Objective: To review the incidence and clinical management of local adverse events (AEs) in patients with knee osteoarthritis (OA) treated with hylan G-F 20 (Synvisc). Design: Retrospective review of prospective data. Setting: Large, knee-only, orthopedic practice. Participants: Patients (N=1047; 1489 knees) who received at least 1 injection of hylan G-F 20 for osteoarthritic knee pain treatment from October 3, 1997 to April 30, 2002. Interventions: Intra-articular hylan G-F 20 (2mL; 3 weekly injections) using fluoroscopy for accurate needle placement. Main Outcome Measures: Local AEs (defined as pain and swelling), their severity, timing postinjection, duration, effusion presence, and treatment were recorded. Physician visual analog scale (VAS) measured osteoarthritic knee pain in knees that had AEs. Results: 54 AEs were experienced by 49 patients (53 knees) who were treated with hylan G-F 20: 5.2% per patient, 3.6% per knee, and 1.2% per injection. Most events were mild or moderate in nature (77.7%). Based on their timing postinjection, AEs were categorized as definitely (<12h; 0 [0.0%]), probably (12–48h; 2 [3.7%]), possibly (3–7d; 35 [64.8%]), or not (>7d; 17 [31.5%]) related to treatment. The duration of AEs varied: 23 (42.6%) occurred within ≤48h; 11 (20.4%) occurred within 3–7d; and 20 (37.0%) occurred >7d later. Events that lasted ≥3 days were considered not related (40.0%; >7d postinjection) or only possibly related (60.0%; 3–7d postinjection) to treatment. Effusions occurred with most events (90.6%); more than half (55.6%) were mild to moderate in nature. Local AEs were transient and resolved spontaneously, or with aspiration and/or oral, intra-articular, or intramuscular corticosteroids. Efficacy for treating osteoarthritic knee pain was not affected by the local AEs experienced (VAS scores: week 0, 75.6±14.3; week 12, 25.6±23.8; week 26, 31.3±27.8; P<.0001). Conclusion: Local AEs with hylan G-F 20 were of low incidence, were clinically manageable, and did not affect treatment efficacy.