The medial parapatellar approach has become the standard technique in total knee arthroplasty (TKA). However, recent studies have reported superior results regarding functional recovery when using the midvastus approach. It was the aim of this study to evaluate the early functional outcome of both surgical techniques. In a prospective, double-blinded, randomized study, 50 patients for TKA were consecutively operated on either by the medial parapatellar or the midvastus approach. Exclusion criteria were defined as previous open knee surgery and leg deformity of more than 10 degrees varus or valgus. TKA was performed in all patients by one surgeon using the same type of implants in both groups. Pain scores (VAS) were documented and follow-up data including quadriceps strength and proprioception were obtained 3 weeks and 6 weeks postoperatively. Both groups were comparable in preoperative demographic data. Postoperatively, patients in the midvastus group demonstrated significantly lower pain in rest (VAS: mean 2.25 vs. 3.03) and under movement (VAS: mean 2.92 vs. 3.13). Further, they showed superior isometric quadriceps strength at 3 weeks (41.4 vs. 27.6 Nm) and 6 weeks (47.6 vs. 35.5 Nm). Moreover, this group showed a superior postoperative proprioception, while the range of motion was similar in both groups. The midvastus approach offers advantages over the standard parapatellar arthrotomy, in the early rehabilitation period. No adverse effects were observed associated with this approach. Therefore, the midvastus approach should be considered as a valuable alternative in TKA.