Background: The purpose of this study was to compare the short-term survival rate of total stemless, metaphyseal fixated, shoulder arthroplasty with that of total stemmed shoulder arthroplasty in the treatment of osteoarthritis. Methods: Data were collected by the national arthroplasty registries in Denmark, Finland, Norway, and Sweden and merged into 1 dataset under the umbrella of the Nordic Arthroplasty Register Association. For the present study, we included all patients with osteoarthritis treated with either stemless (n = 761) or stemmed (n = 4398) shoulder arthroplasty from 2011 to 2016. Results: A total of 21 (2.8%) stemless and 116 (2.6%) stemmed shoulder arthroplasties were revised. The 6-year unadjusted cumulative survival rates were 0.953 for stemless shoulder arthroplasty and 0.958 for stemmed shoulder arthroplasty, P = .77. The most common indication for revision of both arthroplasty types was infection. Five (0.7%) stemless and 16 (0.4%) stemmed shoulder arthroplasties were revised because of loosening of either the glenoid or the humeral component. In the multivariate cox regression model, which included age, category, gender, year of surgery, previous surgery, and arthroplasty type, the hazard ratio (HR) for revision of the stemless shoulder arthroplasty was 1.00 (95% confidence interval [CI], 0.63-1.61), P = .99, with the stemmed shoulder arthroplasty as reference. Male gender (HR = 1.50 [95% CI, 1.06-2.13], P = .02) and previous surgery (HR = 2.70 [95% CI, 1.82-4.01], P < .001) were associated with increased risk of revision. Conclusion: The short-term survival of total stemless shoulder arthroplasty appears comparable with total stemmed shoulder arthroplasty, but longer observation time is needed to confirm whether they continue to perform equally.