Dietary supplements for chronic gout

Abstract

Background

Dietary supplements are frequently used for the treatment of several medical conditions, both prescribed by physicians or self administered. However, evidence of benefit and safety of these supplements is usually limited or absent.

Objectives

To assess the efficacy and safety of dietary supplementation for people with chronic gout.

Search methods

We updated the original search by searching CENTRAL, MEDLINE, Embase, CINAHL, and four trials registers (August 2020). We applied no date or language restrictions. We also handsearched the abstracts from the 2010 to 2019 American College of Rheumatology and European League against Rheumatism conferences, and checked the references of all included studies.

Selection criteria

We considered all published randomised controlled trials (RCTs) or quasi‐RCTs that compared dietary supplements with no supplements, placebo, another supplement, or pharmacological agents for adults with chronic gout for inclusion. Dietary supplements included, but were not limited to, amino acids, antioxidants, essential minerals, polyunsaturated fatty acids, prebiotic agents, probiotic agents, and vitamins. The major outcomes were acute gout flares, study withdrawal due to adverse events (AEs), serum uric acid (sUA) reduction, joint pain reduction, participant global assessment, total number of AEs, and tophus regression.

Data collection and analysis

We used standard methodological procedures expected by Cochrane.

Main results

Two previously included RCTs (160 participants) met our inclusion criteria; we did not identify any new trials for this update. As these two trials evaluated different diet supplements (enriched skim milk powder (SMP) and vitamin C) with different outcomes (gout flare prevention for enriched SMP, and sUA reduction for vitamin C), we reported the results separately.

One trial (120 participants), at unclear risk of selection and detection bias, compared SMP enriched with glycomacropeptides (GMP) with un‐enriched SMP, and with lactose, over three months. Participants were predominantly men, aged in their 50s, who had severe gout. The results for all major outcomes were imprecise, except for pain. None of the results were clinically significant.

The frequency of acute gout attacks, measured as the number of flares per month, decreased in all three groups over the three‐month study period. The effects of enriched SMP (SMP/GMP/G600) compared with the combined control groups (SMP and lactose powder) at three months in terms of mean number of gout flares per month were not clinically significant (mean (standard deviation (SD)) flares per month: 0.49 (1.52) in SMP/GMP/G600 group versus 0.70 (1.28) in the control groups; absolute risk difference: mean difference (MD) ‐0.21 flares per month, 95% confidence interval (CI) ‐0.76 to 0.34; low‐quality evidence).

The number of withdrawals due to adverse effects was similar between groups (7/40 in SMP/GMP/G600 group versus 11/80 in control groups; (risk ratio (RR) 1.27, 95% CI 0.53 to 3.03); there were 4% more withdrawals in the SMP/lactose groups (10% fewer to 18% more; low‐quality evidence).

Serum uric acid reduction was similar across groups (mean (SD) ‐0.025 (0.067) mmol/L in SMP/GMP/G600 group versus ‐0.010 (0.069) in control groups; MD ‐0.01, 95% CI ‐0.04 to 0.01; low‐quality evidence).

Pain from self‐reported gout flares (measured on a 10‐point Likert scale) improved slightly more in the GMP/G600 SMP group compared with controls (mean (SD) ‐1.97 (2.28) in SMP/GMP/G600 group versus ‐0.94 (2.25) in control groups; MD ‐1.03, 95% CI ‐1.89 to ‐0.17). This was an absolute reduction of 10% (95% CI 20% to 1% reduction; low‐quality evidence), which may not be of clinical relevance.

The risk of adverse events was similar between groups (19/40 in SMP/GMP/G600 group versus 39/80 in control groups; RR 0.97, 95% CI 0.66 to 1.45); the absolute risk difference was 1% fewer adverse events (1% fewer to 2% more), low‐quality evidence). Gastrointestinal events such as nausea, flatulence and diarrhoea were the most commonly reported adverse effects.

Data for participant global assessment were not available for analysis; the study did not report tophus regression.

One trial (40 participants), at high risk of selection, performance, and detection bias, compared vitamin C alone with allopurinol, and with allopurinol plus vitamin C, in a three‐arm study. We only included data from the vitamin C versus allopurinol comparison in this review. Participants were predominantly middle‐aged men, and their severity of gout was representative of gout in general.

Allopurinol reduced sUA levels more than vitamin C (MD 0.10 mmol/L, 95% CI 0.06 to 0.15), low‐quality evidence. The study reported no adverse events; none of the participants withdrew due to adverse events.

The study did not assess the rate of gout attacks, joint pain reduction, participant global assessment, or tophus regression.

Authors' conclusions

While dietary supplements may be widely used for gout, this review found no high‐quality that supported or refuted the use of glycomacropeptide‐enriched skim milk powder or vitamin C for adults with chronic gout.

Plain language summary

Dietary supplements for chronic gout

What is gout, and what are dietary supplements?

Gout is caused by crystal formation in the joints, due to high uric acid levels in the blood. People have attacks of painful, warm, and swollen joints, often in the big toe. Some people develop a large buildup of crystals just beneath the skin, known as tophi. A cure can be achieved if uric acid levels in the blood return to normal for a prolonged time, making the crystal deposits dissolve.

Dietary supplements are preparations, such as vitamins, essential minerals, and probiotics. Few studies evaluate their benefits, fewer evaluate their harmful side effects.

Study characteristics

This review is an update of the original review, published in 2014. After searching the medical literature up to August 2020, we found no new studies this time. We kept the two studies from the original review.

One study (120 participants) compared enriched skim milk powder (with peptides that are thought to probably have an anti‐inflammatory effect) to standard skim milk, and to lactose powder, to try to reduce the frequency of gout attacks. The second study (40 participants) compared vitamin C with allopurinol – a drug commonly used in gout in an effort to reduce the uric acid levels in blood. Both studies enrolled people with gout who were predominantly middle‐aged men. In the skim milk study, participants appeared to have severe gout, as they had frequent attacks, and 20% to 43% presented with tophi. Participants in the vitamin C study appeared similar to ordinary people with gout.

Key results – what happens to people with gout who drink enriched skim milk powder?

Gout attacks

People who drank enriched skim milk powder for three months had 0.21 fewer gout attacks per month (from 0.76 fewer to 0.34 more), or 2.5 fewer gout attacks per year:

‐ People who drank enriched skim milk powder had 0.49 gout attacks per month (or six gout attacks per year).

‐ People who drank standard skim milk powder or lactose had 0.70 gout attacks per month (or eight gout attacks per year).

Withdrawals due to adverse events

Four more people out of 100 who drank enriched skim milk powder discontinued the supplement by three months (4% more withdrawals).

‐ 18 out of 100 stopped drinking enriched skim milk powder.

‐ 14 out of 100 stopped drinking standard skim milk powder or lactose.

Number of adverse events

People who drank enriched skim milk powder for three months had lesser adverse events (1% lesser adverse events)

‐ 47 out of 100 people who drank enriched skim milk powder had an adverse event.

‐ 48 out of 100 people who drank standard skim milk powder or lactose had an adverse event.

Effects on serum uric acid levels, joint pain, and overall assessment by participants were uncertain. The effect on the disappearance of tophus was not measured.

Key results – what happens to people with gout who take vitamin C?

Serum uric acid levels

‐ People who took vitamin C showed a reduction in serum uric acid levels of 0.014 mmol/L after eight weeks (or 2.8% reduction)

‐ People who took allopurinol showed a reduction in serum uric acid levels of 0.118 mmol/L after eight weeks (or 23.6% reduction).

There were no reports of side effects or withdrawals due to side effects in either treatment group.

Effects of vitamin C on gout attacks, pain reduction, overall assessment by participants, and tophus disappearance were not measured.

Quality of the evidence

Overall, we found low‐quality evidence in both trials. We reduced the quality of the evidence because the trials were poorly conducted and reported, the sample sizes were small, and the results suggested both benefits and harms for most outcomes. Low‐quality evidence from one study indicated that enriched skim milk, compared with standard skim milk or lactose powder, may not reduce the frequency of gout attacks or improve uric acid levels, but may reduce pain. Further research is likely to change these estimates. We do not have precise information about side effects and complications, but possible side effects may include nausea, bloating or diarrhoea.

Compared with the commonly used medicine, allopurinol, low‐quality evidence from one study showed that vitamin C reduced serum uric acid levels less; the difference was probably clinically unimportant. Other possible benefits of vitamin C are uncertain, as they were not evaluated in the study. No side effects were reported. Further research is likely to change these estimates.

Summary of findings

Summary of findings 1. Skim milk enriched with GMP/G600 compared with skim milk, plus lactose powder for chronic gout.

Skim milk enriched with GMP/G600 compared with skim milk, plus lactose powder for chronic gout
Patient or population: people with chronic gout Settings: outpatients, community Intervention: skim milk enriched with glycomacropeptides (GMP)/G600 Comparison: skim milk or lactose powder (data from 2 groups combined)
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Skim milk and lactose powder	Skim milk enriched with GMP/G600
Acute gout flares Participant self‐report, using gout flare diary Follow‐up: 3 months	The mean acute gout attack frequency in the control groups was 0.70 gout flares per month	The mean acute gout attack frequency in the intervention group was 0.21 flares per month lower (0.76 lower to 0.34 higher)	‐	120 (1 study)	⊕⊕⊝⊝ lowa	Absolute effect 21% lower (76% lower to 34% higher)
Study withdrawals due to adverse events Follow‐up: 3 months	138 per 1000	175 per 1000 (73 to 417)	RR 1.27 (0.53 to 3.03)	120 (1 study)	⊕⊕⊝⊝ lowa	Absolute risk difference: 4% more withdrawals (10% fewer to 18% more).
sUA acid reductionb (in mmol/L) Follow‐up: 3 months	The mean sUA level reduction in the control groups was ‐0.010 mmol/L	The mean sUA level reduction in the intervention group was 0.01 mmol/L lower (0.04 lower to 0.01 higher)	‐	120 (1 study)	⊕⊕⊝⊝ lowa	Absolute reduction 0.01 mmol/L (0.04 lower to 0.01 higher)
Joint pain reduction 10‐point Likert scale (0 = no pain) Follow‐up: 3 months	The mean joint pain reduction in the control groups was ‐0.94 points	The mean joint pain reduction in the intervention group was 1.03 points lower (1.9 lower to 0.1 lower)	‐	120 (1 study)	⊕⊕⊝⊝ lowa	Absolute risk difference: 10% lower (1% lower to 20% lower). NNTB = 10 (5 to 100)c
Participant global assessment 5‐point Likert scale (5 = excellent) Follow‐up: 3 months	See comment	See comment	Not estimable	‐	See comment	Measured and reported in text as improved similarly between groups, but unable to obtain data
Total adverse eventsd Follow‐up: 3 months	487 per 1000	472 per 1000 (321 to 706)	RR 0.97 (0.66 to 1.45)	120 (1 study)	⊕⊕⊝⊝ lowa	Absolute risk difference: 1% less adverse events (1% fewer to 2% more).
Tophus regression	See comment	See comment	‐	‐	See comment	Not measured
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; HAQ: Health Assessment Questionnaire; RR: risk ratio; sUA: serum uric acid.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.

^aThere was a possible influence of selection and detection bias. In addition, the imprecision of the results (in relation to the small size study) contributed to downgrading the quality of the results. Selective reporting of post hoc comparisons between skim milk powder enriched with GMP/G600 and one of the two study controls (lactose) in relation to change in gout attack frequency from baseline were noted, but it was controlled after the raw data were provided by the study authors.
^bData regarding sUA normalisation not reported.
^cNumber needed to treat for an additional beneficial outcome (NNTB) for continuous outcomes calculated using the Wells calculator software available from the Cochrane Musculoskeletal Group editorial office.
^dGastrointestinal adverse events (diarrhoea, nausea and flatulence) reported most commonly. There were no serious adverse events that resulted in hospital admissions due to the study products.

