Prognostic Value of Multiple Circulating Biomarkers for Ventricular Arrhythmias in Left Ventricular Hypertrabeculation　― Longitudinal Cohort Study ―

Abstract

Background: Ventricular arrhythmia (VA) is an independent risk factor for adverse outcomes in patients with left ventricular hypertrabeculation (LVHT). This study explored the predictive value of biomarkers for VAs in LVHT.

Methods and Results: This cohort study retrospectively enrolled 265 LVHT patients (mean [±SD] age 44.2±17.0 years, 65.7% male) with data available for N-terminal pro B-type natriuretic peptide, big endothelin-1, high-sensitivity C-reactive protein, uric acid, and free fatty acid. The primary outcome was a composite of non-sustained ventricular tachycardia, sustained ventricular tachycardia, ventricular fibrillation, and appropriate implantable cardioverter defibrillator therapy. Over a median follow-up of 4.34 years, 82 (30.9%) patients experienced VAs. Multivariable Cox regression analysis revealed that baseline concentrations of big endothelin-1 were independently associated with the occurrence of VAs (hazard ratio 1.513; 95% confidence interval 1.136–2.013; P=0.005). Restricted cubic spline analysis showed that susceptibility to VAs increased markedly with increases in big endothelin-1 concentrations. Subgroup analysis revealed that LVHT patients with big endothelin-1 concentrations >0.63 pmol/L should be closely monitored for VAs, particularly when higher concentrations are accompanied by cardiomyopathies, left ventricular (LV) end-diastolic diameters ≥60 mm, or LV ejection fraction <50%. Individuals with elevated big endothelin-1 concentrations and isolated hypertrabeculation in the LV lateral wall had a significantly greater risk of VAs (log-rank P=0.002).

Conclusions: Big endothelin-1 concentrations and the location of hypertrabeculation can help with risk stratification for VAs in LVHT.

Left ventricular hypertrabeculation (LVHT) is a distinctive morphologic phenotype of the heart with heterogeneous complications and variable prognosis. LVHT is characterized by a bilayered myocardial structure consisting of a thick, excessively trabeculated endocardial layer and a thin, compacted epicardial layer.^1–3 Affected patients have diverse clinical presentations, varying from asymptomatic to life-threatening events, such as end-stage heart failure, malignant arrhythmia, systemic thromboembolism, and sudden cardiac death.^2,4,5 According to recent guidelines on cardiomyopathies,³ LVHT has been described as a phenotypic trait that can exist in isolation or in combination with specific cardiomyopathies and congenital disorders, or develop in response to hemodynamic overload among healthy individuals.^1,2,4,6 Currently, the widely accepted approach to diagnose LVHT is based on the ratio of the hypertrabeculated endocardial layer to the compacted epicardial layer measured by transthoracic echocardiography (TTE) and cardiac magnetic resonance imaging (CMR).^7,8 The apex and lateral wall of the left ventricle (LV) are the most frequently involved regions in LVHT.^2,4,9,10

Ventricular arrhythmia (VA) is an independent risk factor for adverse clinical outcomes among patients with LVHT. Malignant ventricular tachyarrhythmias, including ventricular fibrillation (VF) that causes cardiac arrest, are reported in 38–47% of adult LVHT patients and in 13–18% of LVHT patients with sudden cardiac death.⁴ However, knowledge regarding arrhythmogenesis and factors predictive of VA in LVHT is limited. Previous studies have suggested that cardiovascular biomarkers reflecting cardiac function, endothelial function, inflammation, and oxidative stress are correlated with the occurrence of VA events,^11–16 but the association between these biomarkers and VAs in LVHT is far from clear. Therefore, in this study we investigated long-term follow-up data of a large, morphologically diagnosed, LVHT cohort with the aim of exploring the prognostic value of various biomarkers, namely N-terminal pro B-type natriuretic peptide (NT-proBNP), big endothelin-1 (ET-1), high-sensitivity C-reactive protein (hs-CRP), uric acid (UA), and free fatty acid (FFA), for VAs in this particular population.

Methods

Study Design and Population

This was a retrospective observational longitudinal cohort study that followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.¹⁷ The study was conducted in full compliance with the Declaration of Helsinki, Good Clinical Practice guidelines, and relevant local laws and regulations. It was registered with the China Medical Research Registration and Information Recording System (Registration no. MR11-23-023712) and was approved by the Ethics Committee of Fuwai Hospital (Approval no. 2022-1837). Written informed consent was obtained from all patients.

Consecutive inpatients who were diagnosed with LVHT based on the Jenni criteria via TTE (defined as any segment of the ventricular wall with a maximum end-systolic hypertrabeculated: compacted thickness ratio ≥2)¹⁸ and/or the Petersen criteria via CMR (defined as an end-diastolic hypertrabeculated: compacted thickness ratio ≥2.3)⁹ at Fuwai Hospital between January 1, 2012 and December 31, 2020 were recruited to the study. The medical records of each individual were scrutinized and all individuals underwent comprehensive cardiovascular evaluations. Of the 338 original patients who met the imaging diagnostic criteria for LVHT, we excluded participants who had previously documented major VAs (non-sustained ventricular tachycardia [NSVT], sustained VT, or VF; n=23), lacked baseline data for circulating NT-proBNP, big ET-1, hs-CRP, UA and FFA concentrations (n=46), and were lost to follow-up and/or unable to provide details on the occurrence and date of clinical outcomes (n=4; Figure 1).

Figure 1.

Flowchart showing patient inclusion and exclusion. ET-1, endothelin-1; CMR, cardiac magnetic resonance imaging; FFA, free fatty acid; hs-CRP, high-sensitivity C-reactive protein; LVHT, left ventricular hypertrabeculation; NSVT, non-sustained ventricular tachycardia; NT-proBNP, N-terminal pro B-type natriuretic peptide; TTE, transthoracic echocardiography; UA, uric acid; VF, ventricular fibrillation; VT, ventricular tachycardia.

