Preoperative Digital Templating as a Prediction Tool for Component Size in Total Knee Replacement

Background:
Total knee replacement (TKR) is one of the most commonly performed Orthopedic procedures. In preparation for surgery, pre-operative templating has been mostly described on acetate paper but has not been shown to be efficient in reproducing intraoperative measurements. Recently, digital templating has largely replaced other forms of templating. However, studies on its reliability and accuracy are scarce.

Purpose:
This study aims to determine the reliability and accuracy of digital templating predicting tibial and femoral component sizes in TKR surgery.

Materials and Methods:
We retrospectively retrieved pre-operative data points from patients who underwent TKR at our institution. Pre-operative data points included standard Anterior-Posterior (AP) and Lateral radiographs of the affected knee using a metal ball of a known size as a calibrating tool for the digital templating software (TraumaCad®, Brainlab, Westchester, IL, USA). Surgeries were performed without standard templating and component sizes were measured and chosen at the time of surgery by the attending surgeon. Postoperative radiographs were then reviewed by an attending surgeon who assessed the alignment and sizing of implants. Independently, digital templating was performed post-operatively by a surgeon who did not participate in the index surgery. Upon completion of the digital templating, we analyzed how pre-operative digital templating can serve to predict the actual femoral and tibial component size.

Results:
From 2015-2016, we retrospectively retrieved 93 patients who underwent elective TKR at our institution. After comparing component sizes chosen by the attending surgeon and those predicted by the digital templating, we found that femoral component prediction rate was 44% (n=41). By allowing a size difference of ± 1, prediction rate was found to increase to 87% (n=81). When analyzing the tibial component, we found that prediction rate was 57% (n=53), whereas when the size difference was allowed to be within ± 1, rate had increased to 97.8% (n=91).

Conclusion:
Preoperative digital templating seems to be a reliable method for assessing component size when allowing a size deviation of ± 1. We therefore suggest that it can serve as a valuable guide for the size range needed at the time of surgery.
Nonetheless, preoperative templating alone is probably not accurate enough in order to fully replace intraoperative measurement in the prediction of component size.