THE ACCURACY OF DIGITAL TEMPLATING IN PRIMARY TOTAL KNEE REPLACEMENTS

Abstract

Surgeon’s ability to anticipate the implant size requirements for total knee replacements is important to the success of the procedure. Previously, this has been inconsistently accomplished using plain radiographs. The purpose of this study is to assess the accuracy of digital templating software in predicting the size requirements of the femoral and tibial implants in total knee replacements.

Thirty consecutive PFC (DePuy) total knee replacements were templated preoperatively using digital templating software (TraumaCad, Orthocrat Ltd). The knees were templated by two surgeons working independently using standardised digital AP and lateral radiographs. All films were magnification-calibrated using markers of known size. Postoperatively, the predicted implant size was compared to the actual components selected at the time of surgery.

The size of the femoral prosthesis was accurately selected on the AP view in 53.5% and on the lateral in 66% of cases. The size of the tibial implant was correctly selected on the AP view in 65.5% and on the lateral in 70.5% of cases. The tibial prosthesis was always templated within one size. The femoral prosthesis was predicted within two sizes (93% on AP + 98% on lateral +/− 1 size). There was no correlation with failure of the software to recognise the metal marker and inadequate lateral x-rays.

The lateral x-ray was found to be more reliable than the AP on predicting both the femoral and tibial implants. The tibia was more accurately templated than the femur on average. There was good inter-observer and intra-observer reliability for both prostheses (0.75 – 0.85). Discrepancies in templating may have been due to inaccuracies in placement of the metal marker at the time of x-raying or due to fixed flexion deformities, which may have affected the magnification of the x-ray. Overall, templating using digital software was marginally superior to the standard acetate method.