Right ventricular function, exercise capacity and quality of life in adults with congenital heart disease and the effect of a home-based exercise program Authors: Inne Vanreusel, Daniel J Bowen, Vicky Van Der Meiren, Jacob J Briedé, Annemien E van den Bosch, Bernard P Paelinck, Emeline Van Craenenbroeck, Vincent FM Segers, Wendy Hens, An Van Berendoncks Published: 2026 ABSTRACT AbstractBackgroundRight ventricular (RV) function is crucial in the prognosis of adults with congenital heart disease (ACHD). However, its relationship with exercise capacity (EC) and quality of life (QoL) in ACHD remains underexplored. Despite the established benefits of physical activity in CHD, there is limited knowledge regarding the cardiac effects of exercise training in ACHD.AimTo assess the relationship between RV function, comprehensively evaluated using two-dimensional multiplane echocardiography (2D MPE), and both EC and QoL in adults with various types of CHD. Additionally, to investigate the effect of a 16-week home-based aerobic and strength exercise program on these parameters.MethodsFifty-five ACHD underwent transthoracic echocardiographic assessment of RV function, including conventional parameters and longitudinal RV strain, both measured from four RV walls using 2D MPE. The patients also completed a cardiopulmonary exercise test and a QoL questionnaire, and were randomized to either conventional care or a 16-week home-based rehabilitation program combining aerobic interval cycle training and dynamic strength exercises, followed by retesting.ResultsAt baseline, RV function correlated with EC, measured by percent-predicted peak oxygen consumption. Moreover, significant correlations were observed between RV and QoL parameters, including physical functioning, general health perception and visual analogue scale score. EC, muscle strength and QoL improved following the exercise program, while 39% of patients in the intervention group were non-adherent.ConclusionRV function correlates with EC and QoL in a diverse ACHD population. ACHD patients should be encouraged to engage in regular exercise, and exercise training should be integrated into CHD treatment.Keywords: Congenital heart disease, Right ventricular multiplane echocardiography, Multiview strain, Exercise capacity, Quality of life, Exercise training ABBREVIATIONS1RMone-repetition maximum2D MPEtwo-dimensional multiplane echocardiography4Cfour chamber (view)ACE Iangiotensin-converting enzyme inhibitorACHDadults with congenital heart diseaseAoaortic (view)ARBangiotensin receptor blockerAvgaverageBPblood pressureCHDcongenital heart diseaseCPETcardiopulmonary exercise testingCScoronary sinus (view)CVcoronal viewECexercise capacityFACfractional area changeGLSglobal longitudinal strainHBASEhome-based aerobic and strength exerciseHRheart ratehs-CRPhigh sensitive C-reactive proteinLSlongitudinal strainLV(EF)left ventricle/ventricular (ejection fraction)MWAmultiwall averageNT-proBNPN-terminal-pro hormone B-type natriuretic peptideNYHANew York Heart AssociationPETCO2fraction of carbon dioxide in the expired airPETO2fraction of oxygen in the expired air(%p)pVO2(percent-predicted) peak oxygen consumptionQoLquality of lifeRERrespiratory exchange ratioRVright ventricle/ventricularTAPSEtricuspid annular plane systolic excursionTDI S′tissue Doppler imaging systolic velocity(EQ-)VAS(EuroQol) visual analogue scaleVE/VCO2minute ventilation/carbon dioxide productionVE/VO2minute ventilation/oxygen consumptionVT1/2first/second ventilatory thresholdWattmaxmaximal wattageWLworkload ABBREVIATIONS Introduction Nowadays, the majority of individuals with congenital heart disease (CHD) survive into adulthood, necessitating lifelong management [1]. Exercise intolerance is a dominant symptom in adults with CHD (ACHD) [1]. These patients exhibit reduced exercise capacity (EC) comparable to patients with heart failure due to other etiologies, even in those with simple lesions or who consider themselves asymptomatic [2,3]. Diminished EC in ACHD is linked to higher risks of hospitalization and mortality [3]. While the underlying CHD contributes to exercise limitations, deconditioning and a sedentary lifestyle also play significant roles [4]. Recent evidence indicates that approximately 54% of ACHD patients report a lack of regular exercise [5]. Unfortunately, patients are often encouraged towards a sedentary lifestyle due to ignorance or uncertainty, or overprotection from family and healthcare professionals. Regular exercise benefits physical fitness, diminishes future risk of developing acquired heart disease, stimulates social interaction and enhances psychological well-being in CHD patients [1]. Despite growing research on quality of life (QoL) in CHD [6], factors significantly affecting self-reported health status in ACHD are still inadequately explored [7]. In CHD, the right ventricle (RV) often experiences pressure and/or volume overload, depending on the specific anomaly [8]. Continuous assessment and protection of the RV are crucial for prognosis, complementing left ventricle (LV) assessment [1,2,9]. While the relationship between RV function and EC is well-studied in other conditions [[10], [11], [12], [13]], it has received limited attention in CHD. Nevertheless, existing research suggests a nuanced interplay between RV function, EC, and QoL in CHD that merits deeper investigation. Unfortunately, quantifying RV function remains challenging [14], as established parameters like tricuspid annular plane systolic excursion (TAPSE), fractional area change (FAC) and tissue Doppler imaging systolic velocity (TDI S′) only reflect a small region of the RV [8]. Recently, two-dimensional multiplane echocardiography (2D MPE), which visualizes different RV walls from a single apical view, has emerged as a reproducible, easily applicable and comprehensive method for RV assessment in ACHD [8]. To date, evidence indicates that most children and adults with CHD can safely benefit physically and psychologically from exercise, despite diverse training regimens [4,15]. However, few studies have investigated the direct effects of exercise training on the heart [15], specifically the RV. Moreover, exercise programs are often supervised and rehabilitation center-based, which poses time and transportation barriers for this younger patient population. Home-based rehabilitation offers flexibility in time, location, and choice of exercise partner, and existing literature supports its safety, feasibility, and benefits across all age groups with CHD [16]. The aim of this explorative study is to assess the relationship between RV function, comprehensively assessed via 2D MPE, and EC as well as QoL in a heterogeneous ACHD population. Additionally, the impact of a 16-week home-based aerobic and strength exercise (HBASE) program on these parameters is investigated. Methods Study population Between 2021 and 2023, 55 adult (age range 18–65 years) consecutive CHD patients, with New York Heart Association (NYHA) class I–II and able to comply with exercise training, were prospectively enrolled at the out-patient clinic of the Antwerp University Hospital. Exclusion