RESUMEN
This study determined the accuracy of standard radiographs in observing radiolucency adjacent to the tibial component in total knee arthroplasty. A model of the proximal tibia was used. Plastic sleeves of different heights were glued under the tibial tray to imitate radiolucency. By tilting the x-ray tube 1degrees at a time, an investigation was performed to determine how many degrees it would take to cover the heights of the different plastic sleeves. In a clinical study of patients at standard follow-up, 21 knees were examined using standard radiographs. Radiographs using image intensification were then taken of the same knees to obtain radiographs with views parallel to the plane of the tibial tray. The results of the study on the phantom showed that a 1degrees deviation of the x-ray beam from the optimal position would obscure a 0.5-mm "radiolucency." This was true for radiolucencies of 2, 3, 4, and 5 mm. This was confirmed in the clinical study where < or = 4 mm of radiolucency was obscured on standard radiographs. Therefore, standard radiographs cannot be used to determine radiolucency. The 2-mm radiolucency normally associated with loosening is not clinically valid.