Purpose The aims of this study were to assess the function and quality of life after the Total Evolutive Shoulder System (TESS) reverse shoulder arthroplasty (RSA), to evaluate the radiological stability of the stemless version and to address the effect of arm lengthening and scapular notching (SN) on the outcome. Methods This was a prospective comparative nonrandomised study. A total of 37 consecutive patients (40 shoulders) underwent TESS RSA between October 2007 and January 2012; 16 were stemless and 26 were stemmed. At a mean follow-up of 39 months (15-66), we evaluated range of motion (ROM), pain and functional outcome with QuickDASH and quality of life with EQ-5D score. Radiologically, component positioning, signs of loosening, SN and arm length difference were documented. Results We found a significant improvement in functional outcome and reduction of pain in both stemmed and stemless groups. No humeral loosening was evident, but there were four glenoid loosenings. In 12 shoulders that developed SN, seven already had scapular bone impression (SBI) evident on initial post-operative radiographs. Glenoid overhang seemed to decrease the risk of SN. Arm lengthening was associated with better EQ-5D but did not influence ROM or functional outcome.Conclusions Reverse shoulder arthroplasty markedly improved shoulder function. SN is of concern in RSA, but proper positioning of the glenoid component may prevent its development.

TESS prosthesis showed promising short-term results with few complications. The reverse version could be implanted without stem if initial stability was adequate. Long-term follow-up is required to confirm the results of this innovative system in the long run.

Level 2.

Stemless implants could be of help to reconstruct the shoulder anatomy. This study shows that there are some challenges to be addressed when attempting to ensure optimal implant positioning. The critical step is to determine the correct level of bone cut to avoid varus or valgus humeral head inclination and ensure correct head size.