Background Ankle arthrometers are available to objectively assess ankle joint laxity and range-of-motion, and have been useful in the research environment. This study is unique in that our institution has access to two of the same ankle arthrometers, which enabled us to efficiently study both the reliability and validity of the derived joint laxity measurements.

Objective To estabilish the reliability and validity of instrumented ankle arthrometer measurements from two identical ankle arthrometers.

Design A crossover study design where both examiners tested each subject with one arthrometer the first session and the other arthrometer the second session.

Setting Athletic training research laboratory.

Participants Nineteen participants (6 male, 13 female, age = 21.4 ± 3.3 years, height = 168.0 ± 8.8 cm, body mass = 66.4 ± 12.0 kg) with no history of ankle injuries.

Interventions Ankle arthrometry measurements were performed on two separate occasions, separated by at least seven days. Two identical ankle arthrometers (Blue Bay Research Inc., Milton, FL) were used by two different examiners. Both examiners tested each participant with one arthrometer in the first session and with the other arthrometer in the second session. Anterior displacement (mm) and inversion/eversion rotation (deg) were assessed utilising both ankles. Measurements were analysed using LabVIEW software (National Instruments, Austin, TX). Examiners were blinded to the results of their own measurements to reduce bias.

Main outcome measurements Dependent variables included anterior displacement and inversion/eversion rotation. The independent variables were arthrometer and examiner. Reliability and validity were determined using interclass correlation coefficients (ICC).

Results ICC’s for inter-rater reliability (left) and intra-rater reliability (right) are provided in the table below. The highest inter-rater reliability was 0.97 (left inversion). The lowest inter-rater reliability was 0.80 (left anterior displacement). Intra-rater reliability values varied from 0.25 (left anterior) to 0.92 (right inversion).

Conclusions This study confirms previous reports supporting the reliability of ankle arthrometry measurements. However, inconsistencies in the ICC values, especially those for anterior displacement, question the validity of the measurements derived from the two identical arthrometers.