details are reported in Table 1. To provide the best image quality for post-processing analyses, contrast agent administration occurred after the acquisition of SA cine images [
For post-processing analysis, we used Medis Suite software (Medis Suite, version 4.0, Medis Medical Imaging Systems, Leiden, the Netherlands).
During the CC technique, semiautomatic tracing was applied with manual correction of the endo- and epicardial borders in end-diastole and end-systole from base to apex on SA cine images (Fig. 1). Myocardial mass was counted between the epi- and endocardial borders, blood volume was counted within the endocardial contour, and TPMi was counted as part of the blood volume.
Fig. 1Different methodological post-processing CMR techniques in an LVNC patient, short-axis images in end-diastole.
a) conventional contouring method, b) threshold-based technique with 70% threshold setup, c) threshold-based technique with 50% threshold setup.
Within the green (LV) and blue (RV) epicardial contour the program identifies each voxel as blood or myocardium. The voxels identified as myocardium within the red (LV) and yellow (RV) endocardial border represent the TPMi, which is green in the LV and purple in the RV.
CMR = cardiac MRI, LVNC = left ventricular non-compaction, LV = left ventricle, RV = right ventricle, TPMi = trabeculae and papillary mass index. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
For the assessment of TPMi and volumetric and functional values, the TB algorithm of the software (MassK module of the Medis Suite program) was applied. The TB method is based on the different signal intensities of the blood and myocardium. The program identifies each voxel within the epicardial border as blood or myocardium. The voxels identified as myocardium within the endocardial border represent the TPMi (Fig. 1). The threshold was set to 70% and 50%.
The LV and RV EDV, ESV, stroke volume (SV), EF, end-diastolic TM and TPM were calculated. All parameters were indexed to body surface area (i).
Manual correction of the semiautomatically traced epi-and endocardial contours was performed by three observers with 11 years (ASZ), 7 years (ARK) and 6 years (ZSG) of experience.
2.4 Clinical impact evaluation
For the evaluation of the CMR data, we used the Alfakih normal values for all populations, except the athletes, where the 95th percentile range of the normal values for athletes was considered [
The impact of sex, age and training on biventricular cardiac adaptation in healthy adult and adolescent athletes: cardiac magnetic resonance imaging study.
]. Regarding the evaluation of clinical decisions using the CC and TB methods, we applied the currently available guidelines and recommendations for patients with cardiomyopathies (LVNC [
Recommendations for participation in competitive and leisure time sport in athletes with cardiomyopathies, myocarditis, and pericarditis: position statement of the sport cardiology section of the European Association of Preventive Cardiology (EAPC).
Continuous parameters are described as the mean and standard deviation (SD), and discrete parameters are described as numbers and percentages. The homogeneity of variances was assessed with Levene's test, and the normality of the distribution was assessed with the Shapiro–Wilk test. The differences were analyzed using one-way analysis of variance (ANOVA) and Tukey's post hoc test for normally distributed variables with equal variances, the Welch test and Games-Howell post hoc test for variables with unequal variances and the Kruskal–Wallis test for nonnormally distributed data. Bonferroni correction for multiple comparison was used. To compare the TPMi values with 70% and 50% TB thresholds, the paired Student's t-test or the Wilcoxon test was used.
The interobserver agreement of the three observers was tested using the intraclass correlation coefficient (ICC). A p value <0.05 was considered indicative of statistical significance. IBM SPSS Statistics (Version 28.0, Armonk, NY) was used for the analyses.
3. Results
The interobserver agreement between the three observers was excellent (Supplementary Table 1).
Among the hypertrabeculated conditions, LVTPMi was significantly higher with both 70% and 50% TB setups in all cardiomyopathies and healthy athletes compared to healthy volunteers. Values are shown in Fig. 2.
Fig. 2LV and RV TPMi with TB 70% and 50% threshold setups in different conditions.
Analyzing the CC and TB techniques, the LV and RV EDVi and ESVi were lower and the TMi was higher with the TB method using either 70% or 50% thresholds than the CC method for all groups. Both LV and RV SVi were lower in all groups assessed with the TB technique, but significant differences were observed only for the LVSVi in patients with LVNC, athletes and healthy volunteers and for the RVSVi in patients with DCM and LVNC and healthy volunteers. The LV and RV EF were significantly higher when measured with the TB method using both 70% and 50% thresholds; however, the RVEF in the DCM group was not significantly different between the two methods. The data are shown in Table 2.
