Cognitive dysfunction is a common postoperative neurological complication in patients undergoing valve replacement surgery. This study aimed to compare the effects of sevoflurane versus propofol‐based total intravenous anesthesia on the incidence of cognitive dysfunction following valve replacement surgery. This multicenter, randomized, controlled double‐blinded study was conducted in three teaching hospitals in China. Patients receiving on‐pump valve replacement surgery were enrolled. Stratified block randomization was used to randomly assign patients 1:1 to receive sevoflurane (1.0–1.5 MAC) or propofol (2.0–3.0 mg/kg/h) for anesthesia maintenance. The primary outcome was the incidence of cognitive dysfunction assessed by four cognitive tests before, as well as 7–14 days after surgery. Patients were randomly assigned to receive sevoflurane anesthesia (n = 144) or propofol‐based total intravenous anesthesia (n = 145). The incidence of postoperative cognitive dysfunction in the sevoflurane anesthesia group (31.9%) was significantly lower than that in the total intravenous anesthesia group (43.4%; relative risk 0.61, 95% confidence interval CI: 0.38–0.97, p = 0.044). There was no difference in the incidence of delirium between patients receiving sevoflurane and total intravenous anesthesia (27.8% 35/144 vs. 25.9% 35/145, 1.10, 95% CI: 0.64 to 1.90, p = 0.736). There was a significant difference in the Katz Index on day 3 after surgery (3 0.9) vs. 3 (1.0, 0.095, 95% CI: 0.05 to 0.43, p = 0.012). No difference was observed in other outcomes between the two groups. For patients undergoing on‐pump valve replacement surgery, sevoflurane anesthesia had a smaller effect on cognitive function and independence in daily life activities compared with propofol anesthesia.