POS0276 EFFICACY OF KRILL OIL IN THE TREATMENT OF KNEE OSTEOARTHRITIS: A 24-WEEK MULTICENTRE RANDOMISED DOUBLE-BLIND CONTROLLED TRIAL

Background: Elevated levels of systemic inflammation are common in people with osteoarthritis and predict both pain and structural outcomes. Krill oil has anti-inflammatory properties and reduces severity of inflammatory arthritis in mice by 50% compared to controls. 1 In humans, krill oil reduced knee pain and function in two short, moderate quality randomised controlled trials (RCTs) in people with osteoarthritis. However, evidence from longer trials with imaging data is lacking. Objectives: The aim of this study was to compare the efficacy of krill oil (2g / day) vs. placebo for treating knee pain in patient with clinical knee osteoarthritis who have significant knee pain and effusion-synovitis. Methods: KARAOKE was a 24-week multicentre, randomised, double-blind, placebo-controlled trial conducted at five Australian sites. Participants aged ≥40 years with symptomatic knee OA (according to ACR criteria), significant knee pain (pain score ≥40mm on a 100mm visual analogue scale [VAS]), and effusion-synovitis present on MRI (grade ≥1 according to modified Whole-Organ Magnetic Resonance Imaging Score (WORMS) scoring) were eligible. The study protocol has been published previously. 2 Participants were randomised to receive 2g/day of krill oil, (350 mg/g omega-3 content, 12 mg/g total omega-6 content) or inert placebo (vegetable oil, no EPA or DHA, <5 mg/g (0.05%) other omega-3s). The primary outcome was absolute change in knee pain assessed using a VAS [0-100mm] after 24 weeks. Secondary outcomes were: change in knee pain and function assessed using Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) score [0-500mm]), change in back and hand pain assessed using a VAS [0-100mm], change in lower limb leg strength assessed using a dynamometer, and change in blood parameters (including CRP, triglycerides, fasting glucose and total, HDL, LDL cholesterol), after 24 weeks. Linear mixed-models were used, using patient identification as random intercepts and trial centre and treatment month as random effect to adjust for correlated data within trial centres and repeated measures and to allow different treatment effects among patients over time, respectively. Results: 262 participants were randomised (mean age 61.5 years, 53% females) to receive krill oil (n=130) or placebo (n=132). A total of 85% completed the trial. Knee pain improved in both groups over 24 weeks, but with no between-group difference (krill oil, -20.1mm; placebo, -19.3mm, p=0.81). Secondary outcomes: knee pain and function score improved in both groups, but with no between-group difference (WOMAC pain: krill oil, -86.7; placebo, -82.5mm, p=0.81; WOMAC function: krill oil, -245.3; placebo, -184.3, p=0.14 at 24 weeks). The same applies for hand pain and back pain. No significant changes were seen in leg strength or any of the blood parameters at 24 weeks). Incidence of one or more adverse events was 50% in the krill oil group (n=66) and 55% in the placebo group (n=71). There were 8 serious adverse events in the krill oil group 6 in the placebo group, all considered unrelated to treatment. Conclusion: Krill oil was safe and well tolerated, but did not significantly reduce knee pain in patients with clinical knee osteoarthritis, significant knee pain and effusion-synovitis after 24 weeks compared to placebo. These findings do not support use of krill oil for alleviating knee pain in clinical knee osteoarthritis. References: [1]Ierna M, et al. BMC Musculoskelet Disord 2010;11:136. [2]Laslett L, et al. Trials 2020;21:79 Outcomes Absolute between group difference at 24 weeks P value Primary Knee pain 0.8 (-5.6 to 7.2) 0.81 Secondary Knee pain (WOMAC) 4.2 (-29.1 to 37.5) 0.81 Knee function (WOMAC) 61 (-19.2 to 141.3) 0.14 Hand pain 2.8 (-2.6 to 8.3) 0.31 Back pain 1.9 (-3.9 to 7.8) 0.46 Leg strength -2.59 (-9.41 to 4.23) 0.52 Metabolic factors Total Cholesterol 0.09 (-0.1 to 0.29) 0.34 HDL Cholesterol -0.03 (-0.1 to 0.03) 0.35 LDL Cholesterol 0.05 (-0.12 to 0.22) 0.57 Triglycerides 0.12 (-0.09 to 0.33) 0.27 Fasting glucose 0.01 (-0.26 to 0.29) 0.93 hsCRP 0.64 (-0.56 to 1.84) 0.30 Disclosure of Interests: Laura Laslett Speakers bureau: once, several years ago, and unrelated to this topic, Grant/research support from: Yes, received funding from Aker Biomarine to conduct this trial, Lieke Scheepers Shareholder of: AstraZeneca, Grant/research support from: Pfizer, unrelated to this topic, Employee of: Previously employed by AstraZeneca, Benny Antony Speakers bureau: Zydus, Grant/research support from: Grant support for investigator-initiated trial from NR Ltd for unrelated research, Anita Wluka: None declared, Catherine Hill: None declared, Lyn March Speakers bureau: Speaker fees from Pfizer Australia Ltd, Bristol Myer Squibb Australia, Abbvie Australia, Grant/research support from: Grant support for my institution from Janssen for unrelated research, Helen Keen: None declared, Petr Otahal: None declared, Flavia Cicuttini: None declared, Graeme Jones: None declared