Methods

We retrospectively included 48 patients who undergone THA revisions in our center between Aug, 2013 and Mar, 2014. For patients with hybrid fixation as their primary prosthesis, the femoral or acetabular components with cement fixation were excluded. There were 41 cementless femoral stems and 35 cementless acetabular cups remained for evaluation. All patients took anterior-posterior and lateral view x-ray examination, anterior-posterior T-smart tomosynthesis scan, and computed tomography before revision surgery. As the gold standard, intraoperative pull-out tests and twisting tests were done for every patient to examine the stability of all implants. 7 orthopedic surgeons evaluated the preoperative images independently, who were divided into the senior group (3 doctors with 6∼13 years’ clinical experience) and the junior group (4 doctors with 2∼4 years’ clinical experience). The x-rays were evaluated first, followed by computed tomography 4 weeks later, and after another 4 weeks’ interval the T-smart tomosynthesis were assessed. All doctors used the same criteria for diagnosis. Diagnostic accuracy for each imaging examination was calculated by comparing with the results of intraoperative tests. The diagnostic accuracy, kappa values between 3 imaging techniques were calculated, and chi-square tests were conducted to examine the difference between the senior and junior groups for each technique.