Mylène Jansen

216 Chapter 11 respectively. For all patients, the KL grade was determined before treatment. All details with regard to inclusion criteria and treatment have been described in detail previously for all 3 studies. 12–15 All trials were approved by the medical ethical review committee of the University Medical Center Utrecht (protocol numbers 04/086, 10/359/E, and 11/072) and registered in the Netherlands Trial Register (trial numbers NL419, NL2680, and NL2761). All patients gave written informed consent, which included further use of their data for additional research. Radiography Standardized, semi-flexed PA radiographs were performed under full weight-bearing according to the Buckland-Wright protocol. 16,17 An aluminum step wedge was placed alongside the knee, against the detector and within the field of exposure, in order to quantify bone density and determine the pixel size corrected for possible magnification. Radiographs were taken pre- treatment (baseline) and 1 and 2 years post-treatment. KIDA analysis The KIDA analysis method has not changed since the original publication in 2008. 5 First the aluminum step wedge is identified by the user by indicating the 4 corners of the wedge, after which the program automatically draws the outline of the entire wedge and the different steps (Figure 1). From this, it calculates the pixel size and the reference mm aluminum equivalent (mm Al eq) with which subchondral bone density can be expressed. Next, the user places a framework of 4 lines around the joint, that touch on the medial and lateral side of the joint (2 longest vertical lines), and on the distal femur and proximal tibia (horizontal lines; Figure 1). From these last 2 lines, perpendiculars are calculated, 4 on each area (medial and lateral femur and tibia) at predefined calculated positions; 1 circle along each perpendicular can be moved by the user to place the edge of the circle at the bone-‘cartilage’ interface (16 smallest circles in Figure 1). The distance between each pair of circles is calculated to measure the JSW in mm, at 4 locations of the medial and the lateral compartment. These 4 distances can be averaged to obtain a mean medial and mean lateral JSW, and all 8 distances can be averaged for a mean JSW of the whole joint, all in mm. The mean intensity in each circle is calculated as well, and can be averaged to obtain the subchondral bone density at the medial and lateral tibia and femur, expressed in mm Al eq.