The relationship between patient expectations and patient-reported outcome measures (PROMs) after total knee arthroplasty (TKA) is not well understood. The purpose of the study was to test the influence of desired knee function on postoperative perceived knee function 1 year after TKA.

A total of 102 patients undergoing primary TKA were available for data analyses. Preoperatively, patients completed the Oxford Knee Score (OKS) twice, one representing preoperative function (preoperative OKS); the second representing desired function after TKA (desired OKS). Western Ontario and McMaster Universities Arthritis Index (WOMAC), University of California, Los Angeles (UCLA) Activity score, Hospital for Special Surgery Knee Replacement Expectations Survey (HSS-KRES), Patient Health Questionnaire-9, and EuroQol-visual analogue scales were obtained preoperatively. One year after surgery, all surveys besides the UCLA activity score and HSS-KRES were repeated. The associations between postoperative OKS and WOMAC versus desired OKS and HSS-KRES were assessed using multivariable linear regression models, wherein linear regression coefficients represent the additive effect on the mean postoperative PROM.

The desired OKS was independently associated with the postoperative OKS (linear regression coefficient = 0.43; P = .011), that is, each point increase in desired OKS yielded a 0.43 increase in postoperative OKS. The preoperative OKS showed no association with postoperative PROMs. Desired OKS was correlated with postoperative WOMAC (coefficient = −0.67; P = .014). The HSS-KRES was not associated with the postoperative OKS (coefficient = −0.005; P = .965) or WOMAC (coefficient = 0.18; P = .288).

In TKA patients, higher preoperative desired function predict greater postoperative PROMs. Effects of preoperative expectations on outcomes are independent of patient demographics or preoperative function.