Table 2The measured LV and RV parameters with CC and TB methods using 70% and 50%.
LVNC
ACM
DCM
Athletes
Healthy volunteers
CC
TB 70%
TB 50%
p
CC
TB 70%
TB 50%
p
CC
TB 70%
TB 50%
p
CC
TB 70%
TB 50%
p
CC
TB 70%
TB 50%
p
LVEDVi (ml/m2)
120.57 ± 32.23¶#
86.29 ± 21.32¶
86.75 ± 21.31#
<0.001
106.32 ± 15.54¶#
79.92 ± 12.93¶
80.80 ± 13.04#
<0.001
153.57 ± 40.92¶#
111.82 ± 33.36¶
112.45 ± 33.47#
<0.001
117.55 ± 11.06¶#
87.58 ± 9.55¶
87.82 ± 9.44#
<0.001
84.80 ± 12.23¶#
65.77 ± 9.58¶
65.88 ± 9.54#
<0.001
LVESVi (ml/m2)
71.74 ± 32.99¶#
43.07 ± 21.81¶
43.17 ± 21.82#
<0.001
53.24 ± 12.90¶#
32.26 ± 10.53¶
32.26 ± 10.53#
<0.001
111.99 ± 37.89¶#
75.14 ± 28.97¶
75.32 ± 29.12#
<0.001
53.92 ± 7.67¶#
31.70 ± 6.15¶
31.80 ± 6.19#
<0.001
33.67 ± 6.73¶#
20.59 ± 4.74¶
20.72 ± 4.75#
<0.001
LVSVi (ml/m2)
48.83 ± 8.28¶#
43.23 ± 7.73¶
43.58 ± 7.72#
0.011
53.07 ± 10.01
47.65 ± 9.49
48.53 ± 9.57
0.73
41.58 ± 8.98
36.68 ± 8.56
37.13 ± 8.64
0.61
63.63 ± 9.75¶#
55.88 ± 8.99¶
56.02 ± 8.80#
0.002
51.13 ± 8.51¶#
45.18 ± 6.79¶
45.17 ± 6.73#
0.005
LVEF (%)
43.33 ± 12.94¶#
52.90 ± 14.88¶
53.02 ± 14.84#
0.011
50.15 ± 8.14¶#
59.92 ± 9.74¶
60.36 ± 9.62#
<0.001
28.27 ± 7.29¶#
34.24 ± 7.94¶
34.47 ± 7.93#
0.003
54.03 ± 5.57¶#
63.70 ± 6.27¶
63.70 ± 6.22#
<0.001
60.34 ± 4.86¶#
68.81 ± 4.71¶
68.67 ± 4.69#
<0.001
LVTMi (g/m2)
53.97 ± 15.77¶#
89.96 ± 28.45¶
89.48 ± 28.50#
<0.001
42.92 ± 6.70¶#
70.60 ± 10.58¶
69.68 ± 10.60#
<0.001
71.09 ± 21.32¶#
114.92 ± 29.64¶
114.27 ± 29.51#
<0.001
56.31 ± 8.97¶#
87.80 ± 10.75¶
87.17 ± 10.64#
<0.001
45.64 ± 7.76¶#
65.62 ± 11.11¶
65.50 ± 11.12#
<0.001
RVEDVi (ml/m2)
82.54 ± 13.86¶#
62.41 ± 10.66¶
62.77 ± 10.69#
<0.001
121.70 ± 20.77¶#
101.15 ± 15.78¶
102.02 ± 15.92#
<0.001
82.85 ± 24.12¶#
60.34 ± 17.24¶
60.76 ± 17.40#
<0.001
108.54 ± 12.00¶#
88.20 ± 11.46¶
88.74 ± 11.64#
<0.001
84.09 ± 12.26¶#
66.21 ± 10.70¶
66.31 ± 10.69#
<0.001
RVESVi (ml/m2)
36.65 ± 9.13¶#
23.47 ± 6.46¶
23.53 ± 6.46#
<0.001
75.75 ± 24.49¶#
56.69 ± 22.25¶
56.69 ± 22.25#
<0.001
39.38 ± 17.15¶#
25.69 ± 11.59¶
25.75 ± 11.56#
<0.001
53.61 ± 7.34¶#
33.95 ± 5.04¶
34.27 ± 5.22#
<0.001
36.05 ± 8.22¶#
24.93 ± 6.41¶
25.01 ± 6.40#
<0.001
RVSVi (ml/m2)
45.89 ± 6.52¶#
38.94 ± 5.56¶
39.24 ± 5.55#
<0.001
47.02 ± 10.48¶#
45.49 ± 11.07¶
46.40 ± 11.10#
0.86
41.25 ± 8.95¶#
32.94 ± 8.86¶
33.25 ± 9.01#
<0.001
54.94 ± 8.42
54.25 ± 9.23
54.47 ± 9.06#
0.955
48.67 ± 8.94¶#
41.98 ± 8.73¶
42.00 ± 8.75#
0.005
RVEF (%)
56.09 ± 5.46¶#
62.94 ± 5.74¶
63.06 ± 5.62#
<0.001
38.82 ± 10.47¶#
45.50 ± 13.49¶
45.96 ± 13.41#
0.01
52.30 ± 12.59
56.79 ± 14.15
56.92 ± 14.09
0.332
50.54 ± 4.81¶#
61.31 ± 4.82¶
61.21 ± 4.70#
<0.001
57.58 ± 6.10¶#
62.77 ± 6.47¶
62.70 ± 6.50#
0.002
RVTMi (g/m2)
13.07 ± 2.15¶#
34.20 ± 6.59¶
33.82 ± 6.53#
<0.001
19.32 ± 3.24¶#
40.61 ± 8.36¶
39.69 ± 8.18 #
<0.001
14.18 ± 4.08¶#
36.95 ± 9.75¶
36.51 ± 9.65#
<0.001
22.46 ± 2.22¶#
43.60 ± 4.51¶
42.74 ± 3.88#
<0.001
15.27 ± 3.46¶#
33.97 ± 5.78¶
33.86 ± 5.80#
<0.001
