Metabolic bone disease and arthroplasty loosening

Joint degeneration requiring arthroplasty surgery and the consequences of osteoporosis are the two fundamental pathologies in orthopaedics. There are around 44,000 Medline-indexed journals about osteoporosis, and around 30,000 concerned with arthroplasty. However despite both typically occurring in a similar elderly population, only 350 (less than 0.5%) are cross-indexed. Aseptic loosening is the commonest cause of hip arthroplasty failure, with revision surgery being the only current treatment. Recent work has increased the understanding of the aetiology of aseptic loosening and studies suggest that this process may be inhibited by the use of drugs that are normally used to treat osteoporosis, such as the bisphosphonates. It has also been shown that the occult incidence of metabolic bone disease may be as high as 40% in patients undergoing primary hip arthroplasty. This study is a progression of similar work on the aetiology and control of aseptic loosening done in the same department over the proceeding few years. In the first instance a cellular model of aseptic loosening was investigated by Ong and Taylor [published in 2003]. This laboratory based project used mouse bone, and exposing it to interface membrane tissue sampled at the time of revision arthroplasty surgery. This model was described by Reynolds and Dingle in 1970, and shown to activate osteoclasts. Ong and Taylor demonstrated that osteoclast activation could be inhibited with doxycycline, suggesting that matrix metalloproteinases may be important in the pathophysiology of aseptic loosening, and that the process is potentially preventable. The work was progressed further by Ibrahim and Taylor [2004] who developed a live model of particle induced osteolysis. They measured radio-labelled calcium uptake in mouse femora following implantation of ceramic particles, sham surgery and in controls. This was shown to be a useful model of quantifying osteolysis, although they did not find a difference between the controls and those exposed to ceramic particles. The original aim of this work was to follow on from the previous work and demonstrate that osteolysis could be inhibited or reversed using pharmacological agents. Ideally this would be done in a human clinical model, and a number of drugs were considered, including doxycycline, bisphosphonates and statins. Such a project would have involved recruiting patients to a clinical trial, followed by either randomisation to treatment or control groups before commencing treatment on participants. The ideal end-point would be revision for aseptic loosening (although radiological development of loosening would be an alternative). Because hip arthroplasty is such a successful operation these end-points are both rare and often not seen for many years. Even if we assume a rather optimistic reduction in loosening of 50% using our agent, we would have to recruit several hundred participants and wait at least 10 years to get meaningful results. We therefore have had to sacrifice some of the principles of strong research in favour of a project that could be completed with a limited time-frame and a limited budget. We studied patients that had already had an arthroplasty in situ for a number of years, and in view of the multi-factorial nature of loosening (as discussed below), limited this to one type of arthroplasty. The hypothesis of this study is that patients who have an underlying disorder of bone metabolism (such as osteoporosis or vitamin D deficiency) are more likely to develop aseptic loosening. In addition we hypothesise that there are measurable clinical, radiographic and biochemical markers that help predict those likely to develop loosening. This hypothesis was investigated in 127 patients (78 patients with a loose cemented total hip replacement matched by age, gender, race, prosthesis and time from surgery with 49 patients with a well-fixed stable hip replacement)/ We then conducted four connected studies involving, clinical, radiological, DEXA and biochemical assessment for markers of loosening. The aims are detailed below, but were principally to see whether patients with loosening are more likely to markers of osteoporosis or poor bone health. Unfortunately, this study takes us no further forward with regard to whether aseptic loosening can be inhibited by specific therapeutic agents, but hopefully it helps us to better understand the pathophysiological processes involved with arthroplasty failure. These can be used in future research to help improve arthroplasty function and longevity.