Exercise self-efficacy was assessed using the Exercise Self-Efficacy Scale (17). The total score was calculated as the sum of all questions, with a higher score reflecting greater exercise self-efficacy, which assesses an individual's beliefs in their ability to continue exercising regularly.

Health-related quality of life was measured using the PedsQL Adult Quality of Life Inventory Version 4. The items of each question were reversed scored, and linearly transformed in accordance with the scoring guidelines. In addition to the total score, the physical health summary score and psychosocial health summary score were also calculated, with higher scores suggesting better health-related quality of life.

Center specific CPET protocols were performed on an electronically braked cycle ergometer as part of routine clinical care. In addition to measures of peak VO₂, work rate, and pre-exercise spirometry, CPET parameters including minute ventilation (VE), carbon dioxide production (VCO₂), blood pressure, VO₂ at the anaerobic threshold, heart rate (HR), and arterial oxygen saturation indicated by pulse oximetry were obtained when available. Predicted maximal oxygen pulse (ml/beat)—a surrogate for stroke volume—was calculated by dividing predicted maximal VO₂ by predicted maximal HR (220–age) (18). A cardiovascular limitation to exercise performance was indicated by a chronotropic index (cardiovascular index) above the upper limit of normal (19) or a reduced peak oxygen pulse (<80% predicted). Chronotropic index was calculated as ΔHR (beats/min)/ΔVO₂ (L/min) (19, 20). Maximal voluntary ventilation (MVV) was estimated as forced expiratory volume in 1 s (FEV₁) x 40, and breathing reserve was calculated as MVV–peak VE or (MVV–peak VE)/MVV x 100. A mechanical ventilatory limitation to exercise was suggested by a breathing reserve of <15% or <11 L/min (18, 21). Peak circulatory power was calculated as peak VO₂ (mL/kg/min) x peak systolic blood pressure (mm Hg). Ventilatory inefficiency was suggested by a peak VE/VCO₂ ratio of >40. A significant fall in oxygen saturation was considered as a decrease of ≥5%.

Spirometry parameters were considered as abnormal if values were below the lower limit of normal calculated from the Global Lung Initiative regression equations (22). Lung function was defined as normal, obstruction, restriction, or mixed defect in accordance with the American Thoracic Society/European Respiratory Society algorithm (23). In the absence of total lung capacity measured by plethysmography, ventilatory restriction was suggested if forced vital capacity (FVC) was below the lower limit of normal. Mixed defect was suggested if the FEV₁/FVC ratio and FVC were below the lower limit of normal.

To assess structured sport and physical activity participation across the lifespan, we used a modified version of the Kriska long-term recall physical activity questionnaire (24). Participants were asked to recall the sports and physical activities they participated in across multiple age ranges. For each sport and physical activity reported, the years in each age range, duration (hours) per month, and months per year were recorded. Sports and physical activities were categorized as childhood (ages 4–12 years), high school and early adulthood (ages 13–21 years), older adulthood (ages 22+ years), and physical activity across the lifespan (four to the age at questionnaire completion). The total hours of sport or physical activity participation were summated for each category and indexed as an average per week (h/wk).

Self-reported current levels of physical activity and sedentary time were assessed using the International Physical Activity Questionnaire (IPAQ) Long-Form. The IPAQ scores were not truncated, and metabolic minutes per week (MET-min/week) were calculated in accordance with the IPAQ Scoring Manual. Sedentary activity was reflected by sitting time (minutes) per day.

Statistical analysis was performed using IBM SPSS version 26 software (IBM Corp, Armonk, NY, USA). Data are presented as mean ± standard deviation or number (%) unless specified otherwise. The Shapiro-Wilk test or visual inspection of histograms and Q-Q plots were conducted to assess for normal distribution. An independent t-test or Mann–Whitney U was used as appropriate to compare differences between the “Super-Fontan” group and the control group. Proportions were compared using Pearson Chi-Square. A p-value of < 0.05 was considered as statistically significant.

Detailed participant demographics and characteristics are shown in Table 1. Of the 60 people with a Fontan circulation included in the CPET analysis, 15 had a “Super-Fontan” phenotype, and 35 had impaired exercise performance. The average age was 28.7 ± 7.6 years, and 48% were males.

