Digital Templating of THA Using PACS and an iPhone or iPad is as Accurate as Commercial Digital Templating Software

Digital templating is a standard preoperative planning method in THA. Unfortunately, many hospitals cannot afford the software or have accessibility barriers owing to a limited number of installed computer workstations. We created a templating method using a picture archiving and communication system and the Keynote program on a mobile phone. The method's accuracy and reliability have not been compared with those of commercial digital templating software.

(1) How accurate is this novel method on iPhone and iPad mobile devices compared with a commercially available digital templating software program? (2) Is the method reproducible among users with different levels of experience? (3) Are the results similar for different types of femoral prostheses?

Between January 2017 and May 2020, we treated 209 patients for hip disease or trauma with primary cementless THA. We considered patients with a normal contralateral hip as potentially eligible for this retrospective study. Thus, 91% (191 of 209 hips) were eligible; a further 13% (27 hips) were excluded because of postoperative leg length discrepancy > 5 mm, femoral offset discrepancy > 5 mm (9% [18 hips]), intraoperative periprosthetic fracture (4% [eight hips]), and proximal femoral deformity (0.5% [one hip]), leaving 78% (164 hips) for analysis here. Their preoperative radiographs were evaluated by three independent assessors, including one senior orthopaedic surgeon, one senior resident, and one junior resident, using three methods. The first was digital templating using OrthoView software, which is a commercially available digital templating software program used worldwide and known to possess high accuracy. The other two methods were technically similar to one another, with the only difference being the interface: iPhone versus iPad. In both of those approaches, using the picture archiving and communication system measurement tool, we drew a circle on an acetabular radiograph to depict the cup. We took a photograph of the computer display and imported the photograph into slides of the Keynote program, a presentation software application for Mac computers, on both devices. It was then underlaid on transparent digital templates of the femoral stem, which were scanned from plastic templates and positioned at the center of each slide. We scaled the image to the template by adjusting the image size until a 15-cm straight line on the hip photograph was equal to the 15-cm scale markers of the template. All templating results were compared with the actual implanted cementless THA components to assess accuracy. All assessors were blinded to the information about the actual implants, and they were not involved in performing the surgical procedures. The intrarater and interrater reliabilities were analyzed using intraclass correlation coefficients and kappa values. The accuracy for predicting stem size for each type of the four stem designs was compared among the three methods.

We were able to predict the acetabular cup size within one size in 92% of hips (151 of 164) using OrthoView and in 92% (150 of 164) using the novel method (p > 0.99). The accuracies of the three methods were comparable for predicting a femoral stem size within one size (OrthoView: 90% [148 hips], iPhone: 93% [152 hips], and iPad: 91% [149 hips]; p = 0.78), and neck length (OrthoView: 96% [157 hips], iPhone: 96% [158 hips], and iPad: 97% [159 hips]; p = 0.95). Using OrthoView, the neck offset was correctly predicted in 80% (132 hips), compared to 85% (139 hips) when using the iPhone and 82% (134 hips) when using the iPad (p = 0.57). All methods showed substantial or excellent agreement regarding intrarater and interrater reliability. There was no difference in accuracy regarding any of the four femoral stem designs we evaluated (Avenir, Excia, ML taper, and Metha).

The digital templating technique for THA using an iPhone or iPad combined with a picture archiving and communication system demonstrated high accuracy, comparable to that of commercial digital templating software. This technique is reliable and reproducible for predicting a cementless prosthesis size, neck length, and offset in different types of femoral stems. It may be useful as an alternative in resource-constrained centers where commercial software programs are too expensive to be used in practice.

Level IV, diagnostic study.