Dietary and lifestyle interventions in non-infectious uveitis: protocol for a systematic review

Abstract

Introduction

Non-infectious uveitis (NIU) is a significant cause of visual impairment—often requiring long-term immunosuppressive therapy (which carries systemic side effects). There is growing interest in non-pharmacological adjuncts—broadly encompassing: dietary supplements, lifestyle modifications and psychological interventions—as an adjunct to treatment. However, the existing evidence is fragmented and lacks systematic synthesis.

This is the first systematic review to identify and evaluate studies examining the relationship between modifiable lifestyle behaviours or dietary supplements and clinical outcomes in NIU, across both adult and paediatric populations.

Methods

A comprehensive literature search will be conducted across Ovid Medline, Ovid Embase, CENTRAL, CINAHL, Scopus, Web of Science and grey literature—with no date or language restrictions. Eligible studies will include randomised controlled trials and non-randomised controlled studies.

Primary outcomes include flare frequency/severity, visual acuity, treatment response and adverse events. Secondary outcomes include health-related quality of life, inflammatory markers and mental health outcomes. Two reviewers will independently screen studies, extract data and assess risk of bias using Risk of Bias 2 and Risk of Bias in Non-randomised Studies of Interventions tools. Certainty of evidence will be evaluated using Grading of Recommendations, Assessment, Development and Evaluation (GRADE). A narrative synthesis will be conducted, with meta-analysis where appropriate, and heterogeneity assessed using I². Subgroup and sensitivity analyses will explore age, uveitis subtype, intervention type and study quality.

Ethics and dissemination

No ethical approval is required as this review synthesises publicly available data. Findings will be disseminated via peer-reviewed publication, professional conferences, patient advocacy organisations (eg, Uveitis UK). The protocol was prospectively registered with PROSPERO in August 2025 and follows Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) guidelines.

PROSPERO registration number

CRD420241077488.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• A comprehensive search across seven databases with no date or language restrictions will maximise the retrieval of relevant evidence.

• Independent dual-reviewer screening, data extraction and consensus-based discrepancy resolution will minimise selection and extraction bias.

• Risk of bias in randomised and non-randomised studies will be assessed using Risk of Bias 2 and Risk of Bias in Non-randomised Studies of Interventions, respectively, and evidence certainty will be assessed using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) framework.

• Heterogeneity in uveitis subtypes, populations and interventions may limit comparability.

• The inclusion of studies with low methodological quality or incomplete reporting may affect the strength of conclusions.

Introduction

Non-infectious uveitis (NIU) encompasses a group of immune-mediated ocular inflammatory conditions¹ and remains a leading cause of preventable sight loss worldwide.² Incidence varies widely, 3.9–207.8 per 100 000 persons,^3–5 and negatively impacts quality of life—particularly for patients with ocular pain or prior visual impairment.⁶ This burden often extends to work disability, reduced productivity and mental health conditions.⁷

Long-term immunosuppression (eg, corticosteroids, disease-modifying anti-rheumatic drugs (DMARDs)) forms the mainstay of management yet has significant systemic and ocular side effects.^8–10 Although pharmacological strategies are well established, there is growing interest and evidence in adjunctive, non-pharmacological management. These include sleep, dietary supplements (such as omega-3, curcumin), exercise, smoking and psychological stress^{11 12}—many of which have proposed anti-inflammatory or immunomodulatory effects.^13–15

However, the current evidence is methodologically heterogeneous and poorly synthesised, making it difficult to draw reliable clinical conclusions. To date, no prior systematic review has comprehensively evaluated the role of modifiable, non-drug interventions in NIU. This systematic review will critically evaluate the evidence linking lifestyle behaviours and dietary supplements to clinical outcomes in NIU. The results of this systematic review will guide clinical guidelines, promote patient-centred care and influence future research directions.

Review objectives

• To evaluate the role of specific dietary supplements (eg, curcumin, omega-3 fatty acids) in improving clinical outcomes or reducing inflammation in NIU.

• To identify lifestyle behaviours (eg, smoking, diet, sleep, stress, physical activity) associated with disease activity or complications.

• To assess the quality, consistency and strength of evidence supporting these interventions.

Methods and analysis

Population

We will include studies involving human participants with NIU, including both adult and paediatric cohorts. Animal or in vivo studies will be excluded. The systematic review is planned to commence in November 2026, with an anticipated completion date in July 2027, encompassing database searching, screening, data extraction, risk of bias, assessment and synthesis of evidence.

Intervention

Studies investigating modifiable lifestyle behaviours (eg, sleep, exercise, diet, stress management) or dietary supplements (eg, curcumin, omega-3 fatty acids) will be included. Studies will be excluded if their primary stated aim is microbiome modification (eg, faecal microbiota transplantation, probiotic supplementation with microbiome endpoints as the primary outcome), pharmacological treatments or surgical procedures.

Comparator

Eligible comparators include:

1. Placebo or no intervention.

2. Usual pharmacological care for NIU (eg, corticosteroids, DMARDs), where dietary or lifestyle interventions are assessed as adjuncts.

3. Alternative non-pharmacological intervention.

4. Pre-intervention baseline values in single-arm or pre-post study designs.

The presence of pharmacological co-treatment in the control arm will be recorded and explored in sensitivity analyses.

Outcomes

Primary outcomes include:

• Flare frequency and severity: measured using Standardisation of Uveitis Nomenclature criteria and reported as number of flares per year or change in anterior chamber cell grade—assessed at 3, 6 and 12 months.

• Visual acuity: the logarithm of the minimum angle of resolution will be used and findings will be reported as the mean change from baseline, assessed at 3, 6 and 12 months.

