Mylène Jansen

CT subchondral bone changes after KJD 389 18 bone thickness, male patients showed a smaller bone density decrease at the weight-bearing areas (data not shown), which may be associated with differences in response to KJD between male and female patients. It can only be speculated on this sex difference in response, but it is plausible that hormonal controlled bone density changes may be involved. For example, female (mild) OA patients have previously shown periarticular osteoporosis, while this was not seen in male patients. 25 In this respect bisphosphonate treatment in OA is subject of study. 26,27 The observed bone shape changes, although very exploratory and analyzed only visually, may indicate a reversal of typical OA changes, since both compartments of the tibia and femur seem to become less wide and flat. 28 As opposed to subchondral bone results, the bone shape changed the most in the second year after treatment. Radiographically evaluated osteophytes also showed an increased growth especially in this second year. 12 The inward difference that was seen on the outer edges may therefore also be a result of osteophyte-related changes, since increasingly large osteophytes might have affected the bone segmentation at follow-up, and with that its influence cannot be excluded. Irrespectively, it makes sense that shape changes, including osteophyte formation, show a somewhat delayed response, as they are the result of internal processes and remodeling of subchondral bone. 2 Unfortunately the osteophyte size could not be measured automatically on CT with the current analysis method. This study is clearly an explorative study regarding its sample size and the absence of a healthy control group as well as an untreated matched OA group. The sample size was small, which may be why there were only small areas with statistically significant changes, although they were largely in line with the general concept. KJD is still a relatively new treatment, and CT scans are not often included in studies and especially not in regular care. The observed changes agree with those found previously on radiographs. Furthermore, the 2 patients with a lateral MAC could be a mirrored control group, and the fact they showed opposite results (and as such both showed the same effect for the most affected compartment) is supportive to our conclusions. Notwithstanding, a healthy control group and a matched group of OA patients would have strengthened our conclusions significantly, although not treating patients with such severe OA for multiple years is (ethically) impossible. It also would have been worthwhile to include a calibration phantom during the CT scans, to enable measuring cortical bone mineral density, another useful parameter. Future studies should take these points into account to strengthen the concept of bone normalization upon distraction treatment as 1 of the underlying mechanisms of the observed clinical benefit. In conclusion, we have shown that bone changes after KJD treatment include thinning of the subchondral cortical bone plate, decrease of subchondral trabecular bone density, and normalization of bone shape in the first year sustaining towards the second year.