Modified notching classification in inverted-bearing and conventional-bearing reverse total shoulder arthroplasty

Scapular notching is a common complication of reverse total shoulder arthroplasty (RTSA). Although the notching rate has reduced significantly thanks to modifications to the surgical technique and humeral and glenoid components, uncontrollable polyethylene (PE)-induced osteolysis can still occur. In contrast to conventional-bearing (CB-RTSA), inverted-bearing RTSA (IB-RTSA) systems, with PE glenospheres and metal or ceramic humeral liners, avoid PE abrasion through scapulohumeral contact. If present, the resulting notch has a different size and configuration than with CB-RTSA. Thus, a modified notching classification seems reasonable. Even in CB-RTSA, the early stages of notching show a different configuration than implied by the established Nerot-Sirveaux classification; therefore, a modification of the classification system is recommended.

In a prospective multicentre study, 250 cases underwent IB-RTSA. Of these, 39 died, and 28 were lost to follow-up for other reasons, leaving 183 shoulders for the final follow-up examination at a mean of 120.7 months (range 84.1-172.4 months). In the CB-RTSA group, we retrospectively evaluated radiographs of 59 consecutive patients with a follow-up ranging from 2 to 7 years. We analysed the appearance, evolution over time, and location of bone loss on the scapular neck according to the modified Nerot-Sirveaux classification.

In IB-RTSA, notching resulted in a new morphological configuration: In contrast to the Nerot-Sirveaux classification, notching began far from the baseplate as an impression or abrasion of the humeral component in the inferior scapular rim. Due to simultaneous mechanical ablation, the defect gradually enlarged, but usually did not contact the baseplate or extend beyond the inferior peg. No signs of PE-induced osteolysis were found. The notching rate reached 19% after 2 years and approached 36% after 10 years. Most shoulders had grade 0 notching (64%), followed by grade 1 (29%); extensive erosion (grade 4) was not observed. In the CB-RTSA group, a small bony impression or erosion without signs of PE-induced osteolysis was observed medial to the scapular neck, without contact with the metaglene in the early phase, which was similar to the impression of the humeral component in IB-RTSA. The notching rate was 86% for CB-RTSA.

A new classification was developed for IB-RTSA, which corresponds to the pathophysiological processes and is compatible with the existing Nerot-Sirveaux classification. Bone erosion in CB-RTSA is not adequately represented by the established classification in the early stages. For this reason, a supplement to the existing classification is presented.

Level 4, Case Series, Treatment Study.