Aim: To compare two approaches to the knee joint for total knee arthroplasty with regard to patient comfort and return of knee function.

Methods: Sixty-six consecutive patients were randomised prospectively to undergo either a medial parapatellar or midvastus approach for total knee arthroplasty. All patients received the same type of prosthesis under the guidance of the same surgeon. A comparison was made of operative time, number of soft tissue releases required, the number of days to reach 60 and 80 degrees of flexion, the time to achieve a straight-leg raise, and the time to discharge from hospital. A subgroup of 34 patients who received the same type of anaesthetic (spinal and femoral nerve blocks) and who also received the same type of post-operative analgesia, were compared for pain and analgesic requirements in the first 48 hours following surgery.

Results: The two groups were similar for number, age, weight, gender and diagnosis. A statistical analysis showed that the midvastus approach was significantly better, with an earlier straight-leg raise and shorter operative time. There was a trend towards improvement in early range of motion but no difference in post-operative pain, number of soft tissue releases or difficulty with surgery.

Conclusions: This trial has shown that the midvastus approach was an effective alternative to the medial parapatellar approach for knee arthroplasty. It produced less tissue damage to the quadriceps mechanism and medial patellar blood supply. It allowed an earlier recovery of knee function and shorter surgery time with no increase in the difficulty of the surgery.