criteria were the following: professional endurance athlete, body mass index >35 kg/m2, smoking, the presence of macrovascular coronary artery disease, diabetes mellitus or a systemic disease (e.g. malignancies, acute and chronic inflammatory diseases in the preceding 3 months). Study protocol At baseline, all patients underwent fasting blood sampling and transthoracic Doppler echocardiography including a comprehensive RV assessment, completed a QoL questionnaire and performed cardiopulmonary exercise testing along with strength measurements of 4 major muscle groups. By the end of the first visit, patients were randomized (using Qminim [17] with age and gender as stratification parameters) to receive either conventional care or HBASE training. Following a 16-week period, all patients revisited the hospital and the same assessments were repeated. The study adhered to the principles of the Declaration of Helsinki and the Research and Ethics committee of the Antwerp University Hospital approved the study protocol (Belgian number: B3002020000298). Written informed consent was obtained from all subjects prior to any testing. Fasting blood sampling Fasting peripheral blood was collected using serum and heparin tubes (BD Vacutainer®, Canada). Serum samples were analyzed using an Atellica® IM/CH Analyzer (Siemens Healthcare, Germany) to quantify N-terminal-pro hormone B-type natriuretic peptide (NT-proBNP) concentrations and high sensitive C-reactive protein (hs-CRP). Superoxide anion radical levels in whole blood were determined by electron paramagnetic/spin resonance spectroscopy as previously described [18]. Transthoracic Doppler echocardiography Echocardiography acquisition All echocardiographic examinations were performed by two experienced sonographers (AVB and BP) utilizing harmonic imaging on an EPIQ7 ultrasound system (Philips Medical Systems, Best, the Netherlands) equipped with an X5-1 matrix-array transducer (composed of 3040 elements with 1–5 MHz). Patients were positioned in the left lateral decubitus position for a detailed 2D transthoracic echocardiogram. The RV was comprehensively assessed using the novel 2D MPE approach [8]. Initially, a focused, non-foreshortened RV view was obtained with the RV apex and interventricular septum centered along or as close to the midline of the imaging sector as possible, displaying the lateral RV wall (4C, 0°). From this position, the following images were acquired using electronic plane rotation and identification of anatomic landmarks (Fig. 1): the coronary sinus view (CS, ± +40°) showing the anterior RV wall, the aortic view (Ao, ± −40°) revealing the RV inferior wall, and the coronal view (CV, ± −90°) displaying the RV outflow tract and RV inferior wall. All images were acquired at frame rates of at least 50 frames/s for optimal longitudinal strain analysis. The terms "LV" and "RV" refer to the morphological LV and RV, respectively. Multiplane imaging of the RV Views acquired by electronic plane rotation from a single RV focused apical echocardiographic position: the RV focused apical four chamber view visualising the RV lateral wall (0° rotation), the coronary sinus view visualising the RV anterior wall (±+40° rotation), the aortic view visualising the RV inferior wall (±−40° rotation) and the coronal view (±−90° rotation) visualising the RV outflow tract and inferior wall of the RV. • Abbreviations: 4C = four chamber view, Ao = aortic view, CS = coronary sinus view, CV = coronal view, RV = right ventricle. Echocardiographic measurements To assess LV systolic function, LV ejection fraction (LVEF) was measured using the Simpson biplane method and global longitudinal strain (LV GLS) was calculated by combining 3 apical longitudinal LV chamber views (2-, 3- and 4-chamber). RV function was assessed by RV FAC (calculated as: (end-diastolic—end-systolic area)/end-diastolic area x100), RV TAPSE (measured with 2D echocardiography-guided M-mode) and RV TDI S′ recorded in all 4 RV views. A multiwall average (MWA) was computed when measurements from ≥3 walls of one individual were feasible. The 3 LV views and 4 RV views were digitally exported to Enterprise Imaging (version 8.2.0.160, Agfa HealthCare, Mortsel, Belgium) using TomTec software (version TTA2.51.00_562355, TomTec Imaging Systems, Munich, Germany). Data analysis was performed offline by two observers (IV and VVDM) using DICOM greyscale images. Peak systolic global longitudinal RV strain (RV LS) was assessed in 3 RV views (i.e., 4C, CS, Ao) using an automated strain software (AutoStrain; Philips Medical Systems, Best, the Netherlands). The software automatically defined the contours, with manual correction of endocardial tracing at end-diastole and end-systole if needed. Measurements were considered feasible if all portions of the RV free wall tracked accurately throughout the cardiac cycle. In each RV view, a single RV free wall strain value was derived from the average of 3 segments (basal, mid and apical)(Fig. 2). A MWA was calculated if ≥ 2 RV free walls were feasible to measure. RV LS data analysis was performed offline by two observers (IV and DB). Intra- and inter-observer variability test for RV LS was not performed as this has been reported as excellent in previous articles [8,19,20]. Multiplane imaging of the RV free wall LS assessment Abbreviations: 4C = four chamber view, Ao = aortic view, avg = average, CS = coronary sinus view, LS = longitudinal strain, RV = right ventricle. QoL questionnaire All patients completed the RAND 36-Item Health Survey 1.0, covering 8 health domains and a single item for assessing perceived health change [21]. Additionally, they filled out the EuroQol visual analogue scale (EQ-VAS), which measures current health state on a scale from 0 (worst imaginable) to 100 (best imaginable) [22]. Cardiopulmonary exercise testing A cardiopulmonary exercise test was conducted until exhaustion as previously described [18]. Peak oxygen consumption (pVO2) was determined as the mean VO2 during the final 30 s of exercise. Subsequently, percent-predicted peak oxygen consumption (%ppVO2) was computed, using the reference values following Wasserman et al. [23] and the LowLands registry [24]. Peak work rate was defined as the maximal wattage (Wattmax) achieved during the exercise. The minute ventilation/carbon dioxide production (VE/VCO2) slope was calculated using all data points from loaded cycling. The first and second ventilatory thresholds (VT1 and VT2) were determined according to scientific standards (Fig. S1 in supplementary materials) [25]. The same protocol used for the initial test was employed when retesting the patients. Muscle strength measurements The one-repetition maximum (1RM) was measured on fitness machines (BioCircuit Series 4, TechnoGym®, Benelux) for 4 different exercises (chest press, vertical traction, low row and leg press) targeting major muscle groups. Conventional care In the conventional care arm, patients did not receive explicit exercise advice but were provided with general information regarding a healthy lifestyle, in accordance with current physical activity guidelines [26,27] during