Solutions described to limit the risk of scapular notching mainly concern the glenoid. Our hypothesis is that this risk also depends upon the glenoid-humeral relationship when the arm is resting along the body.

This is a retrospective study of a continuous series of 85 reverse shoulder arthroplasties; 62 of these fulfilled inclusion criteria. The following parameters were studied: body mass index (BMI), inferior overhang of the glenosphere, the angles showing the position of the glenoid (GH) and the humerus (MH) in the scapular plane as well as the glenometaphyseal angle (GM=MH−GH), during an initial postoperative follow-up, at 1 and 2years, and at a final follow up of a mean 45months (24–81). The parameters studied were compared in two groups with and without scapular notching.

There were 21 instances of notching at the final follow-up. This rate was significantly correlated to the BMI, which was a mean 27.2 in patients without a notch and 22.6 in patients with a notch, while the preoperative inclination of the glenoid in these groups was respectively 92.3° versus 85° respectively, the inferior overhang of the glenosphere was 4mm versus 2.8mm, the GM angle was 36° versus 47°, the MH angle at one year of follow-up was 135° versus 145° and the GH angle at the final follow-up was 103° versus 94° respectively. The BMI was significantly correlated to the GM angle, and a low BMI was associated with high values of this angle.

The relative position of the glenoid and humeral components, as shown by the GM angle, was an essential factor in the development of a scapular notching. The humeral component of the GM angle evolved in thin patients with progressive adduction of the arm, which is associated with a risk of notching. This should be taken into account when performing reverse shoulder arthroplasties.

Level IV retrospective study.