Summary of findings 2. Vitamin C compared with allopurinol for chronic gout.

Vitamin C compared with allopurinol for chronic gout
Patient or population: peoeple with chronic gout and sUA level > 0.36 mmol/L Settings: outpatients, rheumatology clinics Intervention: vitamin C Comparison: allopurinol
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Quality of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Allopurinol	Vitamin C
Acute gout flares	See comment	See comment	‐	‐	See comment	Not measured
Study withdrawals due to adverse events Follow‐up: 8 weeks	None	None	‐	40 (1 study)	⊕⊕⊝⊝ lowa	‐
sUA reduction Follow‐up: 8 weeks	The mean sUA reduction in the control group was 0.118 mmol/L	The mean sUA reduction in the intervention group was 0.10 mmol/L less reduction (0.06 to 0.15)	‐	40 (1 study)	⊕⊕⊝⊝ lowa	Absolute difference: 10% higher (6% higher to 15% higher)
Joint pain reduction	See comment	See comment	‐	‐	See comment	Not measured
Participant global assessment	See comment	See comment	‐	‐	See comment	Not measured
Total adverse events Follow‐up: 8 weeks	None reported	None reported	‐	40 (1 study)	⊕⊕⊝⊝ lowa	None reported. No participants developed oxalate renal stones or oxalosis
Tophus regression	See comment	See comment	‐	‐	See comment	Not measured
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RR: risk ratio; sUA: serum uric acid.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.

^aPotential selection, performance, and detection bias might influence the results, as it is an open trial; downgraded twice for bias. No imprecision in data reporting was noted.

Background

Description of the condition

Gout is a musculoskeletal disease characterised by recurrent episodes of acute arthritis, which are due to deposition of monosodium urate (MSU) crystals in the joints. It affects up to 1% to 2% of adults, and is the most common inflammatory arthritis in men (Smith 2010). Persistent hyperuricaemia leads to the deposition of MSU crystals in the synovial lining, tendons, ligaments, and other body sites. Most people with gout have recurrent bouts of acute arthritis, with the first metatarsophalangeal joint of the foot being the most frequently affected joint. Attacks usually resolve spontaneously within one week, and the condition is asymptomatic between episodes, although persistent subclinical inflammation has been noted in the synovial fluid from asymptomatic joints (Pascual 1999). People with longstanding disease may develop persistent arthritis, leading to joint damage and disability. Other clinical consequences of crystal deposition are the development of focal crystal nodes, known as tophi (Richette 2010). Tophi are usually found in limb extension surfaces and pressure areas, but they can appear anywhere. In addition, people with gout are at high risk of developing urate renal stones, and a form of kidney damage (urate nephropathy) has been related to the deposit of crystals in the interstitium and medulla; the significance of this condition is unclear (Nickeleit 1997). On top of these consequences of the crystal deposits, gout is strongly associated with an increased cardiovascular risk (Krishnan 2008; Kuo 2010).

To prevent further acute attacks of gout and ongoing subclinical inflammation, the aim of treatment in gout is to reduce serum uric acid (sUA) levels, which ultimately leads to dissolution of crystals in the joints, and reduction in size and eventual disappearance of tophi (Pascual 2007; Perez‐Ruiz 2002). To achieve the best outcomes, experts recommend reaching and maintaining sUA levels below 0.33 mmol/L for most people, and considering stricter thresholds (0.27mmol/L) for people with severe cases (Richette 2017). When MSU crystals are absent from joints and tissues, gout can be considered to be cured (Pascual 2009a). The present review will restrict the assessment to populations with chronic gout; this term, currently not recommended by international experts (Bursill 2019), is intended to cover the disease stage between flares (acute gout), when management is focused on flare prevention and crystal dissolution by normalising sUA levels.

Several drugs have been used since the mid‐1980s to lower sUA levels by two mechanisms of action: 1. reduce uric acid synthesis by blocking xanthine oxidase (Klinenberg 1965; Pascual 2009b), as is the case with allopurinol and febuxostat, and 2. increase renal uric acid excretion through the inhibition of its re‐absorption at the proximal convoluted tubules of the kidney (Richette 2010), as is the case with benzbromarone, sulphinpyrazone, and probenecid. Recombinant uricases, such as pegloticase that convert uric acid into allantoin, have also been approved for refractory cases (Lyseng‐Williamson 2011). Besides pharmacological therapy to reduce sUA, lifestyle changes (e.g. weight loss, exercise, and diet modifications) might also have a role in the management of people with gout, though evidence is lacking to establish the exact impact of this approach in people with gout.

Description of the intervention

Dietary supplements are widely used by the general population to prevent and treat many diseases and conditions (Wu 2011). These are usually obtained over the counter, rather than being prescribed by doctors. There are different types of dietary supplements, including amino acids (e.g. carnitine, glutamine), antioxidant agents (e.g. melatonin, acetylcysteine), essential minerals (e.g. selenium, calcium, phosphorus), polyunsaturated fatty acids (PUFA), probiotics (e.g. lactobacillus, bifidobacterium), vitamins (e.g. vitamin C, vitamin E, folic acid), and many more. People with rheumatic diseases, such as osteoarthritis or inflammatory arthritis, often use dietary supplements (Buchbinder 2002; Vista 2011); people with other chronic musculoskeletal conditions are also reported to be frequent users of homeopathic and complementary medicines (Rossignol 2011). Despite the uncertainty from published studies, it is likely that people with gout also take supplements among other remedies. A quick search on Google with 'uric acid how to reduce' shows more than four million hits, most of which come from non‐medical resources.

However, dietary supplements might be associated with potential adverse effects. Cases of increased bleeding risk related to different herbs and herbal formulae (Wong 2012), liver injury due to camellia sinensis tea extracts or usnic acid (Stickel 2011), or cardiac dysrhythmias due to enzyte (Philips 2010), have been reported. Taking into account the increasing use of dietary supplements, mostly with no medical prescription, these should be considered as potentially harmful.

How the intervention might work

There is minimal evidence of how these supplements might work in gout; some might reduce sUA levels or have an anti‐inflammatory effect, but other unidentified mechanisms are possible. Evidence suggests that diet supplements, such as vitamin C (Juraschek 2011), depurinised milk (Kocic 2012), and casein or soy proteins (Lo 2010), have sUA‐lowering effects. However, most of these studies have been conducted in animals, or in humans without gout. Diet supplements may also have beneficial effects on inflammation or pain, such as vitamin D (Zhang 2012), omega‐3 fatty acid supplements (Mori 2006), or avocado‐soybean unsaponifiables (Christensen 2008).

Why it is important to do this review

Due to the significant population's belief in dietary supplements for the treatment of arthritis, including gout, it is important to assess current evidence that examines these interventions. This would provide clinicians with useful information in order to optimise advice to people with gout. We completed an initial review published in 2014 (Andres 2014). We only identified data for glycomacropeptides and vitamin C, and found an unclear benefit for people with gout. We updated the search to August 2020 in this update, in order to cover potential new studies on the topic.

Objectives

To assess the benefits and safety of dietary supplements for the treatment of chronic gout.

Methods

Criteria for considering studies for this review

Types of studies

We considered all published randomised controlled trials (RCTs) or quasi‐RCTs (clinical controlled trials (CCTs)) that compared dietary supplements with no supplements, placebo, another supplement, or pharmacological agents for chronic gout. As we were primarily interested in the use of dietary supplements to prevent further acute attacks of gout, and reduce levels of serum uric acids (sUA), we excluded trials in acute gout, where the aims of treatment were different, namely to reduce acute inflammation.

We only included trials that were published as full articles, or available as a full trial report.

Types of participants

We included adults (aged 18 years or older) with a diagnosis of chronic gout. We excluded studies that incorporated a mix of people with gout and other musculoskeletal diseases, unless we could separate results for the gout population for analysis.

Types of interventions

We included all trials that evaluated any dietary supplement, including, but not limited to, amino acids, antioxidants, polyunsaturated fatty acids (PUFA), prebiotic agents, probiotic agents, or vitamins, alone or in combination. We examined all doses and routes of administration.

Comparators could have been:

1. placebo;

2. no treatment;

3. a different dietary supplement;

4. standard of care (allopurinol);

5. non‐pharmacological therapy (e.g. diet modification);

6. combination therapy (any of the above in combination).

Types of outcome measures

OMERACT (Outcome Measures in Rheumatology) is an international network, interested in outcome measurement across the spectrum of rheumatology intervention studies (Tugwell 1993). At the OMERACT‐9 conference, the core and discretionary domains for outcome measurement in clinical studies of acute and chronic gout were defined (Schumacher 2009). For acute gout, core domains included pain, joint swelling, joint tenderness, participant global assessment, and activity limitations. For chronic gout, core domains included sUA, acute gout attacks, tophus burden, health‐related quality of life, activity limitation, pain, and participant global assessment.

We included only outcome core domains for chronic gout, plus safety in this review. For the purpose of this review, if feasible, we grouped trials into those of short‐term (less than three months), medium‐term (three to 12 months) and long‐term (greater than 12 months) duration.

Major outcomes

1. Acute gout flares

2. Study withdrawals due to adverse events (AE)

3. sUA reduction

4. Joint pain reduction

5. Participant global assessment

6. Total adverse events

7. Tophus regression

Minor outcomes

1. Physical function (or disability)

2. Serious AEs

Search methods for identification of studies

Electronic searches

We searched a registry of all RCTs in gout, established by Cochrane Musculoskeletal to facilitate the updates of a series of reviews of interventions for gout, including this review update. The search for the gout registry was designed to not include terms for any interventions, in order to establish a registry of all randomised trials in this condition, regardless of the intervention.

The following electronic databases were searched to establish the registry. The search strategy combined standard Cochrane search filters for 'gout' and 'randomised trial', with no language restrictions.

1. The Cochrane Central Register of Controlled Trials (CENTRAL; 2020, Issue 10) in the Cochrane Library (searched 28 August 2020; Appendix 1);

2. MEDLINE Ovid (1950 to 28 August 2020; Appendix 2);

3. Embase (1980 to 28 August 2020; Appendix 3);

4. CINAHL (Cumulative Index to Nursing and Allied Health Literature; 1937 to August 2020; Appendix 4);

5. ClinicalTrials.gov (www.clinicaltrials.gov; searched 27 August 2020);

6. Australian & New Zealand Clinical Trial Registry (www.anzctr.org.au; searched 27 August 2020);

7. Chinese Clinical Trial Registry (http://www.chictr.org.cn/enindex.aspx; searched 27 August 2020);

8. World Health Organization (WHO) Clinical Trials Registry Platform (www.who.int/ictrp/en/; searched 27 August 2020).

We searched the CINAHL database because data regarding dietary supplements might be published in non‐medical journals that were not indexed by the other databases. Details of the search strategies used for the previous version of the review are given in Andres 2014.

Searching other resources

We hand‐searched the abstracts from the two major international rheumatology scientific meetings, the American College of Rheumatology (ACR) and the European League Against Rheumatism (EULAR), from 2010 to 2019.

Data collection and analysis

Selection of studies

Editorial staff from Cochrane Musculoskeletal initially screen titles and abstracts in the gout registry and retrieve the full text for all records they identifiy as RCTs of an intervention for people with gout. They annotate the population, intervention, and comparator for each full‐text article, and assign it to the appropriate gout review.

We downloaded the full search to a reference management library (Endnote® X5.0.1, PDFTron, 2010) and eliminated duplicates. Two review authors (MA, FS) independently selected the studies for detailed review and collected the data, including items to assess risk of bias, of the finally included studies. We resolved any disagreements by discussion with a third review author (LC).

Two review authors (MA, FS) independently assessed the titles and abstracts of all identified registers to identify the trials that fulfilled selection criteria. We retrieved all possibly relevant articles, including those with a related title and without an abstract, in full text. We resolved any disagreements in study selection by consensus or by discussion with a third review author if needed. We translated studies into English where necessary. We identified and excluded duplicates and recorded the selection process in sufficient detail to complete a Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) flow diagram and the 'Characteristics of excluded studies' table (Moher 2009).