Laboratory Measurement of Biomarkers

Various circulating biomarkers were detected at baseline (in the morning on admission) under resting conditions after patients had fasted for 12 h, as follows: NT-proBNP was examined using an electrochemiluminescent immunoassay (Elecsys pro-BNP II assay; Roche Diagnostics GmbH, Germany); big ET-1 was quantified using ELISA (BI-20082H big ET-1; Biomedica, Austria); hs-CRP was detected using a particle-enhanced immunoturbidimetric assay (Ultrasensitive CRP kit; Orion Diagnostica, Finland); UA was analyzed using the uricase colorimetric method (LABOSPECT 008; Hitachi, Japan); and FFA was determined using an enzymatic assay (DiaSys Diagnostic Systems GmbH, Germany).

Cardiac Evaluation

TTE was performed using an iE33 Color Doppler Ultrasound System (Philips Healthcare, Andover, MA, USA). Left atrial diameter and LV end-diastolic diameter (LVEDD) were measured in 2-dimensional and M-mode images through the parasternal long-axis acoustic window. The modified biplane Simpson’s rule was used to calculate LV ejection fraction (LVEF). CMR studies were conducted using a 1.5-T scanner (Magnetom Avanto, Siemens Medical Solutions, Erlangen, Germany). Cine images of the LV were captured using a true fast imaging technique and a steady state precession sequence, allowing for visualization in the long-axis, short-axis, and horizontal long-axis orientations. Positive late gadolinium enhancement (LGE) was determined if myocardial LGE was identified by magnetic resonance images when 0.2 mmol/kg contrast agent was administrated for 10–15 min. To evaluate the location of the hypertrabeculation, the LV was divided into 9 distinct segments (basal anterior, basal inferior, basal lateral, basal septal, mid-anterior, mid-inferior, mid-lateral, mid-septal, and apical) as described previously.¹⁰ Given that the apex and the LV lateral wall (LVLW) are the most frequently affected regions in LVHT,^2,4,9,10 we further categorized patients based on the distribution of hypertrabeculation in either or both of these regions. Patients were defined as having isolated apical or LVLW hypertrabeculation if only the apex or LVLW was involved, respectively. When the hypertrabeculation was present in both the apex and the LVLW, without any other segments being affected, patients were classified as having apical and LVLW involvement. If there were any discrepancies between the TTE and CMR results, priority was given to CMR findings.

Follow-up and Outcomes

Patients were followed up annually via telephone interviews and/or clinic visits by trained clinical staff. At each follow-up, patients underwent electrocardiogram and Holter monitoring to detect arrhythmias. Individuals with an implantable cardioverter defibrillator (ICD) underwent programming assessments every 6–12 months. The final follow-up took place in December 2022. The primary outcome was a composite of new-onset VA: (1) NSVT, defined as an episode of ≥3 consecutive ventricular beats with a rate of more than 120 beats/min, lasting <30 s; (2) sustained VT or VF; and (3) appropriate ICD therapy due to VT/VF events.

Statistical Analysis

Continuous variables are presented as the mean±SD or as the median with interquartile range (IQR), and were compared using an independent sample t-test or the Mann-Whitney U test. Categorical variables are presented as proportions, and were assessed using the Chi-squared test or Fisher’s exact test. Univariable and multivariable Cox regression analyses were used to explore independent predictors for VAs, NSVT, and sustained VT/VF or appropriate ICD therapy in LVHT. Restricted cubic spline (RCS) analysis fitted for the multivariable Cox regression model was used to determine the non-linear correlation between big ET-1 concentrations and hazard ratios (HRs) for VA. The LVHT cohort was stratified into 3 groups according to big ET-1 tertiles, and the cumulative incidence of VAs among the different groups was compared using Kaplan-Meier curves. In multivariable analyses to predict VA according to big ET-1 tertiles, Model 1 was adjusted for age and presyncope/syncope, Model 2 was adjusted for age, presyncope/syncope, LVEDD, LVEF, and LVHT associated with cardiomyopathies, and Model 3 was further adjusted for isolated apical hypertrabeculation, isolated LVLW hypertrabeculation, and positive LGE in addition to the covariates in Model 2. Subgroup analysis was performed for age, sex, LVEDD, LVEF, and LVHT associated with cardiomyopathy to predict the risk of VA based on big ET-1 tertiles. The significance of differences in the cumulative incidence of VAs according to big ET-1 concentrations and the location of the hypertrabeculation were analyzed using Kaplan-Meier curves. All analyses were conducted using SAS V.9.4 (SAS Institute, Cary, NC, USA) and R (R Foundation for Statistical Computing, Vienna, Austria). Statistical significance was defined as 2-tailed P<0.05.

Results

Baseline Characteristics

We consecutively enrolled 265 LVHT patients (mean age 44.2±17.0 years, 65.7% male) with thorough profiles of circulating biomarkers and no previously documented history of major ventricular arrhythmic events (Figure 1). There were no substantial differences between individuals included and excluded from the study with regard to demographics, comorbidity, cardiac function, and key TTE findings (Supplementary Table 1). The baseline characteristics of the LVHT cohort according to the occurrence of VA are summarized in Table 1. Within the enrolled population, 23% exhibited isolated LVHT, whereas the remaining 77% had different forms of cardiomyopathies, with dilated cardiomyopathy being the most prevalent subtype. Patients who developed VAs had elevated levels of NT-proBNP and big ET-1, presented with an enlarged LV and reduced LVEF, and displayed a lower proportion of isolated apical hypertrabeculation but a higher proportion of isolated LVLW hypertrabeculation.

Table 1.

