Tear-fluid-derived biomarkers of ocular complications in diabetes: a systematic review and meta-analysis

Background: Early identification and management of sight-threatening ocular complications of diabetes using imaging or molecular biomarkers could help prevent vision loss. However, access to specialized infrastructure and expertise is limited, especially in remote areas of the world. Tear-fluid may offer an easier, non-invasive, and localized screenshot of ocular disease. To the best of our knowledge, there is no systematic review and meta-analysis on tear-fluid-based biomarkers for ocular complications in diabetes. Methods: Articles were extracted from PubMed, Embase, Medline, and Web of Science using the MeSH and Emtree terms. The keywords include (diabetes), (diabetic retinopathy), (diabetes mellitus, type 1), (diabetes mellitus, type 2), (insulin-dependent diabetes), (insulin resistant diabetes), (tears), (lacrimal fluid), (biological marker), and (biomarker, marker). Concentrations of tear-fluid biomarkers in individuals with diabetes, diabetic ocular complications, and healthy controls were extracted and standardized mean differences (SMDs) and 95% CIs were calculated. Heterogeneity was assessed using subgroup and leave-one-out sensitivity analyses. Publication and risk of bias were performed using the Egger's test and Cochrane guidelines. The quality of evidence was evaluated using the Newcastle-Ottawa scale. Results: Nine hundred eleven papers were identified, 19 of which met the study criteria and were included in the meta-analysis. Participants (n = 1413) belonged to three groups: healthy controls (Controls), diabetes without any complications (Diabetes), and diabetes with ocular complications (Complications). Actual concentrations were reported for TNF-α, VEGF, IL-1RA, IL-1β, IL-6, IL-8, lactoferrin, lysozyme, and MCP-1 in at least three different studies. Meta-analyses demonstrated that TNF-α concentration was significantly higher in the tear-fluid of Complications group when compared to Controls (SMD = − 1.08, 95% CIs = − 1.78, − 0.38, p = 0.003) or when compared to Diabetes (SMD = − 0.78, 95% CIs = − 1.48, − 0.09, p = 0.03). However, it was not different when Controls were compared to Diabetes (SMD = − 1.00, 95% CIs = − 2.27, 0.28, p = 0.13). VEGF demonstrated a similar trend indicating specificity of tear-fluid TNF-α and VEGF for diabetic ocular complications. Conclusions: Across all biomolecules meta-analyzed in this study, TNF-α and VEGF were identified as the most important biomarkers that could potentially offer a non-invasive tear-fluid-based assessment of progression to ocular complications in diabetes, especially in rural and remote areas where diabetes-related expertise and infrastructure are limited. Trial registration: PROSPERO (CRD42023441867) https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=441867.