Redefining Hindfoot Alignment Based on Pathology rather than Morphology: A Prospective Observational Diagnostic Study on 250 Feet

CATEGORY: Ankle; Ankle Arthritis; Bunion; Hindfoot; Midfoot/Forefoot; Other INTRODUCTION/PURPOSE: The analysis of the hindfoot alignment (HA) is important in the management of various foot and ankle pathologies. Foot Ankle Offset (FAO), measured using Weight Bearing CT (WBCT) scans is a 3D biometric measurement of HA described in the literature, however only with relation to the clinical morphology of the hindfoot or previously known 2D biometrics such as the hindfoot angle. The aim of this study was to observe the FAO distribution on a continuous, large population of patients and to analyze its discriminating power between pathological and non-pathological cases. We hypothesized that threshold values could be determined for the normal range of FAO in terms of risk of presenting with different pathologies in varus or valgus configurations. METHODS: Prospective, IRB approved (COS-RGDS-2016-06-008-P-LINTZ-F, Clinical Trials NCT 04134962), level II study. 125 patients (250 feet) with bilateral WBCT as standard follow-up were included at a single institution. Mean age was 54 years (18-84), 58.4 % were female. Each patient was clinically evaluated, scans analyzed by a fellowship-trained foot and ankle surgeon and all degenerative pathologies classified depending on their anatomical location (medial or lateral). HA was measured using FAO on 3D datasets and Tibio-Calcaneal Angle (TCA) on 2D Saltzman-El-Khoury views. All scans were analyzed a second time by a radiology MD. FAO and TCA Intraclass Correlation Coefficient (ICCs) and Spearman’s correlation coefficient were calculated. Mean FAO values were calculated for all, normal, varus and valgus cases, and each pathology group. Receiver operating Curves (ROC), threshold values of FAO and area under the curve (AUC) were established for predicting an increased risk of medial or lateral pathologies. RESULTS: A threshold FAO value of -1.64% was found to best predict the risk of lateral pathology (51.4% sensitivity, 85.1% specificity, AUC=0.72) and 2.71% (95% sensitivity, 82.8% specificity, AUC=0.93) for medial pathology, both versus no pathology. Mean FAO value was 1.65% +-4.72% and mean TCA was 4.15° +-7.67°. Interobserver reproducibility for FAO and TCA was respectively 0.96 [95% CI 0.95-0.97] and 0.95 [95% CI 0.94-0.96]. Spearman’s correlation coefficient between FAO and TCA was 0.697. Clinical assessment rated 167 feet as normal, 33 varus, and 50 valgus. Mean FAO values were respectively 0.99 +-3.26%, - 2.53+-5.05 % and 6.81+-2.70%. We found that the mean FAO was 0.42+-3.19 % for non-pathologic feet, -2.30+-4.58 for lateral pathologies, 6.62+-2.77 for medial pathologies. CONCLUSION: The most important finding in this prospective clinical trial is the description of normal HA as the FAO range in which the risk for degenerative Foot and Ankle pathology is the least: -1.64% to 2.71%. This new, pathology based prognostic approach to HA is different to the traditional morphological approach. The potential of WBCT to provide computerized, data- based and biomechanically meaningful 3D measurement tools could improve the prognostic potential of biometrics such as HA. Results will be refined in the future by increasing the size of study population.