To determine the effectiveness of a double checking method with an extramedullary guide for improving the accuracy of an intramedullary alignment guide in a total knee arthroplasty.

From June 1999 to October 2002. 30 cases of a double checked total knee arthroplasty with an extramedullary guide were classified as Group I, and an other 30 cases of a total knee arthroplasty with an intramedullary guide were classified as Group II. The extramedullary guide was a set 3-finger width medial to the anterior superior iliac spine as a landmark, and was checked by the X-ray findings during surgery. The radiological assessment was carried out by comparing the preoperative and postoperative standing AP view.

The postoperative femorotibial angle in Group I was valgus 5.9° (5-9°) and valgus 7.7° (4-11°) in Group II. The normal anatomical variation (valgus 6±2°) was outranged in 1 case (3%) in group I, and 6 cases (20%) in group II.

Determining the angle of a distal femoral resection only with an intramedullary guide can cause a misalignment by many factors. The double check method with an extramedullary guide can improve the alignment accuracy, and show better results.