the first study visit. HBASE program Participants in the exercise arm received a personalized rehabilitation schedule combining aerobic interval cycling and dynamic strength exercises, developed to meet physical activity guidelines [26,27]. Aerobic training consisted of cycling on an exercise bike at home for 30-45 min, 3 times a week. The heart rate (HR) was continuously monitored using a Polar® chest strap H9 and displayed in the Polar® Beat smartphone application. Training included intervals targeting 90% and 75% of the HR at VT2, aiming for 40 min of intensive aerobic exercise by the end of 16 weeks (Table S1 in supplementary materials). Strength training involved 4 dynamic exercises targeting major muscle groups, performed thrice weekly with progressive intensity adjustments every 2 weeks (Table S2 in supplementary materials). Participants were instructed to avoid the Valsalva maneuver during all exercises. The first session took place at the hospital, discussing the schedule, installing equipment and conducting one supervised training to ensure that the rest of the training could easily take place at home. Exercise data were uploaded to the participant's Polar® Beat account and supervised by a physiotherapist of the Cardiac Rehabilitation Centre. Adherence was defined as completing ≥70% of the prescribed activities. Participants received regular phone calls: 2 weeks after starting and monthly thereafter, with increased frequency if any problems arose. Statistical analysis Statistical analysis was performed using SPSS version 29.0 (IBM Corp. 2023, NY). Normality of continuous variables was evaluated using histograms, Q-Q plots and the Shapiro-Wilk test. Data are presented as mean ± standard deviation or median (Q1-Q3) depending on the normality of the data. If the lab results showed a 'less than' result (NT-proBNP <35), this was replaced by the number (i.e., 35) and considered as non-normally distributed for the entire study group. Pearson and Spearman correlation coefficient were used for univariable correlation analysis, contingent upon the normality of the data. Correlations between the echocardiographic measurements of LV and RV function, on the one hand, and %ppVO2 and QoL questionnaire scores, on the other hand, were investigated. To assess the effect of the training program, repeated measures ANOVA was employed, with group (conventional care vs. exercise training) as the between-subjects factor and time (baseline vs. follow-up) as the within-subjects factor for normally distributed values. For non-normally distributed variables, the change in outcome score from baseline to follow-up was computed separately for both groups (conventional care vs. exercise training) using the paired Wilcoxon signed-rank test. Subsequently, a comparison between the two groups was conducted using the independent samples Mann Whitney U test. A two-tailed p < 0.05 was considered significant. Results Patient population Fifty-five ACHD (median (Q1-Q3) age 31 (24-42) years), 49% male) were included in this study, representing various types of CHD (Table S3A in supplementary materials). According to the ESC classification of CHD complexity [1] 23 patients (42%) had simple CHD, 24 (44%) had moderate complexity CHD and 8 (14%) had great complexity CHD (Table S3B in the supplementary materials). Thirty patients (55%) had a surgical history for CHD, while 8 patients (15%) received exclusively percutaneous treatment. The remaining 17 patients were managed clinically without prior cardiac interventions. Echocardiographic measurements Three patients (2 with congenitally corrected transposition of the great arteries, one of whom also presented with atrial and ventricular septal defects along with subvalvular pulmonary stenosis, and 1 patient with a bicuspid aortic valve and Turner syndrome) were excluded from the echocardiographic assessment due to poor image quality or the inability to visualize the entire RV. Of the remaining 52 ACHD patients, 50 (96%) had sufficient acoustic windows to analyze ≥3 RV views. The feasibility rates of RV FAC, TAPSE and TDI S’ measurements ranged from 85% to 100%. The feasibility of RV LS measurement was 81%, 64% and 77% for the lateral, anterior, and inferior walls, respectively. The parameters of LV and RV function are displayed in Table 1. Reduced ventricular function, characterized by a LVEF <50% and a RV FAC <35% (assessed on a conventional 4C view), was observed in 8 (15%) and 11 (21%) out of 52 patients, respectively. Table 1Feasibility rates and values of the echocardiographic measurements of LV and RV function in ACHD.Table 1Measurement feasibility (n (%))ACHD (n = 52)LVEF (%)52 (100%)58 (52-62)LV GLS (%)44 (85%)−19 ± 3RV FAC (%)Lateral52 (100%)43 (37-49)Anterior47 (90%)41 ± 8Inferior49 (94%)43 ± 9Inferior CV44 (85%)40 ± 8- > MWA49 (94%)44 (38-46)RV TAPSE (mm)Lateral51 (98%)21 (16-25)Anterior50 (96%)21 (16-26)Inferior50 (96%)20 (16-24)Inferior CV49 (94%)20 ± 5- > MWA50 (96%)20 (16-25)RV TDI S’ (cm/s)Lateral50 (96%)10.7 ± 2.8Anterior50 (96%)10.6 ± 3.0Inferior50 (96%)10.4 ± 2.6Inferior CV49 (94%)9.7 ± 2.4- > MWA50 (96%)10.4 ± 2.6RV free wall LS (%)Lateral42 (81%)−24 ± 6Anterior33 (64%)−25 ± 7Inferior40 (77%)−23 ± 6- > MWA40 (77%)−24 ± 6Data are expressed as mean ± standard deviation, median (Q1-Q3) or n (%). • Abbreviations: ACHD = adults with congenital heart disease, CV = coronal view, FAC = fractional area change, GLS = global longitudinal strain, LS = longitudinal strain, LV = left ventricle, LVEF = left ventricular ejection fraction, MWA = multiwall average, RV = right ventricle, TAPSE = tricuspid annular plane systolic excursion, TDI S’ = tissue Doppler imaging systolic velocity. Feasibility rates and values of the echocardiographic measurements of LV and RV function in ACHD. Data are expressed as mean ± standard deviation, median (Q1-Q3) or n (%). • Abbreviations: ACHD = adults with congenital heart disease, CV = coronal view, FAC = fractional area change, GLS = global longitudinal strain, LS = longitudinal strain, LV = left ventricle, LVEF = left ventricular ejection fraction, MWA = multiwall average, RV = right ventricle, TAPSE = tricuspid annular plane systolic excursion, TDI S’ = tissue Doppler imaging systolic velocity. Correlations between ventricular function and EC A reduced EC (%ppVO2 <85%) was found in 27 of 52 patients (52%) included in the baseline analysis following the predicted values from Wasserman et al. [23](%ppVO2(Wasserman)), and in 45 patients (87%) using the LowLands values [24](%ppVO2(LowLands)). Moreover, there were several significant correlations between EC and both LV and RV function (Table 2). Concerning the LV, GLS correlated negatively with EC, but this correlation was only significant for %ppVO2(Wasserman). Conversely, there was no significant correlation between LVEF and EC. With regard to the RV, the MWA of RV free wall LS exhibited significant negative correlations with both %ppVO2(Wasserman) and %ppVO2(Lowlands) (Fig. 3A). There was also a significant but positive correlation between %ppVO2(Wasserman) and the MWA of TDI S′ and TAPSE, as well as between %ppVO2(Lowlands) and