The average body mass index (BMI) was 25.9 kg/m², and 20 people (33%) were overweight or obese. None who had the “Super-Fontan” phenotype were obese based on BMI compared to 22% in the control group (p = 0.046). The majority (70%) had a total cavopulmonary connection, and the average age at Fontan procedure was 6.4 ± 5.0 years. The age at Fontan procedure was lower in the “Super-Fontan” group compared to the control group (4.0 ± 2.9 vs. 7.2 ± 5.3 years, p = 0.002). Thirty-seven (62%) had a dominant left ventricle, 16 (27%) had a dominant right ventricle, 3 (5%) had biventricular morphology, and 4 (7%) had indeterminate ventricles. A dominant right ventricle was associated with impaired exercise performance (p = 0.043). One person (7%) in the “Super-Fontan” group had a Fontan conversion from an atriopulmonary connection to an extracardiac conduit type circulation. There were no statistically significant differences in age, sex, BMI, type of Fontan procedure, or patent fenestration between groups (p ≥ 0.05 for all).

Twenty-three Fontan participants completed the questionnaires. Of the 23 Fontan study group participants, 10 (43%) were in the “Super-Fontan” group, and 13 (57%) were in the control group. There were no differences between groups in baseline demographics for the subset of study participants who completed the questionnaires (p ≥ 0.05 for all).

The average total exercise self-efficacy score was higher in the “Super-Fontan” group compared to the control group (34.2 ± 3.6 vs. 27.9 ± 7.2, p = 0.02).

There was no statistically significant difference in total health-related quality of life score between the “Super-Fontan” group and the control group (78.9 ± 13.0 vs. 68.2 ± 18.8, p = 0.14). There was also no statistically significant difference between the “Super-Fontan” and control groups in the physical health summary score (80.3 ± 11.7 vs. 67.1 ± 21.9, p = 0.1) or psychosocial health summary score (78.2 ± 15.4 vs. 68.8 ± 19.6, p = 0.23).

Detailed CPET and spirometry results are shown in Tables 2, 3. In the study group, the average time from the CPET to the questionnaires was 2.1 ± 1.9 years, and there was no difference between groups (p = 0.5).

The average percent predicted peak VO₂ and work rate for the entire cohort was 67 ± 17% and 72 ± 22%, respectively. There was no statistically significant difference in percent predicted maximum HR between the “Super-Fontan” and control groups (83 ± 9% vs. 76 ± 15%, p = 0.09). Peak circulatory power, HR reserve (HRR), peak VE, and VO₂ at anaerobic threshold (percentage of predicted VO₂) were higher in the “Super-Fontan” group (p < 0.05 for all). There was no difference in exercise-induced desaturation between groups (p = 0.5).

Of the 40 participants (“Super-Fontan,” n = 11; control, n = 29) where the chronotropic index could be calculated, 7 (18%) participants had values outside the normal range. All participants in the “Super-Fontan” group had a chronotropic index within the normal range. In the control group, 4 (14%) participants had a high chronotropic index, and 3 (10%) had a low chronotropic index suggesting cardiovascular limitation and chronotropic insufficiency as inhibitors to exercise performance, respectively. The “Super-Fontan” group also had a higher percent predicted oxygen pulse compared to the control group (109 ± 16% vs. 80 ± 20%, p < 0.001). When markers of cardiovascular limitation were combined [low peak oxygen pulse (<80% predicted) or a high chronotropic index], no patient with the “Super-Fontan” phenotype had evidence of a cardiovascular limitation to exercise capacity compared to 62% in the control group (p < 0.001).

Thirty-seven people with a Fontan circulation (“Super-Fontan,” n = 10; control, n = 27) had baseline spirometry recorded, 16 (43%) had normal spirometry function, 20 (54%) had evidence of ventilatory restriction, and 1 (3%) had a pattern suggestive of mixed defect. The “Super-Fontan” group tended to have higher percent predicted FVC compared to the control group (85 ± 8% vs. 78 ± 10%, p = 0.05). Lung function abnormalities at rest were associated with impaired exercise performance; 2 (20%) patients in the “Super-Fontan” group had ventilatory or mixed defect compared to 19 (70%) in the control group (p = 0.006).

