In a retrospective review of eighty-two intertrochanteric fractures (twenty-nine stable and fifty-three unstable) in seventy-nine elderly, debilitated patients with associated advanced osteoporosis (Grade III or less by the system of Singh et al.), fifty-six were available for follow-up: twenty-eight that had been treated at the University of Illinois with an approximately anatomical reduction and compression-screw fixation and twenty-eight (in twenty-seven patients) that had been treated at the University of Chicago with an approximately anatomical reduction, compression-screw fixation, and adjunctive methylmethacrylate bone cement in the head-neck fragment. Follow-up analysis after an average of thirty-four months for the group that had augmentation with cement and an average of twenty-six months for the uncemented group showed that for the eighteen stable fractures that could be followed the rates of complications of fixation were the same in the two groups, while for the thirty-eight unstable comminuted fractures that were followed the rate of complications of fixation was lower when adjunctive methylmethacrylate cement was used. Among the unstable fractures, one failure (in twenty-one fractures) in the cement-augmented group and ten failures (in seventeen fractures) in the uncemented group were due to failure of fixation (p less than 0.01). For reasons that are not clear, when the thirty-two patients with a healed fracture who could be evaluated for function were rated using the Iowa hip score, the nineteen who were treated with adjunctive cement had significantly lower scores than did the thirteen who were treated without cement (76 +/- 16.5 compared with 92 +/- 12.1, p less than 0.01).