Abstract
We thank Drs Daungsupawong and Wiwanitkit for their positive response to our research, 1 their suggestion of potential confounding factors for consideration, and their interesting ideas for future avenues of related research. In the following, we will respond in detail to the concerns and suggestions raised by the authors.
In our study, we quantified the association of morphology and alignment metrics with cartilage damage progression over 2 years. Metrics were obtained from baseline imaging in a cohort with mild or no osteoarthritis (OA) and compared with longitudinal data over 2 years. While longer-term follow-up data are available in the Multicenter Osteoarthritis Study (MOST), we intentionally chose a 2-year follow-up to limit the focus to OA changes related to the metrics while minimizing the impact of confounding factors. We did not look at longer-term data, but certainly agree this could be a fruitful area for future research.
We agree with the authors that natural cubic splines can be hard to interpret in their numerical form; therefore, we included graphs and an explanation for model selection in the supplementary information. We hope the authors, as well as any other interested readers, are able to access the supplement (https://journals.sagepub.com/doi/suppl/10.1177/03635465251367716/suppl_file/sj-pdf-1-ajs-10.1177_03635465251367716.pdf), available in the online version of the American Journal of Sports Medicine.
The MOST study recorded a multitude of cofounding factors—including many of those suggested by the authors. We selected body mass index, age, and sex as the factors most likely to be relevant based on the literature and our previous work. While other confounding factors might indeed play a role, we believe the relatively short follow-up time of 2 years and relatively large sample size minimize their potential effects.
We agree that measurements are just markers of underlying morphological features, in our case, trochlea morphology and leg alignment. Their interpretation of the entry point to the transition point (EP-TP) could have come directly out of our previous work. 2 We thank them for their understanding.
Our work addressed whether patellofemoral morphology can predict OA progression in asymptomatic patients. We wholeheartedly agree with the authors that our study lays a foundation for further investigation, and we hope to answer their questions, among others, with our ongoing and future work.
