40 patients scheduled for staged bilateral total knee arthroplasty one week apart were randomized to have surgery on one knee through a minimally invasive midvastus approach and on the other knee through a minimally invasive medial parapatellar approach. Parameters of surgical outcome were monitored intra- and perioperatively, and clinical outcome was assessed over 2 years of follow-up. Estimated blood loss, incision length, incidence of lateral release, duration of surgery, and length of hospital stay were all statistically similar between groups. Few differences were observed in clinical outcome between groups, with no consistent patterns established. Knee Society Score improved in both groups within 6 weeks, and continued improving throughout follow-up.