Data extraction and management

We prepared pre‐piloted forms in advance to help collect the relevant data. Two review authors (MA, FS) independently extracted relevant information from the included trials, extracting study design, characteristics of study population, treatment regimen and duration, outcomes, and timing of outcome assessment.

Two review authors (MA, FS) independently extracted outcome data from included studies. We extracted the number of events and number of participants per treatment group for dichotomous outcomes, and means and standard deviations and number of participants per treatment group for continuous outcomes. We noted in the 'Characteristics of included studies' table if outcome data were not reported in a usable way, and when data were transformed or estimated from a graph. We resolved disagreements by consensus or by involving a third person (LC). One review author (MA) transferred data into the Review Manager 5 file (Review Manager 2020). We double‐checked that data were entered correctly by comparing the data presented in the systematic review with the study reports.

Assessment of risk of bias in included studies

Two review authors (MA, FS) independently assessed the potential for bias using Cochrane's RoB 1 tool for assessing risk of bias (Higgins 2011). This included assessing the potential for bias in the following domains:

1. Random sequence generation

2. Allocation concealment

3. Blinding of participants and personnel

4. Blinding of outcome assessment

5. Incomplete outcome data

6. Selective outcome reporting

7. Other bias

We judged each of these domains as low risk of bias, high risk of bias, or unclear risk of bias (either lack of adequate information or uncertainty over the potential for bias). We resolved disagreements by consensus; a third review author (LC) acted as arbiter. When considering treatment effects, we took into account the risk of bias for the studies that contributed to that outcome. We presented the figures generated by RoB 1 to provide summary assessments of the risk of bias.

Measures of treatment effect

We used Cochrane's statistical software, Review Manager 5, to analyse data (Review Manager 2020). We planned to perform meta‐analyses only if the data of the included studies were sufficiently clinically homogeneous. We presented the results of dichotomous data as risk ratios (RRs) with the corresponding 95% confidence intervals (CIs). We presented the results of continuous variables as mean differences (MDs) between the intervention and comparator groups with the corresponding 95% CIs. When different scales were used to measure the same conceptual domain, we calculated standardised mean differences (SMDs) with the corresponding 95% CIs instead. For the calculation of SMD, we divided the MD by the SD, resulting in a measure without units of treatment effect. SMDs larger than zero indicated a beneficial effect of the dietary supplementation. As described in Cohen 1988, an SMD of 0.2 indicated a small beneficial effect, 0.5 a medium effect, and 0.8 a large effect in favour of dietary supplementation. We considered a SMD to be clinically relevant if it was greater than 0.5. SMD was translated back into the original units of the particular instrument, in order to facilitate appraisal by clinicians. We considered the results of count data (such as the number of gout flares) as continuous outcome data if events were common, and as RRs if they were rare events (Higgins 2021a).

For dichotomous outcomes, the number needed‐to‐treat for an additional beneficial outcome (NNTB), or the number needed to treat for an additional harmful outcome (NNTH) was calculated from the control group event rate and the relative risk using the Visual Rx NNT calculator (Cates 2008). For dichotomous outcomes, we calculated the absolute percent change from the difference in the risks between the intervention and control group using GRADEpro, and expressed as a percentage (GRADEpro GDT 2015). For continuous outcomes, we calculated the absolute percent change by dividing the mean difference by the scale of the measure, and expressing it as a percentage. In the Effects of interventions section, and the What happens column of the summary of findings table, we provided the absolute percent change and the NNTB or NNTH (the NNTB or NNTH provided only when the outcome showed a clinically significant difference).

Unit of analysis issues

For studies containing more than two intervention groups, making multiple pair‐wise comparisons between all possible pairs of intervention groups possible, we planned to include the same group of participants only once in the meta‐analysis. In the event that we had identified cross‐over trials, in which the reporting of continuous outcome data precluded paired analysis, we planned to include these data in a meta‐analysis, in order to avoid unit of analysis error. When we thought carry‐over effects existed, and when sufficient data existed, we intended to only include data from the first period in the analysis. When outcomes were collected at multiple follow‐up times (within the short‐term, medium‐term, and long‐term time frames), we extracted the last outcome.

Dealing with missing data

When data were missing or incomplete, we contacted study authors for further information.

In cases where individual data were missing from the reported results, and no further information was forthcoming from the study authors, we assumed the missing values to have a poor outcome. For dichotomous variables that measured AEs, we calculated the withdrawal rate using the number of participants who received treatment as the denominator (worst‐case analysis). For dichotomous outcomes that measured benefits, we calculated the worst‐case analysis using the number of randomised individuals as the denominator. For continuous variables, we calculated the MD or the SMD, based on the number of participants analysed at the time point. If the number of participants analysed was not available, we would have used the number of randomised participants in each group at baseline.

Where possible, we planned to calculate missing SDs from other statistics, such as standard errors, CI, or P values, according to the methods recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2021). If we could not calculate missing SDs, and there were very few missing, we planned to impute them from other studies in the meta‐analysis.

Assessment of heterogeneity

We planned to assess both clinical and statistical heterogeneity between studies. First, we planned to assess studies for clinical homogeneity with respect to study participants, intervention groups, outcome measures, and timing of outcome. For studies judged as clinically similar, we assessed statistical heterogeneity using the I² statistic (Deeks 2021). We used the following approximate thresholds for the interpretation of I²: 0% to 40% heterogeneity might not be important; 30% to 60% represented moderate heterogeneity; 50% to 90% represented substantial heterogeneity; and greater than 75% represented considerable heterogeneity. In cases of substantial heterogeneity, we planned to explore the data further, including subgroup analyses, to explain the heterogeneity.

Assessment of reporting biases

To assess whether reporting bias was present, we determined whether the protocols of included trials were published before recruitment of participants began. For trials published after 1 July 2005, we searched the WHO International Clinical Trials Registry Platform (apps.who.int/trialsearch/).

We planned to evaluate whether selective reporting of outcomes was present.

We planned to compare the fixed‐effect estimate against the random‐effects model to assess the possible presence of small sample bias in the published literature (i.e. in which the intervention effect was more beneficial in smaller studies). In the presence of small sample bias, the random‐effects estimate of the intervention is more beneficial than the fixed‐effect estimate (Page 2021).

We planned to explore the potential for reporting bias using funnel plots if at least 10 studies were available for meta‐analysis.

Data synthesis

We presented the characteristics of included studies in the Characteristics of included studies table, results from the search and selection of studies in a flowchart, risk of bias in a figure, and specific results by type of study, and by outcome. If we considered studies sufficiently homogeneous, we planned to pool data in a meta‐analysis using a random‐effects model, regardless of the I² statistic results. We planned to perform analyses using Review Manager 5, and produce forest plots for all analyses (Review Manager 2020).

Subgroup analysis and investigation of heterogeneity

In order to explore the heterogeneity of the included studies, we planned a subgroup analysis (if sufficient data were available) for the effect of gender, as different levels of sUA have been reported in men and women (Stamp 2011). Ideally, we would extract the main outcomes separately for men and women from each trial.

We planned to use the formal test for subgroup interactions in Review Manager 5, and cautiously interpret the subgroup analyses, as advised in chapter 10 of the Cochrane Handbook (Deeks 2021). We planned to compare the effect sizes in the subgroup analyses informally to assess possible differences in response to treatment, by considering the overlap of the CIs of the summary estimates in the two subgroups – non‐overlap of the CIs indicated statistical significance. As we had anticipated, the outcomes were not reported by gender in the trials, precluding this planned analysis.

However, one trial did report data by whether or not participants were allopurinol naive, and we reported this subgroup analysis.

Sensitivity analysis

Where sufficient studies existed, we planned sensitivity analyses to assess the impact of including data in a meta‐analysis from trials that were susceptible to selection bias (i.e. with inadequate or unclear allocation concealment), and trials that were susceptible to detection bias (i.e. with inadequate or unclear outcome assessment blinding). However, as there were only two included studies and pooling was not possible, we did not perform sensitivity analyses.

Interpreting results and reaching conclusions

We followed the guidelines in the Cochrane Handbook for Systematic Reviews of Interventions, chapter 15, for interpreting results, and were careful to distinguish a lack of evidence of effect from a lack of effect (Schünemann 2021a). We based our conclusions only on findings from the quantitative or narrative synthesis of included studies. We avoided making recommendations for practice. Our implications for research suggest priorities for future research, and outline the remaining uncertainties in the area.

Summary of findings and assessment of the certainty of the evidence

We presented key findings using summary of findings tables. These tables provide key information about the quality, or certainty of the evidence, the effect size of the interventions examined, and the available data on the outcomes (Schünemann 2021). They include an overall grade of the evidence related to each of the main outcomes using the GRADE approach (GRADEpro GDT 2015). For the summary of findings table, we included the seven major outcomes:

• acute gout flares;

• study withdrawals due to serious adverse events (AEs);

• sUA reduction;

• joint pain reduction;

• participant global assessment;

• total number of AEs; and

• tophus regression.

Two people (MA, FS) independently assessed the certainty of the evidence. We used the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias) to assess the certainty of the body of evidence as it related to the studies that contributed data for the prespecified outcomes, and reported the certainty of evidence as high, moderate, low, or very low. We used methods and recommendations described in Chapter 14 of the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2021). We used GRADEpro GDT software to prepare the summary of findings tables (GRADEpro GDT 2015). We justified all decisions to downgrade the certainty of evidence for each outcome in footnotes, and made comments to aid the reader's understanding of the review where necessary. We provided the absolute and relative per cent change, and the number needed to treat for an additional beneficial or harmful outcome in the What happens column of the summary of findings table, as described in the Measures of treatment effect section above, with the exception of the absolute difference for dichotomous outcomes, which is displayed by default GRADEPro GDT.

Results

Description of studies

Results of the search

The updated search retrieved 5088 references, 813 of which were duplicates. We excluded 3046 records after title and abstract screening, and 1229 after assessing the full‐text report (340 because of wrong study design, 334 because of wrong patient population, 451 because of wrong intervention, 3 because of wrong comparator, and 7 because the full report was not available, or data were not shown (classified as awaiting assessment or ongoing). Eighty‐five were duplicate reports of the same studies and nine reports (from two studies) were already included in our previous review (Andres 2014; Figure 1). We did not identify any new eligible studies for this review update from the updated search. We also failed to identify any additional studies from ACR and EULAR abstracts or from a handsearch of the reference lists of included studies.

1.

Study flow diagram

Included studies

Two studies included in our previous review also met our inclusion criteria for this updated review (Dalbeth 2012; Stamp 2013). Both studies were randomised controlled trials (RCT); and both were completed in New Zealand. See Characteristics of included studies for detailed information.

Dalbeth 2012 was a randomised double‐blind trial that tested enriched skim milk powder (SMP) for the prevention of gouty flares during a three‐month period. The study enrolled 120 participants with recurrent gout flares, from primary‐ and secondary‐care clinics, and from public advertisement. Participants were required to be at least 18 years old, fulfil the ACR criteria for gout and have had at least two gout flares in the preceding four months. People with lactose intolerance or severe renal failure (estimated glomerular filtration rate less than 30 mL/minute) were excluded. Enrolled participants were predominantly middle‐aged white men, with a long duration of gout (about 15 years). Gout was poorly controlled, as participants reported frequent gout flares, and one‐third of participants had tophaceous disease. Mean sUA was 0.42 mmol/L (slightly over normal). Mean serum creatinine was 91 µmol/L (normal). Half of the participants were taking allopurinol; this was not modified throughout the study. The proportion of participants receiving background non‐steroidal anti‐inflammatory drugs (NSAIDs) was 27%, colchicine 27% and prednisolone 13%. Participants were randomised to receive 1. lactose powder, 2. SMP or 3. SMP enriched with glycomacropeptides (GMP) 1.5 g and G600 milk fat extract 0.525 g. The primary end point was the change in frequency of gout flares, both defined as pain at rest greater than 3 on a 10‐point Likert scale and as participant self reported flare, registered by the participant in a daily flare diary, measured monthly for three months. Other outcomes measured were AEs rate, reduction in pain during the acute gout flares, participants global assessment, physical function evaluated through the Health Assessment Questionnaire (HAQ)‐II and the reduction of sUA. Other outcomes specified as being of interest in our review were not measured in this trial. Outcomes were reported at one, two and three months after exposure to the dairy interventions.

