TEMPORAL DETERIORATION IN OUTCOME MEASURES AFTER KNEE ARTHROPLASTY REFLECTS NORMAL AGEING

Patient-reported outcome measures (PROMs) deteriorate over time after total knee arthroplasty (TKA). This study aimed to determine whether this represents joint-specific deterioration or age-related changes in health-related quality of life (HRQoL). A prospective cohort of 462 patients (68.7 years, 62.8% female) undergoing primary TKA were followed-up at 1- (n=214), 5- (n=317), 10- (n=299) and 15-years (n=116) postoperatively. Oxford Knee Score (OKS) and SF-6D (HRQoL) were completed at all timepoints. Linear mixed models assessed temporal change in OKS and HRQoL compared to matched population norms, and the association between HRQoL and OKS change. OKS declined from 39 (IQR 31–44) at 1-year to 35 (IQR 26–42) at 10-years postoperatively (p=0.002). SF-6D declined from 0.807 (IQR 0.645-0.887) at 1-year to 0.696 (IQR 0.557-0.858) at 10-years (p<0.001). However, this deterioration in OKS (−0.2, p=0.746) and SF-6D (−0.021, p=0.072) was comparable to that expected of a matched healthy population. After adjustment for observed postoperative SF-6D, the OKS was greater at 5-years (+1.4, p=0.010), 10-years (+0.5, p=0.454) and 15-years (+0.4, p=0.279) compared to population norms. At 15-year follow-up, 222 patients were deceased (46.8%), and OKS improved to 37 (IQR 27–44, p=0.002). Patients lost to follow-up after 10-year assessment had worse 10-year OKS and SF-6D (p<0.001). Temporal deterioration in the OKS after TKA reflects normal age-related HRQoL changes rather than a worsening in knee function. Attrition bias from revision, mortality, and dropout may result in artificially inflated PROMs at 15-year follow-up. These findings question the utility of long-term PROM collection after TKA.