The Oxford hip score demonstrates moderate ceiling effects at one and two years after total hip arthroplasty: which patients are at risk and does it matter?

Abstract

Background: The aim was to assess whether the postoperative Oxford Hip Score (OHS) demonstrated a ceiling effect at 1 or 2 years after total hip arthroplasty (THA) and to identify which patients are more likely to achieve a ceiling score and whether this limits assessment of their outcome.

Methods: A retrospective cohort of 7871 patients undergoing primary THA was identified from an established arthroplasty database. Patient demographics, ASA grade, socioeconomic status, OHS and EuroQol questionnaire were collected preoperatively and at 1 and 2 years postoperatively. Regression analysis was used to identify independent preoperative factors associated with achieving a ceiling score. Receiver operating characteristic curves were used to identify preoperative OHS's that predicted a postoperative ceiling score.

Results: The ceiling effect (proportion achieving the maximal score) at 1 year was 21.8% (n = 1372) which increased significantly (p < 0.001) to 26.6% (n = 1569) at 2 years. Female gender (p ≤ 0.028), younger age (p < 0.001), decreasing socioeconomic deprivation (only for 2-year OHS), a better preoperative OHS (p < 0.001) or EQ-VAS (p < 0.001) were independently associated with a ceiling OHS postoperatively. The preoperative OHS was demonstrated to be a  poor discriminator of achieving postoperative ceiling score at 1 year (AUC 62.4%, 95% CI 60.7 to 64.1, p < 0.001) and 2 years (AUC 61.5%, 95% CI 60.0 to 63.2). Those achieving a postoperative ceiling OHS at 1 and 2 years had statistically significant (p < 0.001) greater improvements in their OHS, EQ-5D and EQ-VAS and were more likely to have achieved a minimal important change in their OHS relative to their preoperative baseline and a postoperative patient acceptable symptom state.

Conclusion: The OHS demonstrated moderate ceiling effects at both 1 and 2 years following THA, and the preoperative score was a predictor of achieving a ceiling score. However, it would seem the ceiling effect did not limit the potential for improvements relative to baseline and achieving clinically meaningful values in the OHS.