Background: Stemless shoulder arthroplasty systems with uncemented metaphyseal fixation have been used for glenohumeral osteoarthritis since 2004 (Hawi, et al. BMC Musculoskelet Disord 17:376, 2016). The stemless design has several theoretical advantages compared with the stemmed shoulder arthroplasty systems: restoring patients' anatomy; preserving humeral bone stock; and few complications in component removal if the need for a revision arthroplasty arises. The purpose of the study is to compare the short-term, patient-reported outcome of stemless and stemmed total shoulder arthroplasty (TSA). Materials and methods: A randomized clinical trial will be conducted. Eighty patients with clinical and radiological signs of primary or post-traumatic glenohumeral osteoarthritis, computed tomography (CT) scan-verified adequate glenoid bone stock, and no total rupture of rotator cuff tendons verified by a magnetic resonance imaging (MRI) scan will be randomly allocated to a stemless or stemmed TSA. The primary outcome will be the Western Ontario Osteoarthritis Shoulder (WOOS) score at 12 months. Secondary outcomes are the WOOS score at three months and the Oxford Shoulder Score (OSS) and EQ-5D at 3 and 12 months. All complications, including glenoid and humeral component loosening, instability, rotator cuff tear, intraoperative and postoperative periprosthetic fracture, nerve injury, infection, deltoid injury, and symptomatic deep venous thrombosis, will be reported. Discussion: Findings will provide patients with better information about the potential benefits and harms of stemless and stemmed TSA and will assist shoulder surgeons and patients in decision-making. Trial registration: Clinicaltrials.gov, NCT03877315. Registered on 13 March 2019.