Clinical Validation of Model-based RSA for a Total Knee Prosthesis Bart L. Kaptein, MSc, PhD * ; Edward R. Valstar, MSc, PhD *,† ; Berend C. Stoel, PhD ‡ ; Hans C. Reiber, MSc, PhD ‡ ; and Rob G. Nelissen, MD, PhD * Roentgen stereophotogrammetric analysis generally is ac- cepted as the most accurate method to measure prosthesis migration. A disadvantage of the method is it requires mark- ers in the bone and the prosthesis. Model-based roentgen stereophotogrammetric analysis circumvents the need for markers on prostheses by fitting virtual projections of a three-dimensional surface model of a prosthesis to its actual projections in the roentgen image. We confirmed a model- based roentgen stereophotogrammetric analysis for a tibial component. Using implants with attached markers, we com- pared model-based roentgen stereophotogrammetric analy- sis with marker-based roentgen stereophotogrammetric analysis. In addition, we assessed precision of the model- based roentgen stereophotogrammetric analysis with a phan- tom experiment. The precisions for translations of marker- based and model-based roentgen stereophotogrammetric analysis were 0.06 and 0.11 mm, respectively, and for rota- tions, the precisions were 0.20° and 0.23°, respectively. The precisions of model-based roentgen stereophotogrammetric analysis calculated from the phantom data were 0.08 mm for translations and 0.13° for rotations. Although model-based roentgen stereophotogrammetric analysis is less precise than marker-based roentgen stereophotogrammetric analysis, its precision is still acceptable for most clinical applications, especially where marker-based roentgen stereophotogram- metric analysis has practical limitations. Level of Evidence: Level II, prognostic study. See the Guide- lines for Authors for a complete description of levels of evi- dence. In 1974, Selvik 12 introduced roentgen stereophotogram- metric analysis (RSA) for measuring micromotion of joint implants relative to the host bone. Ryd et al 11 suggested micromotion greater than 0.2 mm per year after 2 years postoperatively predicted long-term survival of tibial knee components. Subsequently, RSA as a predictive tool was confirmed for hip implants, 2,7,8 making RSA valuable for implant evaluation. The reported accuracy of RSA ranges between 0.05 and 0.5 mm for translations and between 0.15° and 1.15° for rotations (95% confidence inter- vals). 6,14 To guarantee the high accuracy of RSA, until recently it was necessary to place markers (usually tanta- lum beads) on the implant and in the surrounding bone. From the projections of these markers, detected in stereo- roentgen images, their three-dimensional (3-D) positions are reconstructed, resulting in an accurate calculation of the pose of the implant relative to the bone. 1,14,16 As of September 1, 2007, hip, shoulder, and knee im- plants were reclassified in the European Union from Class IIB to Class III, 15 the highest risk category for medical devices. Because an implant with RSA markers is consid- ered a new design, requiring an additional CE marking procedure, one of the major problems of placing beads on the implant is that it increases the costs and lengthens the start-up period of an RSA study. Our discussions with implant manufacturers suggest markers on the implant may jeopardize its strength and cause local stress raisers in the bone cement. The latter may result in cement cracks that decrease the strength of the fixation of the implant in the bone. Another problem is markers attached to the im- plant often are overprojected by the implant. Received: March 25, 2007 Revised: June 6, 2007; July 18, 2007; July 24, 2007 Accepted: August 1, 2007 From the * Department of Orthopaedics, Leiden University Medical Center, Leiden, The Netherlands; the † Department of Biomechanical Engineering, Faculty of Mechanical, Maritime, and Materials Engineering, Delft Univer- sity of Technology, Delft, The Netherlands; and the ‡ Division of Image Processing, Department of Radiology (LKEB), Leiden University Medical Center, Leiden, The Netherlands. The institution of the authors has received funding from Zimmer, Inc, War- saw, IN. Each author certifies that his or her institution has approved the human protocol for this investigation, that all investigations were conducted in con- formity with ethical principles of research, and that informed consent for participation in the study was obtained. Correspondence to: Bart L. Kaptein, MSc, PhD, Department of Orthopae- dics, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands. Phone: 31-71-526-4542; Fax: 31-71-526-6743; E-mail: B.L.Kaptein@lumc.nl. DOI: 10.1097/BLO.0b013e3181571aa5 CLINICAL ORTHOPAEDICS AND RELATED RESEARCH Number 464, pp. 205–209 © 2007 Lippincott Williams & Wilkins 205 Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.