IMPORTANCEOsteoarthritis (OA) is a prevalent and disabling joint disorder. Minimally invasive interventions targeting the genicular nerves are widely used for the clinical management of knee OA, but evidence on their efficacy and safety remains uncertain.OBJECTIVETo assess the efficacy and safety of minimally invasive interventions targeting the genicular nerves in knee OA.DATA SOURCESWe searched MEDLINE, Embase, CENTRAL and clinical trial registries, from inception to August 2024.STUDY SELECTIONWe included randomised controlled trials evaluating minimally invasive interventions targeting the genicular nerves in patients with knee OA. Comparators included sham/placebo (main comparison), intra-articular injections, and usual care.DATA EXTRACTION AND SYNTHESISData were extracted independently by two reviewers following the PRISMA guidelines. Risk of bias was assessed using the Cochrane RoB-2 tool, and evidence certainty was evaluated according to the GRADE approach. A random-effects meta-analysis was conducted.MAIN OUTCOME(S) AND MEASURES(S)The primary outcomes were pain intensity, physical function, and serious adverse events. Secondary outcomes included quality of life and patient-reported global perceived effect.RESULTSTwenty-five trials were included (n= 2049) evaluating radiofrequency ablation (RFA), genicular nerve block (GNB) and cryoneurolysis. The certainty of evidence was consistently low to very low across all comparisons. RFA may provide moderate short-term pain relief (0 - 10 scale) compared to sham at 4 weeks (MD -1.70, 95% CI -3.03 to -0.36) and 12 weeks (MD -1.86, 95% CI -2.82 to -0.89), but there was no benefit at 24 and 48 weeks, and no improvements in function at any time point. The evidence is very uncertain for GNB, with only a single trial suggesting small pain reduction and moderate functional improvements at 4 weeks. For cryoneurolysis, very low certainty evidence suggests minimal short-term effects on pain and function. No differences in serious adverse events were observed between minimally invasive interventions and sham/placebo.CONCLUSIONS AND RELEVANCEThe evidence supporting the use of RFA for the management of knee OA is highly uncertain, with modest, short-term benefits that are not sustained. Given the very low certainty of current data, we advise against its routine use until more robust evidence is available.PROTOCOL REGISTRATIONPROSPERO - CRD42023454756.