We retrospectively reviewed the results of primary total elbow arthroplasty for the treatment of an acute fracture of the distal aspect of the humerus in twenty consecutive patients (twenty-one elbows) who had a mean age of seventy-two years (range, forty-eight to ninety-two years) at the time of the injury. The patients were managed between November 1982 and October 1992. The presence of rheumatoid arthritis in nine patients (ten elbows) influenced the choice of treatment. The mean interval between the injury and the total elbow arthroplasty was seven days (range, one to twenty-five days). The mean duration of postoperative hospitalization was seven days (range, four to thirteen days). The mean duration of follow-up was 3.3 years (range, three months to 10.5 years). All patients were followed for a minimum of two years or until the time of death; the duration of follow-up was less than two years for three patients who died. None of the patients were lost to follow-up. Twenty implants were intact at the latest follow-up examination. One patient had a revision total elbow arthroplasty twenty months after the index procedure because of a fracture of the ulnar component sustained in a fall on the outstretched arm.On the basis of the Mayo elbow performance score, fifteen elbows had an excellent result and five had a good result; there were inadequate data for one elbow. There were no fair or poor results. The mean arc of flexion was 25 to 130 degrees. There was no evidence of loosening on the radiographs. Postoperative complications included fracture of the ulnar component in one patient, ulnar neurapraxia in three, and reflex sympathetic dystrophy in one. The results suggest that total elbow arthroplasty can be an alternative form of treatment of a severely comminuted fracture of the distal aspect of the humerus in older patients even in the presence of rheumatoid arthritis. This procedure is not an alternative to osteosynthesis in younger patients.