Objective To investigate the operative methods, cl inical outcomes and compl ications of total hi p arthroplasty (THA) in the treatment of patient with hi p joint flexion rigidity due to ankylosing spondyl itis (AS). Methods From May 1992 to July 2004, 56 patients (32 left hips and 39 right hips) with AS received THA through a modified anterolateral approach, including 52 males (67 hips) and 4 females (4 hips) aged 17-48 years with an average of 35.5 years. All the hips were ankylosed in (43.1 ± 7.2)° of flexion and 15 patients had bilaterally ankylosed hips. Preoperatively, Harris hip score was (42.6 ± 5.3) points and all the hips were classified as stage IV according to the standard of American College of Rheumatology (ACR). And the course of disease was 3-11 years. Results Intraoperatively, 1 patient suffering from proximal femur fracture due to severe osteoporosis was treated with titanium wire fixation, and the fracture was healed 6 weeks later. All the patients were followed up for 3-15 years (average 5.3 years). Postoperatively, 1 patient (1 hip) got subcutaneous soft tissue infection at 8 days, 1 patient (1 hip) got wound disunion at 11 days, 2 patients (2 hips) got infection at 11 months and 3 years, respectively. All the infections were healed after symptomatic treatment. The wounds of the rest 52 patients were healed by first intention without joint infections. The postoperative X-rays demonstrated that 4 hips (5.6%) had loose acetabulum prosthesis, 3 hips (4.2%) had loose femoral prosthesis and 5 hips had loose acetabulum and femoral prosthesis (7.0%), and the total loosening rate was 16.8%. Among which, 8 hips received revision resulting in satisfactory therapeutic effects, and the rest 4 hips had no further treatment. Fifteen hips (21.1%) had heterotopic ossification, which was rel ieved after taking nonsteroidal anti-inflamatory drugs. Harris hip score at final follow-up was (82.7 ± 4.1) points, indicating there was a significant difference between before and after operation (P lt; 0.05). Ten hips were evaluated as excellent, 43 hips good, 14 hips fare, and 4 hips bad, and the excellent and good rate was 74.7%. Conclusion THA through the anterolateral approach is effective for the treatment of patient with hip joint flexion rigidity caused by AS.