Abstract
Research Type:
Level 3 - Retrospective cohort study, Case-control study, Meta-analysis of Level 3 studies
Introduction/Purpose:
Ankle sprains are common injuries, with 5 to 33% progressing to chronic conditions, including chronic ankle instability (CAI). CAI involves dysfunction in at least one of the three primary ankle ligament complexes (medial, lateral collateral, and syndesmosis), disrupting biomechanical stability and altering stress distribution. These changes may affect local bone density, particularly in the talus and mortise. We utilized weightbearing CT osteoabsorptiometry, employing Hounsfield units (HU) as a proxy to assess bone density as an indicator of biomechanical stress alterations across the joint.
Methods:
Twenty-seven ankles with chronic ankle instability (CAI), with at least six months of physical therapy, were compared to their contralateral uninjured ankles (control group). Using 3D Slicer software, the ankle mortise was segmented into six regions of interest: lateral malleolus, superior and inferior lateral talus, superior and inferior medial talus, medial malleolus (Figure 1). A mid-tibial diaphysis slice was used as a common denominator to normalize HU values for each ankle, as chronic changes in load bearing on each side would affect mid tibia and ankle bone similarly, as opposed to potential local changes due to the injury. Normalized HU values were compared between injured and uninjured segments across the six anatomical regions. Mean HU values were normalized against the ipsilateral tibial shaft. Paired t-tests, bootstrap resampling, and Bayesian t-tests assessed differences and significance, with statistical thresholds set at p < 0.05 and Bayesian factors (BF) interpreted as evidence strength.
Results:
No significant differences were found in normalized HU values between injured and uninjured segments (p > 0.05). A trend toward significance was observed for tibia values (p = 0.1), with lower injured values. Bayesian t-tests yielded BF < 1, indicating weak evidence against differences. Overall, results showed no consistent or significant differences in HU values between injured and uninjured ankles across all segments.
Conclusion:
These negative findings may indicate that CAI does not consistently alter bone density, potentially due to compensatory mechanisms or insufficient stress to induce detectable remodeling. Alternatively, the changes might be too subtle for detection with the sensitivity of our study. Furthermore, using contralateral ankles as controls may have limitations, due incipient changes in the "uninjured" ankle, potentially confounding results. Despite earlier studies using simpler methods reporting differences, this study found no significant subchondral bone density differences between injured and uninjured ankles in CAI. Further research with larger cohort is needed to better understand the relationship between CAI and bone remodeling.
