Accuracy of the preoperative planning for cementless total hip arthroplasty. A randomised comparison between three-dimensional computerised planning and conventional templating.

INTRODUCTION A high accuracy was recently reported for the three-dimensional (3D) computerised planning of total hip arthroplasty (THA), comparing well with navigation regarding leg length and femoral offset. However, there is no randomised study comparing 3D preoperative planning with conventional 2D templating in terms of accuracy and clinical relevance. HYPOTHESIS The 3D preoperative planning has a higher accuracy than the conventional 2D preoperative templating regarding the implants size and their positioning. PATIENTS AND METHODS A prospective comparative randomised study was carried out from 2008 to 2009, including two groups of 30 patients who underwent THA for primary osteoarthritis. One surgeon performed all the surgical procedures using a minimally invasive direct anterior approach. In one group, the planning was made on calibrated X-rays using 2D templates. In the other group, a CT-scan based 3D computerised planning was performed with dedicated software. The reconstructed hip final anatomy was compared postoperatively to the preoperative planning and the accuracy was expressed as the mean difference (±SD) between the planned positioning and the final positioning of the implants. RESULTS The prediction rate for the stem and the cup sizes were respectively of 100% and 96% in the 3D group versus 43% for both components in the 2D group. When combining both components, the prediction rate was 96% in the 3D group versus 16% in the 2D group. In the 3D group, a high accuracy was achieved for the planning of the leg length (-1.8±3.6 mm ranging from -8 to+4mm) and the femoral offset (-0.07±2.7 mm ranging from -5 to+4mm) versus 1.37±6.4mm ranging from -9 to 13 mm and 0.33±5.7 mm (-16 to 11 mm) in the 2D templating group (P<0.0001). DISCUSSION The 3D planning gives a higher accuracy than conventional 2D templating in forecasting the size of cup and the stem. This contributes to the prediction for leg length and offset that is more reliable with the 3D technique. This study suggests that 3D planning CT-scan data is an attractive alternative to navigation to restore these parameters. The high accuracy achieved by a low-experience surgeon suggests that 3D planning may help shorten the learning curve when using the minimally invasive direct anterior approach. LEVEL OF EVIDENCE Level III low-powered prospective randomized trial.