Accuracy of digital templating for total hip arthroplasty: android smartphone and tablet computer versus commercial templating software.

Abstract

Background: Preoperative radiographic templating plays an important role in optimizing total hip arthroplasty (THA). Digital templating software ensures precise implant selection, but can be costly and limited to select workstations. A new method using an iPhone/iPad with the picture archiving and communication system (PACS) offers comparable accuracy but is restricted by Apple's ecosystem. To improve accessibility, we adapted this method for Android smartphones and tablet computers, enabling broader use among surgeons. This study aimed to compare the accuracy and reproducibility of this novel method with a commercial digital templating software.

Methods: Radiographs of 124 hips were retrospectively templated by two independent assessors using three methods. The first used OrthoView® digital templating software. The other two, performed on an Android smartphone and tablet, utilized the PACS measurement tool. A circle was drawn on the acetabular radiograph to represent the cup, then a photograph of the display was imported into Microsoft PowerPoint 365®, where transparent femoral stem templates, scanned from plastic templates, were overlaid. Templating results were compared with implanted cementless THA components for accuracy. Intra-rater and inter-rater reliabilities were analyzed to assess consistency between and within assessors.

Results: Predicting the acetabular cup ± 1 Size could be achieved in 91.1% of cases (113 hips) by OrthoView® and 88.7% (110 hips) by the novel method (P = 0.674). The accuracies of three methods were comparable to predict ± 1 size of femoral stem [OrthoView® 90.3% (112 hips), smartphone 85.5% (106 hips), and tablet 87.9% (109hips), P = 0.526], and neck length [OrthoView® 94.4% (117 hips, smartphone 91.9% (114 hips), and tablet 93.5% (116 hips), P = 0.571]. The neck offset was correctly predicted using OrthoView® in 83.1% (103 hips), comparable with 81.4% (101 hips) using a smartphone and 85.5% (106 hips) using a tablet (P = 0.717). No different accuracy was found in each type of the 4 designs of the implanted femoral stems. All methods showed substantial and excellent agreement for intra- and inter-rater reliabilities.

Conclusions: Digital templating for THA using an Android smartphone, tablet, and PACS provides accuracy comparable to commercial software. It is reliable and reproducible for predicting cementless prosthetic size, neck length, and offset across femoral stem types.