Stamp 2013 was a randomised open trial assessing the effects of vitamin C compared with allopurinol in sUA reduction during an eight‐week period. Forty participants with gout fulfilling the ACR classification criteria with an sUA over 0.36 mmol/L at baseline were included. Enrolled participants were predominantly middle‐aged white men (mean 58 years of age, 90% men), who were overweight (mean body mass index: 31.2 kg/m²). Mean sUA at enrolment was 0.50 mmol/L, with normal renal function (estimated glomerular filtration rate over 66.7 mL/minute). Background use of diuretics was 27.5% and aspirin 30%. No data regarding disease duration or acute attacks rate were provided. Participants were stratified by allopurinol use at enrolment. Twenty participants were already on allopurinol (50%, mean dose at trial entry 345 mg daily) and were randomised to receive either an increase in allopurinol dose (10 participants) 50 to 100 mg daily or to receive vitamin C 500 mg/day (10 participants). Twenty participants were not taking allopurinol (50%) and were randomised to receive either allopurinol starting at 50 to 100 mg daily, with further dose adjustment at four weeks based on sUA level (10 participants) or vitamin C 500 mg daily (10 participants). Mean reduction of sUA at eight weeks was the primary end point of the trial. AE rate was also reported although it was not a pre‐specified end point. None of the other outcomes specified as being of interest in our review were measured in this trial.

Studies awaiting classification

We identified one trial that was published as an abstract in 2010, which is awaiting classification. It investigated artichoke verus placebo (Arych 2010). See Characteristics of studies awaiting classification for details.

Ongoing studies

We identified six ongoing trials assessing krill oil versus colchcine (ACTRN12618000770268); celery seed extract versus vitamin C versus starch (ChiCTR1900020614); omega‐3 fatty acids versus placebo (ISRCTN79392964); tart cherry juice (which contain anthocyanins) versus placebo (NCT03621215); 60 mL versus 120 mL oral dose of tart cherry extract (NCT03650140); and probiotics versus placebo (NCT04199325). See Characteristics of ongoing studies for details. 

Excluded studies

Yamanaka 2019 was a randomised placebo‐controlled trial of eight weeks duration, completed in Japan, which assessed the effect of lactobacillus gasseri PA‐3 (provided in yoghurt at 8.5 x10^7 CFU/g) on sUA levels in 25 participants with hyperuricemia or gout. The study was deemed to be ineligible as study results for participants with gout were not presented separately. The corresponding author of the study was contacted in order to obtain raw data and results from participants with gout, but no answer was received. See Characteristics of excluded studies.

Risk of bias in included studies

We assessed Dalbeth 2012 at unclear risk of bias, and Stamp 2013 at unclear to high risk of bias (Figure 2). See Characteristics of included studies for detailed information about the risks of bias.

2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study

Allocation

Both trials were at low risk of bias for random sequence generation. Dalbeth 2012 used a random block randomisation algorithm to generate the randomisation schedule. In Stamp 2013, the randomisation list was computer generated by an independent statistician. Randomisation was stratified by allopurinol use in permuted blocks of size four.

Dalbeth 2012 was at unclear risk of bias for allocation concealment; it stated that participants and study staff were blinded to treatment allocation, but no details of the actual method of allocation concealment was provided. Stamp 2013 was at high risk of bias; there was no information provided about intervention concealment, and it was an open trial.

Blinding

Dalbeth 2012 was at low risk of performance and detection bias (self‐reported outcomes) and unclear risk of detection bias for assessor‐reported outcomes as it was unclear whether outcome assessors were blinded or not. In Dalbeth 2012, all study agents were a cream‐coloured powder. This was administered daily as a 250 mL vanilla flavoured shake, so participants were unable to distinguish the type of milk they were taking. Study staff were also blinded to treatment allocation throughout the study. All study participants were treated in the same way, and examined according to the same protocol.

Stamp 2013 was an open‐label study, so both participants and physicians were aware of treatment allocation. Hence there was high risk of performance bias and detection bias for self‐reported outcomes, but its influence on the measurement of laboratory outcomes (sUA levels) was probably unimportant, therefore we judged detection bias due to assessor‐reported outcomes at unclear risk.

Incomplete outcome data

Both trials were judged to be at low risk of attrition bias. In Dalbeth 2012, of the 120 participants who started, 102 (85%) completed the three‐month study (two discontinued for AEs, eight were lost to follow‐up, and eight continued the study without taking the study medication after an AE). No significant differences were noted between groups in the dropout rate.

All 40 randomised participants completed the eight‐week study period in Stamp 2013.

Selective reporting

In Dalbeth 2012, all outcomes planned in the protocol that was registered at the Australian & New Zealand Clinical Trial Registry were reported in the paper. Dalbeth 2012 also reported the findings of a post hoc comparison between two interventions (GMP/G600 SMP and lactose control) and their effects on the study's primary end point (change in gout flare frequency), although not the results of the third intervention (standard SMP). The same selective reporting occurred in the study authors' discussion of GMP/G600 SMP effect on lowering diastolic blood pressure; they only reported the results of the post hoc comparison with the lactose control group. Hence, this study was judged to be at unclear risk of selective reporting bias.

In Stamp 2013, all pre‐defined outcomes in the protocol that was registered at the Australian & New Zealand Clinical Trial Registry were reported in the results, and there was low risk of reporting bias.

Other potential sources of bias

In Dalbeth 2012, groups were well matched at baseline, except for higher diuretic use in the GMP/G600 SMP group. The commencement or discontinuation rates of diuretics, allopurinol, colchicine, prednisone, or non‐steroid anti‐inflammatory drugs (NSAIDs) did not differ between groups throughout the follow‐up period. Compliance was acceptable in all groups (rate of participants who dropped the study or discontinued the dairy product ranged between 13% and 18%). It is noteworthy that study funding was provided by LactoPharma (a joint venture between Fonterra Ltd, Fonterra R&D Ltd and Auckland UniServices Ltd) and the New Zealand Government Foundation for Research Science and Technology. The role of the company in the study was not clearly addressed. The review authors BK‐S, AM, and KP are employees of Fonterra Co‐operative Group Ltd. The review authors AM, ND, and KP are named inventors on a patent application related to milk products and gout. We judged this study to be at an unclear risk of other bias.

Participant adherence to medications was not clearly described in Stamp 2013, but oxypurinol and ascorbic acid levels were measured in all study participants, which could be considered an indirect measure of adherence. This study was at low risk of other bias.

Effects of interventions

See: Table 1; Table 2

We did not pool data from the two included studies, since they each evaluated different supplements. Therefore, we have two comparisons, with one trial in each one. See Table 1 and Table 2 for more details.

Skim milk enriched with GMP/G600 versus standard skim milk or lactose powder

Major outcomes

Acute gout flares

Dalbeth 2012 found a significant reduction in the rate of gout flares with all three dairy products (skim milk enriched with GMP/G600, standard skim milk (SMP), and lactose powder) over the three‐month study period; however, there were no differences in the number of gout flares between groups.

After combining the two control groups (standard SMP and lactose powder) – using the raw data provided by the study authors – and calculating the standard deviation (SD) from the 95% confidence interval (CI), we found that the rate of gout flares in the GMP/G600 SMP group was 0.49 ± 1.52 flares per month compared with 0.70 ± 1.28 flares per month in the combined control group.

After three months of treatment, the change in the number of gout flares per month between the GMP/G600 SMP and the control groups was inconclusive (mean difference (MD) ‐0.21, 95% CI ‐0.76 to 0.34; 1 study, 120 participants; low‐quality evidence; Analysis 1.1); the absolute difference was 21% lower in the GMP/G600 SMP group (76% lower to 34% higher). We downgraded the evidence for bias and imprecision. The difference between groups was inconclusive, and from the review team's perspective, was not clinically relevant.

1.1. Analysis.

Comparison 1: Skim milk powder (SMP; GMP/G600) versus control (SMP/lactose), Outcome 1: Acute gout flares (per month, after 3 months of treatment)

Study withdrawals due to adverse events (AE)

Withdrawals due to adverse events were similar between the three study groups, 7/40 participants dropped out because of an AE in the GMP/G600 SMP group and 11/80 participants dropped out from the joint control groups (risk ratio (RR) 1.27, 95% CI 0.53 to 3.03; 1 study, 120 participants; low‐quality evidence; Analysis 1.2). The absolute difference was 4% more withdrawals from the GMP/G600 SMP group (10% fewer to 18% more). We downgraded the evidence for bias and imprecision.

1.2. Analysis.

Comparison 1: Skim milk powder (SMP; GMP/G600) versus control (SMP/lactose), Outcome 2: Study withdrawals due to adverse events

Serum uric acid (sUA) reduction

sUA reduction was similar between groups (‐0.024 ± 0.066 in GMP/G600 SMP versus ‐0.010 ± 0.068 in control group; MD ‐0.01, 95% CI ‐0.04 to 0.01; 1 study, 102 participants analysed; low‐quality evidence; Analysis 1.3). The absolute change was 0.01 mmol/L more reduction in the GMP/G600 SMP group (0.04 lower to 0.01 higher). We downgraded the evidence for bias and imprecision. Data regarding sUA normalisation were not provided.

1.3. Analysis.

Comparison 1: Skim milk powder (SMP; GMP/G600) versus control (SMP/lactose), Outcome 3: Serum uric acid reduction

Joint pain reduction

GMP/G600 SMP reduced pain (measured on a 10‐point Likert scale) during self‐reported gout flares more than the control treatments (‐1.97 ± 2.28 in GMP/G600 SMP group versus ‐0.94 ± 2.25 in controls, using a 0‐ to 10‐point Likert scale; MD ‐1.03, 95% CI ‐1.89 to ‐0.17; 1 study, 120 participants; low‐quality evidence; Analysis 1.4). The absolute difference was 10% less pain during gout flares in the GMP/G600 SMP group (1% lower to 20% lower). We downgraded the evidence for bias and imprecision. This finding might have a limited interest from a clinical perspective.

1.4. Analysis.

Comparison 1: Skim milk powder (SMP; GMP/G600) versus control (SMP/lactose), Outcome 4: Joint pain reduction (during gout flares)

Participant global assessment

The study authors reported that scores improved similarly between groups throughout the trial, but data were not given. We were unable to obtain the raw data from the study authors.

Total AEs

The rate of adverse events were similar between the three study groups (19/40 in the GMP/G600 SMP group versus 28/80 in the control groups; RR 0.97, 95% CI 0.66 to 1.45; 1 study, 120 participants; low‐quality evidence; Analysis 1.5); absolute risk difference: 1% less adverse events (1% fewer to 2% more). Gastrointestinal events such as nausea and diarrhoea were the most commonly reported adverse effects.

1.5. Analysis.

Comparison 1: Skim milk powder (SMP; GMP/G600) versus control (SMP/lactose), Outcome 5: Total number of adverse events

Tophus regression

The study did not measure this outcome.

Minor outcomes

Physical function (or disability)

In physical function, assessed using HAQ‐II, no differences were found between groups at three months (0.08 ± 0.23 in the GMP/G600 SMP group versus 0.11 ± 0.32 in the control groups; MD ‐0.03, 95% CI ‐0.14 to 0.08) (Analysis 1.6).

