Why locking plates for the proximal humerus do not fit well

Introduction: We compared the angle of the humerus and plate and to assess compatibility of a plate to the proximal humerus using three-dimensional (3D) printed models. Materials and methods: A total of 120 cases were included, who underwent anteroposterior shoulder radiographs. From these, 30 cases with 3D shoulder computed tomography scans were randomly selected to print 3D model. The lateral angle between the lateral cortex of the humeral shaft and lateral border of the greater tuberosity (GT), neck-shaft angle, and height from the most proximal point of the GT to the angular point were measured. When the plates were applied on the 3D models, the gap from the most proximal point of the GT to the proximal rim of the plate was measured. Results: The mean lateral angle in plain radiographs was 12.9 ± 2.2° and height from the most proximal point of the GT to the angular point was 44.4 ± 4.7 mm. The bending angles of the three plates were 8° and 10°. Height from the proximal rim of the plate to the bending point was 42.4, 42.0 and 43.8 mm. In 98% of cases, the lateral angle of the humerus was larger than all three plates. In 43% of cases, height of the GT was smaller than height of plates. When plates were applied to the 3D model, the mean gap from GT to plate was 4.8 ± 2.8 mm. Conclusions: There was large variation in the lateral angle of the proximal humerus, which was not correlated with the neck-shaft angle. The lateral angle of the humerus was larger than the plates and prone to varus reduction and medial collapse. Level of evidence or clinical relevance: Basic science study.