Baseline Characteristics of Patients With LVHT Overall and According to the Occurrence of Ventricular Arrhythmias

	Overall (n=265)	VAs (n=82)	No VAs (n=183)	P value
Demographics
Age (years)	44.2±17.0	47.9±16.2	42.6±17.2	0.019
Male sex	174 (65.7)	53 (64.6)	121 (66.1)	0.814
Body mass index (kg/m2)	23.32±4.26	23.83±4.20	23.10±4.28	0.197
Cardiovascular risk factors
Hypertension	73 (27.5)	19 (23.2)	54 (29.5)	0.286
Hyperlipidemia	77 (29.1)	25 (30.5)	52 (28.4)	0.731
Diabetes	46 (17.4)	19 (23.2)	27 (14.8)	0.094
Smoking	98 (37.0)	31 (37.8)	67 (36.6)	0.852
Clinical features
Dyspnea	210 (79.2)	70 (85.4)	140 (76.5)	0.100
Chest pain	46 (17.4)	15 (18.3)	31 (16.9)	0.788
Palpitation	139 (52.5)	50 (61.0)	89 (48.6)	0.063
Presyncope/syncope	35 (13.2)	17 (20.7)	18 (9.8)	0.015
Edema	61 (23.0)	23 (28.0)	38 (20.8)	0.193
NYHA Class III/IV	146 (55.1)	55 (67.1)	91 (49.7)	0.009
AFl/AF	47 (17.7)	16 (19.5)	31 (16.9)	0.612
III° AVB	12 (4.5)	3 (3.7)	9 (4.9)	0.760
CRT-D implantation	1 (0.4)	1 (1.2)	0 (0.0)	0.134
Medical treatment
β-blockers	215 (81.1)	66 (80.5)	149 (81.4)	0.858
ARNI/ACEi/ARB	182 (68.7)	58 (70.7)	124 (67.8)	0.630
Diuretics	197 (74.3)	65 (79.3)	132 (72.1)	0.219
MRA	192 (72.5)	62 (75.6)	130 (71.0)	0.441
Antiarrhythmic drugs	27 (10.2)	12 (14.6)	15 (8.2)	0.126
Oral anticoagulants	63 (23.8)	21 (25.6)	42 (23.0)	0.638
Blood tests
NT-proBNP (pg/mL)	1,113.0 [283.2–2,595.8]	1,479.5 [441.4–3,510.5]	953.9 [246.7–2,221.3]	0.043
Big ET-1 (pmol/L)	0.41 [0.22–0.86]	0.58 [0.30–1.20]	0.34 [0.22–0.76]	0.003
hs-CRP (mg/L)	1.66 [0.64–4.41]	2.31 [0.90–7.55]	1.30 [0.55–3.93]	0.060
UA (μmol/L)	395.9 [320.2–505.9]	408.9 [351.7–547.6]	393.9 [314.8–482.7]	0.069
FFA (mmol/L)	0.48 [0.30–0.69]	0.51 [0.32–0.73]	0.47 [0.30–0.68]	0.226
Alanine aminotransferase (IU/L)	24.0 [16.0–37.0]	24.0 [17.8–41.3]	24.0 [16.0–37.0]	0.564
Creatinine (μmol/L)	80.0 [67.4–96.0]	83.5 [68.1–103.0]	78.6 [67.0–94.0]	0.118
Glucose (mmol/L)	4.99 [4.54–5.70]	4.96 [4.48–5.63]	5.00 [4.56–5.73]	0.992
ECG
QRS Duration (ms)	115.60±31.41	112.19±25.58	117.16±33.69	0.193
QTc (ms)	450.3±46.9	454.8±38.4	448.2±50.3	0.293
LBBB	38 (14.3)	8 (9.8)	30 (16.4)	0.154
TTE
LAD (mm)	41.7±8.3	42.1±7.7	41.5±8.5	0.569
LVEDD (mm)	61.0±10.2	62.9±10.5	60.1±10.0	0.037
LVEF (%)	40.4±14.4	37.8±14.7	41.6±14.2	0.052
Cardiac thrombosis	29 (10.9)	8 (9.8)	21 (11.5)	0.679
LVHT subtype
Isolated LVHT	61 (23.0)	13 (15.9)	48 (26.2)	0.064
LVHT with cardiomyopathies	204 (77.0)	69 (84.1)	135 (73.8)	0.064
LVHT associated with DCM	168 (63.4)	58 (70.7)	110 (60.1)	0.097
LVHT associated with NDLVC	21 (7.9)	5 (6.1)	16 (8.7)	0.461
LVHT associated with HCM	11 (4.2)	3 (3.7)	8 (4.4)	0.788
LVHT associated with RCM	3 (1.1)	2 (2.4)	1 (0.5)	0.227
LVHT associated with ARVC	1 (0.4)	1 (1.2)	0 (0.0)	0.309
Location of hypertrabeculation
Isolated apical involved	54 (20.4)	9 (11.0)	45 (24.6)	0.013
Isolated LVLW involved	40 (15.1)	19 (23.2)	21 (11.5)	0.025
Apical and LVLW involved	93 (35.1)	30 (36.6)	63 (34.4)	0.781
CMR (n=224)
LGE(+)	134 (59.8)	48 (68.6)	86 (55.8)	0.079

Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). ACEi, angiotensin-converting enzyme inhibitor; AF, atrial fibrillation; AFl, atrial flutter; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor-neprilysin inhibition; ARVC, arrhythmogenic right ventricular cardiomyopathy; AVB, atrioventricular block; CMR, cardiac magnetic resonance imaging; CRT-D, cardiac resynchronization therapy with a defibrillator; DCM, dilated cardiomyopathy; ECG, electrocardiogram; ET-1, endothelin-1; FFA, free fatty acid; HCM, hypertrophic cardiomyopathy; hs-CRP, high-sensitivity C-reactive protein; LAD, left atrial diameter; LBBB, left bundle branch block; LGE(+), positive late gadolinium enhancement; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; LVHT, left ventricular hypertrabeculation; LVLW, left ventricular lateral wall; MRA, mineralocorticoid receptor antagonist; NDLVC, non-dilated left ventricular cardiomyopathy; NT-proBNP, N-terminal pro B-type natriuretic peptide; NYHA, New York Heart Association; QTc, corrected QT; RCM, restrictive cardiomyopathy; TTE, transthoracic echocardiography; UA, uric acid; VA, ventricular arrhythmia.

Clinical Outcomes

Over a median follow-up of 4.34 years (IQR 0.91–7.37 years), 82 (30.9%) patients with LVHT experienced VA events. NSVT and sustained VT/VF occurred in 66 (24.9%) and 19 (7.2%) individuals, respectively. Twenty-four (9.1%) patients underwent ICD implantation (single-chamber ICD: 7 patients; dual chamber ICD: 6 patients; biventricular ICD: 11 patients), with 10 implanted for primary prevention and 14 for secondary prevention. Appropriate ICD therapies due to VT/VF events were observed in 6 (25.0%) patients (VT: 5 patients; VF: 3 patients; VT+VF: 2 patients).

