[Accuracy of total knee replacement bone cuts using a conventional ancillary system: 300 Innex total knee...

Short-term functional results and long-term outcome in terms of stability and wear greatly depend on the precision of the bone cuts. We wanted to know whether conventional ancillaries are still competitive in terms of accuracy in comparison with computer-assisted navigation systems. A few comparative studies favor navigation, but have generally only included a small number of patients. We studied radiographically a prospective consecutive series of 300 total knee prostheses (Innex, Zimmer) implanted with the conventional technique by the same operator.

A telegonometric view in the upright position and a short lateral view were obtained in all patients. In the AP view, implants were measured in comparison with the mechanical axis of the femur (F) and the tibia (T). On the lateral view, the prosthesis-tibial shaft angle (PT) was measured from the proximal portion of the tibial shaft and the prosthesis-femoral shaft angle (PF) from the distal portion of the femur. The same operator made all measurements using the same optimal conditions. The series included 178 women and 122 men, mean age 72 +/- 8 years who presented genu varum (n = 248 knees) and genu valgum (n = 52 knees): degenerative disease (n = 238), polyarthritis (n = 4), hemophilic arthropathy (n = 3), necrosis (n = 3), revision of unicompartmental prosthesis (n = 8), and osteotomy (n = 44).

The standard x-ray protocol was performed at two months in all patients. The mechanical axis (HKA) was 179.4 +/- 2.4 degrees (range 173-186 degrees) and was +/- 3 degrees in 87% of knees with no difference for varum and valgum. F was 90.1 +/- 1.4 degrees (87-95), with +/- 3 degrees for 98.7%. T was 89.3 +/- 1.5 degrees (85-94) with +/- 3 degrees for 95.6%. PF was 88.6 +/- 1.6 degrees (84-93) for 87%, PT was 87 +/- 2 degrees (81-93) with +/- 3 degrees for 94%. The four cuts were within +/- 3 degrees for 227 prostheses (77%), within +/- 2 degrees for 156 (52%) and within +/- 1 degrees for 56 (18%). Measurements made again one year after implantation for 203 knees gave the same results. Operative time for implantation was 68 +/- 23 minutes for implantation and 85 +/- 23 minutes including complete closure (less than 60 minutes for 68 knees).

The accuracy of each cut was satisfactory on average with a small standard deviation. Recent data in the literature show that the accuracy in our series is comparable with that obtained currently with navigation systems. In light of this experience, it can be seen that better precision can be achieved for each of the cuts.

The accuracy of conventional instrumentation systems is still comparable with that obtained with computer-assisted surgery. The purpose of this study was not to question the benefit of navigation, but to establish a basis upon which progress can be measured. The results enabled a more realistic comparison of the precision of navigation systems and also can be comforting for operators still using conventional ancillaries.