Table of Contents:
Total endoprosthesis surgery of the knee and hip joints are established procedures and routine surgeries in Germany. The high number of annually conducted surgeries is indicative for the importance of continuously monitoring and improving the repertoire of procedures associated with such surgery. In particular, pre-surgery prosthetic templating is an essential factor for minimising the rate of complications. Foremost due to the digitalisation of medical X-ray diagnostics, the approach in the pre-surgery templating stages of the procedure has evolved. A digital system based on a reference object (RK), which determines a specific magnification factor (VF) for each X-ray image was developed to compensate for beam divergence. This system has since replaced template-based, analogue prosthetic planning.
The aim of our study was to compare both templating techniques with respect to their accuracy and to illustrate their respective strengths and weaknesses. Further, im-provements and specifications of the templating techniques were evaluated. To this end, we conducted a series of X-ray experiments focusing on the effects of incorrect positioning of the RK. Subsequently the personal details and pre-surgery X-ray im-ages of 300 patients, who received a knee endoprosthesis (n = 150) or a hip endo-prosthesis (n = 150) at the Clinic for Special Orthopaedic and Trauma Surgery of the Auguste-Viktoria-Clinic Berlin between October 2010 and September 2011 were collected retrospectively. The images were used to determine the respective VF, the diameter of the acetabulum and the diameter of the tibial joint surface of the knee. For comparison with analogue templating, we calculated which diameters would have been measured in the analogue technique using the fixed VF of 107.5%, 110%, 115%, 118% and 120%.
Irrespective of the applied templating technique, it became apparent that also the X-ray set-up and the resulting distances (i.e. focus area to detector and object to detec-tor distances) may play a role in relation to the VF. The advantage of the RK-based approach, namely to be able to adjust for changed distances, is of greater relevance when taking overview images of the pelvis due to the smaller X-ray distances. Our X-ray series shows an increased propensity for error with respect to the placing of the RK at the hip joint. In particular, when taking overview images of the pelvis the exact position of the RK in relation to the centre of the hip joint is difficult to deter-mine, which, in turn, may favour erroneous placing. Erroneous placing leading to a vertical displacement of 3 cm can already lead to a templating error of almost +/- one prosthetic size (+/-1.8 mm). In case of the knee joint, a vertical displacement of 10 cm would be required to result in a templating error of one prosthetic size (+/-3.0 mm).
Our data relating to patients with hip joint treatment show superiority of digital tem-plating versus the analogue technique. The analogue templating factor of 118% cor-responded to our average (MW) but lead to worse results as individual adjustments were not possible. We recommend using the templating factor of 118% if digital templating is not available. In such cases the X-ray distances used in the present study should be adhered to. In our set-up for taking overview images in a.p., the dis-tances were 110 cm between focus and detector and 7 cm between table level and detector.
Based on images of the whole leg in a standing position the match of measurement of implanted prosthetics was better when using fixed analogue templating factors. The measurements of the tibia plateau showed a greater distance to the implanted tibia components as it would have been the case using the fixed templating factors of 107.5% or our average of 106.9%. We recommend the use of a fixed VF of 106.9%, if the distances of our X-ray set-up are being used (focus - detector: 270 cm, table level - detector: 7 cm). If it is possible to manually adjust the VF in the templating software, it would make the complicated positioning of the RK and its measurement superfluous. In individual cases with significantly increased or reduced distances be-tween joint centre and detector a RK-based calibration could be reverted to.