The average breathing reserve was 36 ± 18%, and five out of the 37 patients had a mechanical ventilatory limitation to exercise performance. The majority (80%) who had a mechanical ventilatory limitation also had evidence of ventilatory restriction at rest.

One of the participants in the control group had an incomplete questionnaire and was excluded from the analysis. Results for the Kriska physical activity questionnaire are shown in Table 4; Figure 1. Childhood physical activity was higher in the “Super-Fontan” group compared to the control group (3.9 ± 3.3 h/wk vs. 2.0 ± 3.4 h/wk, p = 0.04). The average h/wk of sports and physical activity participation during high school and early adulthood was 5.2 ± 4.4 h/wk in the “Super-Fontan” group and 2.1 ± 3.1 h/wk in the control group, p = 0.04. There was no statistically significant difference in sport and physical activity participation during older adulthood. The overall average duration of sport and physical activity participation indexed per week was higher in the “Super-Fontan” group compared to the control group (4.3 ± 2.6 h/wk vs. 2.0 ± 3.0 h/wk, p = 0.003).

The detailed IPAQ results are shown in Table 4. The average self-reported MET-min/wk was higher at all physical activity intensities and sub-domains, except for the transport domain in the “Super-Fontan” group, although this did not achieve statistical significance (Figure 2). Sitting time tended to be lower in the “Super-Fontan” group compared to the control group (308 ± 123 vs. 453 ± 175 min/day, p = 0.07).

In this study, the age at Fontan completion was lower in those with a “Super-Fontan” phenotype, and the absence of obesity or a dominant right ventricle were associated with normal exercise performance. This contrasts with previous findings that showed no differences between exercise performance Fontan phenotypes with these factors (6, 7). Our study describes an older Fontan cohort compared to previous series, and the conflicting findings might be attributed to an era effect. Alternatively, later age at Fontan completion, dominant right ventricular morphology, and obesity may manifest as important factors that impair the ability to achieve “normal” exercise performance later in life when circulatory function is more susceptible to compromise and maladaptation.

Similar to our findings, a review by Daley et al. found that earlier age at Fontan completion is associated with preserved long-term exercise capacity (25). A large multi-center series in a contemporary Fontan cohort also corroborates this; each year Fontan completion was delayed, percent predicted VO₂ and HRR decreased by 1.5 percentage points and 4.1 beats/min, respectively (26). This association may be explained by enhanced reversal of adverse cardiac remodeling and offsetting volume overload with earlier age at Fontan completion (27, 28). Indeed, patients with a later age at Fontan completion present with evidence of greater ventricular dysfunction and atrioventricular valve insufficiency (29). Conversely, one study found a positive correlation between age at Fontan completion, and percent predicted peak VO₂ (30). Although later age at Fontan completion is accompanied by an extended period of volume overload and cyanosis, it may allow for pulmonary artery catch-up growth and a larger conduit to optimize flow (31, 32); however, this theory requires verification.

We have previously reported increased adiposity is associated with a higher risk of adverse outcomes in people with a Fontan circulation (33). None of the people with the “Super-Fontan” phenotype were obese based on BMI compared to 22% in the comparator group with impaired exercise performance. The latest Pediatric Heart Network results also support this finding (34). This may be related to the impact of excess adiposity on respiratory muscle pump function and increased mechanical loading, which may impair exercise performance. In addition, the importance of maintaining a healthy body weight to preserve low pulmonary vascular resistance is increasingly recognized (32). Importantly, higher BMI often co-exists with increased visceral adiposity (epicardial and intra-abdominal fat), which may be particularly pathological—visceral adiposity is positively associated with pulmonary vascular resistance and inversely associated with ejection fraction and cardiac index in the Fontan circulation (35, 36). Of course, BMI may not be a robust measure of adiposity in the setting of complex congenital heart disease, where myopenia is common (33, 37, 38). The association between adiposity and pulmonary vascular resistance may be attributed to the adverse effects of pro-inflammatory adipokines (39), co-existing obstructive sleep apnea (40, 41), or decreased adiponectin (39, 42). The mechanisms underlying the association between adiposity and Fontan hemodynamics warrant further investigation.