Prognostic Variables for VA Events in LVHT

Univariable and multivariable Cox regression analyses to explore correlations between clinical parameters and the risk of VAs, NSVT, and sustained VT/VF or appropriate ICD therapy in LVHT are presented in Table 2 and Supplementary Tables 2 and 3, respectively. Variables with P<0.1 in the univariable analyses were examined for collinearity (Supplementary Tables 4–6) before being integrated into the multivariable Cox regression models. After adjusting for potential confounders, baseline big ET-1 concentrations, isolated LVLW hypertrabeculation, presyncope or syncope, and age at diagnosis were found to be independently associated with VAs in LVHT. Meanwhile, elevated big ET-1 concentrations were found to be a significant predictor for both NSVT and sustained VT/VF in individuals with LVHT. According to receiver operating characteristic curve analysis, the optimal cut-off point for big ET-1 to predict VAs in LVHT was 0.345 pmol/L, with a specificity of 72.0% and a sensitivity of 51.4% (Supplementary Figure).

Table 2.

Cox Regression Analysis of Risk Factors for Ventricular Arrhythmias in Patients With LVHT

	Ventricular arrhythmias (n=82)
	Univariable analysis			Multivariable analysis
	HR	95% CI	P value	HR	95% CI	P value
AgeA	1.017	1.004–1.030	0.013	1.019	1.004–1.034	0.013
Male sex	0.940	0.598–1.478	0.789
Presyncope/syncopeA	1.811	1.061–3.091	0.029	1.849	1.007–3.394	0.047
Ln[NT-proBNP]A	1.133	0.981–1.309	0.089	–	–	–
Ln[big ET-1]A	1.518	1.186–1.942	0.001	1.513	1.136–2.013	0.005
Ln[hs-CRP]	1.120	0.971–1.292	0.119
Ln[UA]	1.425	0.768–2.643	0.261
Ln[FFA]	1.263	0.818–1.950	0.292
LAD (TTE)	1.009	0.984–1.035	0.471
LVEDD (TTE)A	1.022	1.001–1.043	0.036	–	–	–
LVEF (TTE)A	0.984	0.969–1.000	0.051	–	–	–
LVHT associated with cardiomyopathiesA	1.715	0.948–3.102	0.075	–	–	–
Isolated apical hypertrabeculationA	0.458	0.229–0.915	0.027	–	–	–
Isolated LVLW hypertrabeculationA	1.803	1.079–3.013	0.024	1.810	1.023–3.202	0.041
LGE(+)A	1.584	0.956–2.624	0.074	–	–	–

^AVariables included in the multivariable Cox analysis. CI, confidence interval; HR, hazard ratio. Other abbreviations as in Table 1.

Dose-Response Association Between Big ET-1 Concentrations and VAs in LVHT

The adjusted dose-response associations between plasma big ET-1 concentrations and HRs for VAs were assessed using RCS regression analysis. The L-shaped curve (Figure 2A) indicated that elevated baseline big ET-1 concentrations were strongly correlated with increased susceptibility for VAs in patients with LVHT.

Figure 2.

Prognostic value of big endothelin (ET)-1 for the occurrence of ventricular arrhythmias in patients with LVHT. (A) Restricted cubic spline analysis between big ET-1 concentrations and hazard ratios (HRs) for the occurrence of ventricular arrhythmias in patients with left ventricular hypertrabeculation (LVHT). CI, confidence interval. (B) Kaplan-Meier curves showing the cumulative incidence of ventricular arrhythmias in patients with LVHT according to big ET-1 tertiles. Tertile 1, big ET-1 <0.26 pmol/L; Tertile 2, big ET-1 0.26–0.63 pmol/L; Tertile 3, big ET-1 >0.63 pmol/L.

Clinical Features and VA Onset in LVHT Patients Stratified by Big ET-1 Tertiles

The LVHT cohort was stratified into 3 groups according to big ET-1 tertiles (T1–3): T1, <0.26 pmol/L; T2, 0.26–0.63 pmol/L; and T3, >0.63 pmol/L. Given that the reference ranges for plasma big ET-1 concentrations were 0.00–0.25 pmol/L, all patients in T1 had normal big ET-1 concentrations at baseline. Compared with T1, individuals in the other 2 groups had larger left chambers, lower LVEF, and higher concentrations of NT-pro BNP, hs-CRP, UA, and FFA (Table 3). Kaplan-Meier curves demonstrated a significant increase in the cumulative incidence of VAs for patients in T3 (total log-rank P=0.040 among the 3 groups; log-rank P=0.013 for T3 vs. T1; Figure 2B). In the multivariable analyses to predict the risk of VAs in LVHT according to big ET-1 tertiles, the crude HR for T3 vs. T1 was 2.003 (95% confidence interval [CI] 1.147–3.498; P=0.015). After accounting for multiple potential confounders in Model 3, the adjusted HRs for VAs in T3 remained higher at 1.911 (95% CI 1.004–3.640; P=0.049; Table 4). Subgroup analysis revealed that, for particular LVHT patients with associated cardiomyopathies, LVEDD ≥60 mm, or LVEF <50%, the risk of experiencing VAs was markedly higher for those in T3 than in T1 (Figure 3).

Table 3.

