C148 Poster Presentations 263 THE EFFICACY OF A SINGLE DOSE OF LUMIRACOXIB 400 MG GIVEN PREEMPTIVELY VERSUS POST-OPERATIVELY AFTER AMBULATORY ARTHROSCOPIC KNEE SURGERY J. Grifka 1 , J. Zink 2 , J. Hugot 3 , A. Kreiss 4 , U. Arulmani 3 , V. Yu 5 , G. Krammer 3 1 Orthopädische Klinik der Univ, Bad Abbach, Germany; 2 Orthopädische Fachpraxisgemeinschaft, Jena, Germany; 3 Novartis Pharma AG, Basel, Switzerland; 4 Novartis Pharma GmbH, Nuremberg, Germany; 5 Novartis Pharma AG, East Hanover, NJ Purpose: The purpose of this study was to investigate whether preemptive dosing with lumiracoxib is more effective than post- operative dosing in reducing post-operative pain associated with ambulatory arthroscopic knee surgery. Methods: A multicenter, randomized, double-blind, placebo con- trolled, parallel group study comparing lumiracoxib 400 mg given preemptively versus post-operatively in adult patients undergo- ing minor ambulatory arthroscopic knee surgery (e.g. removal, trimming, shaving or repair of meniscal or articular cartilage including lavage). Eligible patients were randomized to either lumiracoxib 400 mg given preemptively (1 hour before start of surgery); lumiracoxib 400 mg given post-operatively (15 minutes after the completion of surgery); and placebo in a 2:2:1 ratio. The primary endpoint was the pain intensity (PI) after movement in the target knee 2 hours after surgery based on a 0-100 mm Visual Analog Scale (VAS) using Wilcoxon rank sum test. The secondary endpoints included time to rescue medication intake, PI at 1, 2, 3, 4 and 24 hours while at rest, and PI at 1, 3, 4 and 24 hours after movement, and both measured on a 0-100 mm VAS using Logrank test. Results: In total, 110 patients were randomized (lumiracoxib pre- emptive n=45, lumiracoxib post-operative n=44, placebo n=21), and all patients completed the study. The difference in me- dian PI after movement at 2 hours in the lumiracoxib preemp- tive group (PI=2.75) compared to the lumiracoxib post-operative group (PI=3.50) in the intent-to-treat (ITT) population was 0.0 (95%CI: -2,1), p=0.602; while lumiracoxib preemptive and post- operative groups compared to the placebo (PI=9) were -4.0 (95%CI: -9,-1), p=0.007 and -3.5 (95%CI: -8.5,0), p=0.052 re- spectively. The time to rescue medication intake was significantly longer for lumiracoxib preemptive (p=0.003) and post-operative groups (p=0.001) compared to placebo using Logrank test (Figure 1). The difference in median PI (at rest) post-surgery was not statis- tically significant between the lumiracoxib preemptive group and Figure 1. Time-to-first rescue medication (ITT population). the post-operative group at any timepoints (1, 2, 3, 4 and 24 hours). The difference in median PI (at rest) post-surgery for lumiracoxib preemptive and post-operative groups compared to the placebo was statistically significant at 2, 3, 4 and 24 hours. The difference in median PI (after movement) post-surgery was not statistically significant between the lumiracoxib preemptive group and the post-operative group at any timepoints (1, 3, 4 and 24 hours). The difference in median PI (after movement) post-surgery for lumiracoxib preemptive and post-operative groups compared to placebo was statistically significant at 1, 3, 4 and 24 hours. The proportion of patients with adverse events (AEs) were similar for lumiracoxib preemptive (4.4%), post-operative (4.5%) and placebo (4.8%), with no serious AEs reported. Conclusions: Preemptive or postoperative dosing of lumiracoxib 400 mg od had similar efficacy in reducing post-operative pain following ambulatory arthroscopic knee surgery 264 CLINICAL BENEFITS OF USING AN AUTOLOGOUS PREPARATION RICH IN GROWTH FACTORS TO TREAT KNEE OSTEOARTHRITIS M. Sánchez 1 , E. Anitua 2 , J. Azofra 1 , M. de la Fuente 2 , M. Zalduendo 2 , J.J. Aguirre 2 , I. Andia 2 1 Unidad de Cirugía Artroscópica, UCA, Vitoria, Spain; 2 Biotechnology Institute, Vitoria, Spain Purpose: An autologous preparation rich in growth factors (PRGF) secretes a complex mixture of biological mediators es- sential for natural repair, including TGF-ß1, PDGF, VEGF, HGF and IGF-I. Due to the localized nature of OA, the possibility of intra-articular administration of PRGF, along with its biocompati- bility and non-immunogenicity, may make this unique molecular mixture an attractive treatment for OA. Purpose: (1) To explore the potential clinical benefits of PRGF injections for the treatment of OA in a retrospective observational study. (2) To characterize the PRGF treatment in OA patients. Methods: A total of 112 patients with symptomatic knee OA were treated with a series of three weekly intra-articular injections of PRGF and studied retrospectively. Anterio-posterior weight- bearing radiographs were scored for Ahlbäck severity grade. In bilateral patients, the Ahlbäck grade used for analyses was the higher of the two knees. The primary efficacy criteria were mean change from baseline through two and six months in the WOMAC index pain and physical function subscale scores and global WOMAC assessment at month two and six. A repeated measures ANCOVA model was used to compare clinical parame- ters at baseline and at follow-up. Age, gender, Ahlbäck grade and BMI were included in the models. Success rates were calculated as a reduction in the WOMAC pain score of at least 40% from baseline and a 35% reduction in the physical function and global WOMAC scores. For preparation of the PRGF treatment, unco- agulated peripheral blood was collected from the patient; after one single-step centrifugation, the plasma fraction located just above the buffy coat was aspirated and dispensed into an empty tube under vertical air flow conditions. Platelet activation was in- duced by CaCl 2 addition a few minutes prior to PRGF infiltration. Enzyme-linked immunosorbent assays were used for determin- ing levels of PDGF-AB, VEGF, HGF, IGF-I and TGF-β1 in PRGF. Results: The mean age of the participants was 67 years (range: 42-91). Of the participants, 63% were female and the average BMI was 28 kg/m 2 (range: 21-40). Of the 112 knee radiographs included in the study, twenty-three (20%) were graded Ahlbäck I, 50 (45%) were graded Ahlbäck II, 29 (26%) were graded Ahlbäck III and 10 (9%) were graded Ahlbäck IV. The differences between pain subscale scores at baseline and two or six months were