To preserve humeral bone during RTSA, stems have been made shorter and cement avoided whenever possible. However, with the increased use of uncemented RTSA, a phenomenon comparable to the stress shielding of the hip has been described for the proximal humerus. The aim of this study was to investigate the influence of stem length and width on proximal humeral bone resorption after primary uncemented RTSA. The prospective shoulder arthroplasty database of our institution was reviewed for all primary uncemented RTSAs from 2017 to 2020 in osteoarthritis and cuff tear arthropathy cases with > 2-year follow-up. We compared the clinical and the radiographic 2-year outcome of the short and standard length stems of the same prosthesis design. This allowed us to assess the effects of stem length and width with regard to stress shielding. Furthermore, we defined a cut-off value for the filling ratios to prevent stress shielding. Fifty patients were included in the analysis, nineteen were in the short stem group (SHORT) and thirty-one in the standard stem group (STANDARD). After 2years, SHORT showed a relative Constant Score of 91.8% and STANDARD of 98.3% (p = 0.256). Stress shielding was found in 4 patients (21%) in SHORT and in 16 patients (52%) in STANDARD (p = 0.03); it occurred more frequently in patients with higher humeral filling ratios (p < 0.05). The calculated cut-off to prevent stress shielding was 0.7 (± 0.03) for the metaphyseal and distal filling ratio. While short and standard stems for RTSA have good results after 2years, we found a significant negative effect of higher length and width of the stem with regard to stress shielding. Even though the clinical effects of stress shielding have to be assessed, short stems should be chosen with a filling ratio at the metaphyseal and distal position below 0.7. III.