1.6. Analysis.

Comparison 1: Skim milk powder (SMP; GMP/G600) versus control (SMP/lactose), Outcome 6: Physical function

Serious AEs

The risk of serious adverse events (2/40 in the GMP/G600 SMP group versus 3/80 in the control groups) was also similar between groups (RR 1.33, 95% CI 0.23 to 7.66). We downgraded the evidence for bias and imprecision (Analysis 1.7).

1.7. Analysis.

Comparison 1: Skim milk powder (SMP; GMP/G600) versus control (SMP/lactose), Outcome 7: Serious adverse events

Vitamin C versus allopurinol

Major outcomes

Acute gout flares

The study did not measure this outcome.

Study withdrawals due to adverse events (AE)

The study did not measure this outcome.

Serum uric acid (sUA) reduction

Stamp 2013 found that at the end of the eight‐week study period, the level of sUA was reduced more in the allopurinol group than in the vitamin C group (MD 0.10, 95% CI 0.06 to 0.15; 1 study, 40 participants; low‐quality evidence; Analysis 2.1). The absolute risk difference was 0.10 mmol/L more reduction in the allopurinol group (0.06 mmol/L higher to 0.15 mmol/L higher). We considered the higher reduction of sUA observed with allopurinol over vitamin C to be clinically relevant. We downgraded the evidence twice for serious bias.

2.1. Analysis.

Comparison 2: Vitamin C versus allopurinol, Outcome 1: Serum uric acid reduction

The study authors conducted a subgroup analysis, based on whether participants were taking allopurinol at study entry; they reported that overall, it did not affect treatment response, but there were different responses between those who were already taking it, and those who started once in the study. Those in the allopurinol group who had not been taking it when they entered the study had a greater reduction in sUA than those in the vitamin C group (MD 0.15, 95% CI 0.08 to 0.22; 20 participants). Those in the allopurinol group who were on it when they entered the study had similar reductions in sUA than those taking vitamin C (MD 0.06, 95% CI ‐0.01 to 0.13; 20 participants; Analysis 2.2; Figure 3).

2.2. Analysis.

Comparison 2: Vitamin C versus allopurinol, Outcome 2: sUA reduction (subgroup analysis)

3.

Forest plot of comparison: Vitamin C versus allopurinol, outcome: sUA reduction; subgroup analysis [mmol/L]

Joint pain reduction

The study did not measure this outcome.

Participant global assessment

The study did not measure this outcome.

Total AEs

No AEs were detected in either group.

Tophus regression

The study did not measure this outcome.

Minor outcomes

Physical function (or disability)

The study did not measure this outcome.

Serious AEs

The study did not measure this outcome.

Discussion

Summary of main results

We only identified two published randomised controlled trials (RCTs) that tested the benefits of two different dietary supplements in people with chronic gout. One trial assessed enriched skim milk (120 participants); the other evaluated vitamin C against allopurinol (40 participants).

Skim milk enriched with GMP/G600

One trial, including 120 participants and judged to be of low‐quality due to bias and imprecision, compared skim milk (SMP) enriched with GMP/G600 against standard SMP or lactose powder. There results were similar between groups for frequency of gout flares, rate of adverse events (AE; gastrointestinal disturbances were the most commonly reported) and serum uric acid (sUA) reduction. Participants in the GMP/G600 SMP group reported a greater reduction in pain from self‐reported gout flares, but the amount of reduction might be of doubtful clinical relevance. Withdrawal due to AEs did not differ between groups. The study did not assess tophus regression or health‐related quality of life.

Compared with standard skim milk or lactose powder, skim milk enriched with GMP and G600 was of unclear benefit in reducing flares of gout based on a single trial impacted by risk of bias and imprecision.

Vitamin C

One trial, with 40 participants, and judged to be of low‐quality being downgraded twice for bias, evaluated the effect of vitamin C on sUA levels compared with allopurinol, over eight weeks. Participants were stratified according to their use of allopurinol at the entry of the study. Overall, the reduction of sUA with vitamin C was less than with allopurinol. There were no AEs reported in either group. This study did not measure any other outcomes of interest for this review.

Compared with allopurinol (the standard agent used in clinical practice for people with gout), vitamin C showed no benefit in reducing sUA levels in people with gout, a conclusion based on a single trial that was impacted by selection. performance and deteciton bias.

Overall completeness and applicability of evidence

Enriched SMP and vitamin C were the only dietary supplements tested in gout, based on our search.

Participants in Dalbeth 2012 did not seem very representative of the average gout population, as they reported very recurrent flares (3.9 to 5.1 flares in the four months preceding the enrolment), and 20% to 43% showed tophi. This might be attributable to ethnic issues, as the study took place in New Zealand, and Maori and Pacific people have been shown to have a severe form of gout (Rose 1975); however, over 70% of the participants in Dalbeth 2012 were white. So, the recruited participants would be representative of a severe gout population, more often seen in a hospital gout clinic. Other characteristics, such as being middle aged (mean range 56 to 57 years), and predominantly male (88% to 93%), are typical of the disease. The low use of allopurinol (53% to 55%) or colchicine (18% to 33%) is comparable to other studies, in which many participants in the clinical setting are either never prescribed, or do not adhere to urate‐lowering therapy, or flare prophylaxis is seldom considered (Perez‐Ruiz 2011). Participants recruited in Stamp 2013 appeared representative of people with gout seen commonly in clinical practice (Yu 1984).

The dietary supplements identified in this review comprised two different targets of the management of chronic gout (Richette 2017). Enriched milk powder was tested as an anti‐inflammatory agent to reduce the occurrence of acute gout flares, while vitamin C acted as an sUA‐lowering agent to pursue crystal dissolution. However, the effect of these supplements appears negligible, and a similar conclusion was reported by a recent systematic literature review (Yang 2020). Moreover, considering the need to achieve a persistent reduction in sUA levels to achieve monosodium urate (MSU) crystal dissolution, the effect noted with vitamin C seems trivial in terms of reaching this target (Richette 2017). Perhaps, the dose of vitamin C tested in Stamp 2013 was insufficient, as previous reports noted a significant sUA reduction (through increasing uric acid urinary excretion) with larger doses, in hyperuricaemic participants without gout (Berger 1977; Mitch 1981; Stein 1976). Other supplements have been tested in people with asymptomatic hyperuricaemia, but not in gout. Reducing sUA levels to control hyperuricaemia will lead to MSU crystals being dissolved from joints or tissues, but other aims should be considered in people with gout, such as a reduction in the number of acute attacks, an improvement in health‐related quality of life, or regression of the tophi. An anti‐inflammatory effect to control acute episodes or the persistent subclinical inflammation would also be suitable aims for people with gout. Based on this, we deem that the studies in people with hyperuricaemia are not entirely applicable to people with gout, and thus, we did not include them in this review.

Other supplements, such as amino acids, polyunsaturated fatty acids (PUFA), or antioxidants, are widely used for the treatment of several disorders, but our search did not identify any studies of these interventions. In addition, some dietary interventions, such as black tea or cherries, have shown some effect in reducing sUA levels or controlling acute attacks in people with hyperuricaemia or gout. However, these types of interventions could not really be considered supplements, unless specific components, considered to have potential benefits, were to be extracted and purified so they could be individually tested, similar to the evaluation of the glycomacropeptides in Dalbeth 2012. Therefore, they were outside the scope for this review.

Lifestyle interventions were addressed by another Cochrane Review (Moi 2013).

Population‐based studies have identified a number of diet patterns that may be associated with an increased risk of developing hyperuricaemia and gout, such as the intake of alcohol or high fructose‐enriched beverages. However, giving diet advice based on these results, with no confirmation through intervention studies, warrants caution (Rose 1985). The effect of these interventions in the individual participant is likely to be small or lacking, especially in people with gout, where this effect appears to us to be incapable of reducing sUA to the level that is required to make all MSU crystal dissolve. In addition, the effect could vary as the targeted populations are different; risk factors for developing hyperuricemia, and eventually gout, are reported in at‐risk populations (thus, they are still healthy), while the use of dietary supplements proposed and tested here are aimed at people who are already suffering from chronic gout.

Quality of the evidence

Our search identified only two studies, enrolling 160 participants. It is likely that the small number of trials identified reflects the lack of high‐quality research in this area, rather than publication bias. We judged the included studies to be of low quality; issues regarding the study design and imprecision of results reduced the quality of the results, with particular concerns in selection bias and absence of blinding. Relevant outcomes for gout management, such as prevention of acute attacks or sUA reduction, were only assessed in a single study each, and tophus regression was not reported in the included studies.

Potential biases in the review process

Despite a broad search retrieving more than 2800 hits, we only identified two trials that fulfilled the inclusion criteria for this review. We included several different terms for gout and for dietary supplementation (as a general intervention and for different individual ones). It could be possible that some relevant RCTs were missed, due to the lag between the search date (August 2020) and review publication, but this possibility seems very low.

Agreements and disagreements with other studies or reviews

The initial version of this review was the first systematic literature review assessing the efficacy and safety of dietary supplementation in people with gout (Andres 2014). The current update did not identify any newly published evidence in recent years on the topic. However, we retrieved a number of ongoing trials for which there are still no available results (see Characteristics of studies awaiting classification and Characteristics of ongoing studies). Considering their registered protocols, they are potential candidates for a further update of this review. The dietary supplements that are being studied are krill oil compared to colchicine to reduce inflammation and pain in chronic gout; celery seed extract compared to vitamin C or starch to reducing sUA levels; omega‐3 fatty acids compared with placebo, designed as a feasible trial; tart cherry juice concentrate compared against a fruit‐flavoured placebo drink to reduce the occurrence of gout flares; and probiotics compared to placebo to assess sUA levels reduction.

One trial identified in our search evaluated the pharmacokinetics and pharmacodynamics of anthocyanins in chronic gout, but recently suspended recruitment due to the COVID‐19 pandemic, with no resuming date yet; another, using artichoke supplements, communicated their results in a 2010 conference, but remains unpublished.

In summary, there are a number of dietary supplements currently being assessed for different outcomes of chronic gout. When the findings of the ongoing trials are published, they should be critically reviewed in a subsequent update of this review, with a chance of modifying the current conclusions.

Several studies, including one meta‐analysis of RCTs (Juraschek 2011), have reported that vitamin C supplementation may lower sUA levels. However, these studies have generally been of low quality, and have largely been performed in either healthy people or people with asymptomatic hyperuricaemia. In addition, the median sUA levels in the included populations were close to the upper limit of normality (Huang 2005). Therefore, we consider it doubtful that these results could be extrapolated to people with gout.

Authors' conclusions

Implications for practice.

The present review found low‐quality evidence that there were no benefits to using vitamin C or glycomacropeptides‐enriched skim milk for people with chronic gout. The literature search did not identify any published intervention studies for other dietary supplements. Findings from this review should be taken into account when considering the prescription of or the advice given regarding dietary supplements for people with chronic gout.

While observational studies have reported some effects of dietary supplements in healthy people or in people with asymptomatic hyperuricaemia (such as a lowering of serum uric acid (sUA) with vitamin C), our review found that dietary supplementation is supported by low‐quality evidence for use in people with gout. When compared with the standard urate‐lowering agent, allopurinol, vitamin C, at a dose of 500 mg daily, did not reduce sUA to a significant degree. Whether the addition of glycomacropeptide to skim milk reduces acute gout flares in people with gout, compared to the use of standard skim milk and lactose powder remains uncertain. Further research is likely to change the estimates.

Implications for research.

Taking into account the paucity of trials of dietary supplement in gout but their common use in general population, further studies are needed to establish the role of these products in the management of people with gout, despite already having licensed pharmacotherapies proven effective to potentially cure chronic gout (Seth 2014). Randomised controlled trials comparing dietary supplements with placebo, no treatment, other supplements, and urate‐lowering medications are needed before any conclusions can be made. However, we acknowledge that short‐term trials may not be the optimal method for assessing the benefits and long‐term sustainability of dietary supplements, and long‐term prospective longitudinal studies or registry data may also be required.