the MWA of TDI S’. The MWA of TAPSE also correlated positively with %ppVO2(Lowlands), although this correlation was borderline significant. Similarly, the correlations between the MWA of RV FAC and both %ppVO2 values were only borderline significant. Importantly, negative values were included in the analysis for strain measurements. Therefore, negative correlations imply that better (and thus more negative) strain values are associated with better EC. A subanalysis, which exclusively incorporated data from patients who performed a maximal metabolic exercise test, defined as having reached a respiratory exchange ratio ≥1.05, yielded results that were generally consistent with those of the overall analysis (Table S5 in supplementary materials). Table 2Correlations between echocardiographic measurements of LV and RV function and EC as well as QoL scores in ACHD.Table 2LVEFa (%)LV GLS (%)RV FAC (%) MWAaRV TAPSE (mm) MWAaRV TDI S’ (cm/s) MWARV free wall LS (%) MWACCpCCpCCpCCpCCpCCpExercise capacity%ppVO2(Lowlands)0.030.85−0.220.160.260.070.270.060.37∗∗0.008−0.36∗0.02%ppVO2(Wasserman)0.110.42−0.35∗0.020.280.060.39∗∗0.0050.47∗∗<0.001−0.36∗0.03Quality of lifePhysical functioninga−0.110.460.110.460.31∗0.030.250.080.200.16−0.38∗0.02Bodily paina−0.100.470.300.050.0060.970.060.700.0070.96−0.090.60Role limitations (physical)a−0.190.190.190.230.110.470.0070.960.040.78−0.080.61Role limitations (emotional)a0.130.370.31∗0.040.080.590.080.60−0.030.860.0040.98General mental healtha0.0080.950.090.56−0.010.95−0.030.83−0.140.32−0.090.56Social functioninga−0.080.590.100.530.160.290.040.81−0.050.74−0.150.35Energy/fatiguea0.080.590.190.220.080.580.060.680.030.84−0.090.57General health perceptiona−0.060.660.0020.990.41∗∗0.0040.200.160.28∗0.05−0.53∗∗<0.001Change in healtha0.050.70−0.110.47−0.080.61−0.050.740.080.61−0.080.64VAS scorea−0.0020.99−0.200.190.31∗0.030.36∗∗0.0090.37∗∗0.009−0.37∗0.02aNon-parametric correlations, ∗/∗∗ Correlation is significant at the 0.05/0.01 level (2-tailed). • Abbreviations: %ppVO2 = percent-predicted peak oxygen consumption, ACHD = adults with congenital heart disease, CC = correlation coefficient, EC = exercise capacity, FAC = fractional area change, GLS = global longitudinal strain, LS = longitudinal strain, LV = left ventricle, LVEF = left ventricular ejection fraction, MWA = multiwall average, QoL = quality of life, RV = right ventricle, TAPSE = tricuspid annular plane systolic excursion, TDI S’ = tissue Doppler imaging systolic velocity, VAS = visual analogue scale. • Correlations between RV function measurements in the different RV views and EC as well as QoL scores are displayed in Tables S4 and S6 in supplementary materials. Correlations between echocardiographic measurements of LV and RV function and EC as well as QoL scores in ACHD. Non-parametric correlations, ∗/∗∗ Correlation is significant at the 0.05/0.01 level (2-tailed). • Abbreviations: %ppVO2 = percent-predicted peak oxygen consumption, ACHD = adults with congenital heart disease, CC = correlation coefficient, EC = exercise capacity, FAC = fractional area change, GLS = global longitudinal strain, LS = longitudinal strain, LV = left ventricle, LVEF = left ventricular ejection fraction, MWA = multiwall average, QoL = quality of life, RV = right ventricle, TAPSE = tricuspid annular plane systolic excursion, TDI S’ = tissue Doppler imaging systolic velocity, VAS = visual analogue scale. • Correlations between RV function measurements in the different RV views and EC as well as QoL scores are displayed in Tables S4 and S6 in supplementary materials. Correlations between RV function, EC and QoL in ACHD Scatterplots with restricted cubic spline curves illustrating the relationship between the MWA of RV free-wall LS and (A) %ppVO2(Lowlands), (B) physical functioning, (C) general health perception and (D) VAS score. Correlation coefficients and p-values are shown within each panel. • Abbreviations: %ppVO2 = percent-predicted peak oxygen consumption, ACHD = adults with congenital heart disease, EC = exercise capacity, LS = longitudinal strain, MWA = multiwall average, QoL = quality of life, RV = right ventricle, VAS = visual analogue scale. Correlations between ventricular function and QoL Several significant positive correlations have been observed between the MWA of RV FAC, TAPSE and TDI S’, on the one hand, and physical functioning, general health perception and VAS score, on the other hand (Table 2). Likewise, the same QoL parameters exhibited significant, albeit negative, correlations with the MWA of RV free wall LS (Fig. 3B–D). Again, a negative correlation implies that better (and thus more negative) strain values are associated with higher QoL scores. Correlations between ventricular function and hematological parameters None of the echocardiographic measurements of ventricular function showed a significant correlation with NT-proBNP, hs-CRP or superoxide anion radical levels (Table S7 in supplementary materials). Effect of the HBASE program Twenty-four patients were randomized to the conventional care arm and 31 patients were assigned to exercise training. The baseline characteristics of both groups are depicted in Table 3. Due to a drop-out of 5 patients, follow-up investigations were conducted in 22 patients from the conventional care group and in 28 patients from the exercise training group. However, 11 patients in the exercise group (39%) did not adhere to the training program (Fig. 4). Table 3Baseline characteristics of the total group, conventional care arm and exercise training arm.Table 3Total group n = 55Conventional care n = 24Exercise training n = 31Age (years)31 (24-42)32 (24-44)31 (24-40)Gender (male/female)27/2813/1114/17Body mass index (kg/m2)24.72 ± 3.9925.25 ± 4.4224.32 ± 3.64Systolic BP (mmHg)115 (107-125)116 (107-129)115 (106-117)Diastolic BP (mmHg)73 (70-77)72 (70-77)73 (70-77)History of:• Surgery 30 (55%)11 (46%)19 (61%)• Percutaneous treatment 8 (15%)5 (21%)3 (10%)Medication:• Antiplatelet/anticoagulant 14 (25%)8 (33%)6 (19%)• ACE I/ARB 11 (20%)6 (25%)5 (16%)• Beta blocker 15 (27%)7 (29%)8 (26%)• Calcium channel blocker 3 (5%)2 (8%)1 (3%)• Diuretics 1 (2%)0 (0%)1 (3%)• Statin 1 (2%)1 (4%)0 (0%)Data are expressed as mean ± standard deviation, median (Q1-Q3) or n (%). • Abbreviations: ACE I = angiotensin-converting enzyme inhibitor, ARB = angiotensin receptor blocker, BP = blood pressure. Baseline characteristics of the total group, conventional care arm and exercise training arm. • Surgery Surgery • Percutaneous treatment Percutaneous treatment • Antiplatelet/anticoagulant Antiplatelet/anticoagulant • ACE I/ARB ACE I/ARB • Beta blocker Beta blocker • Calcium channel blocker Calcium channel blocker • Diuretics Diuretics • Statin Statin Data are expressed as mean ± standard deviation, median (Q1-Q3) or n (%). • Abbreviations: ACE I = angiotensin-converting enzyme inhibitor, ARB = angiotensin receptor blocker, BP = blood pressure. Overview of the study protocol Abbreviations: RV = right ventricle. Table 4 presents the results of the per-protocol analysis. In the patients who adhered to the training schedule (n = 17), EC measured by pVO2 (Fig. 5A) and