• Treatment response: defined as reduction or cessation of immunosuppressive therapy, or corticosteroid-sparing effect. It will be reported as the proportion of patients achieving treatment modification and will be assessed at 6 and 12 months.

• Adverse events: all harms reported in included studies will be extracted, including both intervention-specific adverse effects and disease-related ocular complications. A pre-specified list will not be imposed, and a harm-inclusive approach will be adopted as per Cochrane guidance.

Secondary outcomes include:

• Health-related quality of life: assessed using the National Eye Institute Visual Function Questionnaire-25, reported as the mean change from baseline at 6 and 12 months.

• Inflammatory markers: assessed as the mean change of the white cell count, C reactive protein and erythrocyte sedimentation rate from baseline to 3 and 6 months.

• Mental health outcomes: validated tools (patient health questionnare-9 for depression, generalised anxiety disorder-7 for anxiety) will be used to assess mental health outcomes, with assessments reported at 6 and 12 months.

• Ocular complications, including but not limited to cystoid macular oedema, cataract formation, raised intraocular pressure and optic disc involvement (papillitis).

Search strategy

Databases to be searched include Ovid Medline, Ovid Embase, CENTRAL, CINAHL, Scopus, Web of Science and grey literature sources (eg, trial registries, conference abstracts)—with no date or language restrictions. The search strategy will be developed with an information specialist, and a sample search strategy for Ovid MEDLINE is provided in online supplemental material 1. MeSH keywords and free text terms include uveitis; lifestyle; nutritional supplementation; flare-ups; stress; sleep and dietary supplements.

Results will be managed using Covidence. Two reviewers will independently screen titles, abstracts and full texts. Discrepancies will be resolved by consensus or a third reviewer. A standardised data extraction form will capture study characteristics, populations, interventions/exposures, outcomes and results. The extraction form will be piloted on three studies prior to commencement; where data are unclear, study authors will be contacted.

Study types

Randomised controlled trials (RCTs) and non-randomised controlled studies (including cohort and cross-sectional studies) will be eligible for inclusion.

Case reports, case-control studies and studies focusing exclusively on uveitis-related conditions, such as Behçet’s disease, where an ocular outcome is not the focus of the paper, will be excluded. However, such studies may contribute to narrative synthesis where they provide the only evidence for a given exposure.

Risk of bias and data synthesis

Risk of bias assessments will be conducted independently by two reviewers per study, with disagreements resolved by consensus or by a third reviewer. Risk of bias will be assessed using:

• Risk of Bias 2 (RoB 2) for RCTs.

• Risk of Bias in Non-randomised Studies of Interventions (ROBINS-I) for non-randomised studies.

Summary risk of bias data will be narratively synthesised and, where appropriate, pooled in a meta-analysis using a random-effects model and incorporated into the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) evidence certainty assessment. Sensitivity analyses will exclude studies rated as high risk of bias to assess their influence on pooled estimates.

Subgroup analyses

Dietary supplement and lifestyle behaviour interventions will be regarded as two separate analytical entities given their clinical differences—importantly, results will be reported individually prior to any aggregated analysis. Subgroup analyses will further investigate the following predetermined sources of heterogeneity, depending on the availability of at least two studies per subgroup, as listed below:

• Age: paediatric (<18 years) versus adult (≥18 years).

• Uveitis subtype: anterior, intermediate, posterior, panuveitis.

• Measurement method: self-reported versus objective.

• Disease severity/duration: mild/moderate versus severe or acute versus chronic.

• Geographic context: high-income versus low-/middle-income settings.

Subgroup effects will be analysed using random-effects meta-analysis with interaction tests (eg, Cochrane’s Q). Each subgroup analysis is predetermined based on established sources of clinical heterogeneity in uveitis research, including variations in disease severity, anatomical subtype and age-related immunological differences.

Sensitivity analyses

We will conduct sensitivity analyses by excluding:

• Studies at elevated risk of bias (per RoB 2/ROBINS-I).

• Small sample size (<20 participants).

• Self-reported exposures.

• Short follow-up (<6 months).

Re-analyses will be conducted narratively or statistically, and changes in effect estimates and heterogeneity will be reported in tables or forest plots.

Assessment and management of heterogeneity

Heterogeneity will be evaluated using I² and Cochrane’s Q. Thresholds for I²:

• 0–25%=low.

• 25%–50%=moderate.

• 50%=substantial.

If heterogeneity is substantial (I²>50% as per Cochrane handbook—indicating significant heterogeneity), narrative synthesis will be prioritised. Meta-regression (if ≥10 studies) may explore effects by intervention type, duration, population age or disease severity. Publication bias will be assessed using funnel plots and Egger’s test where ≥10 studies are included in a meta-analysis—selective reporting bias will be assessed by comparing registered outcomes with reported outcomes. All findings will be reported using tables and forest plots.

Ethics and dissemination

Ethical approval is not required as no primary data will be collected, all data will be extracted from previously published studies. Findings will be submitted for peer-reviewed publication, professional forums, patient advocacy groups (eg, Uveitis UK) and presented at national and international conferences. The protocol was prospectively registered with PROSPERO in August 2025 and follows Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) guidelines, prior to commencement of formal systematic title and abstract screening. Any amendments made to this protocol will be documented with date and rationale in the final review and updated on the PROSPERO record.

Findings will be disseminated through peer-reviewed publication in an open-access journal; presentation at relevant professional forums and conferences (including ophthalmology and evidence-synthesis meetings; patient advocacy groups and charities—principally Uveitis UK—through content summaries and educational materials (eg, patient leaflets) to support self-management; and social media and online platforms to maximise reach to both clinical and patient audiences. For matters relating to research integrity, please contact the research integrity team at Imperial College London (rgit@imperial.ac.uk).