Patellar resurfacing and nonresurfacing show similar 5-year outcomes in total knee arthroplasty

BACKGROUND: Despite extensive research, patellar management during total knee arthroplasty (TKA) remains controversial. We therefore asked the following: Is there a difference in clinical and radiological outcomes at a minimum 5-year follow-up with a modern posterior-stabilized TKA performed with or without patellar resurfacing (PR)? MATERIALS AND METHODS: This single-center prospective randomized trial included 250 knees (245 patients) undergoing primary TKA between April 2017 and November 2018. Exclusion criteria were isolated patellofemoral osteoarthritis, constrained TKA, and preoperative flexion less than 90°. Patients were randomized into PR or patellar nonresurfacing (PNR) groups, receiving the same posterior-stabilized "patella-friendly" prosthesis. Clinical evaluation at 5-year follow-up included the Knee Society Score (KSS), Forgotten Joint Score, Kujala and Lille scores for anterior knee pain, and range of motion. Radiographic assessment evaluated patellofemoral osteoarthritis progression and patellar implant loosening. Complications and surgical revisions were recorded. RESULTS: There were 213 knees (109 in PR, 104 in PNR) available with a 5-year follow-up (9% lost to follow-up and 4% deceased). Clinical outcomes showed no significant differences between groups (KSS knee: P = 0.10; KSS function: P = 0.19; Forgotten Joint Score: P = 0.24; Kujala: P = 0.36; Lille: P = 0.43). There were five patients (4.5%) in PNR who required secondary PR, whereas two patients (1.8%) in PR required revision of the patellar button. The revision rate for patellofemoral causes was 3.8%, without significant difference between groups (P = 0.16). At 60 months, implant survival was similar between PR and PNR (93.6 and 91.8%; P = 0.96). CONCLUSION: This large prospective randomized study found no clinical advantage of routine PR at 5-year follow-up. While resurfacing may reduce anterior knee pain, it also introduces risks such as fracture and maltracking. In the context of patella-friendly posterior-stabilized implants, these findings support a more selective approach, which should be guided by objective criteria.