Between April 1984 and December 1987 a Müller acetabular reinforcement ring was employed in 98 revision cases (94 patients). Twenty-two patients died before follow-up evaluation. Forty-seven (66%) of the remaining 72 patients (49 of 75 hips) were examined after a mean of 6.4 years (range 5-9 years). Excluding those who died, the postoperative outcome of 69 patients is known. Migration was measured according to Sutherland by means of a digitising table. Cup position was determined by a new angle (beta) between the sacroiliacal line and a reference line. The reference line connects the centre of the femoral head with the intersection between the sacroiliacal and obturator lines. Aseptic and septic loosening required revision surgery in 2 cases each (4%). The postoperative beta angle showed a high correlation with migration (P = 0.001) as well as with radiolucencies (P = 0.001). In cases which required further revision, beta amounted to 44 deg (craniolateral position), and in stable cases up to 60 deg. A lateral and cranial position of a Müller acetabular reinforcement ring leads to high loosening rates. The beta angle accurately describes cup position, and its postoperative value is highly predictive for loosening.