Effect of Dietary and Supplemental Omega-3 Polyunsaturated Fatty Acids on Risk of Recurrent Gout Flares

Abstract

Objective

To determine the relationship between omega-3 polyunsaturated fatty acid (n-3 PUFA) consumption (dietary or supplemental) and risk of gout flares.

Methods

We used data from the Boston University Online Gout Study, an internet-based, case-crossover study conducted from 2003–2012. Participants completed questionnaires at times of gout flares (hazard periods) and during gout flare-free periods (control periods) regarding exposures (including supplements and diet) during the preceding 48 hours. We examined the relation of self-reported n-3 PUFA-rich supplements and fish intake to the risk of recurrent gout flares using conditional logistic regression, adjusting for total purine intake, diuretic use, and other urate-lowering or flare prophylactic medications (allopurinol, non-steroidal anti-inflammatory drugs, colchicine).

Results

Of the 724 participants, 85% met the 1977 Preliminary ACR classification criteria for acute gout. In the preceding 48 hours, 22% of participants reported some form of n-3 PUFA consumption (supplements: 4.6%; dietary fatty fish: 19%). The adjusted odds ratios (aOR) were 1.01 (95% CI, 0.63–1.60; p=0.98) for all three supplement types combined and 0.74 (95% CI, 0.54–0.99; p=0.04) for ≥2 n-3 PUFA-rich fish servings.

Conclusions

Dietary n-3 PUFA-rich fish consumption, when adjusted for total purine intake, was associated with lower risk of recurrent gout flares, whereas n-3 PUFA supplementation alone, as taken in a self-directed manner, was not. Consumption of specific sources and adequate doses of n-3 PUFA for gout flare prevention warrants further study in an adequately powered clinical trial.

Gout, an inflammatory crystal arthritis caused by monosodium urate deposition within joints, affects roughly 8.3 million adults in the U.S.(1, 2) Although the pathophysiology of this disease is well understood and urate lowering therapies are widely available, recurrence of gout flares remains high.(2, 3) Identifying additional options for gout flare management could alleviate the burden of this common inflammatory disorder.

Omega-3 polyunsaturated fatty acids (n-3 PUFA) have recently garnered interest for their potential anti-inflammatory effects. Fatty fish such as tuna, mackerel, and salmon are a rich natural source of the biologically active n-3 PUFAs eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA).(4) Although another formulation of n-3 PUFA called α-linolenic acid can also be found in plant-based sources such as flax, canola, soybean, and walnuts, α-linolenic acid is biologically inactive and must be consumed in large quantities to achieve the same level of EPA and DHA available in fatty fish.(4, 5) With respect to gout, certain n-3 PUFAs have been shown to have multiple anti-inflammatory effects through rapid and selective inhibition of the NLRP3 inflammasome via G-protein coupled receptors, specifically GPR120 and GPR140 through arrestin beta 2.(6–9) Additional downstream effects include suppression of toll-like receptor activation, neutrophil chemotaxis, NF-kappa B activity, and prostaglandin synthesis.(8–11) These in vitro findings have been supported by in vivo experiments, where rats fed fish and plant seed oil rich diets compared to standard diets developed less inflammation when injected with monosodium urate crystals.(12) Importantly, these anti-inflammatory effects occur independent of enzymatic product generation, a process that relies on the slow integration of n-3 PUFAs into the leukocyte plasma membrane.(9) By bypassing this slower pathway, it is possible that n-3 PUFAs have acute benefits even when consumed over short intervals.

Clinically, n-3 PUFA supplementation has demonstrated anti-inflammatory effects in several arthritic conditions. Multiple randomized controlled trials, systematic reviews and meta-analyses have noted reduced disease activity with n-3 PUFA consumption in patients with rheumatoid arthritis and psoriatic arthritis.(13–20) Fewer studies have explored the impact of n-3 PUFA consumption on gout, however. Abhishek and colleagues showed that low n-3 PUFA serum levels were associated with increased frequency of gout flares, although they were not able to adjust for other dietary confounders.(21) Interestingly, n-3 PUFA rich diets have also been linked to reduced inflammation in the gut microbiome, with clinical implications for metabolic syndrome, cancer, and non-alcoholic fatty liver disease.(22–25) Whether this is the same mechanism by which n-3 PUFA can affect gout remains to be elucidated.