We judged the included trials in this review to be at risk of selection, performance and detection bias. Further trials should consider and report their method of randomisation and treatment allocation concealment; blinding of study participants, study personnel and outcome assessment; follow‐up of all participants who entered the trial, and complete reporting of outcomes. Sample sizes should be reported, and have adequate power to answer the research question; ideally, trials should assess both the benefits and risks of dietary supplements.

Our review highlights the absence of high‐quality evidence to either support or refute the value of dietary supplementation in people with gout. Though vitamin C appeared suboptimal as a urate‐lowering therapy in comparison with allopurinol, this is derived from low‐quality evidence, so further trials might potentially change this conclusion. In the case of milk powder, a potential anti‐inflammatory effect was noted with milk powder enriched with GMP/G600 (Dalbeth 2010). Further trials comparing enriched milk powder with non‐steroidal anti‐inflammatory drugs (NSAIDs) or colchicine in the prevention of acute gout flares may be worthwhile.

What's new

Date	Event	Description
26 January 2022	Amended	Minor typo erros corrected in the abstract.

History

Protocol first published: Issue 11, 2012
Review first published: Issue 10, 2014

Date	Event	Description
8 November 2021	New citation required but conclusions have not changed	Several ongoing trials identified, no new published studies.
28 August 2020	New search has been performed	Updated search 28 August 2020; several ongoing trials identified, but no eligible published studies.
24 September 2019	Amended	Published note added after a request from the funding arbiter.

Appendices

Appendix 1. CENTRAL search strategy

Original search date: 27 August 2020

#1 MeSH descriptor: [Gout] explode all trees

#2 gout*:ti,ab

#3 tophus:ti,ab

#4 tophi:ti,ab

#5 tophaceous:ti,ab

#6 MeSH descriptor: [Uric Acid] this term only

#7 uric next acid:ti,ab

#8 ((sodium or monosodium or potassium or ammonium) near/2 urate):ti,ab

#9 urate next crystal*:ti,ab

#10 MeSH descriptor: [Hyperuricemia] this term only

#11 Hyperuric?emi*:ti,ab

#12 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11

#13 MeSH descriptor: [Dietary Supplements] explode all trees

#14 ((diet* or food) next supplement*):ti,ab

#15 MeSH descriptor: [Food, Fortified] explode all trees

#16 enriched:ti,ab

#17 N?utraceutical*:ti,ab

#18 MeSH descriptor: [Antioxidants] explode all trees

#19 antioxidant*:ti,ab

#20 "free radical*":ti,ab

#21 MeSH descriptor: [Carnitine] explode all trees

#22 Carnitine:ti,ab

#23 MeSH descriptor: [Glutamine] explode all trees

#24 glutamine:ti,ab

#25 oligoelement*:ti,ab

#26 MeSH descriptor: [Amino Acids] explode all trees

#27 "amino acid*":ti,ab

#28 MeSH descriptor: [Minerals] explode all trees

#29 mineral*:ti,ab

#30 MeSH descriptor: [Vitamins] explode all trees

#31 vitamin*:ti,ab

#32 probiotic*:ti,ab

#33 prebiotic*:ti,ab

#34 synbiotic*:ti,ab

#35 yeast*:ti,ab

#36 Fatty acids, unsaturated

#37 PUFA:ti,ab

#38 "unsaturated fatty acid*":ti,ab

#39 #13 or #14 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 or #26 or #27 or #28 or #29 or #30 or #30 or #32 or #33 or #34 or #35 or #36 or #37 or #38

#40 #12 and #39

Appendix 2. MEDLINE search strategy

Original search date: 27 August 2020

1. exp Gout/

2. gout$.tw.

3. tophus.tw.

4. tophi.tw.

5. tophaceous.tw.

6. Uric Acid/

7. uric acid.tw.

8. ((sodium or monosodium or potassium or ammonium) adj2 urate).tw.

9. urate crystal$.tw.

10. Hyperuricemia/

11. Hyperuric?emi$.tw.

12. or/1‐11

13. exp Dietary Supplements/

14. ((diet$ or food) adj supplement$).tw.

15. exp Food, fortified/

16. enriched.ti,ab

17. N?utraceutical$.tw.

18. exp Antioxidants/

19. antioxidant$.tw.

20. Free Radical$.tw.

21. exp Carnitine/

22. Carnitine.tw.

23. exp Glutamine/

24. glutamine.tw.

25. oligoelement$.tw.

26. exp amino acids/

27. amino acid$.tw.

28. exp Minerals/

29. mineral$.tw.

30. exp Vitamins/

31. vitamin$.tw.

32. probiotic$.tw.

33. prebiotic$.tw.

34. synbiotic$.tw.

35. yeast$.tw.

36. exp Fatty acids, unsaturated/

37. PUFA.tw

38. unsaturated fatty acid$.ti,ab

39. or/13‐38

40. 12 and 39

41. randomized controlled trial.pt.

42. randomized.ti,ab.

43. placebo.ti,ab.

44. or/41‐43

45. 40 and 44

Appendix 3. Embase search strategy

Original search date: 27 August 2020

#1 gout/exp

#2 gout*:ti,ab

#3 tophus:ti,ab

#4 tophi:ti,ab

#5 tophaceous:ti,ab

#6 'uric acid'

#7 'uric acid':ti,ab

#8 ((sodium OR monosodium OR potassium OR ammonium) NEAR/2 urate):ti,ab

#9 'urate crystal*':ti,ab

#10 hyperuricemia

#11 Hyperuric?emi*:ti,ab

#12 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11

#13 'diet supplementation'

#14 ((diet* or food) NEXT supplement*):ti,ab

#15 enriched:ti,ab

#16 nutraceutical/

#17 N?utraceutical*:ti,ab

#18 antioxidant/exp

#19 antioxidant*:ti,ab

#20 free radical/

#21 Free Radical*:ti,ab

#22 carnitine/

#23 Carnitine:ti,ab

#24 glutamine/

#25 glutamine:ti,ab

#26 oligoelement*:ti,ab

#27 amino acid/exp

#28 'amino acid*':ti,ab

#29 mineral/

#30 mineral*:ti,ab

#31 vitamin/exp

#32 vitamin*:ti,ab

#33 probiotic agent/

#34 probiotic*:ti,ab

#35 prebiotic agent/

#36 prebiotic*:ti,ab

#37 synbiotic agent/

#38 symbiotic*:ti,ab

#39 yeast/

#40 yeast*:ti,ab

#41 'unsaturated fatty acid'/exp

#42 PUFA:ti,ab

#43 'unsaturated fatty acid$':ti,ab

#44 (#13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 or #26 or #27 or #28 or #29 or #30 or #31 or #32 or #33 or #34 or #35 or #36 or #37 or #38 or #39 or #40 or #41 or #42 or #43)

#45 #12 and #44

#46 (random* or placebo*):ti,ab

#47 ((single* or double* or triple* or treble*) and (blind* or mask*)):ti,ab.

#48 'controlled clinical trial*':ti,ab.

#49 RETRACTED ARTICLE/

#50 #46 or #47 or #48 or #49

#51 (animal* not human*):sh,hw.

#52 #50 not #51

#53 #45 and #52

Appendix 4. CINAHL search strategy

Original search date: August 2020

S1 (MH "Gout")

S2 TI gout* OR AB gout*

S3 TI tophus OR AB tophus

S4 TI tophi OR AB tophi

S5 TI tophaceous OR AB tophaceous

S6 (MH "Uric Acid")

S7 TI uric acid OR AB uric acid

S8 TI sodium OR AB sodium

S9 TI monosodium OR AB monosodium

S10 TI potassium OR AB potassium

S11 TI ammonium urate OR AB ammonium urate

S12 TI urate crystal OR AB urate crystal

S13 TI Hyperuricemi* OR AB Hyperuricemi*

S14 S1 OR S2 OR S3 OR S4 OR S5 OR S6 OR S7 OR S8 OR S9 OR S10 OR S11 OR S12 OR S13

S15 (MH "Dietary Supplements+")

S16 TI diet* supplement* OR AB diet* supplement*

S17 TI food supplement* OR AB food supplement*

S18 (MH "Food, Fortified")

S19 TI enriched OR AB enriched

S20 TI N?utraceutical* OR AB N?utraceutical*

S21 (MH "Antioxidants+")

S22 TI antioxidant* OR AB antioxidant*

S23 TI Free Radical* OR AB Free Radical*

S24 (MH "Carnitine")

S25 TI Carnitine OR AB Carnitine

S26 (MH "Glutamine")

S27 TI glutamine OR AB glutamine

S28 TI oligoelement* OR AB oligoelement*

S29 (MH "Amino Acids+")

S30 TI amino acid* OR AB amino acid*

S31 (MH "Minerals+")

S32 TI mineral* OR AB mineral*

S33 (MH "Vitamins+")

S34 TI vitamin* OR AB vitamin*

S35 TI probiotic* OR AB probiotic*

S36 TI prebiotic* OR AB prebiotic*

S37 TI synbiotic* OR AB synbiotic*

S38 TI yeast* OR AB yeast*

S39 (MH "Fatty Acids, Unsaturated+")

S40 TI PUFA OR AB PUFA

S41 TI unsaturated fatty acid* OR AB unsaturated fatty acid*

S41 S15 OR S16 OR S17 OR S18 OR S19 OR S20 OR S21 OR S22 OR S23 OR S24 OR S25 OR S26 OR S27 OR S28 OR S29 OR S30 OR S31 OR S32 OR S33 OR S34 OR S35 OR S36 OR S37 OR S38 OR S39 OR S40 OR S41

S42 S14 AND S42

S43 PT randomized controlled trial

S44 TI randomized OR AB randomized 

S45 TI placebo OR AB placebo 

S46 S43 OR S44 OR S45

S47 S14 AND S41 AND S46

Data and analyses

Comparison 1. Skim milk powder (SMP; GMP/G600) versus control (SMP/lactose).

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Acute gout flares (per month, after 3 months of treatment)	1	120	Mean Difference (IV, Random, 95% CI)	‐0.21 [‐0.76, 0.34]
1.2 Study withdrawals due to adverse events	1	120	Risk Ratio (M‐H, Random, 95% CI)	1.27 [0.53, 3.03]
1.3 Serum uric acid reduction	1	102	Mean Difference (IV, Random, 95% CI)	‐0.01 [‐0.04, 0.01]
1.4 Joint pain reduction (during gout flares)	1	120	Mean Difference (IV, Random, 95% CI)	‐1.03 [‐1.89, ‐0.17]
1.5 Total number of adverse events	1	120	Risk Ratio (M‐H, Random, 95% CI)	0.97 [0.66, 1.45]
1.6 Physical function	1	104	Mean Difference (IV, Random, 95% CI)	‐0.03 [‐0.14, 0.08]
1.7 Serious adverse events	1	120	Risk Ratio (M‐H, Fixed, 95% CI)	1.33 [0.23, 7.66]

Comparison 2. Vitamin C versus allopurinol.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Serum uric acid reduction	1	40	Mean Difference (IV, Random, 95% CI)	0.10 [0.06, 0.15]
2.2 sUA reduction (subgroup analysis)	1	40	Mean Difference (IV, Random, 95% CI)	0.10 [0.02, 0.19]
2.2.1 Not on allopurinol at study entry	1	20	Mean Difference (IV, Random, 95% CI)	0.15 [0.08, 0.22]
2.2.2 On allopurinol at study entry	1	20	Mean Difference (IV, Random, 95% CI)	0.06 [‐0.01, 0.13]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Dalbeth 2012.

