[Twenty nine shoulder reconstructions after resection of the proximal humerus for neoplasm with mean 7-year follow-up].
PURPOSE OF THE STUDY: Techniques available for shoulder reconstruction after resection of a tumor of the proximal humerus include scapulohumeral arthrodesis, humerus prosthesis with or without an allograft, inverted prostheses, and massive allografts. The purpose of this study was to review clinical and radiological outcomes in a series of 29 patients (20 men and 9 women) who underwent resection-reconstruction of the proximal humerus and to establish from these cases a decision making algorithm for therapeutic indications as a function of tumor invasion. MATERIAL AND METHODS: The tumors were 20 chondrosarcomas, five osteosarcomas, two Ewing sarcomas and one malignant hemangiopericytoma. In 17 patients epiphyso-metaphyseal or epiphyso-metaphyso-diaphyseal resection was performed with preservation of the abductor muscles (type S34A or S345A according to the Musculoskeletal Tumor Society classification). For 12 patients epiphyso-metaphyseal or epiphyso-metaphyso-diaphyseal resection was performed without preservation of the abductor muscles (type S34B or S345B). Reconstruction was achieved using a centromedullary cemented nail in one patient, scapulohumeral arthrodesis in three, a massive humerus prosthesis in 15, and composite humerus prosthesis in three and an inverted prosthesis in seven. The functional score of the Musculoskeletal Tumor Society (MSTS) and standard x-rays were used to assess outcome. RESULTS: Mean follow-up was 85 months (range 16-300). The mean MSTS score was 88% for inverted prostheses, 76% for composite prostheses, 72.6% for massive prostheses, 75% for scapulohumeral arthrodeses, 67% for massive prostheses, and 80% for cemented centromendullary nail. Five patients died from their malignant disease and one from another cause. Four patients are alive but with active disease after a mean follow-up of 108 months and 19 patients (65.5%) are alive and free of locoregional recurrence or metastasis after a mean 83.5 months. We had 28 complications. Glenohumeral instability was the most frequent (11 cases). DISCUSSION: Resection of the upper portion of the humerus should be performed to achieve cancerologically satisfactory tumor resection and enable shoulder resection, if possible, with preservation of a viable and functional abductor system. The functional outcome after such reconstruction depends on the type of bony resection, but also on the sacrifice of the rotator cuff and the deltoid muscle. In light of our experience and results in the literature, we advocate, despite the small number of cases for the different reconstructions, the following decision-making algorithm after resection of the proximal humerus without joint invasion: when the resection removes the rotator cuff and the deltoid (or the axillary nerve), there are two options: scapulohumeral arthrodesis or massive humerus prosthesis for patients who do not desire a complex therapy with a long postoperative period; when the resection preserves the rotator cuff and/or the deltoid muscle, reconstruction can be achieved with a composite (inverted or not) prosthesis with suture of the cuff tendons. We prefer the inverted composite prosthesis; if the deltoid muscle can be preserved but not the rotator cuff, the composite inverted prosthesis appears to be the most logical solution, but scapulohumeral arthrodesis can be proposed in selected cases.