Imageless computer navigation in total knee arthroplasty.

OBJECTIVES To study the difference in clinical and radiological outcomes between imageless computer-navigated and the conventional technique for performing total knee arthroplasty. DESIGN Prospective case-control study. SETTING University teaching hospital, Hong Kong. PATIENTS Forty-seven patients with imageless computer-navigated total knee arthroplasty were matched with 47 patients with total knee arthroplasty using conventional technique over the period 2003 to 2007. MAIN OUTCOME MEASURES Postoperative radiological alignment, tourniquet time, Knee Society knee score and functional score, and range of motion. RESULTS The femoral anteroposterior, tibial anteroposterior, overall anteroposterior alignment of the lower limb, and lateral femoral lateral flexion angle had significantly less deviation from neutral alignment in those having computer-assisted total knee arthroplasty than conventional total knee arthroplasty. The former had significantly fewer outliers (>3 degrees from neutral alignment) than the latter, in both the coronal and sagittal planes of the femoral and tibial sides. The mean tourniquet time was significantly longer in the former (111 minutes) than in those having a conventional arthroplasty (98 minutes). There was no tracker/pin tract complication or wound infection in either group. CONCLUSION Computer-assisted navigation can give a better alignment than the conventional technique, but the tourniquet time was significantly longer. Computer-assisted navigation surgery can help the surgeon in determining limb alignment during the operation.