Preoperative Patient-Reported Outcome Measure Thresholds Should Not be Used for Indicating Total Knee Arthroplasty

While Medicare requires patient-reported outcome measures (PROMs) for many quality programs, some commercial insurers are requiring preoperative PROMs when determining eligibility for total knee arthroplasty (TKA). Concerns exist that these data may be used to deny TKA to patients above a specific PROM score, but the optimal threshold is unknown. We aimed to evaluate TKA outcomes based on theoretical PROM thresholds.

We retrospectively analyzed 25,246 consecutive primary TKA patients from 2016 to 2019. Hypothesized preoperative knee injury and osteoarthritis outcome score for joint replacement cutoffs of 40, 50, 60, and 70 points were used. Preoperative scores below each threshold were considered "approved" surgery. Preoperative scores above each threshold were considered "denied" surgery. In-hospital complications, 90-day readmissions, and discharge disposition were evaluated. One-year minimum clinically important difference (MCID) achievement was calculated using previously validated anchor-based methods.

For "denied" patients below thresholds 40, 50, 60, and 70 points, 1-year MCID achievement was 88.3%, 85.9%, 79.6%, and 77%, respectively. In-hospital complication rates for approved patients were 2.2%, 2.3%, 2.1%, and 2.1%, while 90-day readmission rates were 4.6%, 4.5%, 4.3%, and 4.3%, respectively. Approved patients had higher MCID achievement rates (P < .001) for all thresholds but higher nonhome discharge rates than denied patients for thresholds 40 (P < .001), 50 (P = .002), and 60 (P = .024). Approved and denied patients had similar in-hospital complication and 90-day readmission rates.

Most patients achieved MCID at all theoretical PROMs thresholds with low complication and readmission rates. Setting preoperative PROM thresholds for TKA eligibility can help optimize patient improvement, but such a policy can create access to care barriers for some patients who would otherwise benefit from a TKA.