In contrast to n-3 PUFAs, n-6 PUFAs like arachidonic acid and linoleic acid, which are commonly found in vegetable oils and animal sources, possess greater pro-inflammatory properties mediated by generation of prostaglandins, leukotrienes, cyclooxygenase, and thromboxanes.(5) There is evidence that the ratio of n-3 to n-6 PUFA consumption plays a critical role in the development of inflammation.(8, 26) Diets rich in n-6 PUFA but low in n-3 PUFA have been associated with increased inflammation in rheumatoid arthritis, cardiovascular disease, cancer, and even osteoarthritis.(27–29) No studies to date have examined the effect of n-6 PUFA on gout flare development. Understanding the benefits and potential harms of n-3 and n-6 PUFA consumption with regards to gout may have broad health implications given roughly half of American adults report using one or more dietary supplements, with around 12% using supplements for joint health.(30)

In this study, we sought to determine the relationship between n-3 PUFA consumption (dietary or supplemental) and risk of gout flares in a case-crossover study. We hypothesized that subjects taking n-3 PUFA supplements or consuming n-3 PUFA rich diets would have lower risk of gout flare recurrence compared with subjects who did not consume n-3 PUFAs. We also sought to examine the impact of n-3:n-6 levels on gout flare risk.

PARTICIPANTS AND METHODS

Study Design

The Boston University Online Gout Study was an internet-based, case-crossover study conducted from 2003–2012 originally aimed at identifying risk factors for recurrent gout flares.(31) Each subject served as his or her own control eliminating between-person differences (e.g., genetics, sex, education, and race) to minimize time-invariant confounding.(32) Such an approach allowed for comparison of the relative frequency of risk factors prior to gout flares (hazard periods) and during gout-flare free periods (control periods) within the same subject. Several nutritional factors, including purines, alcohol, cherry, in relation to gout flares have been examined in this study.(33–35) This study was approved by the Institutional Review Board of Boston University Medical Campus.

Study Sample

Potential subjects were recruited through a Google advertisement linked to the search term “gout,” which then directed them to the study website (https://dcc2.bumc.bu.edu/GOUT) on an independent secure server within the Boston University Medical Center domain. At baseline, subjects provided information on sociodemographic characteristics, medical history, past medications, and gout-related history (i.e. diagnosis of initial gout attack, age of onset, medication used for the treatment of attacks, and number of gout attacks in the last 12 months).

Eligibility criteria included subjects who were at least 18 years of age, U.S. residents, had a diagnosis of gout by a physician, had a gout flare within the preceding 12 months, agreed to the release of medical records pertaining to gout diagnosis and treatment, and provided electronic informed consent. Gout diagnoses were confirmed through chart review of participant medical records and/or a checklist of the features in the American College of Rheumatology (ACR) preliminary criteria for the classification of gout completed by the subject’s physician.(36) All medical records and checklists were separately reviewed by two rheumatologists to ensure accurate diagnosis of gout according to the ACR criteria.

Gout Flare Assessment

At the time of a gout flare, subjects were asked to complete online questionnaires regarding the attack including date of onset, anatomic location of the flare, clinical signs and symptoms (e.g., erythema, swelling, or pain within 24 hours), and treatment (e.g., colchicine, nonsteroidal anti-inflammatory drugs [NSAIDs], systemic corticosteroids, and intra-articular corticosteroid injections). This method has been advocated by the ACR/European League Against Rheumatism and utilized in gout trials for defining gout flares based on patient-reported features.(37, 38) To ensure more stringent inclusion criteria, definite flares were further limited to those presenting with erythema, podagra, and maximal pain within 24 hours of onset; requiring treatment with at least one anti-gout medication (mentioned above); and a combination of any of those features (e.g., cases with 2, 3, or all 4 features); 95% of subjects’ reported gout flares met these criteria.

Diet and Supplement Intake Assessment

Participants completed questionnaires every 3 months over a 12-month period during intercritical periods and additionally at times of gout flares in which a series of potential exposures over the prior 24 and 48 hours were assessed (Figure 1). Exposures assessed included dietary factors, medications, trauma/physical activity and geographic location. Subjects could enter additional medications or supplements not specifically queried via free text. n-3 PUFA-rich supplements were identified from among the free text entries as “fish oil,” “cod liver oil,” and “omega-3 fatty acids (n-3 PUFA).” Dosages were not consistently reported. n-3 PUFA consumption related to fatty fish including anchovies, mackerel, salmon, sardines, trout, and herring were identified and quantified as total number of servings in the preceding 24 and 48 hours based on standard serving sizes for cooked fish (5 ounces).

Figure 1.

Case-crossover study design and timing of hazard and control periods with respect to gout flares for an individual participant

Participants were evaluated during control (hatched boxes) and hazard periods (black box) during the span of 1 year. Hazard periods were defined as the 48 hours preceding a gout attack. Control periods, which were defined as the 48 hours preceding gout flare-free periods, were assessed every 3 months during the 1 year follow up for each participant. Exposure to supplemental and dietary n-3 PUFA were compared between control and hazard periods.

