The Oxford Knee Score is a reliable predictor of patients in a health state worse than death and awaiting total knee arthroplasty

N D Clement; I Afzal; P Liu; K M Phoon; V Asopa; D H Sochart; D F Kader

doi:10.1186/s42836-022-00132-9

The Oxford Knee Score is a reliable predictor of patients in a health state worse than death and awaiting total knee arthroplasty

Arthroplasty. 2022 Aug 3;4(1):33. doi: 10.1186/s42836-022-00132-9.

Authors

N D Clement^{1

2}, I Afzal³, P Liu³, K M Phoon³, V Asopa³, D H Sochart³, D F Kader³

Affiliations

¹ Edinburgh Orthopaedics, Royal Infirmary of Edinburgh, Little France, Edinburgh, EH16 4SA, UK. nick.clement@nhslothian.scot.nhs.uk.
² Southwest London Elective Orthopaedic Centre, Epsom, UK. nick.clement@nhslothian.scot.nhs.uk.
³ Southwest London Elective Orthopaedic Centre, Epsom, UK.

Abstract

Background: The health-related quality of life of patients awaiting a total knee arthroplasty (TKA) deteriorates with increasing time to surgery and identification of those with the worst quality of life may help to prioritize patients. The aims were to identify and validate independent variable(s) associated with a health state worse than death (WTD) in patients awaiting a TKA and whether these variables influenced patients-reported outcome measures.

Methods: A retrospective cohort of 5857 patients undergoing a primary TKA was identified from an established arthroplasty database. Patient demographics, body mass index (BMI), index of multiple deprivation, Oxford Knee Score (OKS), EuroQoL five dimension (EQ-5D) 3 level, and visual analogue scale (EQ-VAS) were collected preoperatively and one year postoperatively. An EQ-5D utility of less than zero was defined as WTD. A randomly selected subset of patients (n = 3076) was used to validate the variable that was most predictive of a state WTD and to assess the influence on patient-reported outcomes.

Results: There were 771 (13.2%) patients with a health state WTD. Increasing social deprivation (P = 0.050), worse preoperative OKS (P < 0.001), or EQ-VAS (P < 0.001) were independently associated with a health state WTD. The OKS was the most reliable predictor (area under curve 88.9%, 95% CI 87.8 to 90.1, P < 0.001) of a health state WTD. A threshold value of 16 or less, 80% sensitive and specific, was validated and confirmed to have a negative predictive value of 97.5%. Patients with an OKS of 16 or less had a significantly greater improvement in their OKS (difference 6.9, P < 0.001) and EQ-5D score (difference 0.257, P < 0.001). When adjusting for confounding factors, a health status WTD was not associated with worse postoperative OKS (difference -0.6, 95% CI -1.4 to 0.3, P = 0.177), EQ-5D (difference -0.016, 95% CI -0.036 to 0.003, P = 0.097) or patient satisfaction (difference -1.8, 95% CI -4.3 to 0.7, P = 0.162).

Conclusion: A threshold score 16 or less in OKS was a reliable predictor of a health status WTD and was associated with a greater improvement in knee-specific and health-related quality of life following TKA.

Keywords: Arthroplasty; Knee; Outcome; Oxford knee score; Worse than death.