Wattmax significantly improved. This was not the case in the conventional care arm (n = 22). Additionally, the difference over time was significantly larger in the exercise group. Likewise, neither the ventilatory efficiency measured by VE/VCO2 nor the oxygen consumption (VO2) at VT1 and VT2 improved in the conventional care arm. However, there was a significant improvement in VO2 at VT1 and VT2 in the exercise group, and the evolution over the course of the study was significantly larger in the exercise group for VE/VCO2 and VO2 at VT1. A subanalysis in the patients who performed a maximal metabolic exercise test (respiratory exchange ratio ≥1.05) on both study visits yielded similar results (Table S8 in supplementary materials). Furthermore, the 1RM for chest press, low row and leg press significantly improved in patients who adhered to the training schedule; however, the progression over the study period was only significantly greater in the exercise training group for chest press. Table 4Effects of the exercise program on EC, muscle strength, QoL and ventricular function in ACHD: per-protocol analysis.Table 4Conventional care n = 22Exercise training n = 17P (CC-ET)BaselineFollow-upP (B-FU)BaselineFollow-upP (B-FU)Exercise capacitypVO2 (ml/kg/min)26.65 ± 6.8327.61 ± 6.720.1728.70 ± 6.8532.75 ± 8.24<0.001∗∗0.005∗∗Wattmax (W)177.00 (119.25-246.25)181.00 (137.75-238.00)0.72168.00 (150.00-210.50)186.00 (159.00-250.00)<0.001∗∗<0.001∗∗VE/VCO228.60 (24.58-31.55)27.65 (25.28-30.95)0.2229.70 (24.60-32.75)30.70 (25.45-33.05)0.140.005∗∗VO2@VT1 (ml/kg/min)17.10 (13.80-19.50)16.50 (12.20-19.05)0.93 (n=21)17.50 (13.05-19.70)19.70 (15.40-25.25)0.002∗∗0.003∗∗VO2@VT2 (ml/kg/min)24.11 ± 4.6824.37 ± 5.110.72 (n=18)23.96 ± 5.9526.16 ± 6.040.01∗(n=14)0.08Muscle strength11RM chest press (kg)57.00 (38.75-72.75)58.00 (43.58-70.75)0.46 (n=14)42.00 (31.90-61.00)48.00 (33.00-64.50)<0.001∗∗0.008∗∗1RM vertical traction (kg)55.14 ± 17.6256.47 ± 16.920.36 (n=14)48.03 ± 15.6249.45 ± 16.710.280.961RM low row (kg)62.00 (47.00-70.50)65.50 (52.68-72.00)0.33 (n=14)46.60 (38.50-71.00)50.70 (41.00-74.00)0.002∗∗0.711RM leg press (kg)171.00 (127.00-200.00)175.00 (149.00-215.00)0.22 (n=15)154.70 (119.00-197.00)164.40 (142.00-231.50)<0.001∗∗0.63Quality of lifePhysical functioning (%)90.00 (68.75-100)90.00 (78.75-100)0.1190.00 (70.00-100)90.00 (82.50-100)0.02∗0.27Bodily pain (%)100 (78.00-100)90.00 (68.00-100)0.6778.00 (74.00-100)90.00 (73.00-100)0.880.92Role limitations (physical) (%)100 (93.75-100)100 (87.50-100)0.58100 (87.50-100)100 (100-100)0.330.49Role limitations (emotional) (%)100 (91.75-100)100 (100-100)0.55100 (83.50-100)100 (100-100)0.130.75General mental health (%)78.00 (67.00-88.00)80.00 (69.00-85.00)0.5476.00 (68.00-80.00)76.00 (72.00-84.00)0.03∗0.13Social functioning (%)88.00 (72.00-100)94.00 (75.00-100)0.0688.00 (75.00-100)100 (100-100)0.01∗0.32Energy/fatigue (%)70.00 (55.00-80.00)70.00 (53.75-80.00)0.9060.00 (47.50-70.00)65.00 (55.00-77.50)0.02∗0.05∗General health perception (%)62.50 (55.00-75.00)67.50 (53.75-80)0.3965.00 (40.00-75.00)65.00 (45.00-75.00)0.531.00Change in health (%)50.00 (50.00-75.00)50.00 (50.00-75.00)0.9350.00 (50.00-50.00)50.00 (50.00-75.00)0.03∗0.14VAS score (%)80.00 (73.75-87.75)80.00 (75.00-85.00)0.4475.00 (65.00-85.00)80.00 (75.00-88.50)0.009∗∗0.03∗Ventricular functionLVEF (%)60 (52-64)58 (56-63)0.64 (n=20)57 (51-61)58 (53-61)0.61 (n=16)0.67LV GLS (%)−18 ± 3−19 ± 30.54 (n=17)−20 ± 4−20 ± 30.68 (n=11)0.66RV FAC MWA (%)44 (39-46)42 (40-45)0.63 (n=20)43 (34-46)43 (39-46)0.29 (n=14)0.40RV TAPSE MWA (mm)21 (17-25)22 (18-25)0.63 (n=20)22 (16-26)22 (18-27)0.14 (n=15)0.07RV TDI S′ MWA (cm/s)10.9 ± 2.710.6 ± 2.30.34 (n=20)11.0 ± 2.010.9 ± 2.10.71 (n=14)0.74RV free wall LS MWA (%)−24 ± 6−23 ± 60.38 (n=13)−28 ± 4−26 ± 60.12 (n=8)0.47Data are expressed as mean ± standard deviation or median (Q1-Q3). The mean/median values were only presented for patients for whom both baseline and follow-up values were available. ∗/∗∗ The significance level is 0.05/0.01 (2-tailed). 1 Not all maximal strength measurements could be conducted in all patients due to medical restrictions (e.g., aortic dilatation) or injury limiting articular range of motion. • Abbreviations: 1RM = 1 repetition maximum, ACHD = adults with congenital heart disease, B = baseline, CC = conventional care, EC = exercise capacity, ET = exercise training, FAC = fractional area change, FU = follow-up, GLS = global longitudinal strain, LS = longitudinal strain, LV = left ventricle, LVEF = left ventricular ejection fraction, MWA = multiwall average, pVO2 = peak oxygen consumption, QoL = quality of life, RV = right ventricle, TAPSE = tricuspid annular plane systolic excursion, TDI S’ = tissue Doppler imaging systolic velocity, VAS = visual analogue scale, VE/VCO2 = minute ventilation/carbon dioxide production, VO2@VT1/2 = oxygen consumption at the first/second ventilatory threshold, Wattmax = maximal wattage. • P(B-FU) = change in outcome score from baseline to follow-up per randomization group (CC or ET, respectively), P(CC-ET) = comparison of change in outcome scores, i.e., P(B-FU), between both groups (CC and ET). Effects of the exercise program on EC, muscle strength, QoL and ventricular function in ACHD: per-protocol analysis. Data are expressed as mean ± standard deviation or median (Q1-Q3). The mean/median values were only presented for patients for whom both baseline and follow-up values were available. ∗/∗∗ The significance level is 0.05/0.01 (2-tailed). 1 Not all maximal strength measurements could be conducted in all patients due to medical restrictions (e.g., aortic dilatation) or injury limiting articular range of motion. • Abbreviations: 1RM = 1 repetition maximum, ACHD = adults with congenital heart disease, B = baseline, CC = conventional care, EC = exercise capacity, ET = exercise training, FAC = fractional area change, FU = follow-up, GLS = global longitudinal strain, LS = longitudinal strain, LV = left ventricle, LVEF = left ventricular ejection fraction, MWA = multiwall average, pVO2 = peak oxygen consumption, QoL = quality of life, RV = right ventricle, TAPSE = tricuspid annular plane systolic excursion, TDI S’ = tissue Doppler imaging systolic velocity, VAS = visual analogue scale, VE/VCO2 = minute ventilation/carbon dioxide production, VO2@VT1/2 = oxygen consumption at the first/second ventilatory threshold, Wattmax = maximal wattage. • P(B-FU) = change in outcome score from baseline to follow-up per randomization group (CC or ET, respectively), P(CC-ET) = comparison of change in outcome scores, i.e., P(B-FU), between both groups (CC and ET). Effects of the exercise program on pVO2 (A), energy/fatigue (B) and VAS score (C) in ACHD: per-protocol analysis Data are expressed as mean ± standard deviation or median (Q1-Q3). The mean/median values were only presented for patients for whom both baseline and follow-up values were available. • Abbreviations: ACHD = adults with congenital heart disease, EC = exercise capacity, pVO2 = peak oxygen consumption, QoL = quality of life, VAS = visual analogue scale. With regard to the QoL scores, again, there were no significant improvements