LV = left ventricular, RV = right ventricular, CC = conventional contouring method, TB = threshold-based technique, LVNC = left ventricular non-compaction, ACM = arrhythmogenic cardiomyopathy, DCM = dilated cardiomyopathy, LVEDVi = left ventricular end-diastolic volume index, LVESVi = left ventricular end-systolic volume index, LVSVi = left ventricular stroke volume index, LVEF = left ventricular ejection fraction, LVTMi = left ventricular total mass index, LVTPMi = left ventricular trabeculated and papillary muscle mass index, RVEDVi = right ventricular end-diastolic volume index, RVESVi = right ventricular end-systolic volume index, RVSVi = right ventricular stroke volume index, RVEF = right ventricular ejection fraction, RVTMi = right ventricular total mass index, RVTPMi = right ventricular trabeculated and papillary muscle mass index.
bold values of p indicate statistical significance (p < 0.05).
¶ = p < 0.05 TB method at 70% versus conventional contouring technique.
# = p < 0.05 TB 50% method versus conventional contouring technique.
Comparing the 70% and 50% thresholds of the TB method, we did not find significant differences in the LV and RV volumetric, functional and TMi parameters in any of the groups (Table 2). However, the TPMi values measured with a threshold of 70% were significantly higher than those measured with a threshold of 50% for both the patient groups and the healthy volunteers (Fig. 2).
Regarding the clinical implications of the TB method, the reduction in volumetric parameters, the improvement in the ejection fraction and the quantification of TPMi could result in changes in the diagnostic criteria, risk stratification and therapeutic decisions not only in the DCM, LVNC, and ACM groups but also in healthy athletes. The LVNC group was the only group in which the TB technique influenced all the abovementioned clinical viewpoints: TPMi was required for the diagnosis in each case, and other aspects were influenced in approximately 30% of the cases. Regarding the ACM group, the diagnostic Task-Force criteria could be modified in 67% of the patients, the diagnosis could be denied in 17% of the patients, and the indication of the pharmaceutical therapy could be reduced in 33% of the patients; in one case the indication for implanted cardiodefibrillator (ICD) therapy could become unnecessary. In the DCM group, modification in the indication of pharmaceutic or device therapy could occur in 34% of the patients, and in two patients, the TB method could reduce the estimated risk for mortality or cardiac transplantation. For three asymptomatic athletes who fulfilled the Petersen and Jacquier criteria for LVNC, the TB method normalized the LVEF above 50%. The detailed data are reported in Table 3.
Table 3a. Modification effect of the TB method with both 70% and 50% threshold setups in clinical decision-making – detailed data.
Patient population (n)
What does the TB method modify?
