We hypothesized that constant compression of the knee would mobilize residual synovial fluid and promote successful arthrocentesis. Two hundred and ten knees with grade II-III osteoarthritis were included in this paired design study: (1) conventional arthrocentesis was performed with manual compression and success and volume (milliliters) determined; and (2) the intra-articular needle was left in place, and a circumferential elastomeric brace was tightened on the knee to provide constant compression. Arthrocentesis was attempted again and additional fluid volume was determined. Diagnostic procedural cost-effectiveness was determined using 2017 US Medicare costs. No serious adverse events were noted in 210 subjects. In the 158 noneffusive (dry) knees, sufficient synovial fluid for diagnostic purposes (≥ 2 ml) was obtained in 5.0% (8/158) without compression and 22.8% (36/158) with compression (p = 0.0001, z for 95% CI = 1.96), and the absolute volume of arthrocentesis fluid obtained without compression was 0.28 ± 0.79 versus 1.10 ± 1.81 ml with compression (293% increase, p = 0.0001). In the 52 effusive knees, diagnostic synovial fluid (≥ 2 ml) was obtained in 75% (39/52) without compression and 100% (52/52) with compression (p = 0.0001, z for 95% CI = 1.96), and the absolute volume of arthrocentesis without compression was 14.7 ± 13.8 versus 25.3 ± 15.5 ml with compression (72.1% increase, p = 0.0002). Diagnostic procedural cost-effectiveness was $655/sample without compression and $387/sample with compression. The new technique of constant compression via circumferential mechanical compression mobilizes residual synovial fluid beyond manual compression improving the success, cost-effectiveness, and yield of diagnostic and therapeutic arthrocentesis in both the effusive and noneffusive knee.