We reviewed the long-term results of thirty-seven corrective osteotomies that had been performed for osteonecrosis of the femoral head that was stage II or III according to the classification of Ficat and Arlet. At a mean of 11.5 years (range, five to eighteen years) postoperatively, twenty-eight hips (76 per cent) had a good or excellent result according to the Harris hip-scoring system, and nine (24 per cent) had a fair or poor result and subsequently needed a total hip arthroplasty. Six of the nine failures were in the seventeen hips of patients who had received corticosteroids. Conversely, of the twenty hips of patients who had not received corticosteroids, seventeen (85 per cent) had a good or excellent result as determined by the Harris hip score at the latest follow-up evaluation. Five of the six hips that had had a combined necrotic angle of more than 200 degrees preoperatively had subsequent collapse of the femoral head. Of the thirty-one hips that had had a combined necrotic angle of less than 200 degrees preoperatively, twenty-seven (87 per cent) had a good or excellent clinical result.There were five complications. Three non-unions and one cutout of the compression screw were successfully treated, and these hips had an excellent clinical result at the time of the latest follow-up. The fifth complication was osteomyelitis and led to a poor result.The results of this study suggest that corrective intertrochanteric osteotomy is a successful treatment for Ficat and Arlet stage-II or III disease if the patient has a small or medium lesion (a combined necrotic angle of less than 200 degrees) and is not receiving continuous high doses of corticosteroids.