Angling for Something Safer and Less Costly: Fish Oil for Rheumatoid Arthritis


Written By

Stephanie Kasten, PharmD
Arthur Schuna, MS, BCACP

Reviewed By

Dana A. Brown, PharmD, BCPS
Lisa Charneski, PharmD, BCPS


Proudman SM, James MJ, Llewellyn DS, et al. Fish oil in recent onset rheumatoid arthritis: a randomized, double-blind controlled trial within algorithm-based drug use. Ann Rheum Dis. 2015; 74:89-95.

In an age when expensive biologic agents get all of the attention, could adding something as simple and inexpensive as fish oil to disease-modifying anti-rheumatic drugs (DMARDs) in patients with rheumatoid arthritis make a difference?  Could it delay the progression to more intensive, expensive, and risky therapies?

A number of studies using fish oil have shown it can reduce the number of tender joints and duration of morning stiffness in patients with rheumatoid arthritis (RA).1,2  These effects were observed in patients with long-standing RA who took fish oil for at least three months and at a dose of at least 2.7g daily.3  Moreover, fish oil use allows some patients with rheumatoid arthritis to reduce or discontinuation non-steroidal anti-inflammatory drugs (NSAIDs).1,4,5 Fish oil may have anti-inflammatory properties similar to NSAIDs but a safer option because it does not cause gastrointestinal bleeding or increase cardiovascular risk.3,6

The results of a double-blind, randomized, controlled trial adds fresh new evidence regarding the potential benefits of fish oil.7 Patients were screened and enrolled in this multicenter, investigator-initiated Australian study between September 2001 and December 2008. Patients at least 18 years of age, diagnosed with rheumatoid arthritis in the previous 12 months using the 1987 revised American College of Rheumatology (ACR) criteria, who had polyarthritis, and DMARD naïve were included in the study. Patients were excluded if they had used DMARDs or anti-malarials  for more than 1 month; had recent sero-conversion to parvovirus, Ross River virus, Barmah Forest, or rubella viruses; had an antinuclear antibody titer greater than 1:320; evidence of hepatitis B, C, or HIV; known sensitivity to methotrexate (MTX), sulfasalazine, or hydroxychloroquine (HCQ); or a systemic disease likely to increase the risk of toxicity to these drugs.

Of the 187 patients screened, 140 were randomly allocated in a 2:1 ratio to a high or low-dose fish oil groups. The high dose group took 5.5g every day of eicosapentaenoic acid and docosahexaenoic acid (EPA+DHA) while the low dose group, which served as the control, received only 400mg per day of EPA+DHA.  All patients took 10mL of liquid fish oil daily. Baseline characteristics were generally similar between the two groups except smoking status, steroids use at baseline, and presence of the shared epitope. These group differences were later taken into account and adjusted for during data analysis.

All participants received MTX 10mg weekly, folic acid 500mcg daily, sulfasalazine 500mg daily and HCQ 200mg twice daily. Sulfasalazine was automatically increased over 4 weeks to 1gram twice daily. DMARD therapy was escalated using an established algorithm until maximal tolerated doses of triple therapy were achieved.  If symptoms persisted, leflunomide was added. DMARD therapy was not withdrawn for lack of efficacy. In the event of toxicity or intolerance, the most likely culprit was identified and the dose was reduced, the drug was discontinued, or the formulation was changed at the investigator’s discretion. NSAID use was actively discouraged. Parenterally administered corticosteroids were permitted but oral corticosteroids were tapered and discontinued when possible. Patients were evaluated every 3 weeks for the first 12 weeks and then every 6 weeks thereafter for 52 weeks. Drug toxicity was assessed at every visit and with lab tests every 3 weeks.  

The primary study outcome was “failure” of triple DMARD therapy — defined by progression to leflunomide. Secondary study outcomes assessed included fulfilment of ACR criteria for remission on at least one occasion, disease activity scores using DAS28-EST and EULAR responses, and assessment of physical disability using the modified Health Assessment Questionaire (mHAQ). The study was adequately powered and used an intent-to-treat analysis.

By the end of 52 weeks, the number of patients in the high-dose fish oil (11/87) and control (6/53) groups who withdrew from the study were similar. Nine of 86 patients (10.5%) in the high dose fish oil group and 17 of 53 patients (32.1%) in control group failed triple therapy and progressed to leflunomide (p < 0.01). One patient in the control group progressed to biologic therapy.

The time to first ACR remission was significantly shorter in the high-dose fish oil group and participants were twice as likely to meet ACR criteria for remission (p < 0.05).  EULAR, DAS28, and mHAQ responses were not statistically different. There was no difference between groups in the proportion of participants who received steroids at any time after enrollment and no difference between groups in the cumulative steroid dose. Of the participants taking NSAIDs at baseline, 72% of the fish oil group and 44% of the control group discontinued their NSAIDs by 12 weeks (p=0.07). One person in the control group was taking NSAIDs at the 52 week follow-up.

Adherence with fish oil was lower in the high-dose group when compared to the low-dose group. Adherence was evaluated through participant verbal inquiry and direct measurement of the amount of liquid left in the bottle. The calculated average daily intake of EPA+DHA was 3.7g in the high dose group and 0.36g in the control group. The proportion of participants who experienced any adverse event was similar in the high and low-dose groups (86.0% vs. 92.4%).  Serious adverse events (requiring hospitalization) were somewhat higher in the high-dose group but were not statistically different (11.6% vs. 3.8%; p =0.13). The most common serious adverse events were cardiac in nature and included myocardial infarction, unstable angina, and arrhythmias.  There were no deaths.

Although this study and others suggest there may be a benefit from fish oil in patients with RA, it is not known whether fish oil has disease modifying properties or if it merely improves symptoms.  Also unknown is the dose needed to meet the anti-inflammatory threshold. It is also not known whether similar benefits would be seen in patients with long-standing disease since this study only enrolled patients who were DMARD naïve and diagnosed in the previous 12 months.  Additional studies are needed to assess the benefit of fish oil when added to other conventional treatment strategies, such as DMARD monotherapy or biologic therapy. Lastly, It is unclear whether the adherence issues observed in this study were due the high-dose liquid product causing frequent gastrointestinal side effects and an unpleasant fishy taste.  Perhaps the capsule formulation, which is more commonly used by patients, would have been better tolerated.

This study found that in patients with early onset rheumatoid arthritis receiving triple DMARD therapy, fish oil prevented or delayed the need for additional therapy.  Patients who took fish oil were more likely to achieve disease remission.  Moreover, there was no statistical difference in adverse events.  These findings suggest that fish oil treatment could reduce the need for costly drug therapy and decreased exposure to agents with some very serious adverse effects.  However, while the data are promising, there are simply too many unanswered questions and product quality concerns for us to routinely recommend using high dose fish oil in patients with RA.   What do you think?