Abstract Background The arrhythmia burden in adult patients (Ps) with tetralogy of Fallot (TOF) is considerable, being a major late complication in repaired TOF. The incidence of atrial arrhythmias is relatively high and one of the main causes of morbidity. Our aim was to assess which cardiopulmonary exercise testing (CPET) parameters best correlate with supraventricular arrythmias (SVA) to potentially improve identification of high-risk Ps. Methods A single-center retrospective analysis was performed from 2009 to 2018 on adult Ps with repaired TOF who underwent maximal CPET. Ambulatory ECG monitoring was used to analyze the arrhythmic profile. Demographics, standard measures of CPET interpretation, and major cardiovascular outcomes were collected. Multivariate analysis for the prediction of SVA was performed using Cox Regression, by including all statistically significant variables in the univariate analysis and those considered clinically relevant. Results CPET was performed in 63 repaired TOF Ps (57% male, mean age of 34 ± 9 years), with a mean follow-up of 60 ± 33 months. 56% Ps had severe pulmonary regurgitation and 48% Ps underwent pulmonary valve replacement (PVR). Supraventricular arrhythmias occurred in 12 Ps (19%), mainly atrial fibrillation or atrial flutter (67%). There were no statistically significant differences regarding the clinical profile and baseline CPET parameters between both groups, namely: age (p=0.113), male gender proportion (p=0.581), proportion of Ps who underwent PVR (p=0.578), RV systolic dysfunction prevalence (p=0.854) and peak oxygen uptake (p=0.083). VE/VCO2 slope was an independent predictor of SVA (HR 1.41, CI 95% 1.08-1.84, p=0.012). Exercise capacity assessed by estimated metabolic equivalents (METS), as well as the peak heart rate (PHR) during exercise (HR 0.86, CI 95% 0.76-0.96, p=0.007), revealed a protective effect regarding SVA (HR 0.71, CI 95% 0.51-0.98, p=0.045). Neither peak oxygen uptake (p=0.090) nor cardiorespiratory optimal point (p=0.427) showed statistically significant correlation with SVA during follow-up. Ps with a PHR above 165 bpm had a significantly higher survival free of SVA (log-rank p=0.045). Conclusion In our population, VE/VCO2 slope was an independent predictor of SVA during follow-up, while both a higher PHR and a higher exercise capacity had a protective effect. CPET could be an accessible way of improving risk-stratification of TOP Ps and should therefore be included in their routine assessment.