Clinical Features and Outcomes of Patients With LVHT Overall and According to Tertiles of Big ET-1 Concentrations

	Overall (n=265)	Tertile 1 (n=84)	Tertile 2 (n=91)	Tertile 3 (n=90)	P value
Clinical parameters
Age (years)	44.2±17.0	45.7±15.5	43.0±17.9	44.0±17.6	NS
Male sex	174 (65.7)	53 (63.1)	60 (65.9)	61 (67.8)	NS
Presyncope/syncope	35 (13.2)	14 (16.7)	11 (12.1)	10 (11.1)	NS
NYHA Class III/IV	146 (55.1)	24 (28.6)	52 (57.1)	70 (77.8)	A, a, b, c
NT-proBNP (pg/mL)	1,113.0 [283.2–2,595.8]	272.4 [64.5–786.7]	1,113.0 [465.8–1,900.4]	2,611.3 [1,152.3–3,817.9]	A, a, b, c
hs-CRP (mg/L)	1.66 [0.64–4.41]	1.15 [0.48–2.59]	1.26 [0.48–3.62]	3.58 [1.11–9.94]	A, b, c
UA (μmol/L)	395.9 [320.2–505.9]	351.6 [293.0–414.2]	395.0 [312.6–506.0]	479.2 [382.6–590.1]	A, a, b, c
FFA (mmol/L)	0.48 [0.30–0.69]	0.42 [0.25–0.60]	0.47 [0.30–0.73]	0.55 [0.36–0.75]	A, b
LAD (TTE) (mm)	41.7±8.3	37.8±6.6	42.2±8.7	44.8±7.9	A, a, b, c
LVEDD (TTE) (mm)	61.0±10.2	58.4±8.6	60.7±9.7	63.6±11.5	A, b
LVEF (TTE) (%)	40.4±14.4	47.0±13.1	39.2±14.0	35.4±13.9	A, a, b
Cardiac thrombosis	29 (10.9)	5 (6.0)	12 (13.2)	12 (13.3)	NS
Isolated apical hypertrabeculation	54 (20.4)	25 (29.8)	15 (16.5)	14 (15.6)	A
Isolated LVLW hypertrabeculation	40 (15.1)	15 (17.9)	17 (18.7)	8 (8.9)	NS
LGE(+) (n=224)	134 (59.8)	36 (49.3)	53 (64.6)	45 (65.2)	NS
Clinical outcomes
Ventricular arrhythmias	82 (30.9)	19 (22.6)	27 (29.7)	36 (40.0)	A, b
Sustained VT/VF	19 (7.2)	4 (4.8)	7 (7.7)	8 (8.9)	NS
Non-sustained VT	66 (24.9)	16 (19.0)	21 (23.1)	29 (32.2)	NS
ICD implantation	24 (9.1)	8 (9.5)	10 (11.0)	6 (6.7)	NS
Appropriate ICD therapies	6 (2.3)	2 (2.4)	2 (2.2)	2 (2.2)	NS

Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). ^AP<0.05 among the 3 groups; ^aP<0.05 between Tertile 1 and Tertile 2; ^bP<0.05 between Tertile 1 and Tertile 3; ^cP<0.05 between Tertile 2 and Tertile 3. ICD, implantable cardioverter defibrillator; Tertile 1, big ET-1 <0.26 pmol/L; Tertile 2, big ET-1 0.26–0.63 pmol/L; Tertile 3, big ET-1 >0.63 pmol/L; VF, ventricular fibrillation; VT, ventricular tachycardia.

Table 4.

Multivariable Analysis to Predict Ventricular Arrhythmias in Patients With LVHT According to Tertiles of Big ET-1 Levels

	No. events (%)	Unadjusted		Model 1		Model 2		Model 3
		HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
Ventricular arrhythmias
Tertile 1	19/84 (22.6)	Ref.		Ref.		Ref.		Ref.
Tertile 2	27/91 (29.7)	1.404 (0.781–2.526)	0.257	1.522 (0.845–2.742)	0.162	1.467 (0.803–2.682)	0.213	1.307 (0.690–2.475)	0.411
Tertile 3	36/90 (40.0)	2.003 (1.147–3.498)	0.015	2.152 (1.229–3.768)	0.007	2.005 (1.113–3.611)	0.021	1.911 (1.004–3.640)	0.049

Model 1 was adjusted for age and presyncope/syncope. Model 2 was adjusted for age, presyncope/syncope, LVEDD, LVEF, and LVHT associated with cardiomyopathies. Model 3 was adjusted for age, presyncope/syncope, LVEDD, LVEF, LVHT associated with cardiomyopathies, isolated apical hypertrabeculation, isolated hypertrabeculation of the LV lateral wall, and LGE(+). Other abbreviations as in Tables 1–3.

Figure 3.

Subgroup analysis to predict ventricular arrhythmias in patients with LVHT according to tertiles of big endothelin (ET)-1 concentrations. CI, confidence interval; HR, hazard ratio; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; LVHT, left ventricular hypertrabeculation; Tertile 1, big ET-1 <0.26 pmol/L; Tertile 2, big ET-1 0.26–0.63 pmol/L; Tertile 3, big ET-1 >0.63 pmol/L.

Prognostic Value of Big ET-1 and Hypertrabeculation Distribution for VA in LVHT

The cumulative incidence of VAs in LVHT according to different combinations of big ET-1 concentrations and LVLW hypertrabeculation status is shown in Figure 4. Individuals with elevated (>0.25 pmol/L) big ET-1 concentrations at baseline and isolated LVLW hypertrabeculation had a significantly greater risk of developing VAs compared with those with normal (0.00–0.25 pmol/L) big ET-1 concentrations and non-isolated LVLW hypertrabeculation (log-rank P=0.002, total log-rank P=0.016).

Figure 4.

Kaplan-Meier curves showing the cumulative incidence of ventricular arrhythmias in patients with left ventricular hypertrabeculation (LVHT) according to big endothelin (ET)-1 concentrations and the location of hypertrabeculation. Q1, normal big ET-1+non-isolated left ventricular lateral wall (LVLW) hypertrabeculation; Q2, normal big ET-1+isolated LVLW hypertrabeculation; Q3, elevated big ET-1+non-isolated LVLW hypertrabeculation; Q4, elevated big ET-1+isolated LVLW hypertrabeculation. Note, the normal reference range for big ET-1 is 0.00–0.25 pmol/L.

Discussion

In this study we explored the prognostic value of multiple circulating biomarkers for the onset of VA in patients with LVHT, and found that baseline big ET-1 concentrations were independently associated with the development of major VA events during the long-term follow-up period. RCS analysis showed that the susceptibility to VAs in LVHT was markedly increased with increases in big ET-1 concentrations. Patients with higher plasma big ET-1 concentrations experienced more severe cardiac dysfunction and remodeling. Individuals with elevated big ET-1 concentrations and isolated LVLW hypertrabeculation were at a significantly higher risk of experiencing VA events.