We showed that an absence of a dominant right ventricle was associated with the “Super-Fontan” phenotype. A linear association between ventricular morphology and exercise capacity has previously been reported (43, 44). However, when categorized into exercise performance phenotypes (i.e., normal exercise performance vs. impaired exercise performance), series from Cincinnati Children's Hospital and the Children's Hospital in Philadelphia (a younger population than ours) found no association between ventricular morphology and a superior exercise performance phenotype (6, 7). It would seem plausible that a systemic right ventricle (compared to a systemic left ventricle) would be less likely to adapt to progressive hemodynamic perturbations over time and become more susceptible to circulatory demise and exercise intolerance. Long-term follow-up studies show dominant right ventricular morphology to be associated with worse clinical outcomes (45, 46).

We did not find any differences with regard to type of Fontan circulation, fenestration patency, or sex between the “Super-Fontan” phenotype and those with impaired exercise performance. However, although not statistically significant, there was a higher percentage of females in the “Super-Fontan” group compared to those with impaired exercise performance. Previous studies have also shown that a higher proportion of females are able to achieve better exercise performance (6, 7, 34), and have superior long term-outcomes (47, 48). The mechanisms underlying these sex differences are unclear and warrant further investigation.

Of note, while not statistically significant, and there were no differences in oxygen saturation, 20% of people with impaired exercise performance had a patent fenestration. It is unknown if this is associated with institutional bias toward fenestration or if these patients reflect a higher risk cohort requiring a fenestration at Fontan completion.

There were also no statistically significant differences in health-related quality of life measures between the “Super-Fontan” and control group in this study. However, the associations between quality of life and exercise performance are inconsistent across studies and warrant further investigation (49–51). This may be related to adults with a Fontan circulation accommodating to exercise intolerance over time. Supporting this notion, even asymptomatic people with congenital heart disease (New York Heart Association Functional Class I) have objectively impaired exercise capacity (2). Furthermore, the benefits of physical activity and exercise training on quality of life (particularly in the psychosocial domains) are potentially related to the social engagement that accompanies participation rather than better physical function.

HRR was greater in those with the “Super-Fontan” phenotype compared to the control group. Importantly, diminished HRR may be associated with arrhythmia-related mortality (9). This may explain why the combination of peak VO₂ and HRR has a more substantial prognostic value for 5-year survival (52), with peak VO₂ likely associated with heart-failure-related mortality. We also found higher peak circulatory power on average in people with the “Super-Fontan” phenotype, which is associated with better outcomes (53, 54). Collectively, the “Super-Fontan” exercise response reflects a low-risk Fontan phenotype.

Overall, most of our Fontan cohort had normal range chronotropic index implying an appropriate HR response relative to the metabolic load. Only three patients had a low chronotropic index with impaired exercise capacity indicating chronotropic insufficiency—It has been postulated that an inappropriate HR response is not a primary limiting factor to exercise performance unless severe chronotropic incompetence is present (55, 56). Exercise performance in the Fontan circulation appears predominately limited by preload deprivation that inhibits stroke volume augmentation. Over half of our Fontan patients with impaired peak VO₂ show evidence of a cardiovascular (stroke volume) limitation to exercise performance denoted by an elevated chronotropic response or low peak oxygen pulse. Of course, the reduced oxygen pulse and elevated chronotropic index can also reflect impaired peripheral muscle oxygen extraction, which is also present in the Fontan circulation (57).

Lung function is commonly impaired in people with a Fontan circulation and is associated with prognosis and exercise capacity (58–60). The majority of our Fontan cohort with baseline spirometry results recorded had evidence of ventilatory restriction or mixed defect. Baseline spirometry abnormalities were less prevalent in the “Super-Fontan” phenotype compared to those with impaired exercise performance. However, despite abnormal lung function at rest, most patients in this study had ample breathing reserve, potentially because exercise performance in Fontan patients is predominantly impaired by cardiovascular limitations prior to encroaching upon ventilatory constraints. Only five people had evidence of mechanical ventilatory limitation during exercise, with no difference between the “Super-Fontan” and control groups. Of note, it is likely ventilatory limitations to exercise performance is underappreciated using breathing reserve in isolation as a surrogate. The addition of exercise flow-volume loops will perhaps reveal more Fontan patients with ventilatory limitations to exercise performance.

