Tracking Polyethylene Wear in Total Ankle Replacement: A Longitudinal Study Using Weight-Bearing CT

Research Type: Level 3 - Retrospective cohort study, Case-control study, Meta-analysis of Level 3 studies Introduction/Purpose: Assessing polyethylene (PE) wear in total ankle replacement (TAR) has traditionally relied on PE height and indirect signs such as cyst formation and implant loosening on weightbearing 2D radiographs or non-weightbearing 3D CT scans. On the contrary, functional loading on Weightbearing CT (WBCT), combined with Distance Mapping (DM), offers a novel volumetric method to quantify PE wear over time. This study aimed to assess volumetric PE wear using DM in a series of TAR cases with successive follow up WBCT scans. We hypothesized that PE volume would progressively decrease over time and would correlate with implant alignment metrics. Methods: This retrospective cross-sectional study included 29 ankles (16 male, mean age 57.9 y±10.5, BMI 26.3 Kg/m2±4.1, Foot Ankle Offset (FAO) 4.4%±3.2) with SALTO fixed-bearing implants. Two weight-bearing CT follow-up assessments were performed at minimum six months postoperatively, and minimum two years between the earliest and latest follow-ups. Demographics, alignment metrics (FAO, coronal, axial, sagittal rotation and translation) and clinical outcomes (AOFAS ,Pain and range of motion) were recorded. Polyethylene (PE) volume was measured using segmentation and Distance Mapping (DM) via Bonelogic® software (Paragon28, Englewood, CO, USA). Normality was assessed with the Shapiro-Wilk test. Paired Student’s t-tests were used for comparisons. Correlations were analyzed using Spearman’s coefficient and group differences with a Mann-Whitney U test. Predictors of polyethylene insert wear were identified using LASSO regression (5-fold cross-validation). Selected variables underwent multicollinearity checks (VIF < 5), followed by bootstrap multivariate regression (1000 iterations, sampling with replacement), reporting coefficients and 95% confidence intervals. Results: Mean follow up was 5±1.5 years (minimum 2 years). Mean polyethylene (PE) volumes at the first and last follow-up were 3.92 mL (SD = 1.19) and 3.86 mL (SD = 1.13), respectively. Mean difference was -0.054 mL ±0.11 (p = 0.015). Changes in PE wear correlated with implant positioning, showing a moderate correlation with coronal rotation (ρ = -0.59, p < 0.001), mediolateral translation (ρ = -0.54, p = 0.002), and sagittal tibio-talar approximation (ρ = 0.59, p < 0.001). Outliers did not affect results significance. Increased BMI above 27 was linked to greater wear (p = 0.033). Coronal plane mismatch of the implants, (β=−0.021, CI[−0.029,−0.011]), distal migration of the tibial component (β=0.019, CI[0.002,0.063]) and malrotation (β=−0.007, CI[−0.011,−0.002]) significantly influenced PE wear. Conclusion: These results provide in vivo evidence of PE wear in TAR using WBCT-based DM. Measured wear rate (~25 mm³/year) aligned with prior published estimates, influenced by implant positioning and BMI. Identification of multiplanar implant misalignments as independent predictors of PE wear underscores the importance of precise implant positioning and optimal bilateral component fit for implant longevity. Limitations included inconsistences in tibia positioning relative to the foot which led to outliers, but did not affect result significance. Larger studies with standardized positioning protocols are necessary to validate these findings and establish clinical thresholds.