O14 Predicting chronic ankle instability following a first-time lateral ankle sprain using clinical assessment: a prospective cohort analysis

Study DesignCohort study.ObjectivesTo investigate whether a clinical test battery conducted within 2 weeks of a first-time lateral ankle sprain (LAS) can be used to predict outcome (Chronic Ankle Instability [CAI] vs LAS ‘coper’) 12 months later.BackgroundNo prospective analysis is currently available which has sought to identify whether clinical assessment tools can be used to predict long-term recovery following a first-time LAS.Methods and MeasuresEighty-two individuals with first-time LAS were assessed using a clinical test battery within two-weeks of incurring a first-time LAS. These participants were classified 12 months later as having CAI or as being LAS ‘copers’ using the Cumberland Ankle Instability Tool (CAIT).Outcome measuresScores on the ‘talar-glide’ (deg), anterior-drawer, talar-tilt, figure-of-eight [figure8] for swelling (m), knee-to-wall (m) and hand-held goniometric range-of-motion [inversion; eversion; plantar-flexion {in degrees}] tests within two weeks of a first-time LAS, and scores on the CAIT 12 months later.ResultsSeventy (85%) of the original 82 injured participants completed the 12 month follow-up. Of the final seventy, 28 (40%) were designated as having CAI with 42 (60%) being designated as LAS copers. A logistic regression analysis revealed that a combined model using scores from the talar-glide, talar-tilt and anterior-drawer tests in addition to plantar-flexion ROM was statistically significant (p<0.01) and correctly classified 68.8% of cases. The final model had a sensitivity of 64% and a specificity of 72%.ConclusionsThis is the first analysis in which the predictive value of a clinical test battery for ankle sprain injury for determining CAI has been investigated. While our results showed that some of these clinical tests demonstrate predictive value, the accuracy at which they identify individuals at risk of developing CAI is moderate. Further research is required to determine whether performing these tests in a less heterogenous sample of individuals (perhaps within 48 hours of injury) would improve their predictive value.