Predictors of functional outcome following femoral neck fractures treated with an arthroplasty: limitations of the Harris hip score

INTRODUCTION: To study the association between potential prognostic factors and functional outcome at 1 and 5 year follow-up in patients with femoral neck fractures treated with an arthroplasty. To analyze the reliability of the Harris hip score (HHS). MATERIALS AND METHODS: A multicenter analysis which included 252 patients who sustained a femoral neck fracture treated with an arthroplasty. Functional outcome after surgery was assessed using a modified HHS and was evaluated after 1 (HHS1) and 5 (HHS5) years. Several prognostic factors were analyzed and reliability of the HHS was assessed. RESULTS: After 1 year the presence of co-morbidities was a significant (p = 0.002) predictor for a poor functional outcome (mean HHS1 71.8 with co-morbidities, and 80.6 without co-morbidities). After 5 years none of the potential prognostic factors had significant influence on functional outcome. Internal consistency testing of the HHS showed that when pain and function of the HHS were analyzed together, the internal consistency was poor (HHS1 0.38 and HHS5 0.20). The internal consistency of the HHS solely in function (without pain) improved to 0.68 (HHS1) and 0.46 (HHS5). Analyzing the functional aspect exclusively, age and the existence of co-morbidities could be defined as predictors for functional outcome of femoral neck fractures after 1 and 5 years. CONCLUSION: After using the HHS in a modification, age and the existence of pre-operative co-morbidities appeared to be predictors of the functional outcome after 1 and 5 years. The HHS, omitting pain, is a more reliable score to estimate the functional outcome, than HHS analyzing pain and function in one scoring system.