The purpose of this study was to clarify the factors affecting the results of the endoprosthetic replacement for the comminuted fresh fracture of the proximal humerus. Twenty-one shoulders were investigated with follow up time 44.8 months on the average. The average at age at operation was 67 years old. There was 1 male and 20 females. The right shoulder was involved in 9 patients and left shoulder in 12patients. The dominant shoulder was involved in 8 patients. A 3 part fracture was seen in 6 shoulders and a 4 part fracture in 15shoulders. A monopolar endoprosthesis was used in 9 shoulders and a bipolar one was emplyed in 12 shoulders. Patients were evaluated by Japanese Orthopaedic Association shoulder score ( JOA score ) with reference to the age at operation, duration between fracture and surgery, type of fracture, type of endoprosthesis, utilization of bone cement at operation, and follow-up radiological findings. The range of motion at follow up were 121degrees of flexion, Li of internal rotation and 27 degree of external rotation. The average JOA score at followup was 80.1 points. Whether or not bone cement was used for the humeral stem fixation did not cause any statistically differences. The JOA score was higher in patients without complications (P=0.04), in patients with sufficient bone stock around the fin of the prosthesis (P=0.04), and in patients with humeral offset more than 25mm(P=0.03). This procedure was useful for elderrly patients without major complications if it was performed within 2 weeks of the fracture. Good bone stock around the fin of the prosthesis and a humeral offset of morethan 25mm were important factors for the good prognosis.
