Surgical treatment including shoulder arthroplasty is a treatment option for patients with advanced osteoarthritis of the shoulder who have failed conservative treatment. To determine the benefit and harm of surgery in patients with osteoarthritis of the shoulder confirmed on X-ray who do not respond to analgesics and NSAIDs. We searched: The Cochrane Central Register of Controlled Trials (CENTRAL), via The Cochrane Library; OVID MEDLINE; CINAHL (via EBSCOHost); OVID SPORTdiscus; EMBASE; and Science Citation Index (Web of Science). All randomized clinical trials (RCTs) or quasi-randomized trials including adults with osteoarthritis of the shoulder joint (PICO- patients) comparing surgical techniques (total shoulder arthroplasty, hemiarthroplasty, implant types and fixation- intervention) versus placebo or sham surgery, non-surgical modalities, no treatment, or comparison of one type of surgical technique to another (comparison) with patient-reported outcomes (pain, function, quality of life etc.) or revision rates (outcomes). We reviewed titles and abstracts for inclusion, extracted study and outcomes data and assessed the risk of bias of included studies. For categorical outcomes, we calculated the risk ratio (with 95% confidence interval (CI)) and for continuous outcomes, the mean difference (95% CI). Seven studies (238 patients) were included for analyses. None of the studies compared shoulder surgery to sham surgery, non-surgical modalities or placebo. Two studies compared hemiarthroplasty to total shoulder arthroplasty; three compared keeled and pegged humeral components; and one each compared navigation surgery to conventional and all-polyethylene to metal-backed implant. Two studies (88 patients) compared hemiarthroplasty to total shoulder arthroplasty. Patients who underwent hemiarthroplasty had statistically significantly worse functional scores on American Shoulder and Elbow Surgeons Shoulder Scale (100 point scale; higher = better) at 24 to 34 month follow-up compared to those who underwent total shoulder arthroplasty (mean difference, -10.05; 95% CI, -18.97 to -1.13; 2 studies, 88 patients), but no statistically significant differences between hemiarthroplasty and TSA were noted for pain scores (mean difference, 7.8; 95% CI, -5.33 to 20.93; 1 study, 41 patients), quality of life on short-form 36 physical component summary (mean difference, 0.80; 95% CI, -6.63 to 8.23; 1 study, 41 patients) and adverse events (Risk ratio, 1.19; 95% CI, 0.37 to 3.81; 1 study, 41 patients), respectively. A non-statistically significant trend towards higher revision rate in hemiarthroplasty compared to total shoulder arthroplasty was noted (Risk ratio, 6.18; 95% CI, 0.77 to 49.52; 2 studies, 88 patients; P = 0.09). Total shoulder arthroplasty seems to offer an advantage in terms of shoulder function, with no other clinical benefits over hemiarthroplasty. More studies are needed to compare clinical outcomes of surgery using different components and techniques in patients with osteoarthritis of the shoulder. There is a need for studies comparing shoulder surgery to sham, placebo and other non-surgical treatment options.