Joost Peerbooms

42 Chapter 3 Ethics All patients had to be able to read and understand the protocol and the informed consent. The Medical-Ethical Committee (METC) and the National and Institutional Review Board approved the study. METC protocol number 04-17. Date of approval 27 th October 2004. The trial was performed in compliance with the Helsinki Declaration 2000 and Good Clinical Practice 1997. RESULTS From June 2005 to March 2007 a total of 111 patients with an indication for TKA were included in the study. 9 patients had to be excluded due to inclusion errors. Analysis was per protocol and focused on the remaining 102 patients to a pre-established analysis plan based on the principle of intention to treat: 50 TKA were treated with PG and 52 without PG. Patient characteristics at baseline were similar between the two groups (Table 2). However, due to a reorganization of the patient’s ward, no or partial measurements were recorded in a number of the included patients during hospital stay. Missing data were imputed based on the “last known result carried forward” principle. Characteristics of the patients who were fully recorded compared to the other patients were similar (Table 3). Eventually, full data were recorded for 32 patients in the PG group and 41 patients in the C group. Analysis for possible bias caused due to missing data showed no differences between resulting groups at baseline (Table 3) (Figure 1). Table 2 Patient Characteristics at Inclusion Primary endpoint (wound healing) Both groups had normal wound healing (Table 4). No clinical or statistical differences in wound closure were notable between the two groups during hospital stay (third day, 95% CI: 11% to 25% difference; fourth day, 95% CI: 30% to 10% difference). At 2 weeks