Do radiolucent lines and stress shielding of the humeral shaft really matter in shoulder arthroplasty?

The purpose of this study is to evaluate at a mid-term follow up, the radiological survival of an uncemented humeral stem in shoulder arthroplasty. One hundred and twenty-six replacements including hemi (HA), total (TSA) and reverse (RSA) implanted from 1999 to 2008 were reviewed at a mean follow up of 7.2 years (48-144 months). The same uncemented triconical stem (SMR, Lima Corporate) was implanted. There were: 23 HSA, 43 TSA, 60 RSA. An independent observer evaluated all the patients with Constant Score. A radiologic analysis by an expert radiologist and an orthopaedic surgeon was performed: humeral component-bone interface was divided in seven zones. They judged a mobilisation if a migration or tilt of the humeral implant or if≥ 2 mm radiolucent line in at least three zones was present. Chi-squared test, Fisher test and analysis of variance were performed and a p<0.05 was considered statistically significant. No major radiological signs of loosening and no tilt or migration of the humeral component were found. Only 23 (18.2%) patients had no RL around the humeral implant. In the remaining 103 (81.7%) implants: 96 (76.1%) presented RL less than 2 mm, particularly 75 (59.5%) in less than 3 zones and 21 (16.6%) in more than 3 zones. Of the remaining 7 (5.5%) implants the presence of RL of 2 mm or greater in only one zone was seen. Apart from sepsis no revision was performed for humeral component loosening. Although a high rate of RL, uncemented humeral stem has an excellent survivorship at a mid-term follow up. Relationship between presence, position and depth of RL and internal stress shielding is commonly observed but does not appear t