Statistical Analysis

We categorized n-3 PUFA-rich supplements as “any” versus “none” in the preceding 48-hours for each type individually (fish oil, cod liver oil, n-3 PUFA supplements) as well as for any of the three types combined. For n-3 PUFA consumption related to fatty fish in the preceding 48 hours, we categorized the number of servings as none, 1, and ≥2 servings. Because these fatty fish have relatively high n-3:n-6 PUFA ratios, we also examined foods with differing ratios to further explore the potentially beneficial effects of n-3 PUFA versus the potentially detrimental effects of n-6 PUFA. We used spinach as an example of a food with relatively equivalent n-3:n-6 PUFA ratio, and egg as an example of a food with relatively low n-3:n-6 PUFA ratio. We examined the relation of n-3 PUFA consumption, as defined by these various types of supplements and food sources, to the risk of recurrent gout flares using conditional logistic regression, which takes into account the matching of each subject’s own hazard and control periods.(39) In multivariable regression models, we adjusted for diuretic use, alcohol intake, total purine intake, and gout-related medication use (allopurinol, colchicine, NSAIDs, and other urate-lowering therapies). In a sensitivity analysis, we adjusted for purine intake related to foods other than the fatty fish intake, instead of total purine intake, to examine the full effects of fatty fish intake on risk of gout flare.

RESULTS

Between February 2003 and January 2012, 724 participants completed online questionnaires during control and hazard periods. Participants were recruited from 49 states including the District of Columbia. The mean age was 54.5 years, and the study population was predominantly Caucasian (88.7%), male (78.5%), and obese with a mean BMI of 32.1 kg/m² (Table 1). Approximately 85% met ACR Preliminary Classification Criteria for Gout, and 48% were on urate lowering or flare prophylactic medications (NSAIDs 38.0%, colchicine 25.4%, allopurinol 43.8%). During the 1-year follow-up period, a total of 1434 gout attacks occurred, 89% of which were managed with NSAIDS, colchicine, and/or intra-articular/systemic glucocorticoids.

Table 1.

Baseline characteristics of the study sample

	N= 724
Age- mean (SD)	54.5 (12.5)
Sex- male	568 (78.5)
BMI- kg/m2, mean (SD)	32.1 (6.9)
Race
Caucasian	642 (88.7)
African American	20 (2.7)
Other	53 (7.3)
Refused to answer	9 (1.2)
Education
Less than high school graduate	14 (1.9)
High school graduate	66 (9.1)
Some college/technical school	223 (30.8)
College graduate	175 (24.2)
Some professional/graduate school	78 (10.8)
Completed professional or graduate school	168 (23.2)
Annual household income- $
< 25,000	58 (8.0)
25,000–49,999	141 (19.5)
50,000–74,999	133 (18.4)
75,000–99,999	106 (14.6)
≥100,000	186 (25.7)
Refused to answer	100 (13.8)
Disease duration- mean (SD)	8.0 (9.3)
Alcohol use	316 (43.7)
Treatments
NSAIDs	275 (38.0)
Colchicine	184 (25.4)
Allopurinol	317 (43.8)

Values are the number (%) of patients, except where indicated.

BMI- body mass index; NSAIDs- nonsteroidal anti-inflammatory drugs; SD- standard deviation

Of the 724 participants, 22% reported n-3 PUFA consumption within the prior 48 hours, of which 4.6% was consumed as supplemental forms of n-3 PUFA (self-reported as fish oil, cod liver oil, or n-3 FA), and 19% consumed dietary fatty fish (anchovies, mackerel, salmon, sardines, trout, or herring). Amongst participants who consumed supplemental forms of n-3 PUFA, none of the three supplements individually reduced recurrent gout flare risk significantly. After adjusting for diet (alcohol and total purine intake), diuretic use, and other urate-lowering or flare prophylactic medications (allopurinol, NSAIDs, colchicine), the adjusted odds ratios (aOR) for any consumption of the three supplement types combined in the prior 48 hours was 1.01 (95% CI, 0.63–1.60, p=0.98) compared with no consumption (Table 2).

Table 2.

