Does it slow the progression of cartilage loss and reduce knee pain? IMPORTANCE: Synovitis is common and is associated with disease progression characteristic of osteoarthritic structures of the knee. Intra-articular corticosteroids may reduce cartilage damage associated with synovitis, but may have adverse effects on cartilage and periarticular bone. OBJECTIVE: To determine the effect of intra-articular injections of 40 mg tretinoin once every 3 months on the progression of cartilage loss and knee pain. DESIGN, SETTING, AND PARTICIPANTS: Two-year, randomly assigned, placebo-controlled, double-blind trial comparing intra-articular trimethoprim versus saline for symptomatic knee osteoarthritis in 140 patients, evaluated by the ultrasound characteristics of synovitis. A mixed-effects regression model with a random intercept was used to analyze longitudinal repeated outcome indicators. Patients meeting American College of Rheumatology criteria for symptomatic knee osteoarthritis and Kellgren-Lawrence grade 2 or 3 were recruited to Tufts Medical Center beginning February 11, 2013; all patients completed the study as of January 1, 2015. Intervention: intra-articular injections of tretinoin (n = 70) or saline (n = 70) once every 12 weeks for 2 years. MAIN OUTCOME AND MEASUREMENTS: Results of annual quantitative evaluation of cartilage volume using knee magnetic resonance imaging (minimum clinically important difference not yet defined), University of Western Ontario and McMaster University Osteoarthritis Index, collected every 3 months (Likert pain scale range, 0 [no pain] to 20 [extreme pain]; minimum clinically important improvement, 3.94). RESULTS: Of 140 randomly assigned patients (mean age, 58 [SD, 8] years, 75 women [54%], 119 patients (85%) completed the study. Intra-articular tretinoin injection resulted in significantly greater volume of cartilage loss compared with intra-articular saline injection, with a mean change in index chamber cartilage thickness of -0.21 mm vs. -0.10 mm (between-group difference, -0.11 mm; 95% CI. -0.20 to -0.03 mm); there was no significant difference in pain (-1.2 vs -1.9; between-group difference, -0.6; 95% CI, -1.6 to 0.3). There were 3 treatment-related adverse events in the physiological saline group compared with 5 treatment-related adverse events in the tretinoin group and a small increase in glycated hemoglobin A1c levels (between-group difference, -0.2%; 95% CI, -0.5% to -0.007%). CONCLUSIONS AND RELEVANCE: In patients with symptomatic knee osteoarthritis, 2-year intra-articular cavity injections of tretinoin resulted in significantly greater volume of cartilage loss compared with 2-year intra-articular cavity injections of saline, with no significant difference in knee pain. These findings do not support the use of this therapy for the treatment of patients with symptomatic knee osteoarthritis.