Abstract Aims Controversy remains as to whether the exercise stress test (EST) is sufficient for risk evaluation in patients with pre-excitation. This study aims to clarify the usefulness of EST in risk stratification in both asymptomatic and symptomatic patients presenting with pre-excitation. Methods and results This prospective study includes consecutive asymptomatic and symptomatic patients with pre-excitation referred for risk assessment. All participants performed an incremental EST (bicycle) prior to an electrophysiology study (EPS). Primary data from the EST included loss of pre-excitation during exercise, and primary data from the EPS included the measurement of accessory pathway effective refractory period (APERP), shortest pre-excited RR interval (SPERRI), and inducible arrhythmia with the use of a beta-adrenergic receptor agonist if deemed necessary. One hundred and sixty-four patients (59 asymptomatic, 105 symptomatic) completed an EST and EPS. Forty-five patients (27%) demonstrated low-risk findings on EST, of which 19 were asymptomatic and 26 were symptomatic. Six patients with low-risk EST findings had SPERRI/APERP ≤ 250 ms at EPS, and two of them were asymptomatic. The sensitivity, specificity, positive predictive value, negative predictive value (NPV), and accuracy of low-risk EST for excluding patients with SPERRI/APERP ≤ 250 ms were 40, 91, 87, 51, and 60%, respectively. The number of patients with inducible arrhythmia at EPS was similar in the asymptomatic (36, 69%) and symptomatic (73, 61%) groups. Conclusion Sudden loss of pre-excitation during EST has a low NPV in excluding high-risk APs. The EPS with the use of isoproterenol should be considered to accurately assess the risk of patients with pre-excitation regardless of symptoms (ClinicalTrials.gov Identifier: NCT03301935). Graphical Abstract Graphical Abstract