Early Predictors of Outcomes Following Non-Operative Management of ACL Injury: The Delaware-Oslo ACL Cohort Study

OBJECTIVES: Identification of patient characteristics early after injury that influences long-term outcomes is needed to guide appropriate decision-making with regard to surgical management. The purpose of this study was to determine factors early after ACL injury which predict outcomes following non-operative management of ACL injury. METHODS: 59 (52.5% F; mean age 31.3±10.7 yrs) athletes completed rehabilitation to resolve knee joint range of motion, effusion, pain and gait impairments (quiet knee) after ACL injury. Patients then completed 10 additional rehabilitation sessions over 5 weeks consisting of progressive strengthening and neuromuscular training. Quadriceps strength testing, 4 single-legged hop tests, the Knee Outcome Survey Activities of Daily Living Scale (KOS), Global Rating Scale for Perceived Function (GR), and International Knee Documentation Committee 2000 form (IKDC) were completed following these 10 sessions. Two years after non-operative rehabilitation patients again completed the IKDC. “Normal” knee function was defined as an IKDC score in the top 85% of scores reported by uninjured people of the same age and sex, with “below normal” knee function equal to scores in the bottom 15% of normative data. Fisher's exact tests and Mann-Whitney U tests were used to test differences in baseline and clinical measures after non-operative rehabilitation between those with “normal” and “below normal” knee function at 2 years. A logistic regression model was used to identify factors predictive of 2 year knee function. A prior significance level was set at p≤.05. RESULTS: 11 patients (18.6%) reported knee function <15% on the IKDC at 2 years (Normal: 93.8±4.5%, 95% CI: 92.3-95.1; Below Normal: 72.1±12.7%, 95% CI: 63.6-80.6). No group differences existed for age (p=0.613) or sex (p=0.320) between those who scored in the normal range and those who scored <15%. Patients with self-reported IKDC knee function <15% had significantly lower IKDC scores at baseline (p=0.010; Normal: 83.0±10.0%, 95% CI: 80.1-85.9; Below Normal: 72.7±10.8%, 95% CI: 64.9-80.4) and required a greater number of days to achieve a quiet knee after injury (p=0.005; Normal: 58.5±21.9 days, 95% CI: 52.1-64.8; Below Normal: 77.6±17.7 days, 95% CI: 65.7-89.5). A trend toward significance was present for quadriceps strength (p=0.076; Normal: 93.5±9.1%, 95% CI: 90.8-96.1; Below Normal: 88.4±7.4%, 95% CI: 83.4-93.4) but no group differences were present for hop scores (single: p=0.684; crossover: p=0.630; triple: p=0.724; 6-meter timed: p=0.341), KOS (p=0.119), or GR (p=0.136). A logistic regression model including IKDC, days to achieve a quiet knee, and quadriceps strength was statistically significant with IKDC and days to achieve a quiet knee being significant predictors of knee function at 2 years (p=0.001; R2=0.433; IKDC: p=0.040, OR=0.909; days to achieve a quiet knee: p=0.014, OR=1.054; quad strength: p=0.220, OR=0.941). CONCLUSION: Only 18% of active individuals managed non-operatively after ACL injury had IKDC scores <15% of norms 2 years after injury and rehabilitation. Lower baseline subjective knee function was predictive of poorer self-reported non-operative outcomes despite similar hop scores to those reporting normal knee function at 2 years. Longer time from injury to resolution of initial knee impairments also predicted poorer self-reported knee function, highlighting the importance of early rehabilitation with aggressive intervention to resolve impairments.