LVHT is a poorly understood myocardial phenotype due to its heterogeneous manifestations and variable prognosis. VA, as one of the major complications of LVHT, has been recognized as an independent risk factor for adverse clinical outcomes.^2,4 A large-scale meta-analysis found that the incidence rate of VAs is approximately 2.17 per 100 person-years among patients with LVHT.¹⁹ In line with consistent diagnostic criteria and with participants sharing similar baseline characteristics (including age, cardiac function, LV size, and LVEF) to prior LVHT investigations,^10,20–24 the present study revealed that nearly 30% of enrolled patients experienced VAs over a median follow-up of 4.34 years. Through the multivariable prognostic analyses on a series of biomarkers reflecting cardiac function, endothelial function, inflammation, and oxidative stress, we demonstrated, for the first time, that baseline plasma concentrations of big ET-1 were significantly correlated with the development of VAs in patients with LVHT.

ET-1, as the most potent vasoconstrictor in the endothelial system, contributes to the stimulation of cardiac hypertrophy, reduction of cardiac output, and fibroblast proliferation in the myocardium.²⁵ It also has a direct arrhythmogenic effect, with the ability to prolong the action potential, provoke early afterdepolarization, and increase the duration and dispersion of ventricular repolarization.^26,27 However, due to the rapid clearance of ET-1, it is difficult to detect its concentration in the circulation. Big ET-1, a 39-amino acid precursor of ET-1, has a much longer biological half-life, making it a more stable biomarker for clinical surveillance and prognostic evaluation. The potential link between plasma big ET-1 concentrations and VAs has been suggested by some clinical studies. For example, big ET-1 was found to be a strong marker for ventricular tachyarrhythmias and VT/VF events in patients with postinfarction LV aneurysm.¹² It was also identified as an effective predictor of VAs among patients requiring ICDs for primary prevention.¹³

Although earlier research explored the connection between big ET-1 and major adverse cardiovascular events in LVHT,²⁸ the specific correlation between big ET-1 and VAs within this particular group of patients has not been thoroughly investigated. In the present study we demonstrated that increased plasma concentrations of big ET-1 were an independent risk factor for both NSVT and sustained VT/VF in patients with LVHT. Several underlying mechanisms may account for our findings. LVHT is characterized by excessive trabeculation in the endocardial layer. This abnormal phenotype not only leads to impaired global LV systolic and diastolic function, but also plays a role in the development of subendocardial ischemia due to compromised intramural perfusion.^2,4 Given that the expression of ET-1 stems primarily from cardiac mechanical stress and ischemia, the abovementioned pathophysiological changes in LVHT may explain why nearly 70% of the patients in our cohort had increased endogenous big ET-1 concentrations at baseline. Meanwhile, elevated concentrations of big ET-1, as a strong hemodynamic, neurohormonal, and proinflammatory mediator, have the potential to further aggravate the myocardial dysfunction and fibrosis in LVHT.²⁵ Comparisons among the big ET-1 tertiles in our study revealed that individuals with higher big ET-1 concentrations had more severe cardiac dysfunction and remodeling, with enlarged left chambers, reduced LVEF, activated inflammatory status, and increased proportions of positive LGE. The border zone connecting the hypertrabeculated segments and fibrotic regions in LVHT has the ability to induce re-entrant circuits and focal automaticity, which can provide ideal arrhythmogenic substrates for scar-related and focal VAs.^29–31 In the present study, Kaplan-Meier curves illustrated prominent differences in the cumulative incidence of VAs among plasma big ET-1 tertiles. RCS analysis demonstrated that the risk of VAs increased markedly with increases in big ET-1 concentrations. Multivariable analyses revealed that, even after accounting for age, syncope, LV size, LVEF, LVHT subtypes, hypertrabeculation location, and the presence of LGE, elevated plasma big ET-1 concentrations remained significantly associated with a heightened susceptibility to major VA events, indicating that big ET-1 may be a novel and promising biomarker for VA risk stratification and prognosis among patients with LVHT. Subgroup analysis suggested that LVHT patients with significantly increased (>0.63 pmol/L) big ET-1 concentrations should be closely monitored for VAs, especially those with associated cardiomyopathies, LVEDD ≥60 mm, or LVEF <50%. The clinical outcomes of these particular subgroups of LVHT patients deserve further investigation. Given the relatively limited occurrence of positive endpoints in the isolated LVHT group within our study cohort, we did not draw definitive conclusions for patients without associated cardiomyopathies in our subgroup analysis. A prospective cohort study with a larger sample size should be conducted in the future to further explore the prognostic value of big ET-1 in individuals with isolated LVHT.

Another novel observation from our study was that the hypertrabeculation distribution in LVHT may offer additional predictive value for the occurrence of VAs. The relationship between the location of hypertrabeculation and its effect on clinical outcomes in LVHT has not been extensively investigated, with some studies suggesting that there is no significant correlation between them.^22,32 However, a large long-term follow-up cohort study consisting of 339 LVHT patients found that hypertrabeculation extending to the mid- or basal segments of the LV was associated with increased overall mortality.¹⁰ Cardiac electrophysiological research revealed that the arrhythmogenic substrate of sustained VT in patients with LVHT typically involved the mid-apical segments of the inferior and lateral walls.³¹ In the present study, up to 70% of patients presented with apical and/or LVLW hypertrabeculation. In multivariable analysis, isolated LVLW hypertrabeculation was demonstrated to be a strong predictor for VAs, and this finding was independent of left chamber size, LV systolic function, associated LVHT subtypes, and fibrosis on CMR. Combining the location of hypertrabeculation with big ET-1, we found that patients who had isolated LVLW hypertrabeculation and elevated (>0.25 pmol/L) plasma big ET-1 concentrations were at a significantly higher risk of developing VAs. Incorporating these clinical indicators may be helpful for the risk stratification of VA events in LVHT, allowing for more individualized, precise patient surveillance and management, and ultimately a better outcome.

