Comparison of the vastus-splitting and median parapatellar approaches for primary total knee arthroplasty: a prospective, randomized study.

BACKGROUND

A vastus-splitting approach for total knee arthroplasty has been advocated to preserve function of the extensor mechanism and to decrease the prevalence of lateral release. Critics have claimed that there is greater blood loss and compromised exposure in large patients who are managed with this approach. The purpose of the present study was to compare vastus-splitting and median parapatellar approaches for primary total knee arthroplasty.

METHODS

Forty-two consecutive patients (fifty-one knees) undergoing primary total knee arthroplasty were randomized to treatment with a median parapatellar or vastus-splitting approach. The interval of the vastus muscle split was marked with radiopaque vascular clips. Surgical data, functional parameters, and preoperative and postoperative electromyograms were assessed.

RESULTS

Early (six-month) and intermediate-term (five-year) follow-up showed no differences in functional parameters, tourniquet time, or the frequency of patellar resurfacing. Significantly more lateral releases (p < 0.01) and greater blood loss (p = 0.03) occurred in the median parapatellar group. Nine (43%) of twenty-one knees in the vastus-splitting group had abnormal electromyographic findings at six months postoperatively, whereas all patients in the median parapatellar group had normal findings. Seven knees with abnormal electromyographic findings at six months had normal findings when restudied at five years; in each of these knees, the vastus split had been developed bluntly. The other two knees with abnormal findings at six months had had sharp dissection for the muscle split. Both of these knees had chronic changes, one with changes indicative of reinnervation and the other with ongoing denervation, but neither demonstrated functional compromise.

CONCLUSIONS

The vastus-splitting approach offers a viable alternative to the median parapatellar approach for primary total knee arthroplasty that reduces the need for lateral retinacular release without impairment of quadriceps function. Electromyographic abnormalities in the quadriceps muscle have no functional consequence and most likely represent reversible neurapraxic injury that may be avoided by blunt dissection in the vastus medialis muscle.