Study characteristics
Methods	3‐month randomised controlled double‐blind trial
Participants	N = 120 Inclusion criteria: 1. Adults aged ≥ 18 years 2. Gout diagnosed (according to the American College of Rheumatology (ACR) diagnostic classification, recurrent gout flares (at least 2 flares in the preceding 4 months) 3. Participants experiencing frequent gout flares at the time of study enrolment (≥ 2 flares in the preceding 4 months) Exclusion criteria: 1. Lactose intolerance 2. Severe renal impairment (defined as estimated glomerular filtration rate < 30 mL/minute) Lactose group (N = 40): 1. Males, n (%): 37 (93) 2. Mean age, years (SD): 57 (16) 3. White ethnicity, n (%): 28 (70) 4. Number of self‐reported flares in preceding 4 months, mean (SD): 3.9 (2.7) 5. Number of gout flares in baseline month, mean (SD): 1.3 (1.5) 6. Allopurinol use, n (%): 21 (53) 7. Colchicine use, n (%): 12 (30) 8. Prednisone use, n (%): 4 (10) 9. NSAID use, n (%): 11 (28) 10. Diuretic use, n (%): 2 (5) 11. sUA, mmol/L, mean (SD): 0.44 (0.11) 12. Tophaceous gout, n (%): 8 (20%) 13. Serum creatinine, μmol/L, mean (SD): 91 (18) SMP group (N = 40): 1. Males, n (%): 36 (90) 2. Mean age, years (SD): 56 (12) 3. White ethnicity, n (%): 28 (70) 4. Number of self‐reported flares in preceding 4 months, mean (SD): 4.5 (2.3) 5. Number of gout flares in baseline month, mean (SD): 1.1 (1.4) 6. Allopurinol use, n (%): 22 (55) 7. Colchicine use, n (%): 7 (18) 8. Prednisone use, n (%): 8 (20) 9. NSAID use, n (%): 10 (25) 10. Diuretic use, n (%): 1 (2.5) 11. sUA, mmol/L, mean (SD): 0.41 (0.09) 12. Tophaceous gout, n (%): 17 (43) 13. Serum creatinine, μmol/L, mean (SD): 91 (19) GMP/G600 SMP group (N = 40): 1. Males, n (%): 35 (88) 2. Mean age, years (SD): 56 (13) 3. White ethnicity, n (%): 22 (55) 4. Number of self‐reported flares in preceding 4 months, mean (SD): 5.1 (9.6) 5. Number of gout flares in baseline month, mean (SD): 1.8 (2.4) 6. Allopurinol use, n (%): 22 (55) 7. Colchicine use, n (%): 13 (33) 8. Prednisone use, n (%): 4 (10) 9. NSAID use, n (%): 11 (28) 10. Diuretic use, n (%): 8 (20) 11. sUA, mmol/L, mean (SD): 0.42 (0.11) 12. Tophaceous gout, n (%): 10 (25) 13. Serum creatinine, μmol/L, mean (SD): 93 (20)
Interventions	Intervention 1: lactose powder, active control Intervention 2: SMP, active control Intervention 3: SMP enriched with GMP and G600 (GMP protein 1.5 g (10% total protein) and G600 0.525 g (3.5% of total protein weight))
Outcomes	Outcome assessments at 1, 2, and 3 months: Primary end point: 1. Change in frequency of gout flares Secondary end points: 1. Change in swollen joint count (/66) 2. Change in tender joint count (/68) 3. Pain (10‐point Likert; scored 0 to 10; where 0 = no pain and 10 = severe pain) 4. Participant global assessment (0 to 100; where 0 = very well and 100 = very poor) 5. C‐reactive protein (mg/L) 6. sUA concentration (mmol/L) 7. Fractional excretion of uric acid (%) 8. Health Assessment Questionnaire (HAQ)‐II, 10‐item questionnaire, each item scored from 0 = without any difficulty to 3 = unable to perform. Sum of the scores of each questionnaire item divided by the number of questions answered to obtain a value between 0 (minimal loss of function) and 3 (completely disabled) 9. Open‐ended enquiry to elicit adverse events
Notes	Proof‐of‐concept trial Study end point results not covered in this review: 1. Participants in the GMP/G600 SMP group showed greater reduction in tender joint count from baseline (MD ‐0.49, 95% CI ‐0.85 to ‐0.12). The 95% CI suggested that there could be a benefit, as it could be as much as a 1.96 reduction. 2. No significant differences between groups were detected for change in the swollen joint count from baseline (MD ‐0.23, 95% CI ‐0.61 to 0.16). 3. No significant differences in reduction of the number of self‐reported flares (MD ‐0.49, 95% CI ‐1.08 to 0.09) were detected. 4. The trial authors reported no statistical difference between GMP/G600 SMP and the control group in changes in serum creatinine, sUA concentrations, C‐reactive protein levels, waist circumference, serum lipid profile, or weight over time. Diastolic blood pressure was reported by the trial authors to decrease by a mean of 3.6 mmHg (95% CI 1.8 to 5.4) in the GMP/G600 SMP group over the study period (P = 0.0002), with a greater reduction in diastolic blood pressure recorded when compared with the lactose control (Tukey post hoc test, P = 0.001). Unpublished and raw data kindly supplied by Dr Nicola Dalbeth Funded by LactoPharma and the New Zealand Government Foundation for Research Science and Technology. Most of the authors declared a relevant conflict of interest Australian & New Zealand Clinical Trial Registry number: ACTRN12609000479202
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Patients were randomised using a random block randomisation algorithm"
Allocation concealment (selection bias)	Unclear risk	Quote: "Participants and study staff were blinded to treatment allocation throughout the study..." Comment: no details of the actual method of allocation concealment to intervention was provided
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "The products were dry‐blended and packed into identical, custom‐made aluminium foil sachets... Each intervention was a cream‐coloured powder administered daily as a 250 ml vanilla flavoured shake"
Blinding of outcome assessment (detection bias) Laboratory assessed outcomes	Unclear risk	1. sUA concentration 2. Fractional excretion of uric acid 3. C‐reactive protein 4. Tender joint count 5. Swollen joint count 6. Adverse events Quote: "Study staff were blinded to treatment allocation throughout the study..." Comment: although not explicitly stated, it was implied from the aforementioned statement that outcome assessors were blinded
Blinding of outcome assessment (detection bias) Participant assessed outcomes	Low risk	1. Gout flare frequency 2. Participant global assessment 3. Health Assessment Questionnaire (HAQ)‐II
Incomplete outcome data (attrition bias) All outcomes	Low risk	102 (85%) completed the study per protocol (2 discontinued for adverse events, 8 lost to follow‐up, and 8 continued the study without taking the study product after an adverse event) Comment: the distribution of drop‐outs was even between groups as shown in supplementary figure
Selective reporting (reporting bias)	Unclear risk	Comment: all pre‐specified outcomes reported. There was selective reporting of a post hoc comparison between GMP/G600 SMP and lactose powder control on change in gout flare frequency, and lowering of diastolic blood pressure (an outcome not covered in this review)
Other bias	Unclear risk	Conflict of Interest: "This work was funded by LactoPharma (a joint venture between Fonterra Ltd, Fonterra R&D Ltd and Auckland UniServices Ltd) and the New Zealand Government Foundation for Research Science and Technology. Barbara Kuhn‐Sherlock, Alastair MacGibbon and Kate Palmano are employees of Fonterra Co‐operative Group Ltd. Alastair MacGibbon, Nicola Dalbeth, and Kate Palmano are named inventors on a patent application related to milk products and gout.", although it says that "Data analysis was completed by a biostatistician independent of the study sponsors"

Stamp 2013.

Study characteristics
Methods	8‐week open‐label, parallel, randomised controlled trial
Participants	N = 40 Inclusion criteria: 1. Gout, diagnosed by the ACR preliminary criteria 2. sUA level > 0.36 mmol/L Exclusion criteria: 1. Participants taking non‐prescription vitamin supplements Stratification of participants whether taking allopurinol or not at entry Vitamin C group (N = 20): 1. Males, n (%): 18 (90) 2. Age, mean (range): 61.2 years (39 to 86) 3. Weight, mean (± SEM): 93.1 ± 3.3 kg 4. Body mass index, mean (± SEM): 30.4 ± 0.96 kg/m² 5. New Zealand European, n: 14 6. sUA, mean (± SEM): 0.5 ± 0.11 mmol/L 7. Estimated glomerular filtration rate, mean (± SEM): 65.5 ± 3.5 8. Taking diuretics, no (%): 6 (30) 9. Taking aspirin, no (%): 5 (25) 10. Smoker, no (%): 1 (5) No vitamin C group (N = 20): 1. Males, n (%): 18 (90) 2. Age, mean (range): 55.0 years (27‐78) 3. Weight, mean (± SEM): 100.3 ± 5.9 kg 4. Body mass index, mean (± SEM): 32.0 ± 1.5 kg/m² 5. New Zealand European, n: 11 6. sUA, mean (± SEM): 0.5 ± 0.09 mmol/L 7. Estimated glomerular filtration rate, mean (± SEM): 67.9 ± 4.6 mL/minute 8. Taking diuretics, no (%): 5 (25) 9. Taking aspirin, no (%): 7 (35) 10. Smoker, no (%): 3 (15)
Interventions	Subgroup 1 (participants not taking allopurinol at entry): Intervention 1: vitamin C 500 mg daily Intervention 2: allopurinol 50 mg to 100 mg daily (dose adjustment at 4 weeks based on sUA level) Subgroup 2 (participants already taking allopurinol at entry): Intervention 1: added vitamin C 500 mg daily to the stable dose of allopurinol Intervention 2: increased the allopurinol dose (no specified dose scheme)
Outcomes	Outcome measures at 4 (only for the allopurinol group) and 8 weeks Primary end point: 1. sUA Secondary end points: 1. Plasma ascorbate levels 2. Oxypurinol levels 3. Adverse events
Notes	Allopurinol dose was adjusted at 4 weeks if sUA < 0.36 mmol/L was not achieved, while vitamin C dose was not modified Unpublished and raw data kindly supplied by Dr Lisa Stamp Australian & New Zealand Clinical Trial Registry number: ACTRN12610000545066
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization of patients was stratified according to current allopurinol use, using permuted blocks of size 4..." Comment: we asked the authors, and their answer was that the randomisation list was computer generated by the independent statistician, in advance of recruitment. Randomisation was stratified on allopurinol use, separate lists for both strata and arranged in permuted blocks of size 4
Allocation concealment (selection bias)	High risk	Comment: no information regarding allocation concealment was given by the authors, but it was an open trial
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "This was an 8‐week open‐label, parallel‐group, randomised controlled trial..." Comment: neither participants nor investigators were blinded
Blinding of outcome assessment (detection bias) Laboratory assessed outcomes	Unclear risk	1. sUA reduction Comment: no data regarding study staff blindness to treatment allocation were given in the study, but its influence on the urate level measurement was likely to be irrelevant
Blinding of outcome assessment (detection bias) Participant assessed outcomes	High risk	1. Adverse events Comment: it was an open trial
Incomplete outcome data (attrition bias) All outcomes	Low risk	All randomised participants completed the study, with no dropouts or loss to follow‐up
Selective reporting (reporting bias)	Low risk	All planned outcomes were clearly reported in the results section
Other bias	Low risk	Independent research; no relation to vitamin C or allopurinol producers

CI: confidence interval; GMP: glycomacropeptide; MD: mean difference; N: number; NSAID: non‐steroidal anti‐inflammatory drug; SD: standard deviation; SEM: standard error of the mean; SMP: skim milk powder; sUA: serum uric acid.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Howren 2021	wrong comparator
Stewart 2020	wrong comparator
Yamanaka 2019	Study population not limited to chronic gout, and no separate analysis provided. Authors contacted for raw data, but no reply was received.
Zong 2019	wrong comparator

Characteristics of studies awaiting classification [ordered by study ID]

Arych 2010.

Methods	Randomized controlled trial.
Participants	Gout (details not provided)
Interventions	Artichoke (cynara scolymus 200 mg) or placebo
Outcomes	"The artichoke promotes the hepaticobiliary system dysfunction acceleration regress, normalizes serum bilirubin, uric acid, urea concentration, serum liver enzymes (alanine aminotransferase, alanine aminopeptidase, aldolase, amylase, aspartate aminotransferase, gamma glutamyltransferase, lactate dehydrohenase, alkaline phosphatase)." ... crude or numerically data not given
Notes	Conference abstract, unpublished study. Authors' contact information unavailable.

