Is retinal vein occlusion an underrecognized paraneoplastic sentinel of lymphoma? A systematic review and meta‐analysis of ocular–haematological interactions
Kai‐Yang Chen, Hoi‐Chun Chan, Chi‐Ming ChanAbstract
Topic
This systematic review and meta‐analysis evaluated the association between retinal vein occlusion (RVO) and lymphoma across two complementary evidence components: a qualitative clinical synthesis of RVO occurring in patients with lymphoma and a quantitative synthesis of subsequent cancer outcomes in patients with RVO. For the qualitative synthesis, the population comprised patients with lymphoma and RVO and no comparator was required. For the quantitative synthesis, RVO was the exposure, individuals without RVO formed the comparator, and the prespecified outcomes were overall cancer and lymphoma incidence. Secondary outcomes were ocular, haematological, biochemical, immunological and protein‐related clinical findings.
Clinical Relevance
RVO is a common cause of visual morbidity typically linked to vascular risk factors. Its role as a sentinel manifestation of haematological malignancy remains underrecognized. Early recognition of lymphoma‐associated RVO supports timely systemic diagnosis and coordinated ophthalmic‐haematological evaluation, particularly in atypical or bilateral presentations lacking conventional vascular risk factors.
Methods
A Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA)‐compliant review (PROSPERO CRD420251134925) searched PubMed, Scopus, ScienceDirect, Cochrane Library and Google Scholar from inception to 21 July 2026. Eligible original human studies comprised case‐based reports for qualitative synthesis and observational cohort or registry‐based studies for quantitative synthesis. Case reports were appraised using the Joanna Briggs Institute (JBI) criteria and non‐randomized studies with the Risk Of Bias In Non‐randomized Studies of Interventions (ROBINS‐I) tool. Hazard ratios (HRs), odds ratios (ORs) and standardized incidence ratios (SIRs) were analysed as log‐transformed relative‐effect estimates using fixed‐effects and random‐effects models; heterogeneity was assessed using I 2 and Cochran Q .
Results
Twenty‐six studies were included: 19 case‐based publications describing 24 RVO cases with lymphoma and 7 observational studies contributing to quantitative synthesis. Central RVO (CRVO) predominated, often with bilateral severe visual impairment, optic disc swelling, diffuse haemorrhages and macular oedema. Recurrent systemic findings included anaemia, hyperviscosity, elevated erythrocyte sedimentation rate (ESR) and immunoglobulin M (IgM) paraproteinaemia. Meta‐analysis demonstrated increased overall cancer risk (pooled log effect size 0.198; 95% confidence interval [CI] 0.141–0.255; p < 0.001; I 2 = 21.6%; ≈22% increase) and lymphoma risk (OR 1.50; 95% CI 1.08–2.09; p = 0.015; I 2 = 58.9%). Age was not a significant study‐level moderator ( p = 0.93).
Conclusion
RVO, particularly atypical or bilateral CRVO, represents a clinically informative ocular presentation associated with lymphoma in the reviewed evidence. The combined ocular, haematological and quantitative findings support mechanistic links involving hyperviscosity, thrombosis, leucostasis and infiltration. Targeted systemic evaluation is warranted in high‐risk presentations.