Relation of n-3 PUFA-rich supplements to risk of recurrent gout flares

Exposure	Control Periods	Case Periods	Adjusted OR* (95% CI)	P
Supplements
Fish oil
Not consumed	1870	1381	1.0 (ref)	----
Consumed	77	53	1.04 (0.58–1.87)	0.90
n-3 PUFA
Not consumed	1914	1417	1.0 (ref)	----
Consumed	33	17	1.33 (0.43–4.10)	0.97
Cod liver oil
Not consumed	1940	1432	1.0 (ref)	----
Consumed	7	2	0.48 (0.03–7.65)	0.60
All 3 supplements
Not consumed	1830	1362	1.0 (ref)	----
Consumed	117	72	1.01 (0.63–1.60)	0.98

^*adjusted for diet (alcohol and total purine intake), diuretic use and other urate-lowering or flare prophylactic medications (allopurinol, non-steroidal anti-inflammatory drugs, colchicine)

PUFA- polyunsaturated fatty acid; OR- odds ratio; CI- confidence interval; ref- reference

In contrast, participants who consumed n-3 PUFA-rich fish experienced a significant reduction in recurrent gout flare risk. Consuming any amount of fatty fish in the preceding 48 hours was associated with a 33% lower risk of gout flare compared with no fatty fish consumption (aOR 0.77, 95% CI 0.61–0.96, p=0.02), after adjusting for potential confounders. Moreover, a dose-response relationship was noted. For 1 serving of n-3 PUFA-rich fish consumed, the aOR was 0.79 (95% CI, 0.61–1.03; p=0.08), and for ≥ 2 servings consumed, the aOR was 0.74 (95% CI 0.54–0.99; p=0.04) compared with no consumption in the prior 48 hours (Table 3). When adjusting for purine content other than that derived from fatty fish consumption, the relation of any intake of n-3 PUFA-rich fish in the prior 48 hours was no longer associated with risk of gout flares (aOR 1.09, 95% CI 0.88–1.36).

Table 3.

Relation of n-3 PUFA-rich fish to risk of recurrent gout flares

Exposure	Control Periods	Case Periods	Adjusted OR* (95% CI)	P
n-3 PUFA-rich Fish:
0 servings	1564	1123	1.0 (ref)	----
1 serving	214	165	0.79 (0.61–1.03)	0.08
≥2 servings	166	146	0.74 (0.54–0.99)	0.04

^*adjusted for diet (alcohol and purine intake), diuretic use and other urate-lowering or flare prophylactic medications (allopurinol, non-steroidal anti-inflammatory drugs, colchicine)

PUFA- polyunsaturated fatty acid; OR- odds ratio; CI- confidence interval; ref- reference

Because we hypothesized that n-6 PUFA may counteract beneficial effects of n-3 PUFA in relation to gout flares, we further evaluated the effects of various foods on risk of gout flare based on their n-3:n-6 ratios. In contrast to the findings for fatty fish which possess higher n-3:n-6 ratios, foods with more neutral n-3:n-6 ratios such as spinach did not confer any risk reduction (for ≥ 2 servings consumed, aOR 1.19; 95% CI, 0.85–1.66) (Table 4). Foods with low n-3:n-6 ratios such as egg, significantly increased the risk of gout flare (for ≥ 2 servings consumed, aOR 1.34; 95% CI, 1.10–1.63).

Table 4.

Risk of recurrent gout flare based on variable n-3:n-6 PUFA ratios^*

Exposure	Adjusted OR* (95% CI)
n-3 >> n-6 (fatty fish)	----
1 serving	0.79 (0.61–1.03)
≥2 servings	0.74 (0.54–0.99)
n-3 = n-6 (spinach)	----
1 serving	0.83 (0.62–1.10)
≥2 servings	1.19 (0.85–1.66)
n-3 << n-6 (egg)	----
1 serving	0.85 (0.66–1.08)
≥2 servings	1.34 (1.10–1.63)

PUFA- polyunsaturated fatty acid; OR- odds ratio; CI- confidence interval

^*0 servings was the referent group

DISCUSSION

In this large cohort of U.S. adults with preexisting gout, n-3 PUFA-rich fish consumption was significantly associated with lower risk of recurrent gout attacks. Subjects who consumed n-3-PUFA rich fish during the preceding 48 hours had a 33% lower risk of recurrent gout flare compared with those who did not. Risk reduction also occurred in a dose-dependent manner whereby increases in number of servings were associated with lower risk of gout flares.

