This study aimed to determine whether the exercise capacity of patients with pulmonary atresia/intact ventricular septum (PA/IVS) who have undergone biventricular repair is superior to that of patients with single ventricle repairs and to account for any differences. PA/IVS is generally treated with either biventricular (outflow tract reconstruction) or univentricular (Fontan) palliation. Although biventricular repair is believed to result in superior exercise function, this theory is untested. Symptom-limited programmed bicycle ergonometry with expiratory gas analysis was prospectively performed on all patients with PA/IVS >7 years old seen over 18 months. Nineteen biventricular and 10 Fontan patients (mean age 16.5 +/- 6.5 vs 12.7 +/- 5.0 years, p = 0.12) were enrolled. The exercise capacity of biventricular patients was not statistically superior to that of Fontan patients (predicted peak VO2 83.5 +/- 21% vs 76.0 +/- 17.5%, p = 0.34), although chronotropic function and ventilatory efficiency were significantly better in the former. The peak exercise capacity varied widely within each group, and there was considerable overlap between biventricular and Fontan patients. Within groups, imaging studies did not reliably predict exercise capacity. Most patients in each group had subnormal peak VO2, and there was a trend toward impaired performance with increasing age regardless of type of repair. In conclusion, biventricular repair may not guarantee superior exercise performance over single-ventricle palliation in PA/IVS. Regardless of repair type, aerobic capacity may deteriorate with age and is not reliably predicted by noninvasive imaging. These findings underscore the need for a quantitative, proactive approach to the assessment and preservation of exercise function.