Clinical impact
n / %
LVNC (n = 30)
quantification of LVTPMi
verify Jacquier diagnostic criteria
30 / 100
reduction in LVEDVi
reduction in the risk of complications
10 / 33
improvement in LVEF
reclassification from HFrEF to HFmrEF- pharmacotherapeutic changes
8 / 27
reduction in the primary prophylactic criteria for anticoagulation
8 / 27
reduction of CRT-D implantation indication
9 / 30
ACM (n = 30)
reduction in RVEDVi or/and improvement in RVEF
loss of ACM diagnosis
8 / 27
loss of major Task-Force criteria
6 / 20
reduction of major to minor Task-Force criteria
6 / 20
loss of minor Task-Force criteria
5 / 17
improvement in LVEF
reclassification from HFmrEF to preserved EF- pharmacotherapeutic changes
7 / 23
reclassification from HFrEF to HFmrEF- pharmacotherapeutic changes
3 / 10
reduction of CRT-D implantation indication
1 / 3
DCM (n = 30)
improvement in RVEF
reduction in the risk of mortality and cardiac transplantation
2 / 7
improvement in LVEF
reclassification from HFrEF to HFmrEF- pharmacotherapeutic changes
5 / 17
rejection of CRT-D implantation indication
5 / 17
Athletes (n = 30)
improved LVEF
falls out from partial sport restriction
3 / 10
b. Modification effect of the TB method with both 70% and 50% threshold setups in clinical decision making – summary
Diagnostic criteria evaluation (n/%)
Risk stratification (n/%)
Pharmacotherapy (n/%)
Device therapy (n/%)
LVNC
30 / 100
10 / 33
8 / 27
9 / 30
ACM
17 / 56
0
10 / 33
1 / 3
DCM
0
2 / 7
5 / 17
5 / 17
Athletes
3 / 10
0
0
0
Healthy volunteers
0
0
0
0
TB = threshold-based, LVNC = left ventricular non-compaction, ACM = arrhythmogenic cardiomyopathy, DCM = dilated cardiomyopathy, LVTPMi = left ventricular trabeculated and papillary muscle mass index, LVEDVi = left ventricular end-diastolic volume index, LVEF = left ventricular ejection fraction, RVEDVi = right ventricular end-diastolic volume index, RVEF = right ventricular ejection fraction, HFrEF = heart failure with reduced ejection fraction, HFmrEF = heart failure with mid-range ejection fraction, CRT-D = cardiac resynchronization therapy with defibrillator.
In this study, we compared the CC and TB CMR post-processing methods and two different TB threshold setups to assess the differences in LV and RV volumes, EF, TMi and TPMi and their clinical impact in hypertrabeculated patients and healthy athletes.
According to our results, the TB method calculated lower volumetric parameters and higher EF and TMi values, which is supported by the literature [
Left and right ventricular parameters corrected with threshold-based quantification method in a normal cohort analyzed by three independent observers with various training-degree.
]. Weinsaft et al., Quick et al. and Riffel et al. described similar differences in healthy volunteers and patients with different cardiovascular diseases using other types of manual contouring methods, where TPMi was considered either as a part of the blood volume or as the myocardial muscle mass [
Left ventricular papillary muscles and trabeculae are significant determinants of cardiac MRI volumetric measurements: effects on clinical standards in patients with advanced systolic dysfunction.
Left and right ventricular parameters corrected with threshold-based quantification method in a normal cohort analyzed by three independent observers with various training-degree.
]. Despite the abovementioned differences, when validating the TB method with a dynamic heart phantom model, Jasper et al. found good correlations between the two techniques [
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
]. Regarding the validation of SVi with aortic flow measurements, the TB mode and the aortic flow were comparable, and both of them gave significant differences from the CC method [
Semi-automatic Detection of Myocardial Trabeculation Using Cardiovascular Magnetic Resonance: Correlation with Histology and Reproducibility in a Mouse Model of Non-compaction. 2023
Comparing the TB method with the 70% and 50% thresholds setups, of the measured parameters, only TPMi was significantly higher at 70%. To the best of our knowledge, results comparing the different thresholds are less well understood, and there is no specific recommendation regarding the optimal threshold setup [
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
]. Varga-Szemes et al. validated this method at 50% with manual correction, and Jaspers et al. used 70% with a heart phantom validation model, although they claimed that 70% might be too high, namely, the model cannot entirely mimic normal physiology [
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
Left and right ventricular parameters corrected with threshold-based quantification method in a normal cohort analyzed by three independent observers with various training-degree.
]. Considering this technique, the significant modifications by this technique in LV and RV volumetric parameters, EF and myocardial mass values could cause alterations when using conventional guidelines to evaluate the diagnostic, prognostic and therapeutic decisions.