This study has several limitations. First, owing to its retrospective nature, potential bias caused by incomplete follow-up cannot be excluded. Encouragingly, our study yielded relatively low rates of missing data and follow-up failures. Moreover, the LVHT cohort was enrolled at the national-level cardiovascular center with in-depth clinical, laboratory, and imaging assessments, and all patients were evaluated according to consistent standards. Second, we were unable to apply all established imaging diagnostic criteria for LVHT in our cohort. Consistent with prior high-quality investigations on LVHT,^10,22,33 the Jenni criteria by TTE and Petersen criteria by CMR were used to screen patients. Third, the detection of VA events was based primarily on the results of electrocardiography, Holter monitoring, ICD programming assessments, and detailed review of medical history at annual follow-up. Nevertheless, potential underestimation of the incidence rate of VAs is still possible. Prospective randomized controlled trials with larger sample sizes in a multicenter setting are warranted to further validate our findings.

Conclusions

Through a long-term investigation on a large cohort of morphologically diagnosed LVHT patients, we demonstrated that elevated baseline big ET-1 concentrations and isolated LVLW hypertrabeculation are independent predictors for the occurrence of VA events. Incorporating routine assessments of these clinical variables may provide further assistance in the risk stratification for VAs in this particular population.

Acknowledgments

The authors are grateful to all the LVHT patients and their families who made this study possible.

Sources of Funding

This research was supported by the National Clinical Research Center for Cardiovascular Diseases, Fuwai Hospital, Chinese Academy of Medical Sciences (Grant no. NCRC2020012) and the National Natural Science Foundation of China (Grant no. 82000323).

Disclosures

The authors have no conflict of interests to declare.

Author Contributions

Y.Y. designed the study as the principal investigator. L. Liu and Y.Y. were involved in study conception and design. L. Liu, L. Li, S.C., A.C., M.X., Y.D., L. Zhou, Y.L., and M.L. were involved in the acquisition, analysis, or interpretation of the data. L. Liu and L. Li conducted the statistical analyses. L. Liu drafted the manuscript. L. Zheng, L.D., X.F., and Y.Y. conducted critical revision of the manuscript for important intellectual content. All authors were involved in interpretation of the results and revision of the manuscript. All authors have read and approved the final manuscript. Y.Y. has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of data analysis.

IRB Information

This study was approved by the Ethics Committee of Fuwai Hospital (Approval no. 2022-1837). Written informed consents were collected from all patients before the study began.

Data Availability

The research data for this study will not be shared.