Although self-reported physical activity levels were higher and sedentary (sitting) time was lower in the “Super-Fontan” group compared to the controls, this did not achieve statistical significance. This may be attributed to the duration between CPET and the administration of the questionnaires (~2 years), which likely does not truly reflect their current levels of physical activity. Furthermore, while the IPAQ shows moderate validity compared to accelerometers (61), in the setting of congenital heart disease, patients often overestimate their physical activity levels using the IPAQ long-form (62).

Indeed, Powell et al. reported that 77% of patients with the “Super-Fontan” phenotype participated in regular physical activity compared to 10% in those with impaired exercise performance (6). This is in accordance with our previous reports that show many of those with a “Super-Fontan” phenotype or positive exercise capacity trajectory regularly participate in moderate-to-vigorous sports and physical activity (5, 63). Importantly, increased physical activity levels could be attributed to higher exercise self-efficacy in the “Super-Fontan” cohort. Preceding studies have shown an association between physical activity levels and exercise self-efficacy (64).

We found that participation in regular structured sports and physical activity from childhood to early adulthood was significantly higher in those with a “Super-Fontan” phenotype compared to those with impaired exercise performance. Total overall participation in sports and physical activity was also higher in the “Super-Fontan” group. While exercise training interventions can increase peak VO₂ (65, 66), it appears that participation in regular sports and physical activity from a younger age lays the foundation to achieve a low-risk “Super-Fontan” phenotype. This is consistent with the findings of Ohuchi et al. who showed that increased physical activity levels during childhood—reflected by a positive exercise capacity trajectory—were associated with better adult Fontan physiology, hepatic function, and prognosis (67). In another series, children and adolescents with a Fontan circulation who participated in sports during middle and high school had better lung function and exercise capacity (68). Regular sports, exercise training, or physical activity may be particularly crucial during childhood when organs, and especially the pulmonary vasculature, are likely more sensitive to adaptation in this period of rapid growth and development (67).

The association with regular moderate-to-vigorous physical activity participation and the “Super-Fontan”—superior exercise performance and low-risk—phenotypical expression may be attributed to multiple mechanisms. Regular participation in moderate-to-vigorous intensity sports and physical activity is important for the development of skeletal muscle mass and prevention of Fontan-associated myopenia (37). Deficiencies in skeletal muscle mass have important implications for ventricular function (37), and exercise capacity (57, 69). Higher appendicular muscle mass provides structural support to reduce venous compliance and enhances skeletal muscle pump function, facilitating preload and ventricular filling (70, 71). Indeed, increasing skeletal muscle mass through resistance exercise training improves ventricular filling, cardiac output, peak VO₂, and reduces respiratory dependence in people with a Fontan circulation (72).

A primary constraint to ventricular filling appears to be elevated pulmonary vascular resistance. To maintain low pulmonary vascular resistance, there must be an adequately developed pulmonary circulation. However, pulmonary artery growth essentially ceases after Fontan completion (73), likely due to the combination of reduced pulsatility and pulmonary flow. We have previously shown that lower limb exercise can transiently increase (pulsatile) pulmonary flow (74). It is likely that engaging in regular long-term physical activity or exercise training (particularly during childhood), which transiently increases pulmonary flow (and may also alter the flow profile), facilitates pulmonary vascular development. Furthermore, periodic increases in ventricular filling associated with exercise may augment volume load to the chronically preload deprived ventricle and attenuate progressive “disuse hypofunction”. This phenomenon is observed when volume load is restored following atrial septal defect closure in adults, reversing diastolic dysfunction (31).

Of course, the association between the “Super-Fontan” phenotype and physical activity during childhood, and early adulthood, may simply reflect that those with superior exercise performance are more capable of participating in regular sports and physical activity, especially from an earlier age.