peptides (NPs) and the Kansas City Cardiomyopathy Questionnaire (KCCQ). Moreover, patients with NYHA class I and higher NPs experienced a higher event rate than those with NYHA classes II or III and lower NPs. Thus, as also highlighted by the authors, the current practice of considering patients with NYHA class I as a stable, homogeneous group at low risk of outcomes and crudely excluding them from HF trials might constitute a missed opportunity, affect generalizability, and decrease the feasibility of the enrollment. Beyond the suitability of using NYHA classification as treatment eligibility criteria, we believe that the current study has further implications for trial design.Changes in NYHA class are frequently used as a surrogate end point in HF trials. A meaningful surrogate trial end point ought to either gauge the risk of hard clinical outcomes or detect meaningful changes in clinical status. In this analysis, changes in NYHA class did not significantly predict subsequent harder events when adjusted for the NYHA value at follow-up. 1 Similar findings have been recently reported in 2 independent studies from the CHAMP-HF and Swedish HF Registries. 2,3 This would not necessarily compromise the value of NYHA class changes as trial end points if they were able to detect patients' well-being or objective measures of functional capacity. However, the NYHA classification appears a poor discriminant of patient-reported health status 1,2,4 and functional capacity as assessed by 6-minute walk test 4 when viewed crosssectionally and hardly correlates with changes in KCCQ scores when viewed longitudinally. 2 In contrast, the KCCQ was more sensitive in detecting changes in health status over time and its trajectories were associated with prognosis. 2 The NYHA classification has remained largely unchanged since its introduction a century ago. Meanwhile, the HF clinicians and trialists of today now have access to biomarkers that might better predict clinical events, the validated and more sensitive KCCQ to assess patient-reported symptoms, and simpleto-use objective measures of functional capacity. While the NYHA classification remains a useful tool to assess patients' clinical status in everyday practice, we are left questioning: does it still belong in our trials?