Introduction Congenital heart defects may require surgical intervention such as the Fontan procedure that connects the systemic venous return to the pulmonary arteries. Although this procedure has increased survival, it results in reduced exercise capacity; which is reduced not only due to cardiovascular factors, but respiratory limitations as well. However, there is a lack of evidence outlining ventilatory limitations during constant-load exercise, which better represents exercise in cardiac rehabilitation programs and non-laboratory based exercise. Therefore, the aim of the present study was to compare responses to constant-load exercise in adult Fontan patients with those of healthy well-matched controls. Methods 14 adult Fontan patients (5F, 27 ± 6yrs) were recruited with 14 healthy matched controls. Participants performed forced vital capacity (FVC), as well as maximal inspiratory and and expiratory pressure assessments (MIP and MEP, respectively). Patients performed an incremental cycling test (ICT) to exhaustion to determine peak work rate. Following a period of recovery, patients performed a constant-load cycling test (CLCT) at 70% of peak ICT work rate until exhaustion. Healthy subjects reproduced the exercise of their matched patient. Cardiorespiratory variables and heart rate (HR) were measured using a metabolic cart and a 12-lead electrocardiogram, respectively. Participants were asked to rate their perception of breathlessness and respiratory exertion via a visual analogue scale every 2 min and at peak exercise. Patients without cardiac pacemakers underwent involuntary assessments of respiratory muscle contractility via phrenic (n = 8) nerve magnetic stimulation before and following exercise to quantify respiratory muscle fatigue. Results Patients showed significantly reduced FVC, MIP and MEP compared to controls (all p < 0.025). Patients’ time-to-exhaustion during the CLCT was 7.1 ± 3.3 min. During CLCT vs. the ICT, patients reached maximal HR, respiratory rate (fR), breathlessness, respiratory exertion, and leg exertion. End-exercise V̇O2 during the CLCT did not reach ICT values, with a mean difference of 1.5 ml/kg/min (p = 0.017). Controls did not reach peak ICT responses during the CLCT. During the CLCT, patients displayed significantly elevated minute ventilation (V̇E; mean difference = 21.5 L/min), fR (mean difference = 13.8 breaths-per-minute), breathlessness (mean difference = 3.4 points), and respiratory exertion (mean difference = 2.3 points), along with significantly decreased ventilatory reserve (V̇E/maximal voluntary ventilation; mean difference = 27.5%; all p < 0.002). Following the CLCT, Fontan patients showed a larger decrease in involuntary respiratory muscle contractility (15 ± 12% vs. 2 ± 11%). Finally, a decreased ventilatory reserve was significantly correlated with decreased MIP (r = 0.723, p = 0.003) and MEP (r = 0.623, p = 0.042). Discussion/Conclusion Fontan patients had a lower-than-expected time-to-exhausiton, in part due to their abnormal ventilatory response. First, the increased pulmonary restriction in the Fontan patients likely led to increased V̇E driven by a high fR during exercise. Second, Fontan patients showed decreased ventilatory reserve - which was significantly associated with respiratory muscle weakness. Third, patients showed significantly increased respiratory muscle fatigue following exercise. Collectively, these factors likely contributed to the increased breathlessness and respiratory exertion in patients, leading to increased exercise limitation. Given this, and the fact that patients reached near maximal physiological responses during the CLCT, it’s possible that patients may benefit more from aerobic training at less than 70% of peak word rate, or from interval-training with significant recovery time.