VOL. 99-B, No. 5, MAY 2017 601  HIP Radiographic scoring system for the evaluation of stability of cementless acetabular components in the presence of osteolysis R. Narkbunnam, D. F. Amanatullah, A. J. Electricwala, J. I. Huddleston III, W. J. Maloney, S. B. Goodman From Stanford University Medical Center, California, United States  R. Narkbunnam, MD, Assistant Professor, Siriraj Hospital Mahidol University, 2 Wanglang Road Bangkoknoi, Bangkok 10700, Thailand.  D. F. Amanatullah, MD, PhD, Assistant Professor  W. J. Maloney, MD, Professor  S. B. Goodman, MD PhD FRCSC FACS FBSE, Professor  J. I. Huddleston III, MD, Associate Professor of Orthopaedic Surgery Stanford University Medical Center Outpatient Center , 450 Broadway Street, Redwood City, CA 94063-6342, California, USA.  A. J. Electricwala, MD, Assistant Professor Sancheti Institute of Orthopaedics and Rehabilitation, 16, Shivajinagar, Pune, Maharashtra 411005, India. Correspondence should be sent to R. Narkbunnam; email: mai_parma@hotmail.com ©2017 The British Editorial Society of Bone & Joint Surgery doi:10.1302/0301-620X.99B5. BJJ-2016-0968.R1 $2.00 Bone Joint J 2017;99-B:601–6. Received 27 September 2016; Accepted after revision 26 January 2017 Aims The stability of cementless acetabular components is an important factor for surgical planning in the treatment of patients with pelvic osteolysis after total hip arthroplasty (THA). However, the methods for determining the stability of the acetabular component from pre-operative radiographs remain controversial. Our aim was to develop a scoring system to help in the assessment of the stability of the acetabular component under these circumstances. Patients and Methods The new scoring system is based on the mechanism of failure of these components and the location of the osteolytic lesion, according to the DeLee and Charnley classification. Each zone is evaluated and scored separately. The sum of the individual scores from the three zones is reported as a total score with a maximum of 10 points. The study involved 96 revision procedures which were undertaken for wear or osteolysis in 91 patients between July 2002 and December 2012. Pre-operative anteroposterior pelvic radiographs and Judet views were reviewed. The stability of the acetabular component was confirmed intra- operatively. Results Intra-operatively, it was found that 64 components were well-fixed and 32 were loose. Mean total scores in the well-fixed and loose components were 2.9 (0 to 7) and 7.2 (1 to 10), respectively (p < 0.001). In hips with a low score (0 to 2), the component was only loose in one of 33 hips (3%). The incidence of loosening increased with increasing scores: in those with scores of 3 and 4, two of 19 components (10.5%) were loose; in hips with scores of 5 and 6, eight of 19 components (44.5%) were loose; in hips with scores of 7 or 8, 13 of 17 components (70.6%) were loose; and for hips with scores of 9 and 10, nine of nine components (100%) were loose. Receiver-operating-characteristic curve analysis demonstrated very good accuracy (area under the curve = 0.90, p < 0.001). The optimal cutoff point was a score of ≥ 5 with a sensitivity of 0.79, and a specificity of 0.87. Conclusion There was a strong correlation between the scoring system and the probability of loosening of a cementless acetabular component. This scoring system provides a clinically useful tool for pre-operative planning, and the evaluation of the outcome of revision surgery for patients with loosening of a cementless acetabular component in the presence of osteolysis. Cite this article: Bone Joint J 2017;99-B:601–6. The treatment of periprosthetic osteolysis in patients with a cementless acetabular compo- nent at revision total hip arthroplasty (THA) remains controversial. 1,2 The stability of the component may determine treatment. One of the most accepted treatment algorithms has been described by Maloney et al. 3 This system divides cementless acetabular components into three types. In type 1, the component remains stable and the following six criteria are met: the component was not malpositioned; the locking mechanism is intact; the shell is not damaged by penetration of the femoral head; the replacement polyethylene liner is of ade- quate thickness; the design of the component has an acceptable track record; and is modular. Patients who fulfill these criteria are treated with an isolated exchange of the polyethylene liner, debridement of granulomatous tissue and filling with graft material. In type 2, if the com-