AVASCULAR NECROSIS OF THE SHOULDER. ANALYSIS OF PROSTHETIC SURGERY RESULTS BY RADIOGRAPHIC STAGE: CONTRIBUTION OF THE NEW CLASSIFICATION SYSTEM

Abstract

Purpose: The indication for humeral or total shoulder prosthesis in patients with avascular osteonecrosis of the shoulder depends on the radiographic stage of the necrosis. The purpose of this study was to analyse the prognostic radiographic features enabling proper identification of the indications for humeral or total prostheses.

Material and methods: This retrospective multicentric study included 53 shoulders operated on for prosthetic repair of avascular osteonecrosis. There were 20 men and 29 women, mean age 57 years. A humeral prosthesis was used in 34 cases and a total prosthesis in 19. Clinical assessment was based on the Constant score and the radiographic analysis used the Arlet and Ficat classification. Staging was II (n=4), III (n=16), IV (n=21) and V (n=12). Independently of the radiographic stage of necrosis, three characteristic groups were defined. Group 1 (19 shoulders) included necrosis without bony collapse of the humeral head. Group 2 (20 shoulders) included necrosis with bony collapse of the humeral head with preservation of the anatomic relations with the glenoid cavity. Group 3 (14 shoulders) included necrosis with bony collapse and impaction of the humeral head in the glenoid with medialisation of the humerus.

Results: Preoperatively, presence of bony collapse was associated with decreased pain and motion score. Major joint stiffness was noted in group 3 (8). The humeral prosthesis gave similar results for groups 1 and 2 (Constant score 73 and 75.3). Less satisfactory results was the rule in group 3 (Constant score 51.6). There was a postoperative erosion of the glenoid cavity in 83% of the shoulders in group 3 versus 12.5% in group 1 and 17% in group 2. The total prosthesis optimised results despite persistence of a difference between groups 2 and 3 (Constant score 83 and 64 respectively).

Discussion: Destabilisation of the glenohumeral joint by impaction of the humeral head against the glenoid cavity has an effect on the clinical result of the prosthetic surgery. Preservation of the anatomic relations of the articulation (groups 1 and 2) appears to be necessary for good results. Impaction medialisation of the humerus has a negative effect on the clinical result of humeral prostheses with a risk of glenoid erosion.

Conclusion: Independently of radiographic stage of necrosis, the presence of humeral impaction with medi-alisation should be considered individually as it has a negative effect on the clinical outcome of these prostheses, particularly humeral prostheses.