BackgroundAlthough transcatheter aortic valve replacement (TAVR) is the least invasive treatment for patients with symptomatic aortic stenosis, some patients hesitate to undergo the procedure. We investigated the clinical impact of treatment delay after patient refusal of TAVR.Methods and ResultsWe used the Japanese OCEAN (Optimized Catheter valvular intervention) regsitry data of 1542 patients who underwent TAVR. Refusal was defined as at least 1 refusal of TAVR at the time of informed consent. Patients were separated into 2 groups: refusal (28/1542, 1.8%) and non‐refusal (1514/1542, 98.2%). We compared the baseline characteristics, procedural outcomes, and mortality rates between the groups. Additionally, data on reasons for refusal and those leading to eventually undergoing TAVR were collected. Age, surgical risk scores, and frailty were higher in the refusal group than in the non‐refusal group (P<0.05 for all). Periprocedural complications did not differ between groups, whereas 30‐day and cumulative 1‐year mortality were significantly higher in the refusal group than in the non‐refusal group (7.1% versus 1.3%, P=0.008 and 28.8% versus 10.3%, P=0.010, respectively). Multivariate Cox regression analysis revealed that TAVR refusal was an independent predictor of increased midterm mortality (hazard ratio: 3.37; 95% confidence interval: 1.52–7.48; P=0.003). The most common reason for refusal was fear (13/28, 46.4%), and the most common reason for changing their mind was worsening heart failure (21/28, 75.0%). All patients in the refusal group decided to undergo TAVR within 20 months (median: 5.5 months).ConclusionsRefusing TAVR even once led to poorer prognosis; therefore, this fact should be clearly discussed when obtaining informed consent.