Determining Trendelenburg test validity and reliability using 3-dimensional motion analysis and muscle dynamometry

Background: The hip abductor muscle group stabilises the pelvis during gait to prevent excessive pelvic drop. Hip abductor weakness has been linked to musculoskeletal conditions such as chronic low-back pain. As such, it is important that practitioners can correctly diagnose hip abductor weakness in a clinical setting. Although the Trendelenburg test is commonly used by practitioners, the validity of this test to assess hip abductor weakness in the absence of musculoskeletal injury remains questionable. The aim of this study was to determine the validity of the Trendelenburg test, as observed by a practitioner, to assess frontal plane pelvic motion and hip abductor strength in a population without intra-articular hip disorders. Methods: This study was performed between June 14th and October 16th 2019. Eighteen participants were recruited for this study. Peak normalised isometric and isokinetic hip abductor torque were measured bilaterally (n = 36) using the Biodex System 4 isokinetic dynamometer. Each participant performed the Trendelenburg test bilaterally (n = 36) while a graduate year chiropractic practitioner assessed for a “positive” or “negative” sign. The test was simultaneously recorded using Vicon 3-Dimensional motion capture to measure frontal plane pelvic motion and elevation. Correlation analyses were performed between the measures of peak hip abductor torque and pelvic motion to determine if any relationship existed. Agreement between the practitioner and 3-Dimensional analysis was calculated using the kappa (κ) statistic. Results: Weak, non-significant correlations were found between hip abductor strength and pelvic motion before outlier removal. Significant (p < 0.05) yet weak correlations were found after outlier removal, except for isometric hip abductor strength. Weak agreement was found between the chiropractic practitioner and 3-Dimesnional analysis for the Trendelenburg test assessment (κ = 0.22–0.25). Conclusions: This study found no significan