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-24-0931

References

• 1.   Petersen SE, Jensen B, Aung N, Friedrich MG, McMahon CJ, Mohiddin SA, et al. Excessive trabeculation of the left ventricle: JACC: Cardiovascular Imaging Expert Panel paper. JACC Cardiovasc Imaging 2023; 16: 408–425, doi:10.1016/j.jcmg.2022.12.026.
• 2.   Adabifirouzjaei F, Igata S, DeMaria AN. Hypertrabeculation; a phenotype with heterogeneous etiology. Prog Cardiovasc Dis 2021; 68: 60–69, doi:10.1016/j.pcad.2021.07.007.
• 3.   Arbelo E, Protonotarios A, Gimeno JR, Arbustini E, Barriales-Villa R, Basso C, et al. 2023 ESC guidelines for the management of cardiomyopathies. Eur Heart J 2023; 44: 3503–3626, doi:10.1093/eurheartj/ehad194.
• 4.   Towbin JA, Lorts A, Jefferies JL. Left ventricular non-compaction cardiomyopathy. Lancet 2015; 386: 813–825, doi:10.1016/S0140-6736(14)61282-4.
• 5.   Ozawa SW, Takarada S, Okabe M, Miyao N, Nakaoka H, Ibuki K, et al. Clinical characteristics and prognosis of fetal left ventricular noncompaction in Japan. Circ J 2021; 86: 98–105, doi:10.1253/circj.CJ-20-1148.
• 6.   Dorobantu DM, Radulescu CR, Riding N, McClean G, de la Garza MS, Abuli-Lluch M, et al. The use of 2-D speckle tracking echocardiography in assessing adolescent athletes with left ventricular hypertrabeculation meeting the criteria for left ventricular non-compaction cardiomyopathy. Int J Cardiol 2023; 371: 500–507, doi:10.1016/j.ijcard.2022.09.076.
• 7.   Rao K, Bhaskaran A, Choudhary P, Tan TC. The role of multimodality imaging in the diagnosis of left ventricular noncompaction. Eur J Clin Invest 2020; 50: e13254, doi:10.1111/eci.13254.
• 8.   Kogaki S. Left ventricular noncompaction develops even in late fetal life: The need for fetus-specific diagnostic criteria. Circ J 2021; 86: 106–108, doi:10.1253/circj.CJ-21-0730.
• 9.   Petersen SE, Selvanayagam JB, Wiesmann F, Robson MD, Francis JM, Anderson RH, et al. Left ventricular non-compaction: Insights from cardiovascular magnetic resonance imaging. J Am Coll Cardiol 2005; 46: 101–105, doi:10.1016/j.jacc.2005.03.045.
• 10.   Vaidya VR, Lyle M, Miranda WR, Farwati M, Isath A, Patlolla SH, et al. Long-term survival of patients with left ventricular noncompaction. J Am Heart Assoc 2021; 10: e015563, doi:10.1161/JAHA.119.015563.
• 11.   Levine YC, Rosenberg MA, Mittleman M, Samuel M, Methachittiphan N, Link M, et al. B-Type natriuretic peptide is a major predictor of ventricular tachyarrhythmias. Heart Rhythm 2014; 11: 1109–1116, doi:10.1016/j.hrthm.2014.04.024.
• 12.   Ning X, Yang Z, Ye X, Si Y, Wang F, Zhang X, et al. Big endothelin-1 as a clinical marker for ventricular tachyarrhythmias in patients with post-infarction left ventricular aneurysm. Anatol J Cardiol 2019; 22: 256–261, doi:10.14744/AnatolJCardiol.2019.67862.
• 13.   Li XY, Zhao S, Fan XH, Chen KP, Hua W, Liu ZM, et al. Plasma big endothelin-1 is an effective predictor for ventricular arrythmias and end-stage events in primary prevention implantable cardioverter defibrillator indication patients. J Geriatr Cardiol 2020; 17: 427–433, doi:10.11909/j.issn.1671-5411.2020.07.003.
• 14.   Blangy H, Sadoul N, Dousset B, Radauceanu A, Fay R, Aliot E, et al. Serum BNP, hs-C-reactive protein, procollagen to assess the risk of ventricular tachycardia in ICD recipients after myocardial infarction. Europace 2007; 9: 724–729, doi:10.1093/europace/eum102.
• 15.   Theuns DA, Smith T, Szili-Torok T, Muskens-Heemskerk A, Janse P, Jordaens L. Prognostic role of high-sensitivity C-reactive protein and B-type natriuretic peptide in implantable cardioverter-defibrillator patients. Pacing Clin Electrophysiol 2012; 35: 275–282, doi:10.1111/j.1540-8159.2011.03289.x.
• 16.   Hu X, Fu S, Wang S. Hyperuricemia is associated with an increased prevalence of ventricular tachycardia and fibrillation in patients with ST-elevation myocardial infarction after primary percutaneous coronary intervention. BMC Cardiovasc Disord 2022; 22: 199, doi:10.1186/s12872-022-02635-4.
• 17.   von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. Lancet 2007; 370: 1453–1457, doi:10.1016/S0140-6736(07)61602-X.
• 18.   Jenni R, Oechslin E, Schneider J, Attenhofer Jost C, Kaufmann PA. Echocardiographic and pathoanatomical characteristics of isolated left ventricular noncompaction: A step towards classification as a distinct cardiomyopathy. Heart 2001; 86: 666–671, doi:10.1136/heart.86.6.666.
• 19.   Aung N, Doimo S, Ricci F, Sanghvi MM, Pedrosa C, Woodbridge SP, et al. Prognostic significance of left ventricular noncompaction: Systematic review and meta-analysis of observational studies. Circ Cardiovasc Imaging 2020; 13: e009712, doi:10.1161/CIRCIMAGING.119.009712.
• 20.   Steffel J, Hürlimann D, Namdar M, Despotovic D, Kobza R, Wolber T, et al. Long-term follow-up of patients with isolated left ventricular noncompaction: Role of electrocardiography in predicting poor outcome. Circ J 2011; 75: 1728–1734, doi:10.1253/circj.CJ-10-1217.
• 21.   Tian T, Liu Y, Gao L, Wang J, Sun K, Zou Y, et al. Isolated left ventricular noncompaction: Clinical profile and prognosis in 106 adult patients. Heart Vessels 2014; 29: 645–652, doi:10.1007/s00380-013-0409-z.
• 22.   Andreini D, Pontone G, Bogaert J, Roghi A, Barison A, Schwitter J, et al. Long-term prognostic value of cardiac magnetic resonance in left ventricle noncompaction: A prospective multicenter study. J Am Coll Cardiol 2016; 68: 2166–2181, doi:10.1016/j.jacc.2016.08.053.
• 23.   Sedaghat-Hamedani F, Haas J, Zhu F, Geier C, Kayvanpour E, Liss M, et al. Clinical genetics and outcome of left ventricular non-compaction cardiomyopathy. Eur Heart J 2017; 38: 3449–3460, doi:10.1093/eurheartj/ehx545.
• 24.   Li S, Zhang C, Liu N, Bai H, Hou C, Wang J, et al. Genotype-positive status is associated with poor prognoses in patients with left ventricular noncompaction cardiomyopathy. J Am Heart Assoc 2018; 7: e009910, doi:10.1161/JAHA.118.009910.
• 25.   Jankowich M, Choudhary G. Endothelin-1 levels and cardiovascular events. Trends Cardiovasc Med 2020; 30: 1–8, doi:10.1016/j.tcm.2019.01.007.
• 26.   Yorikane R, Shiga H, Miyake S, Koike H. Evidence for direct arrhythmic action of endothelin. Biochem Biophys Res Commun 1990; 173: 457–462, doi:10.1016/s0006-291x(05)81080-0.
• 27.   Liu T, Qin M, Shi SB, Chen Z, Wang T, Huang CX. Effects of endothelin-1 chronic stimulation on electrical restitution, beat-to-beat variability of repolarization, and ventricular arrhythmogenesis. J Cardiovasc Pharmacol 2013; 62: 549–558, doi:10.1097/FJC.0000000000000015.
• 28.   Fan P, Zhang Y, Lu YT, Yang KQ, Lu PP, Zhang QY, et al. Prognostic value of plasma big endothelin-1 in left ventricular non-compaction cardiomyopathy. Heart 2021; 107: 836–841, doi:10.1136/heartjnl-2020-317059.
• 29.   Sohns C, Ouyang F, Volkmer M, Metzner A, Nürnberg JH, Ventura R, et al. Therapy of ventricular arrhythmias in patients suffering from isolated left ventricular non-compaction cardiomyopathy. Europace 2019; 21: 961–969, doi:10.1093/europace/euz016.
• 30.   Tumolo AZ, Nguyen DT. Spectrum of cardiac arrhythmias in isolated ventricular noncompaction. J Innov Card Rhythm Manag 2017; 8: 2774–2783, doi:10.19102/icrm.2017.080701.
• 31.   Muser D, Liang JJ, Witschey WR, Pathak RK, Castro S, Magnani S, et al. Ventricular arrhythmias associated with left ventricular noncompaction: Electrophysiologic characteristics, mapping, and ablation. Heart Rhythm 2017; 14: 166–175, doi:10.1016/j.hrthm.2016.11.014.
• 32.   Amzulescu MS, Rousseau MF, Ahn SA, Boileau L, de Meester de Ravenstein C, Vancraeynest D, et al. Prognostic impact of hypertrabeculation and noncompaction phenotype in dilated cardiomyopathy: A CMR study. JACC Cardiovasc Imaging 2015; 8: 934–946, doi:10.1016/j.jcmg.2015.04.015.
• 33.   Casas G, Limeres J, Oristrell G, Gutierrez-Garcia L, Andreini D, Borregan M, et al. Clinical risk prediction in patients with left ventricular myocardial noncompaction. J Am Coll Cardiol 2021; 78: 643–662, doi:10.1016/j.jacc.2021.06.016.