Krill Oil Fails in New Osteoarthritis Trial

Whether an omega-3 supplement derived from tiny Antarctic shrimp-like creatures relieves knee pain from osteoarthritis (OA) no longer has a simple answer, with a new randomized trial contradicting results from earlier studies.

Called KARAOKE, the new 262-patient study failed to show any difference in knee pain reductions with Antarctic krill oil versus a vegetable-oil placebo (-19.9 vs -20.2 points on a 100-point scale), according to Laura Laslett, PhD, of the Menzies Institute for Medical Research in Hobart, Australia, and colleagues writing in JAMA.

But the finding is likely to leave many observers scratching their heads. In three previous randomized, placebo-controlled trials, as Laslett’s group put it in their report, krill oil “reduced some aspects of knee pain, functional impairment, and stiffness” in patients with knee OA and knee pain. While two of these studies were considerably smaller than the new trial, with 47 and 56 patients, the third had 235.

The disparate findings may stem from differences in eligibility criteria, with the new trial being more restrictive in who could participate. One of the previous trials recruited patients with knee pain from any source, including rheumatoid arthritis or other inflammatory conditions as well as OA. Another merely required patients to report knee pain without any particular diagnosis. Only the trial with 235 patients stipulated an OA diagnosis, but not necessarily with objective support from imaging.

In contrast, Laslett and colleagues made a point of restricting enrollment to patients showing effusion-synovitis on MRI scans, so that they could evaluate krill oil’s effects on OA structural features as well as pain. Their results didn’t show improvement in this objective outcome. But the researchers also offered another possible explanation for the discrepant findings, invoking subtle differences between the krill oils used in these studies.

“In the current study,” they wrote, “participants received 2 g/d of krill oil containing 380 mg of EPA [eicosapentaenoic acid] and 200 mg of DHA [docosahexaenoic acid], which led to an Omega-3 Index value of 8.0%, whereas participants in the [235-patient trial] received 4 g/d of krill oil containing 880 mg/d of EPA and DHA (600 mg of EPA and 280 mg of DHA), which resulted in an Omega-3 Index value of 9.0%.” In addition, Laslett and colleagues suggested, it’s possible that participants in the new study were taking in more omega-3 fatty acids because, while the earlier trial limited dietary omega-3 intake, the new one did not.

On the strength of those earlier trials, Antarctic krill oil has become a popular dietary supplement, sold at major store chains and all over the internet. Krill are found in all the world’s oceans and are the chief food for many whale species. As Laslett’s group explained, their omega-3-rich oil is similar to other fish oils but with the addition of astaxanthin, a carotenoid antioxidant, that is also common in salmon (and gives salmon flesh its distinctive pink-orange color) but not other fish oils.

Study Details

Patients with knee pain rated at 40 or higher on a 100-point scale, and who had a clinical OA diagnosis and MRI evidence of effusion-synovitis, were enrolled at five centers in Australia. People with other diagnoses or a history of knee injury were excluded. Participants also had to consent to temporarily stopping krill/fish oil supplements that they were already taking for an initial washout period. Patient-reported pain was the primary outcome measure; change in effusion-synovitis volume was the main secondary endpoint.

Participants were randomized 1:1 to krill oil or placebo taken for 24 weeks. The latter consisted of a mix of olive, corn, and palm kernel oils, with no EPA/DHA and less than 0.5% other omega-3 fatty acids. Mean patient age was about 61 and they were about evenly split between men and women. Baseline knee pain scores averaged just under 50. Most participants were taking some type of over-the-counter medication or supplement to relieve their knee pain.

After 24 weeks, changes from baseline in median effusion-synovitis scores were 0.81 in the krill oil arm compared with -0.94 with placebo; the between-group difference of 1.75 points reached statistical significance (P=0.01). “The reason for this unexpected outcome remains unclear,” Laslett and colleagues wrote. They were skeptical that the vegetable-oil placebo could have had a genuine effect, concluding, “this finding may have been due to chance.”

That was the only measure to show even a hint of a between-group difference. Specific aspects of pain, such as total versus weight-bearing, as well as functional assessments largely improved in both groups and to essentially the same degree.

Limitations included the possibility of inadequate krill oil dosing and the use of MRI without contrast. Also, the researchers acknowledged that dietary intake of omega-3s could have differed between groups and influenced the results.

Please enable JavaScript to view the

comments powered by Disqus.