Distribution of Bone Mineral Density in the Ankle Joint: Correlation with Hindfoot Alignment

Abstract

Category: Ankle; Basic Sciences/Biologics; Hindfoot Introduction/Purpose: Abnormal Hindfoot Alignment (HA) has been correlated with increased failure rates in ankle fusion or replacement for osteoarthritis (OA). An altered stress distribution in the surrounding bone and abnormal Bone Mineral Density (BMD) around the native ankle may be predisposing factors for those unwanted outcomes. Cone Beam Weight Bearing CT (WBCT) has recently been used to investigate BMD and to correlate the localization of periprosthetic cysts and alignment in ankle arthroplasty. The objective of this study was to assess the spatial distribution of BMD around the ankle joint in patients with normal or abnormal HA. We hypothesized that BMD would be evenly distributed in normally-aligned ankles whilst increased medially in varus and laterally in valgus configurations. Methods: In this retrospective comparative Level III study, 60 ankles (41 adults),without any trauma or surgery affecting HA, with WBCT datasets (PedCat, Curvebeam LLC, PA-USA) were allocated to 3 groups (comparable by age, p=0.79; BMI, p=0.24; and side, p=0.93), based on the Foot Ankle Offset (FAO) values: 20 normal (0%-2), 20 varus (FAO5%; 40% female, age 58.9+-14.6 years, BMI 28.6+-4.2 kg.m-2). Semi-Automatic Segmentation (BoneLogic, Disior Oy, Helsinki-Finland) was applied to identify bones of interest. The tibia and talus were digitally compartmented in medial (M) and lateral (L) volumes relative to the median sagittal plane. Mean Hounsfield Unit (HU) value per compartment was used to assess BMD. The primary outcome measure was the Medial over Lateral HU ratio (M/L-HU). Comparisons were performed using one-way ANOVA, Kruskal-Wallis and Chi2 tests. Results: All values of BMD were normally distributed but M/L-HU ratios were not. Mean +- standard deviation HU values in the compartments in normal cases were 523+-103 (medial tibia), 519+-115 (lateral tibia) 421+-81 (medial talus), 470+-92 (lateral talus) and 725+-109 (fibula). The mean BMD was significantly lower in all compartments in valgus cases compared to normal (all p<0.05). It was decreased in valgus vs varus in the talus (p<0.04), specifically in the medial compartment (p<0.01). The tibia M/L- HU ratio was decreased in valgus vs normal (0.87+-0.16 vs 1.01+-0.07; p=0.001) and vs varus (1.04+-0.09; p<0.001). The talus M/L-HU ratio was increased in varus cases vs normal (1+-0.22 vs 0.83+-0.09; p<0.01) and valgus (0.07+-0.19; p<0.001). Conclusion: We found that BMD in distal tibia, fibula and the talus varies with hindfoot alignment. In valgus configuration, all bone compartments were less dense compared to normal and varus. The medio-lateral ratio increased in the tibia and in the talus in varus cases, suggesting medial concentration of bone, and decreased in valgus cases, suggesting a more lateral concentration. This data supports the role of WBCT in analyzing BMD distribution. This method could be clinically useful in ankle OA to evaluate bone quality for such considerations as surgical indications or implant positioning.