Re: Association between bisphosphonate use and implant survival after primary total arthroplasty of the knee or hip: population based retrospective cohort study
CCBYNCOpen access
Rapid response to:
Research
Association between bisphosphonate use and implant survival after primary total arthroplasty of the knee or hip: population based retrospective cohort study
Re: Association between bisphosphonate use and implant survival after primary total arthroplasty of the knee or hip: population based retrospective cohort study
Bisphosphonates ‘extend hip replacement life’ according to a paper from Oxford[1], and the BBC [2]. Any reasonable physician would be therefore be inclined to consider bisphosphonates for patients who are due to undergo joint replacement. As none of the authors of this paper actually treat fractures, they may have missed steady stream of atypical fractures that are actually caused by this class of drug in the longer term, as the over 80 publications in the last two years on this subject suggest.
From a dataset spanning 20 years, which would include interesting data on both typical and atypical fractures, the authors choose to report a bizarre and blinkered view of ‘results’ over a very short time frame, focusing soley on revision. Their mean time to revision of 4 years (1.6-6.3 interquartile range) for 11 early revisions in 1912 (0.6%) patients on bisphosphonates are compared with 511 early revisions in 41,995(1.3%) ordinary patients whose principal complaint was osteoarthritis. They choose to report neither readmission for fractures at a lower rate for typical fractures, possibly prevented by bisphosphonates, nor at a higher rate for atypical fractures, potentially caused by them. One can only surmise why they passed up this opportunity to provide such information from such a large and interesting dataset.
The patients given bisphosphonates in this study from 1986 onwards were, not surprisingly, lighter, and more female, and thus probably less active, with less muscle and bone mass. The operations were probably easier to perform for the surgeons, and were most unlikely to have been used harshly following surgery, as both high levels of activity and obesity protect against osteoporosis, and thus prevent patients being prescribed bisphosphonates. The assumption that the mode of failure is aseptic loosening as a result of a chronic inflammatory response that recruits osteoclasts is pure conjecture and rather less likely at 4 years than technical errors by the surgeons, which are more likely in stronger boned, heavy, active people who are much more difficult to operate upon.
Revision is a very poor metric of outcome, as one of the authors of this paper recently highlighted [3]. The conclusions he discredited from a very large dataset had been used to promote a large and less successful procedure over a cheaper and more successful one. This paper invites similar criticism. The pressure to promote the wide prescription of expensive drugs, which have a half-life of decades and no known antidote, in a paper whose senior authors have received financial support from every bisphosphonate manufacturer has to be balanced by reporting of their primary efficacy and well documented and serious side effects. By neglecting to report on the elephant in the room, and focussing solely and oddly on early revision of joint replacement, the authors potentially create a much larger problem than the one that they are addressing.
1. Prieto-Alhambra, D.e.a., Association between bisphosphonate use and implant survival after primary total arthroplasty of the knee or hip: population based retrospective cohort study. BMJ, 2011. 343.
2. Gallagher, J. Bisphosphonates 'extend hip replacement life'. 2011; Available from: http://www.bbc.co.uk/news/health-16051151.
3. Goodfellow, J.W., J.J. O'Connor, and D.W. Murray, A critique of revision rate as an outcome measure: re-interpretation of knee joint registry data. J Bone Joint Surg Br. 92(12): p. 1628-31.
Rapid Response:
Re: Association between bisphosphonate use and implant survival after primary total arthroplasty of the knee or hip: population based retrospective cohort study
Bisphosphonates ‘extend hip replacement life’ according to a paper from Oxford[1], and the BBC [2]. Any reasonable physician would be therefore be inclined to consider bisphosphonates for patients who are due to undergo joint replacement. As none of the authors of this paper actually treat fractures, they may have missed steady stream of atypical fractures that are actually caused by this class of drug in the longer term, as the over 80 publications in the last two years on this subject suggest.
From a dataset spanning 20 years, which would include interesting data on both typical and atypical fractures, the authors choose to report a bizarre and blinkered view of ‘results’ over a very short time frame, focusing soley on revision. Their mean time to revision of 4 years (1.6-6.3 interquartile range) for 11 early revisions in 1912 (0.6%) patients on bisphosphonates are compared with 511 early revisions in 41,995(1.3%) ordinary patients whose principal complaint was osteoarthritis. They choose to report neither readmission for fractures at a lower rate for typical fractures, possibly prevented by bisphosphonates, nor at a higher rate for atypical fractures, potentially caused by them. One can only surmise why they passed up this opportunity to provide such information from such a large and interesting dataset.
The patients given bisphosphonates in this study from 1986 onwards were, not surprisingly, lighter, and more female, and thus probably less active, with less muscle and bone mass. The operations were probably easier to perform for the surgeons, and were most unlikely to have been used harshly following surgery, as both high levels of activity and obesity protect against osteoporosis, and thus prevent patients being prescribed bisphosphonates. The assumption that the mode of failure is aseptic loosening as a result of a chronic inflammatory response that recruits osteoclasts is pure conjecture and rather less likely at 4 years than technical errors by the surgeons, which are more likely in stronger boned, heavy, active people who are much more difficult to operate upon.
Revision is a very poor metric of outcome, as one of the authors of this paper recently highlighted [3]. The conclusions he discredited from a very large dataset had been used to promote a large and less successful procedure over a cheaper and more successful one. This paper invites similar criticism. The pressure to promote the wide prescription of expensive drugs, which have a half-life of decades and no known antidote, in a paper whose senior authors have received financial support from every bisphosphonate manufacturer has to be balanced by reporting of their primary efficacy and well documented and serious side effects. By neglecting to report on the elephant in the room, and focussing solely and oddly on early revision of joint replacement, the authors potentially create a much larger problem than the one that they are addressing.
1. Prieto-Alhambra, D.e.a., Association between bisphosphonate use and implant survival after primary total arthroplasty of the knee or hip: population based retrospective cohort study. BMJ, 2011. 343.
2. Gallagher, J. Bisphosphonates 'extend hip replacement life'. 2011; Available from: http://www.bbc.co.uk/news/health-16051151.
3. Goodfellow, J.W., J.J. O'Connor, and D.W. Murray, A critique of revision rate as an outcome measure: re-interpretation of knee joint registry data. J Bone Joint Surg Br. 92(12): p. 1628-31.
Competing interests: No competing interests