ACR: American College of Rheumatology

CRP: C‐reactive protein

PD: Power Doppler

US: ultrasound

VAS: visual analog scale

EULAR: European Alliance of Associations for Rheumatology

GP: General Practitioner

ULT: urate‐lowering therapy

Characteristics of ongoing studies [ordered by study ID]

ACTRN12618000770268.

Study name	A randomized controlled trial of of krill oil on ultrasound detected inflammation in people with gout.
Methods	Open (masking not used) randomised controlled trial
Participants	Over 18 years 1. A diagnosis of gout according to 2015 American College of Rheumatology (ACR) criteria 2. Evidence of inflammation as defined as a CRP > 5, or PD signal on US 3. Able, and willing to commence either krill oil or colchicine 4. Currently taking, or willing to commence allopurinol therapy 5. Happy to consent to krill oil therapy for 12 weeks, and review at 12 weeks
Interventions	Krill oil 1 gm twice daily versus oral colchicine 500 mcg daily
Outcomes	Intra‐articular inflammation, defined as ultrasound detected power, doppler signal in the joints of people with gout (primary outcome); change in serum CRP; change in participant global pain VAS; change in participant global health VAS; change in global health of the participants; participant‐reported number of acute flares; change in function, assessed by the Health Assessment Questionnaire (HAQ); change in disease activity, assessed by the gout assessment questionnaire (GAQ); quality of life (AQoL); safety (participant‐reported symptoms, biochemistry)
Starting date	1 June 2018
Contact information	Dr Helen Keen, + 61 8 61522222, helen.keen@uwa.edu.au
Notes	Not yet recruiting (at 31 August 2021)

ChiCTR1900020614.

Study name	The efficacy of celery seed extract in patients with primary gout: a randomized controlled trial
Methods	Single‐center, randomized, double‐blind, controlled design
Participants	(1) Aged 18 to 75 years old males, consistent with the clinical classification and diagnosis criteria for gout in 2015 ACR/EULAR; (2) uric acid value is 420 microns/L (7 mg/dL); (3) obtain the consent and sign the informed consent form of the subject, or family members, or guardians, or legal agents
Interventions	Celery seed extract versus vitamin C versus starch
Outcomes	Uric acid compliance rate (primary outcome); percentage of participants with uric acid at least once to SUA < 360 μmol/L; percentage of participants with uric acid < 300 μmol/L at the last visit; frequency of the percentage of participants with acute gout; others (body composition indexes, laboratory values)
Starting date	1 March 2019
Contact information	Ting Zhao, +86 18661809907, ztxhhh@sina.com
Notes	Not yet recruiting (at 31 August 2021)

ISRCTN79392964.

Study name	Omega–3 fatty acids for the prophylaxis of acute attacks of gout on initiating urate‐lowering treatment – a feasibility study for a randomised controlled trial (SOGAS)
Methods	Interventional (randomised controlled trial)
Participants	1. GP/physician‐diagnosed gout 2. Meets the American College of Rheumatology/European League against Rheumatism (EULAR) classification criteria for gout 3. Willing to commence on urate‐lowering treatment 4. Serum uric acid ≥ 360 μmol/L at the screening visit 5. Age ≥ 21 years at the screening visit 6. Able to give informed consent 7. No change in the average weekly dose of analgesics for at least 4 weeks prior to the screening visit 8. Be able to adhere to the study visit schedule and other protocol requirements 9. Subjects able to communicate well with the investigator or designee, to understand and comply with the requirements of the study, and to understand and sign the written informed consent
Interventions	Intervention: omega‐3 fatty acid capsules (4 g/day; containing 3.36 g of omega‐3 fatty acids) Control: placebo capsules (4 g olive oil/day)
Outcomes	Primary outcome measure: dropout rate, defined as number of participants randomised that remain in the study until end of follow‐up at 28 weeks Secondary outcomes: 1. Recruitment rate, recorded as the number of eligible participants, based on self‐reported number of gout flares in initial gout questionnaire who are randomised in the study 2. The proportion of people approached by the GP with information about the study and requested to return the reply slip to the study team, who complete the following by the end of the 28‐week study: 2.1. Reply to Academic Rheumatology, University of Nottingham 2.2. Meet the eligibility criteria on telephone screening, and based on reply to the initial gout questionnaire 2.3. Agree for screening visit 2.4. Meet the eligibility criteria after checking serum uric acid, full blood count, liver and kidney function tests 2.5. Are randomised into the trial 3. The number of gout flares measured using gout diary between weeks 5 and 28 of the study 4. The severity of gout flares defined as an average of daily pain visual analogue scale (0 to 100) during a gout flare between weeks 5 and 28 of the study. Calculated by sum of daily pain scores divided by number of days on which the gout flare is present 5. The number of days gout flares were experienced by each participant, measured using gout diary between weeks 5 and 28 of the study 6. The number of participants who withdrew themselves from the study due to side effects related to omega‐3 fatty acids between weeks 0 and 28 of the study 7. The number of participants not on ULT at each GP surgery (office) at the start of recruitment when surgeries complete the mail‐out 8. Compliance with study drug measured by capsule count at weeks 14 and 28
Starting date	5 February 2019
Contact information	Dr Abhishek Abhishek, University of Nottingham, City Hospital, Room A27, Clinical Sciences Building, Nottingham, NG5 1PB, United Kingdom
Notes	Ongoing, no longer recruiting (at 31 August 2021)

NCT03621215.

Study name	The effect of tart cherry juice on risk of gout attacks
Methods	Participants will be randomly allocated to consume either tart cherry juice or a fruit‐flavoured placebo drink (matched as closely as possible for energy and sensory characteristics) daily for 12 months
Participants	18 to 85 years, existing diagnosis of gout, with at least one gout flare in the previous 12 months
Interventions	30 mL tart cherry juice concentrate versus fruit‐flavoured placebo drink
Outcomes	Change in gout flare frequency (primary outcome); change in gout flare intensity; changes in serum urate; change in fractional excretion of uric acid; change in inflammatory markers; change in oxidative damage; change in antioxidant status; change in blood pressure; change in arterial stiffness; change in lipid profile; difference in body mass index
Starting date	15 May 2019
Contact information	Contact: Tony Lynn, PhD 0114 2252065 T.Lynn@shu.ac.uk Contact: Margo Barker, PhD 0114 2560765 Margo.Barker@shu.ac.uk
Notes	Active, not recruiting (at 31 August 2021)

NCT03650140.

Study name	Pharmacokinetics and pharmacodynamics of anthocyanins
Methods	Interventional (clinical trial)
Participants	18 years to 65 years, diagnosis of gout
Interventions	Single 60 mL versus 120 mL oral doses of tart cherry extract
Outcomes	Reduction in hs‐CRP (primary outcome); the area under the plasma anthocyanin concentration‐time curve; fold change in mRNA expression of Nrf2; peak plasma concentration of anthocyanins; time to achieve peak plasma anthocyanin concentration; plasma anthocyanin half‐life
Starting date	1 February 2019
Contact information	Naomi Schlesinger, MD. 732‐235‐7217. schlesna@rwjms.rutgers.edu
Notes	Recruitment suspended due to COVID‐19 (at 31 August 2021)

NCT04199325.

Study name	Probiotics for gout/hyperuricemia: randomized, double‐blind, intervention, parallel controlled, multicenter
Methods	Interventional (clinical trial), randomized, parallel assigment
Participants	Adults (18 to 70 years old), gender unlimited, previous history of gout according to the 2015 EULAR/ACR classification criteria, and fasting serum uric acid ≥ 480 μmol/L (8mg/dL)
Interventions	Probiotics versus placebo
Outcomes	Serum urate at 24 weeks (% of SU below 360 μmol/l) and other time‐points, and number of gout flares that occurred
Starting date	1 November 2018
Contact information	The Affiliated Hospital of Inner Mongolia Medical University. qpfff@126.com
Notes	Recruiting (at 31 August 2021)

Differences between protocol and review

In addition to the pre‐planned screening of 2010 to 2011 meeting abstracts of American College of Rheumatology (ACR) and European League against Rheumatism (EULAR), we also screened abstracts from the 2012 and 2013 meetings in order to increase the chance of identifying studies for our review.

We added methods describing how we planned to analyse cross‐over trials, if we had identified any, and described how we calculated absolute effects for dichotomous and continuous data, and the estimate of the minimal clinically important difference.

We had not pre‐planned a subgroup analysis to investigate whether response to treatment with vitamin C varied in people who were allopurinol naive or already taking allopurinol. However, one trial did report such data. This might be a relevant issue, as some authors pointed out that the response to urate‐lowering therapy, in terms of serum uric acid (sUA) reduction for people who are allopurinol naive may be significantly different from people who are already taking allopurinol and have their dosage increased (Pascual 2009a). Further increases in drug doses may not result in a comparable sUA reduction; this might be related to an increase in sUA as crystalline deposits start to dissolve. Following this reasoning, we decided to include the analysis in our review.

In the 2014 protocol, we planned that the pharmacological comparators could comprise any agent with urate‐lowering properties (xanthine‐oxidase inhibitors, uricosurics, recombinant uricases, etc.). For the 2020 update of the review, the pharmacological comparisons for diet supplements were limited to allopurinol, as this remains the standard‐of‐care agent for urate management in people with gout worldwide (Richette 2017). Nevertheless, no study comparing a diet supplement with other urate‐lowering agents were retrieved in the updated search,

Contributions of authors

MA wrote the first draft of the review.

FS, RB, and LC contributed to the final version of the review by providing comments and suggestions. JPP ran the searches.

All authors approved the final version of the review.

Sources of support

Internal sources

• Hospital General Universitario de Alicante, Alicante, Spain

  In‐kind support

• Hospital General Universitario de Elda, Alicante, Spain

  In‐kind support

• Columbia University Medical Center, New York, USA

  In‐kind support

• Cabrini Hospital, Malvern, Australia

  In‐kind support

• Monash University, Melbourne, Australia

  In‐kind support

• Universidad Camilo José Cela, Madrid, Spain

  In‐kind support

External sources

• National Health and Medical Research Council (NHMRC), Australia

  Funding for Cochrane Musculoskeletal – Australia and Cochrane Australia

Declarations of interest

MA: from Grunenthal, honoraria for consultancy (January 2017), lectures (November 2018), and travel expenses to attend 2017 EULAR meeting; from Menarini, honoraria for lecture (2019); from Horizon, honoraria for consultancy (June 2017); and from Astra‐Zeneca, honoraria for consultancy (2015), and travel expenses to attend 2015 ACR meeting. None of these laboratories manufacture diet supplements or allopurinol.

FS: received honoraria for consultancy from Abbvie, Pfizer, Grunenthal, and MSD; payment for lectures from Grunenthal; and travel expenses from Novartis, Pfizer, and Abbvie. Her institution also received grants from Roche, Novartis, and Sandoz.

LC: has no conflicts of interest with the producers or providers of products appearing in the review; however, her institution provided consultant services for studies and educational activities to all the producers of potential drugs to include in the review: Grunenthal (to analyse a database and draft a manuscript on gout mortality in 2018), SOBI (The Spanish Society of Paediatric Rheumatology, via a grant from SOBI, in 2016 to 2017 to run a national registry on JIA‐Still's), and Novartis (in 2018 to 2020 to finance online courses to medical specialists on research methodology). LC has not personally benefited from these funds.

RB: is the Co‐ordinating Editor of Cochrane Musculoskeletal but is not involved in editorial decisions regarding this review. She is the recipient of a National Health and Medical Research Council (NHMRC) Cochrane Collaboration Round 7 Funding Program Grant, which supports the activities of Cochrane Musculoskeletal Australia and Cochrane Australia, but the funding source did not participate in the conduct of this review.

JPP: none known

Edited (no change to conclusions)