Abstract
Introduction
Aseptic loosening is the main reason for total knee arthroplasty (TKA) failure, responsible for more than 25% of the revision procedures, with most of the problems occurring with the tibial component. While early loosening can be attributed to failure of primary fixation, late implant loosening is associated with loss of fixation secondary to bone resorption due to altered physiological load transfer to the tibial bone. Several attempts have been made to investigate these changes in bone load transfer in biomechanical simulations and bone remodeling analyses, which can be useful to provide information on the effect of patient, surgery, or design-related factors. On the other hand, these factors have also been investigated in clinical studies of radiographic changes of bone density following TKA. In this study we made an overview of the knowledge obtained from these clinical studies, which can be used to inform clinical decision making and implant design choices.
Methods
A literature search was performed to identify clinical follow-up studies that monitored peri-prosthetic bone changes following TKA. Within these studies, effects of the following parameters on bone density changes were investigated: post-operative time, region of interest, alignment, body weight, systemic osteoporosis, implant design and cementation.
Moreover, we investigated the effect of bone density loss on implant survival. Results
A total of 19 studies was included in this overview, with a number of included patients ranging from 12 to 7,760. Most studies used DEXA (n=16), while a few studies performed analyses on calibrated digital radiographs (n=2), or computed tomography (n=1). Postoperative follow-up varied from 9 months to 10 years.
Studies consistently report the largest bone density reduction within the first postoperative year. Bone loss is mainly seen in the medial region. This has been attributed to the change in alignment following surgery, during which often the pre-operative varus knee is corrected to a more physiological alignment, resulting in a load shift towards the lateral compartment. Measurements in unoperated contralateral legs were performed in 3 cases, and two studies performed standardized DEXA measurements to provide information on systemic osteoporosis. While on the short term no changes were observed, significant negative correlations have been found between severity of osteoporosis and peri-prosthetic bone density. No clear effects of bodyweight and cementation on bone loss have been identified. Although some studies do find differences between implant types, the variation in the data makes it difficult to draw general conclusions from these findings.
Several studies reported no effect of bone loss on implant migration. In another study, a medial collapse was associated with a medial increase in density, suggesting that altered loading and increased stresses are responsible for both bone formation and the overload leading to collapse.
Discussion
There are important lessons to be learned from these clinical studies, although generally the large spread in the DEXA data restricts strong conclusions. There is a large variation in used ROI definitions, complicating direct comparisons. Finally, most studies report density changes of well-functioning reconstructions, since only very large studies are able to gather enough failed cases.