in the conventional care group. In contrast, several QoL parameters, namely physical functioning, general mental health, social functioning, energy/fatigue, perceived change in health and VAS score, improved in the exercise group. However, the improvement over the course of the study was only significantly larger in the exercise group for self-reported energy/fatigue (Fig. 5B) and VAS score (Fig. 5C). Concerning ventricular function, none of the parameters showed a significant increase in either group. However, the change over time was borderline significantly greater in the exercise group for the MWA of RV TAPSE. Among the hematological parameters (Table S9 in supplementary materials), superoxide anion radical levels significantly increased in the conventional care group (p = 0.02), but not in the exercise group (p = 0.27). Importantly, in both groups, there were no adverse events. The intention-to-treat analysis (Table S10 in supplementary materials) demonstrates significant improvements in pVO2, Wattmax, VO2 at VT2, 1RM for chest press, low row and leg press (p < 0.001 for all), as well as in VAS score (p = 0.05) in the patients assigned to exercise training, even if not fully adherent to the training regimen. Additionally, superoxide anion radical levels significantly increased in both groups (p = 0.02 for conventional care, p = 0.04 for exercise training) when non-adherent patients were included in the analysis. Discussion This exploratory study investigated the relationship between RV function, comprehensively assessed using 2D MPE, and both EC and QoL in a heterogeneous ACHD population. Additionally, the effects of a 16-week HBASE program were evaluated. In addition to the function of the more extensively studied LV, RV function was correlated with EC. Furthermore, several significant correlations were observed between the MWAs of the RV parameters and QoL, particularly physical functioning, general health perception and VAS score. In the patients who adhered to the training schedule, EC significantly improved together with several QoL parameters. The RV in CHD To the best of our knowledge, this study is the second to perform a comprehensive RV analysis with 2D MPE in a spectrum of ACHD patients, building upon previous research demonstrating its feasibility in healthy subjects [19] and ACHD [8,28]. This new technique offers advantages such as a short learning curve and limited additional acquisition time [8,28] and represents an attractive tool for both research and daily clinical practice [8]. In the current study, reduced LV and RV function were observed in 15% and 21%, respectively, with several significant correlations between EC and both LV and RV function. Notably, for the LV, only the correlation between GLS and %ppVO2(Wasserman) was significant. These findings underscore the clinical relevance of RV assessment in ACHD, particularly given the vulnerability of the RV to pressure and volume overload, and its unique susceptibility to oxidative stress and microvascular dysfunction. This also aligns with previous research in patients with heart failure, where EC has little relation to the extent of LV systolic dysfunction as measured by LVEF [29]. Regarding the RV, prior research has already demonstrated a notable decrease in RV function across a whole spectrum of ACHD [8]. Several other studies have also identified associations between RV function and EC. Specifically, reduced EC has been related to impaired RV function in patients with a systemic RV [[30], [31], [32], [33]], while RV free wall strain has been independently predictive of low EC in patients with repaired Tetralogy of Fallot [34]. Moreover, systemic and sub-pulmonic ventricular function, assessed semi-quantitatively with transthoracic echocardiography, determined EC in a heterogeneous group of adults with diverse types of CHD [3]. In patients with CHD-related pulmonary arterial hypertension, the ratio of TAPSE to pulmonary artery systolic pressure may serve as an echocardiographic indicator for evaluating exercise tolerance [35]. In the present study the observed correlations between RV function parameters and EC were generally small to moderate in magnitude. These findings support a potential role of ventricular function in determining exercise performance in ACHD, although they also suggest that exercise limitation is likely multifactorial. The precise mechanism behind RV dysfunction in ACHD remains incompletely understood [8,36]. In CHD, the RV often faces pressure and/or volume overload [8]. The thin-walled RV is particularly sensitive to elevations in afterload [37], initially responding with hypertrophy, but rapidly followed by dilation and ultimately failure [38]. Increased afterload and myocyte hypertrophy disrupt RV oxygen supply-demand balance, with microvascular dysfunction playing a significant contributory role [8,37]. Moreover, despite similarities in cellular remodeling under pressure overload between the RV and LV, the RV is more prone to oxidative stress [36,39,40], suggesting a distinct pathophysiological response that warrants further investigation. Interestingly, RV function parameters did not correlate with NT-proBNP, hs-CRP or superoxide anion radical levels in our cohort. This may be related to the relatively preserved functional status of the current study population, its heterogeneous composition and relatively small sample size, or to distinct and not yet fully elucidated pathophysiological pathways underlying RV dysfunction in ACHD. A deeper understanding of the pathophysiology of RV dysfunction in CHD could pave the way for the development of targeted preventive strategies and therapies tailored specifically to the RV. This is crucial because conventional heart failure treatments have proven ineffective for RV failure and have not improved survival rates in these patients [36,39]. Exercise training should also be considered alongside pharmaceutical treatments to address RV dysfunction in ACHD. Assessment of EC To assess correlations with EC, the %ppVO2 was calculated in the present study using reference values provided by Wasserman et al. [23] and the LowLands registry [24]. Since the globally utilized reference values by Wasserman et al. [23] are derived from an American cohort and are not applicable to female subjects, we also considered the reference values from the LowLands registry [24] for comparison. These values originate from a more recent Dutch and Flemish cohort that includes both men and women and are remarkably higher than other reference values. Consequently, these reference values can facilitate a more accurate interpretation of maximal oxygen uptake in the West-European population [24]. QoL in CHD In this study on a diverse ACHD population, RV function also correlated with self-reported physical functioning, general health perception and VAS score. These associations were generally small to moderate in magnitude, suggesting a potential but modest relationship between ventricular function and patient-reported health status. Patient-reported outcomes provide insight into overall well-being and disease burden and may