Mylène Jansen

MRI cartilage thickness after KJD and HTO 289 14 axis shift that is induced with HTO treatment and not due to actual cartilage thickness gain. Despite the increased loading on the LAC, it did not show significant changes in cartilage thickness. Any changes in cartilage thickness after HTO may therefore be solely natural progression, which means HTO treatment does not (quantitatively) affect the tibiofemoral cartilage 2 years after treatment. The axis shift appeared to offer enough relieve in itself, as these same HTO patients have previously shown a significant increase in clinical patient- reported outcomes over the 2-year period after treatment. 14,21,22 The results for HTO patients contradict those found in literature, where quantitative cartilage restoration was seen, although most studies used second-look arthroscopy to visually score cartilage restoration after HTO and discrepancies may be expected between such different scoring/measurement methods. 38 Two studies have suggested cartilage gain to occur after HTO using the same MRI measurement method as used in the current study, although in both cases the increases were not statistically significant. 39,40 Also, those studies measured the cartilage thickness 1 year after treatment instead of after 2 years. In the current study, MRI measurements were not performed at 1 year after treatment for HTO patients as the metal plate and screws were still present in the knee. A previous study demonstrated that while 2-year values were significantly improved compared to baseline, even better results were seen 1 year after treatment. 18,19 This observation suggests that the cartilage restoration in the current study could be an underestimation of the true initial restorative effect caused by KJD treatment. Similarly, it might be that the HTO patients would have shown a slight increase in cartilage thickness after 1 year, but subsequent loss of cartilage (natural progression) in the second year causes an overall negative 2-year effect. This also brings forward a limitation of our study in that it did not include a natural progression group with similar baseline characteristics as the HTO- and TKA-indicated patients. The question is whether such a population exists for TKA-indicated patients, because indication for TKA needs a past of ineffective conservative treatment. This may make it unethical to keep these patients on conservative treatment for an additional 2 years to serve as a control group. Despite severe versus mild OA being the strongest predictor of cartilage thickness changes after KJD, it should be noted that the R 2 value of this model was only 0.32, so only 32% of the group variance could be explained by the baseline OA severity. This might be the result of the small number of patients included in the analysis, so it would be worthwhile to perform these analyses in more KJD patients in the future. However, despite the small patient number, the between-group effect sizes for almost all comparisons were moderate to large when dividing groups by severity. This could mean that there are other important factors involved that determine the amount of cartilage restoration that were not included in this study, such as baseline cartilage quality measurements or metabolic joint condition reflected in e.g. synovial fluid marker levels. The fact that the significance of difference in denuded