Abstract
Research Type:
Level 3 - Retrospective cohort study, Case-control study, Meta-analysis of Level 3 studies
Introduction/Purpose:
Bone mineral density (BMD) is an important factor that can influence decision making in the operative treatment of ankle arthritis. Therefore, it is crucial to understand patterns of BMD in arthritis so surgeons can choose strategies and implants to optimize success for each patient. To date, no study has addressed BMD differences in arthritic and non-arthritic ankles. The goal of this study was to 1) analyze BMD in varus arthritic patients in comparison with controls and 2) investigate how ankle and foot deformity impact BMD. We hypothesized that in ankles with varus alignment, BMD will be increased on the medial side (where bone is overloaded) and decreased on the lateral side (where bone is underloaded) for tibias and taluses when compared to controls.
Methods:
This is a retrospective review of a primary total ankle replacement (TAR) database of patients who underwent preoperative weight-bearing CT (WBCT) using a standardized patient-specific instrumentation. Ankle deformity was assessed using the tibiotalar alignment angle (TTA), while foot deformity was calculated using foot and ankle offset (FAO). Tibial and talar BMD were measured via a standardized technique by two observers using Hounsfield units (HU) (Figure). Contralateral non-arthritic ankles were used as controls. Differences between BMD of arthritic and control ankles were evaluated using independent samples t tests. To assess the role of deformity in BMD, two comparisons were made. First, 59 patients with varus foot alignment (FAO <-0.6) were compared to 34 controls with neutral FAO. Second, 37 patients with varus ankle alignment (TTA < -5°) were compared to 34 controls with neutral TTA. Multiple linear regression models were used to investigate the relationships of BMD to FAO and TTA.
Results:
We found no difference between tibial or talus BMD comparing varus FAO arthritic patients (mean FAO – 6.57) and control patients (mean FAO – 0.19). Similarly, no difference was noticed in BMD comparing varus TTA arthritic patients (mean – 12.1°) with neutral TTA controls (mean 0.01). Pairing per patient, talar BMD is higher than tibial BMD by up to 300HU. By linear regression, we found significant statistical correlation for talar BMD and varus alignment (based in both FAO and TTA; R2 0.461 and 0.649, respectively) (figure 1C-1F). No correlation between tibial BMD and FAO or TTA was noticed.
Conclusion:
This study reveals relevant information for understanding how arthritis and deformity interferes with bone quality. Interestingly, there was no significant association between tibial BMD and markers of ankle or foot deformity. However, there was a statistically significant correlation between talar BMD and varus deformity. To our knowledge, this is the first study not only to compare BMD between arthritic ankles and controls, but also to demonstrate that talar BMD is correlated with varus deformity, which doesn’t seem to be the case for tibial BMD.
Figure 1
A and B – Tibial and talar BMD were measured 10mm above and 6mm below the ankle joint, respectively, where most of the total ankle arthroplasty resections are performed. Both bones were divided into quadrants based on the center of the bone and the syndesmotic axis for the tibia and the lateral gutter axis for the talus. C – F shows linear regression models demonstrating the correlation between varus alignment (FAO – C and D; TTA – E and F) and medial and lateral talar bone density.