In terms of diagnostic revisions, in addition to the existing Petersen diagnostic criteria (TPMi/compact myocardium >2.3), the use of the additional Jacquier criteria (TMPi/TMi > 20%) could help to distinguish LVNC from physiological hypertrabeculation. By measuring the TPMi, the TB method offered an optimal option to assess the Jacquier criteria in all patients included in the LVNC group [
Regarding the revised Task-Force criteria in the ACM group, the TB method defined decreased RVEDVi and improved RVEF, which modified the major and minor CMR criteria. In some cases, these changes led to reevaluation of the definite diagnosis, especially in cases with fewer clinical features of ACM [
]. Thus, this technique could have a role in diagnosing borderline cases and in patients with less clinically concerned diseases, although to the best of our knowledge, no clinical data are available in the literature. Similar to ACM, the corrections in LVEDVi and LVEF might also influence the diagnosis of DCM, although no changes were found in our study [
Classification of the cardiomyopathies: a position statement from the European Society of Cardiology working group on myocardial and pericardial diseases.
According to Pelliccia et al., in athletes with a diagnosis of LVNC and near-normal LVEF, restriction is advised from extremely high-intensity sports, while incidentally discovered hypertrabeculation with normal EF should not be diagnosed as LVNC [
Recommendations for participation in competitive and leisure time sport in athletes with cardiomyopathies, myocarditis, and pericarditis: position statement of the sport cardiology section of the European Association of Preventive Cardiology (EAPC).
]. In our study, four athletes fell into this abovementioned restriction zone, but for three of them, the TB technique normalized the LVEF, which resulted in no restrictions [
Recommendations for participation in competitive and leisure time sport in athletes with cardiomyopathies, myocarditis, and pericarditis: position statement of the sport cardiology section of the European Association of Preventive Cardiology (EAPC).
]. Moreover, using the TB method, athlete's volumetric and functional values approached and reached those of the normal population, which could help to understand the physiological hypertrabeculation mechanism and to distinguish it from the pathological one.
Considering the prognostic impact of the TB method in the LVNC group, the increased LVEF and decreased LVEDVi might reduce the estimated risk of adverse cardiovascular events and the need for further investigations or genetic testing [
]. Similarly, changes in the RVEF in the DCM group altered the estimation of mortality or cardiac transplantation, as an important element of risk stratification [
Regarding the therapeutic aspects, the TB method could result in reclassifications of the heart failure (HF) staging: from HF with reduced (HFrEF) to mid-range EF (HFmrEF) and from HFmrEF to preserved EF in the LVNC, ACM and DCM groups. This modification could affect the latest pharmaceutical heart failure therapy, as new drugs (e.g., dapagliflozin, empagliflozin, vericiguat) are dedicated to HFrEF patients [
]. On the other hand, in subjects who were recategorized from HFmrEF to preserved EF, could postpone the drug therapy. In addition, in LVNC patients with an LVEF below 40%, the initiation of oral anticoagulation therapy is recommended due to blood stasis in the intertrabecular recesses [
Finally, LVEF below 35% is one of the main criteria for ICD and cardiac resynchronization therapy implantation in HF patients. Other aspects also affect the decision to implant an ICD; however, the improvement in the LVEF using the TB method in 15 cardiomyopathy patients could modify the indication of cardiac device therapy [
Left ventricular papillary muscles and trabeculae are significant determinants of cardiac MRI volumetric measurements: effects on clinical standards in patients with advanced systolic dysfunction.
For summary, although, the TB method is not a routinely used technique in the daily clinical practice, we hope that our results will help to clarify these technical questions. The abovementioned cases emphasize the usability of TB method in hypertrabeculated populations, however present normal values based on CC technique should not be considered in case of TB method and therefore modified normal values and criteria would be necessary. We should keep in mind that the choice of CC or TB method may influence the diagnosis, risk stratification and therapeutic decision-making of patients with hypertrabeculation. However, using both as a complementary scheme could improve the quality of clinical decisions.
5. Conclusions
Comparing the post-processing CMR techniques, we found that the TB method with both thresholds resulted in significantly lower LV and RV volumes and higher EF and TMi values than the CC technique. The 70% and 50% TB setups gave different results only in the value of TPMi. Regarding the clinical relevance of the changes, all the studied cardiomyopathy groups were affected when using the TB method and the LVNC was the only group where the TB method influenced all the observed clinical aspects. In athletes, this technique can play a role in the interpretation of borderline cases and in the evaluation of partial sport restrictions.
