Comparison of total hip arthroplasty surgical approaches by Principal Component Analysis

Abstract

Gait adaptations are persistent after total hip arthroplasty and can depend on the type of surgery. This study focused on two surgical approaches: anterior and lateral. To analyze gait adaptations, biomechanical analyses usually employ an a priori selection of the parameters that leads to incomplete or redundant information. In contrast, Principal Component Analysis (PCA) provides an efficient transformation of the dataset by automatically identifying the major sources of variability. The purpose of this study was to investigate differences in level-walking among three groups of participants using PCA: patients undergoing an anterior surgical approach, patients undergoing a lateral surgical approach, and healthy controls. Biomechanical descriptions of the extracted principal components aided in the interpretation of the statistically significant results obtained from multivariate analysis of covariance (MANCOVA) tests. A point system was introduced to summarize the results and guide the interpretation. PCA captured reduced magnitude in sagittal and frontal moments in the anterior approach group, and reduced sagittal peaks angle in the lateral group, as previously found with traditional analyses. PCA also identified significant pattern delays in the anterior group, unnoticed in previous studies. In conclusion, neither surgical approach restored normal gait functionality because lower extremity kinetics and kinematics alterations persisted at 300-day follow-up after the surgery, regardless of the technique.