contribute to clinical decision-making by offering a more holistic, patient-centered approach. Nevertheless, they are still not routinely incorporated as primary endpoints in large cardiovascular clinical trials [41]. Notably, in a previous study of ACHD patients with atrial arrhythmias, lower SF-36 physical component scores as well as higher mEHRA scores were associated with an increased risk of adverse clinical outcomes, thereby underscoring the prognostic value of health status measures in ACHD [42]. Over the past decades, the number of articles on QoL in CHD has increased substantially [6]. However, the available evidence remains heterogeneous and sometimes contradictory, making firm conclusions difficult due to conceptual issues and methodological differences between studies [6,43]. Overall, QoL in CHD patients is generally considered good [6]. For example, a meta-analysis reporting QoL as a single index score found no reduction in QoL among adolescents and young adults with CHD compared with healthy age-matched controls [44]. In contrast, a meta-analysis of studies using the SF-36 questionnaire demonstrated significantly lower physical functioning and general health perception scores in patients with moderate or complex CHD, whereas scores in patients with mild defects were comparable to those of controls [43]. These findings suggest that impairments in QoL are mainly driven by physical health domains and are more evident in patients with greater anatomical or functional disease complexity. Consistent with this, RV function in our cohort was associated with self-reported physical functioning and general health perception, suggesting a contribution of ventricular performance to the physical dimension of QoL in ACHD. Factors influencing QoL in CHD patients have been extensively studied [6], yet the factors having the greatest impact on self-reported health status remain insufficiently explored [7]. Lower academic performance and education, lower employment rates, history of cardiac surgery, fewer daily activities, poorer functional status, lower social support, being given an implantable cardioverter defibrillator, physical limitations and a distressed personality type are among the identified correlates of poorer QoL [6]. Additionally, newer factors like sense of coherence, general self-efficacy, empowerment, illness perception, personality traits, parental support, and feelings of loneliness have emerged as predictors of QoL in CHD [6]. In addition to our study, other research has explored the relationship between RV function and QoL in CHD. For instance, NT-proBNP levels alongside various clinical indicators of RV failure such as EC, have been found to correlate with QoL in grown-up patients with congenital right heart disease [45]. Furthermore, RV systolic dysfunction, assessed by magnetic resonance imaging, is linked to lower scores in physical functioning in individuals who underwent repair for Tetralogy of Fallot [46]. However, such a relationship with QoL appears absent in patients with mild pulmonary insufficiency following Tetralogy of Fallot repair or after balloon dilation for isolated valvar pulmonary stenosis [47]. The effect of exercise training in CHD With our training regimen, we observed improvements in EC, muscle strength and QoL. However, a positive effect on ventricular function was not evident. Current studies do not provide sufficient evidence to determine the optimal training intensity for CHD patients [15]. Two systematic reviews, one on exercise training in children and young adults with CHD [15] and another on home-based exercise interventions in CHD patients [16], indicate that most training programs last 12 weeks with (at least) 3 sessions per week. Our training program was slightly longer. Home-based exercise session in ACHD ranged from 5 to 150 min [16]. Training intensity typically relies on percentages of peak HR or HR at the anaerobic VT [15,16]. While initially relatively uncommon [15], interval training has been prescribed several times in CHD [16]. Consistent with our study, outcomes primarily focused on EC measured as peak oxygen uptake [16], which is widely regarded as the best measure of cardiorespiratory fitness and EC [24]. As is often the case in studies concerning CHD, most research has focused on children [15,16]. In clinical practice, it is recommended to tailor the nature and intensity of exercise training individually based on the type of CHD and the specific surgical correction undergone [4]. The primary objectives of exercise training for CHD patients are to enhance overall health, increase exercise tolerance and improve QoL [4]. Nevertheless, our study's baseline findings suggest that the impact of exercise training on RV function should be included as an outcome measure in exercise training studies. To our knowledge, this is the first time it is being investigated in CHD using 2D MPE. Our training regimen lead to enhancements in pVO2 (+4.05 ± 2.94 ml/kg/min, or +13.89 ± 8.35%), similar to those reported in other studies on home-based exercise interventions in CHD (+3.6 ml/kg/min (15%) [48]; +2.2 ml/kg/min (7%) [49]; +1.8 ml/kg/min (7.7%) [50]). The effect of exercise training on muscle strength and submaximal exercise measures in CHD has been examined in a restricted number of studies [15]. Those that were not underpowered showed a significant increase in muscle strength [15]. Additionally, in our study, oxygen consumption improved at both VT1 and VT2, indicating advancements in oxygen delivery and consumption, thereby shifting the threshold for anaerobic metabolism upwards [15]. Furthermore, ventilatory efficiency significantly improved in the exercise training group compared to the control group. Similar to findings in other studies [15,51], we observed no statistically significant change in serum NT-proBNP following exercise training. Additionally, hs-CRP did not show improvement in either group. An interesting observation, however, is the apparent increase in superoxide anion radical levels, except in the patient group adherent to the training schedule. The underlying reason for this rise in oxidative stress levels cannot be explained based on our study; however, our results allow speculation that our training regimen may trigger adaptive mechanisms protective against oxidative stress, a finding that warrants further exploration in future studies. Finally, we also observed significant improvements in several QoL parameters following exercise training. Evidence regarding the effect of exercise interventions on QoL in CHD remains inconsistent. A systematic review of home-based exercise interventions in CHD [16] reported that health-related QoL generally remained unchanged following training in ACHD populations, irrespective of the QoL instrument used (including EQ-VAS [48], SF-36 [49] and the Kansas City Cardiomyopathy Questionnaire [50]), although one study using a structured 10-week walking program demonstrated significant improvements in QoL [52]. Fortunately, QoL has gained importance in recent years, shifting the focus of CHD management from mere survival to enhancing QoL [53]. Assessing QoL provides complementary insights to clinical data, enabling identification of CHD patients who are at risk for health problems and require professional support [53]. Furthermore, evaluation QoL is fundamental to understand the impact of the disease and its treatment on patients’ lives [54]. A crucial direction for future research is to explore strategies for optimizing QoL in patients with CHD [6]. Promoting physical activity among these patients seems to be a promising approach in this regard. Adherence to exercise training in CHD A notable finding of the present study was that 39% of patients allocated to exercise training were classified as non-adherent. This highlights the challenge of maintaining engagement with exercise programs in ACHD, even within a structured monitored environment of a clinical trial. As adherence in routine clinical practice may be even lower, this finding has important implications for the real-world implementation of home-based exercise interventions in ACHD. Nevertheless, our study highlights that EC, muscle strength and VAS score improve even with partial adherence to the training regimen. This indicates that even physical activity to a lesser extent may be beneficial and supports the recommendation that all adults with CHD, except those with critical unaddressed defects, should be encouraged to exercise regularly, along with following standard healthy lifestyle recommendations [4]. Further research is necessary to determine the most effective exercise protocols for individuals with CHD. Additionally, strategies must be developed to efficiently promote and integrate exercise training as a fundamental part of CHD treatment [4]. Regrettably, as also previously documented in other studies, adherence and compliance remain a challenge [16]. While home-based programs offer flexibility, they may reduce compliance and adherence due to lack of supervision [16]. Gaining a deeper understanding of the factors that affect adherence can help motivate ACHD patients towards a more active lifestyle. Limitations The current study has several limitations. First, this investigation was designed as an exploratory study and therefore no formal a priori sample size calculation was performed. Consequently, the findings should be considered hypothesis-generating until confirmed in larger prospective studies. The relatively small sample size also precluded subgroup analyses according to specific CHD diagnoses or disease complexity, leaving uncertainty about potential differences in the relationships between RV function, EC and QoL, as well as differences in trainability among CHD subgroups [15]. Furthermore, inclusion was restricted to NYHA class I–II patients, which may limit generalizability to the broader CHD population. Second, the feasibility of TAPSE, TDI S’ and RV LS measurements was somewhat lower than reported previously in a broad ACHD patient spectrum [8]. While correlations between RV parameters and both EC and QoL were observed, causality cannot be inferred. Given the exploratory nature of the study, no adjustment for multiple testing was applied, which may increase the risk of type I error. Moreover, due to the limited sample size, multivariable analyses adjusting for established clinical and echocardiographic determinants could not be performed, and it therefore remains unclear whether RV function provides independent or incremental information beyond conventional markers. Third, the intervention duration of 16 weeks may have been insufficient to induce detectable changes in ventricular function. In addition, superoxide anion radical and muscle strength measurements were only available for a subset of patients, further limiting the analysis. Despite randomization, some baseline differences between groups were present, particularly for muscle strength, which may have at least in part influenced the observed training effects. Fourth, the study design also made it impossible to objectively measure total physical activity in both study groups or detect lifestyle changes in the conventional care group. Finally, with a single reassessment after 16 weeks, no conclusions can be drawn regarding the long-term effects or sustainability of the observed changes. Future perspectives We aim for this randomized trial to stimulate further research, particularly large-scale trials in CHD, investigating the feasibility, safety, and efficacy of various training regimens. Additionally, there is a need to determine the barriers and motivation for training in CHD, the underlying pathophysiological mechanisms of RV dysfunction and adaptation mechanisms of exercise training in CHD and its long-term effects. Conclusion In a heterogeneous ACHD population, RV function, assessed comprehensively via 2D MPE and RV strain, was significantly correlated with exercise capacity. Multiwall averages of RV parameters were also associated with QoL, particularly physical functioning, general health perception and VAS scores. Additionally, our 16-week HBASE program improved EC, muscle strength and QoL, without adverse events. However, adherence remains a challenge, emphasizing the importance of encouraging regular physical activity and integrating exercise training into standard ACHD care. CRediT authorship contribution statement Inne Vanreusel: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft. Daniel J. Bowen: Formal analysis, Investigation, Methodology, Software, Writing – review & editing. Vicky Van Der Meiren: Investigation. Jacob J. Briedé: Conceptualization, Supervision, Writing – review & editing. Annemien E. van den Bosch: Conceptualization, Supervision, Writing – review & editing. Bernard P. Paelinck: Conceptualization, Supervision, Writing – review & editing. Emeline Van Craenenbroeck: Conceptualization, Supervision, Writing – review & editing. Vincent F.M. Segers: Conceptualization, Supervision, Writing – review & editing. Wendy Hens: Conceptualization, Supervision, Writing – review & editing. An Van Berendoncks: Conceptualization, Funding acquisition, Supervision, Writing – review & editing. Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Inne Vanreusel reports financial support was provided by Fonds Wetenschappelijk Onderzoek - Vlaanderen (FWO). Vincent F.M. Segers reports financial support was provided by Fonds Wetenschappelijk Onderzoek - Vlaanderen (FWO). Emeline Van Craenenbroeck reports financial support was provided by Fonds Wetenschappelijk Onderzoek - Vlaanderen (FWO). If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements Sources of funding/support: This work was supported by Aspirant Fundamental Research (to IV) and Senior Clinical Investigator fellowships (to VFS and EVC) from the Fund for Scientific Research Flanders (Application numbers 11L3522N, 1501118N and 1842219N). The authors thank Kristien Wouters for statistical support. Footnotes Supplementary data to this article can be found athttps://doi.org/10.1016/j.ijcchd.2026.100694. Supplementary data The following is the Supplementary data to this article: Multimedia component 1 References Associated Data Supplementary Materials Multimedia component 1