We evaluated the results of Cementless Surface Replacement Arthroplasty (CSRA) of the shoulder in 67 patients with advanced glenohumeral destruction who have an intact rotator cuff.

Between november 2002 and december 2008, 70 CSRA (32 Copeland/ Biomet and 38 SMRR/ Lima) were implanted in 67 patients. A deltopectoral approach was used in 34 cases and an anterosuperior approach in 36 cases. Patients were assessed using Constant score, a patient satisfaction score and a detailed radiographic analysis. The mean follow up was 3.4 years (range 1 to 7.5 years).

The mean Constant score improved preoperatively from 17.6 points (range 2–55) to an average postoperative score of 66.1 (range 13–91). The pain score improved from 1.13 points (range 0–6) to 12.3 points (range 3–15). The forward flexion and external rotation improved from 71° (range 20 to 140) and 0° (range −40 to +45) to 143° (range 60 to180) and 34.4° (range −20 to +60) respectively.

Complications included: 1 subscapularis detachment, 5 secondary rotator cuff tear, 1sepsis, 3 patients with shoulder stiffness.

No shift in implant position was observed. 11 humeral components developed radiolucencies at the prosthesis-bone interface. The radiographic analysis involved a system of dividing the prosthesis/bone interface into 5 zones.

The best clinical results were significantly achieved in patients with necrosis compared with osteoarthritis (Constant Score, ant. elevation, ext. Rotation).

Using regression analysis we found that changes in the head-shaft angle position of the implant (valgus/varus placement of the CSRA) significantly predicted the age and sex adjusted Constant score. When the inclination angle of the humeral head decreases, the adjusted Constant score increases. In the same model, we also found that the lateral offset of the humerus significantly predicted the adjusted Constant score. When the lateral offset of the humerus decreases, the adjusted Constant score increases.

The medialization of the glenoid significantly and negatively predicted the Constant score.

Purpose: We assessed mid-term outcome of mobile cup shoulder prostheses for the treatment of degenerative shoulders with a destroyed cuff.

Material and methods: Forty-four biopolar prostheses were assessed with a mean follow-up of 32 months (range 13 – 50 months). The SOFCOT scoring chart was used. The morphological features of the different forms of shoulder degeneration with cuff destruction were analysed on standard x-rays to define anatomic conditions favourable for his type of implant.

Results: At last follow-up 86% of the shoulders exhibited little or no pain. Mean active anterior elevation was 59.6° – 82.84°, external rotation in position 1 was 8.86° to 301.68°, and internal rotation was 3.13 points to 5.68 points. The mean Constant score was 48.86 points (weighed score 69.13%). Seventy-seven percent of the patients were satisfied and had resumed their former activities. We had one anterior displacement of a shoulder with deltoid palsy and three complications requiring revision: one polyethylene insert which slipped out of the mobile cup, one conflict between the tendon of the long head of the biceps and the cup, and one anterioposterior conflict because of an oversized cup. One non-cemented prosthesis exhibited mobility at 33 months follow-up. One case of excessive lateralisation of the humerus, one mediocre joint congruency, and atrophy of the deltoid were also recorded.

The best outcomes were observed in joints with “centred” degeneration and cuff destruction. Preservation of a satisfactory centring of the cup avoided pain and enabled a mean anterior elevation of 1112.3° with a weighted Constant score of 90.5%.

For early excentred degeneration, preservation of the glenoid bone stock prevented perfect joint congruency leading to excessive lateralisation of the humerus: 20% of the shoulders remained painful and anterior elevation reached 68° with a weighted Constant score of 56%. Paradoxically, results were better in patients with severely excentered degeneration. “Acetabulation” of the shoulder preserved joint congruency and lengthened the lever arm of the deltoid by medialisation and lowering of the centre of rotation. Full pain relief was achieved in all such cases, with anterior elevation at 86° and a weighted Constant score at 78%.

Discussion: Biopolar arthroplasty is effective in shoulders with centred degeneration and cuff destruction. It is an interesting salvage solution for advanced-stage excentric degenerated shoulders beyond the limits of other implants. We propose a decisional tree based on an anatomic classification of the shoulder degeneration with cuff destruction.

Purpose: The purpose of this work was to study the biomechanical properties of mobile cup shoulder prostheses and factors affecting their kinetics.

Material and methods: Bipolar shoulder prostheses were implanted in 39 patients with degenerative shoulders and a destroyed cuff. Radiocinematic recordings of anterior elevation and active rotation were made at a mean 32 months follow-up (13 months–550 months).

Results: Three types of biomechanical behaviour were observed for elevation movements. “Normal” behaviour was observed in 17 prostheses with preservation of the scapulohumeral rhythm and chronological participation of the three articular interfaces [intraprosthetic (head/cup), extra-prosthetic (cup/glenoid), scapulothoracic]. Mean anterior elevation was 114.7° for these shoulders. A “paradoxical” behaviour was observed in ten prostheses. Anterior elevation depended entirely on the scapulorthoracic joint, and was limited on the average to 42.5°. An “intermediate” behaviour was observed in 12 prostheses with inversion of the scapulohumearl rhythm. The glenohumeral mobility was decreased due to the absence of extraprosthetic mobility (eight cases) or intraprosthetic mobility (four cases). Mean anterior elevation in these shoulders was 80.83°. When the elevation behaviour was “normal”, the Constant score at last follow-up was significantly better compared with “intermediate” (p = 0.008) or “paradoxical” (p = 0.0001) behaviour.

Three types of biomechanical behaviour were also observed for rotation movements: a “chronological “ behaviour was observed for 15 prostheses, via extraprosthetic mobility in all. Mean external rotation was 37.33° and mean internal rotation was 6.53 points. An “anarchic” behaviour was observed in 16 prostheses with a random proportion of intra- and extraprosthetic mobility. Mean external rotation was 8.75° and mean internal rotation was 4.25 points. For shoulders with “chronological” or “anarchic” behaviour, the mean external rotation (p = 0.002) and the mean internal rotation (0.04) were statistically better than shoulders with “truncated” behaviour.

Discussion: An atrophic deltoid, mediocre joint congruency, and early-stage excentred scapular degeneration with preserved glenoid bone stock are factors favouring “paradoxical” elevation. Deltoid atrophy alone favours “truncated” rotation. This study was helpful in identifying conditions most appropriate for implanting this type of prosthesis and factors predicting postoperative outcome.