Objective To investigate extracorporeal shock wave therapy effect on knee osteoarthritis compared to a sham or kinesiotherapy by a systematic review and meta-analysis of randomized clinical trials. Data Sources The search was performed in: Cochrane Library, PubMed, PEDro, Web of Science, EMBASE, Scopus, LILACS, and Scielo. Review methods We performed the online search until October, 2022. The following terms were used (Osteoarthritis) AND (“knee joint”) AND (“Extracorporeal Shockwave Therapy”). Eligibility criteria: (1) randomized clinical trials; (2) effects comparison of shockwave therapy to a sham or kinesiotherapy in individuals with knee osteoarthritis; (3) pain and physical function as outcome variables. Risk of bias assessed using the PEDro scale. PROSPERO registration (CRD42021235597). Results We identified 4217 studies, and 12 were included in the qualitative synthesis and the meta-analysis, totaling 403 individuals submitted to the intervention and 331 control individuals. Compared to sham, shockwave was favored in short-term for the function outcome (SMD = −1.93; 95%CI: [−2.77; −1.09]; I² = 83%; P < 0.01). For the pain outcome, the shockwave was favored in the short (MD = −2.05; 95%CI: [−2.59; −1.51]; I² = 84%; P < 0.01), medium (MD = −3.46; 95%CI: [−4.03; −2.89]; I² = 0%; P < 0.01) and long-term (MD = −2.01; 95%CI: [−3.36; −0.65]; I² = 98%; P < 0.01). The association with kinesiotherapy was favored in the short term for the function outcome (SMD = -1.88; 95%CI: [−2.98; −0.78]; I² = 94%; P < 0.01) and favored for the pain outcome in the short (MD = −1.44; 95%CI: [−1.81; −1.07]; I² = 37%; P = 0.14), medium (MD = −1,31; 95%CI: [−1.76; −0,85]; I² = 0%; P = 0.41), and long terms (MD = −1.63; 95%CI: [−1.73; −1.52]; I² = 0%; P = 0.43). Conclusion Shockwave therapy may improve functionality in patients with knee osteoarthritis in the short term and pain in all follow-up moments, compared with sham. When associated to kinesiotherapy, it may improve function in the short term and pain in all follow-up time points, although improvement in pain may not be clinically significant.