Abstract

The clinical and radiological evolution of the glenoid socket in total shoulder arthroplasty is well documented, whereas evaluation of the humeral component has received much less attention. The outcome of 40 humeral replacements performed in 39 patients was studied with a minimum follow-up of one year. There were 26 Neer prostheses, 9 Modular Shoulder prostheses and 5 Aequalis prostheses. Twenty-nine prostheses were implanted with cement and 11 were cementless. The clinical and radiological results were assessed according to the scoring system used for the 1994 symposium of SO.F.C.O.T. X rays in the coronal plane were used to assess the position of the stem, radiolucencies around the humeral component and the percentage of the cross-section of the diaphysis filled by the humeral stem. The average follow-up period was respectively 55.6 months for the Neer prostheses, 17.9 months for the Modular Shoulder prostheses and 12.4 months for the Aequalis prostheses. Radiolucent lines were noted in 20 cases, all with uncemented humeral components. Two prostheses were loose; however, none required revision. The underlying pathology, surgical approach, position of the humeral component, implant design or association with glenoid resurfacing were not correlated with the presence of radiolucent lines or with a loose implant. Cementless fixation was the only parameter statistically related with the presence of radiolucent lines. This report confirms the good results obtained with cemented humeral prostheses. Radiological evaluation of the smooth cementless humeral implant showed many progressive radiolucent lines, without clinical expression. The authors analyze the various studies in the literature and discuss the interest of cementless implants with ingrowth surface.