Abstract
Introduction
The application of digital radiography in orthopaedic settings has facilitated the improvement in the retention and utilization of these images in pre and post-operative assessments [1]. In addition to the cost-effectiveness of such technology the use of digital imaging combined with advanced computer image processing software such as TraumaCadTM software system (TraumaCad, BRAINLAB, Westchester, IL, USA) can provide more accurate details about patients in total hip replacement arthroplasty (THA), a process traditionally called preoperative templating [2] by which intraoperative complications are minimized and overall surgical time is reduced[3]. In a study of 486 patients we demonstrated that patients demographic had significant effect on the outcome of the measurement and utilizing them in a predictive model had helped with improving the results [4]. In this study, we aimed to improve and optimize the proposed algorithm by utilizing more patients’ information and improving the model by using a nonlinear relationship. Our main hypothesis in this study was that the model would significantly predict the actual implant size based on the preoperative assessments.
Method
We analyzed the outcome of digital radiographs of 1018 patients who were treated with THA.
| Minimum | Maximum | Mean | Std. Deviation | ||
| Templated Acetabulum Size | 44.00 | 64.00 | 54.12 | 4.05 | |
| Height (m) | 147.32 | 202.20 | 172.02 | 10.73 | |
| Weight (kg) | 39.10 | 139.10 | 84.44 | 19.67 | |
| BMI | 15.48 | 43.06 | 28.33 | 5.18 | |
| Acetabular Size | 44.00 | 64.00 | 54.25 | 3.75 |
Digital radiographs were acquired in the anteroposterior view of the pelvis centered over the pubic symphysis. The hip was internally rotated 10° to 15°. We evaluated multiple interactions and nonlinear models and developed the most significant model based on the available clinical data.
Results
We derived the following equation based on the model presented by the multiple regression analysis
Act[estimated]=−9.0467–0.35*Act[temp]+34.79*Height−0.35*Weight+1.32*BMI+0.01* (Act[temp])∘2–3.56*Height∘2–0.01*BMI∘2
In which Act[estimated] is the estimated size of acetabulum cup, Act[temp] is the preoperative templated acetabulum size from digital radiography, Height was in m and Weight was in kg. Figure 1 and Figure 2 depicts the residual assessments of the model. Figure 3 depicts the range of effects by each factor.
Discussion
Patients’ specific data would improve the preoperative accuracy by more than 5% within one size of the actual acetabular component size. This improvement in accuracy translates into significant cost saving in THA cases as the cost of implant inventory could be significantly minimized. In our practice based on these assessments we use customized patients trays to reduce intraoperative costs.
