To systematically review the prognosis of transcatheter aortic valve replacement (TAVR) versus surgical aortic valve replacement (SAVR) in patients with severe aortic stenosis. A systematic search of PubMed, EMBASE, Scopus, Cochrane Library, China biomedical literature database, China journal full text database (CNKI), Wanfang database and VIP database from January 2012 to February 2022 was conducted for randomized controlled trial (RCT) that comparing TAVR and SAVR in the treatment of severe aortic stenosis. The primary outcomes were the incidence of all-cause mortality, stroke incidence, reoperation rate and complications (pacemaker implantation, atrial fibrillation) at 1 month and 1, 2, 5 years after operation. Jadad scale was used to evaluate the literature quality of RCTs. All statistical analyses were performed using the standard statistical procedures provided in RevMan 5.4.1. A total of 17 studies including 11 712 patients were identified, including 6 007 patients treated with TAVR and 5 705 patients treated with SAVR. There were 4 high-quality studies and 13 medium-quality studies. The results of meta-analysis showed that the rate of new onset atrial fibrillation was lower in TAVR group than that in SAVR group (=0.28, 95% 0.21-0.38, <0.001), and there was no significant difference in all-cause death, stroke, pacemaker implantation and reoperation rate (all >0.05) at 30 days follow-up. At one year after TAVR and SAVR treatment, all-cause mortality (=0.85, 95% 0.74-0.97, =0.01) and new onset atrial fibrillation (=0.28, 95% 0.20-0.39, <0.001) were lower in TAVR group than SAVR group. However, the pacemaker implantation rate was higher in TAVR group than that of SAVR group (=1.79, 95% 1.11-2.89, =0.02), while there was no significant difference in the incidence of stroke and reoperation between the two groups (>0.05). At two years after TAVR and SAVR treatment, the pacemaker implantation rate was higher in TAVR group than that in SAVR group (=2.23, 95% 1.28-3.86, =0.004), and the rate of new atrial fibrillation was lower in TAVR group than that in SAVR group (=0.46, 95% 0.38-0.56, <0.001). There was no significant difference in all-cause death, stroke and reoperation rates between the two groups (>0.05). At five years after TAVR and SAVR treatment, the pacemaker implantation rate (=1.89, 95% 1.13-3.17, =0.02) and reoperation rate (=3.64, 95% 1.75-7.58, =0.000 5) were higher in TAVR group than those in SAVR group, while the rate of new onset atrial fibrillation was lower in TAVR group than that in SAVR group (=0.45, 95% 0.37-0.55, <0.001). There was no significant difference in all-cause death and stroke incidence between the two groups (all >0.05). The all-cause mortality and the incidence of new onset atrial fibrillation after TAVR are lower than SAVR, and TAVR is a preferred therapy for patients with aortic stenosis.