Original Article
Templating in uncemented total hip arthroplasty—on intra- and interobserver reliability and professional experience
Abstract
Background: This study examines the intra-, and interobserver reliability of digital templating in uncemented total hip arthroplasty (THA), and assesses whether these values are dependent on professional experience.
Methods: Three independent observers retrospectively examined digital X-rays of 34 consecutive hips scheduled for uncemented THA. These were templated using templating software. Evaluations were carried out on two occasions at least 6 weeks apart. Findings were compared to each surgeon’s own findings, and then to the other surgeons’ findings. Data underwent statistical analysis to assess and describe reliability.
Results: The intraobserver reliability of the method was found to be good. The intra-class correlation coefficient (ICC) for individual surgeons ranged from 0.81 to 0.87 for acetabular components and 0.74 to 0.91 for femoral components. However, it was somewhat lower for neck length with kappa statistics (κ) from 0.41 to 0.51 with agreement in about 70% of the cases. Interobserver reliability was similar, with an ICC of 0.87 for the acetabular component and 0.79 for the femoral component, but somewhat lower for neck length with κ of 0.27 and agreement in 41% of the cases. We found no association between increasing experience and increasing precision, as the least experienced observer showed the highest intraobserver reliability.
Conclusions: The reliability of digital templating of uncemented THA is good for acetabular and femoral components, but inferior for neck length. Precision does not rely on professional experience. Digital templating provides surgeons with a valuable tool for preoperative planning, but cannot supersede the intraoperative assessment and final decision.
Methods: Three independent observers retrospectively examined digital X-rays of 34 consecutive hips scheduled for uncemented THA. These were templated using templating software. Evaluations were carried out on two occasions at least 6 weeks apart. Findings were compared to each surgeon’s own findings, and then to the other surgeons’ findings. Data underwent statistical analysis to assess and describe reliability.
Results: The intraobserver reliability of the method was found to be good. The intra-class correlation coefficient (ICC) for individual surgeons ranged from 0.81 to 0.87 for acetabular components and 0.74 to 0.91 for femoral components. However, it was somewhat lower for neck length with kappa statistics (κ) from 0.41 to 0.51 with agreement in about 70% of the cases. Interobserver reliability was similar, with an ICC of 0.87 for the acetabular component and 0.79 for the femoral component, but somewhat lower for neck length with κ of 0.27 and agreement in 41% of the cases. We found no association between increasing experience and increasing precision, as the least experienced observer showed the highest intraobserver reliability.
Conclusions: The reliability of digital templating of uncemented THA is good for acetabular and femoral components, but inferior for neck length. Precision does not rely on professional experience. Digital templating provides surgeons with a valuable tool for preoperative planning, but cannot supersede the intraoperative assessment and final decision.
