Mylène Jansen

114 Chapter 6 Under trial conditions The above described treatment was used for all patients included in the RCTs as well. However, the patients treated in the OPS received 8 instead of 6 weeks of distraction and returned to the hospital every 2 weeks, where the tubes were temporarily removed and the knee was flexed and extended by use of continuous passive motion device for 3 to 4 hours. Pain at the pin sites determined the maximum degree of flexion (mean 25°; range 15°–80°). 8 Follow-up In regular care, weight-bearing PA radiographs were taken and the range of motion (ROM) was measured presurgery and at 4 and 12 months after frame removal in the outpatient clinic. A standard registry for all orthopedic patients provided data on patient-reported outcome measures (PROMs). Patients were requested to fill out several PROMs by questionnaires, before surgery and 3, 6, and 12 months after surgery, and every year thereafter. This is done automatically by e-mail, without reminder, causing relatively high numbers of missing data. Trial patients were seen at comparable time points (6 and 12 months after frame placement) where the ROM was measured and questionnaires were filled out on paper, causing limited missing data. One-year follow-up results have been published previously for each trial separately. 8,13,14 No standardized radiographs were made in regular care and for that reason in clinical practice, the in previous trials reported cartilaginous tissue repair could only be confirmed quantitatively. Since this outcome is a major benefit of the distraction treatment, 2 representative sets of pre- and 1 year post-treatment radiographs of a regular care patient and clinical trial patient have been provided. Data collection All regular care KJD patients treated in our hospital before 2018 were included and thus provided 1-year follow-up. Electronic charts of these patients were evaluated to check essential baseline characteristics. The ROM, measured by the orthopedic surgeon, and complications as a result of treatment had been registered for these patients, data which was also available from the OPS and RCT patients. Only data collected for both regular care and clinical trial patients were compared. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC, version 3.1) questionnaire was used for evaluation of clinical efficacy, as this questionnaire was available for all patient groups. Since regular care patients filled out their questionnaires online, a relatively large amount of missing data is expected. To limit bias, only patients who filled out the questionnaires both before and 1 year after treatment were included in the analysis of clinical