In contrast, n-3 PUFA supplements such as fish oil and cod liver oil as used in the community with varying regimens and doses, did not appear to have a protective effect against flare occurrence. One possible explanation for this discrepancy between dietary and supplemental n-3 PUFA is insufficient dosing. Prior population studies have showed that median total daily intake of fish oil in the community is approximately 0.3g of EPA and DHA.(40) In past rheumatoid arthritis studies, anti-inflammatory effects of n-3 PUFA consumption were observed at daily doses of 3.4g and higher, roughly 10 times greater than the typical over-the-counter formulation.(14) Depending on the type of fish and method of preparation, EPA and DHA levels in fatty fish sources can range from 0.7 to 5.6g per serving size.(41) As such, subjects taking supplemental n-3 PUFA are more likely to be consuming a level of EPA and DHA far below that which is required for significant anti-inflammatory effect. The results of our study suggest that n-3 PUFA supplements, when administered at adequate doses, may provide a prophylactic effect. In addition to this potential therapeutic benefit, high dose n-3 PUFA has a relatively benign side effect profile. Although increased bleeding times have been observed in Greenland Inuits due to their n-3 PUFA-rich diet, no studies have shown significant bleeding risk with high dose fish oil supplementation.(42, 43) GI side effects such as nausea, eructation, and loose stools are much more common with increased dosing of supplements, but whether this is true for dietary fatty fish is not clear.

The risk reduction observed in this study remained after adjustment for total purine content consumed, including that contributed by fatty fish consumption. It is well established that purine-rich diets, which include most seafood, lead to increased risk of gout flares.(44) Our study demonstrated that isolation of dietary n-3 PUFA independent of purine content, led to protective anti-inflammatory effects. However, when we evaluated the association of n-3 PUFA-rich fish consumption that adjusted for purine consumption not related to the fish intake, there was no longer an association noted. Thus, this study provides a proof-of-concept regarding the potential for n-3 PUFA to provide a beneficial effect in reducing gout flare risk in the absence of concomitant purines. While isolation of n-3 PUFA is difficult to achieve with dietary fatty fish since it has concomitant purines, supplements would bypass this obstacle offering a pharmacological benefit without the gout-inducing features of purines.

We also observed an association between relative n-3:n-6 PUFA levels and gout flare risk. Higher n-3:n-6 ratio foods such as fatty fish was associated with lower risk of gout flares, more neutral n-3:n-6 ratio foods such as spinach had no effect on gout flare risk, and lower n-3:n-6 ratio foods such as egg increased the risk of gout flare. Although our study did not assess whether the beneficial effects of dietary and supplemental n-3 PUFA were truly related to mechanisms affecting inflammation, our results align with prior studies that suggest n-3:n-6 PUFA levels can influence the degree of inflammation.(27, 29)

This study comprised a large adult cohort with extensive data on a wide range of triggers and confounding variables. The case-crossover design allowed for analysis of multiple exposures and their acute effects while avoiding control selection bias through self-matching of subjects. Internet recruitment enabled access to a large participant population, and real time self-reporting via online questionnaires reduced recall bias. Further, our results were fairly robust to unmeasured confounding, as the observed aOR of 0.77 for “any fatty fish” consumption in the preceding 48 hours compared with no such consumption could only be explained away by an unmeasured confounder that was associated with both the exposure and the outcome by a risk ratio of 1.92-fold each, above and beyond the measured confounders, while weaker confounding could not do so.(45)

There were some limitations to our study. Crystal proven diagnosis of gout was not common, reflecting the general community pattern of gout diagnosis; in the Health Professionals Follow-Up Study, only 7% had a crystal-proven diagnosis.(46) Nonetheless, all subjects in this study met the ACR Preliminary Classification criteria for gout or had their gout diagnosis confirmed by chart review, and we implemented additional criteria to define “definite flare”. Additionally, the study design required self-reporting of exposures, including dietary intake, though any misclassification was likely to be non-differential, particularly as study participants were not apprised of study hypotheses. As well, the various types of fish evaluated do not have equal amounts of n-3 PUFA, and therefore this misclassification would likely have biased our results towards the null. Further, our study was not able to assess supplementation dosage or serum n-3 PUFA levels; therefore, we cannot rule out an effect of potentially adequately dosed supplements. Only a few subjects reported use of supplemental n-3 PUFA (only 4.6% of participants), which limited precision. Our study also did not assess serum urate levels, so it is unclear if n-3 PUFA effects were mediated through effects on urate or through other mechanisms. Finally, it is possible that the beneficial effects of fatty fish intake may be, in part, related to not ingesting other potentially risk-inducing foods (e.g., red meat); we attempted to account for this with purine intake adjustment.

Our study highlights the potentially beneficial effect of n-3 PUFA for limiting the symptom burden from acute flares in gout. Although supplemental n-3 PUFA as taken in a self-directed manner was not found to be protective and fatty fish is not the ideal source of n-3 PUFA due to its concomitant purines, these results provide support for future clinical trials examining n-3 PUFA supplementation at appropriate anti-inflammatory doses for gout flare prevention.