[Objective] To compare the fast track rehabilitation effects of total knee arthroplasty(TKA) with subvastus approach(SVA) versus medial parapatellar approach(MPA). [Methods] This prospective study included 84 consecutive patients with primary knee osteoarthritis who underwent TKA surgery. They were randomly divided into SVA group (subvastus approach) and MPA group (medial parapatellar approach) with 42 cases in each group. All patients were first-time, unilateral TKA operation. The incision length, operation time, postoperative drainage volume, additional dose of Etoricoxib tablets, straight leg elevation time, bed time, VAS score under exercise and rest, knee joint mobility and HSS score were compared between the two groups. [Results] 2 patients were detached in each of the two groups with eighty-four patients, and the remaining 80 patients were followed up. The operation time of SVA group was prolonged (P=0.036), but it was better than MPA group in incision length, additional dose of Etoricoxib tablets, straight leg elevation time, bed time (P < 0.05), no drainage tube was placed in SVA group, and the drainage volume in MPA group was 174.6 ± 16.1 ml. There was no significant difference in VAS score under exercise, knee joint mobility and HSS score between the two groups before operation (P > 0.05). The VAS scores of SVA group were lower than those of MPA group at 6 hours, 1 day, 3 days, 1 week, 1 month under exercise and 6 hours, 1 day and 3 days under rest after operation (P < 0.05). There was no significant difference in VAS scores between the two groups at 1 week under rest, 3 months under exercise after operation (P > 0.05). The knee joint mobility and HSS score in SVA group were higher than those in MPA group at 3 days, 1 week and 1 month after operation (P < 0.05), but there was no significant difference in knee joint mobility and HSS score between the two groups at 3 months after operation (P > 0.05). [Conclusion] TKA surgery via SVA approach is more conducive to postoperative fast track rehabilitation.