The TB method can be a useful tool in hypertrabeculated patients, however updated normal values and criteria could be necessary.
6. Limitations
We need to mention that this is a cardiac imaging-focused study; thus, the included participants' symptoms, medical therapy, and follow-up are limited. Although this study was performed on a large population, the smaller sample size of each patient group may have affected the statistical findings.
We also have to mention the limitations of the threshold-based software. The software currently quantifies ejection fraction and volume using short-axis slices, with an 8 mm standard for spatial resolution in the Z-direction. Trabeculae and the papillary muscles do not cross the slice perfectly perpendicularly, resulting in partial volume effects. Depending on the actual path of the trabeculae, this could have influenced the threshold-based quantification.
Authors' contributions
KG: methodology, formal analysis, database, investigation, resources, data curation, writing - original draft and visualization, ZSG, ARK: methodology, data curation, writing - review and editing, MH, BM: data curation, CC: methodology, AT, LS, ZD: writing – review and editing, HV: writing – review and editing and supervision, BM: supervision, AS: conceptualization, methodology, investigation, resources, writing – review and editing, supervision, project administration. All authors approved the submitted manuscript.
Declaration of Competing Interest
The authors have no conflict of interests.
Acknowledgment
We would like to thank the technicians who helped during the cardiac magnetic resonance imaging examinations.
The research was financed by the Thematic Excellence Programme (Tématerületi Kiválósági Program, 2020-4.1.1.-TKP2020) of the Ministry for Innovation and Technology in Hungary within the framework of the Therapeutic Development and Bioimaging Programs of Semmelweis University; by the Development of Scientific Workshops of Medical, Health Sciences and Pharmaceutical Education (Project identification number: EFOP-3.6.3-VEKOP-16-2017-00009); by the TKP2021-NKTA-46 program provided by the Ministry of Innovation and Technology of Hungary from the National Research, Development and Innovation Fund; and by the Ministry of Innovation and Technology NRDI Office within the framework of the Artificial Intelligence National Laboratory Program. Project no. NVKP_16-1-2016-0017 (National Heart Program) has been implemented with the support provided by the National Research, Development, and Innovation Fund of Hungary, financed under the NVKP_16 funding scheme. KG received funding from the ELTE Márton Áron Research Fellowship Program.
Formation and malformation of cardiac trabeculae: biological basis, clinical significance, and special yield of magnetic resonance imaging in assessment.
Left ventricular papillary muscles and trabeculae are significant determinants of cardiac MRI volumetric measurements: effects on clinical standards in patients with advanced systolic dysfunction.
Increased left ventricular trabeculation in highly trained athletes: do we need more stringent criteria for the diagnosis of left ventricular non-compaction in athletes?.
Reversible de novo left ventricular trabeculations in pregnant women: implications for the diagnosis of left ventricular noncompaction in low-risk populations.
Improving the reproducibility of MR-derived left ventricular volume and function measurements with a semi-automatic threshold-based segmentation algorithm.
Left and right ventricular parameters corrected with threshold-based quantification method in a normal cohort analyzed by three independent observers with various training-degree.
Standardized image interpretation and post-processing in cardiovascular magnetic resonance - 2020 update : Society for Cardiovascular Magnetic Resonance (SCMR): Board of Trustees Task Force on standardized post-processing.
Recommendations for participation in competitive and leisure time sport in athletes with cardiomyopathies, myocarditis, and pericarditis: position statement of the sport cardiology section of the European Association of Preventive Cardiology (EAPC).
Classification of the cardiomyopathies: a position statement from the European Society of Cardiology working group on myocardial and pericardial diseases.
The impact of sex, age and training on biventricular cardiac adaptation in healthy adult and adolescent athletes: cardiac magnetic resonance imaging study.
Semi-automatic Detection of Myocardial Trabeculation Using Cardiovascular Magnetic Resonance: Correlation with Histology and Reproducibility in a Mouse Model of Non-compaction. 2023
Over the last decades, clinical guidelines and established cut-off have been proposed to guide the diagnosis and treatment of many congenital and acquired heart and vessels diseases and have increasingly became a part of clinical practice and potentially citable evidence for malpractice litigation [1]. However, disparities in the measurements of vessels dimensions and ventricular volumes and function are not uncommon between different imaging modalities and even within the same modality.
00462-X&id=gr1.jpg){kind=link}
00462-X&id=gr2.jpg){kind=link}