Subchondral Knee Injections: A Comprehensive Systematic Review of Current Clinical Outcomes, Safety, and Arthroplasty Delay

Introduction Osteoarthritis is considered a whole-joint disease involving subchondral bone. Intraosseous therapies such as calcium phosphate (CaP), platelet-rich plasma (PRP), and mesenchymal stem cells (MSCs) offer joint-preserving options for knee osteoarthritis (OA). Purpose To critically appraise and update the clinical evidence on intraosseous injections for knee osteoarthritis, focusing on safety, functional outcomes, need for retreatment, and conversion rates to total knee arthroplasty (TKA). Methods A systematic review of PubMed, Embase, and Cochrane was conducted following PRISMA guidelines and PROSPERO registration (CRD420251104989). Clinical studies reporting outcomes of intraosseous injections with CaP, PRP, or MSCs for knee osteoarthritis with ≥5 patients and ≥6 months of follow-up were included. Methodological quality was assessed using the modified Coleman Methodology Score and Cochrane Risk of Bias 2.0 tool. Results Twenty-four studies involving 1,109 patients (mean age, 55 years; mean follow-up, 38 months) met inclusion criteria: 10 on CaP, 6 on PRP, and 8 on MSCs. Five were randomized controlled trials (RCTs). Most studies reported significant improvements in pain and function. CaP injection outcomes were variable, with TKA conversion rates ranging from 1.3% to 45%. PRP and MSCs studies showed favorable safety profiles and lower conversion rates. Long-term MSCs data indicated sustained relief and delayed TKA over up to 15 years. However, overall study quality was modest, with only one RCT rated as low risk of bias. Conclusion Intraosseous injections may improve symptoms and delay arthroplasty in selected patients with knee OA, with MSCs showing the most favorable long-term results. PRP appears to be a safe option whereas CaP outcomes are more variable. Standardized protocols and high-quality RCTs with long-term follow-up are needed to optimize patient selection and treatment efficacy.