**Background:** Proliferative diabetic retinopathy (PDR) is an advanced stage of diabetic retinopathy characterized by retinal neovascularization, vitreous hemorrhage, and tractional retinal detachment. While numerous systemic factors such as hypertension, obesity, and glycemic control are known to influence diabetic retinopathy development and progression, their relationship with intraocular inflammatory and angiogenic cytokines in PDR remains poorly understood. This study aimed to examine correlations between systemic factors and vitreous fluid (VF) cytokines and analyze their contributions to PDR pathology.
**Methods:** This retrospective observational study included 26 PDR eyes from 25 patients with type 2 diabetes who underwent pars plana vitrectomy for vitreous hemorrhage and/or tractional retinal detachment at National Defense Medical College between January 2014 and August 2021. The control group comprised 30 eyes (14 idiopathic macular hole, 16 idiopathic epiretinal membrane). Exclusion criteria included history of intraocular inflammatory diseases, previous vitrectomy, ocular trauma, prior intravitreal therapies, and cataract surgery within 6 months. Systemic factors collected at admission included age, body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse pressure difference (PPD), heart rate (HR), PT-INR, APTT, random blood glucose (RBG), HbA1c, and estimated glomerular filtration rate (eGFR). Undiluted VF samples (approximately 0.5 mL) were collected at the beginning of vitrectomy and analyzed using a multiplex bead platform (Bio-Plex Human Cytokine 27-plex panel) for 27 cytokines. Statistical analyses included Mann-Whitney U test, Spearman's rank correlation, hierarchical cluster analysis, and principal component analysis (PCA).
**Key Results:** The PDR group was significantly younger than controls (58.6 ± 13.5 vs. 70.7 ± 7.94 years, p=0.001). VF levels of 16 cytokines were significantly higher in PDR than controls, including IL-6 (133.2 ± 159.1 vs. 8.53 ± 19.4 pg/mL, p=0.001), IP-10 (4568.3 ± 4746.5 vs. 979.0 ± 810.6 pg/mL, p=8.70×10⁻⁶), MCP-1 (717.4 ± 460.1 vs. 209.0 ± 86.1 pg/mL, p=7.10×10⁻⁷), and VEGF-A (870.0 ± 1680.8 vs. 0.16 ± 0.89 pg/mL, p=1.55×10⁻⁶). In PDR patients, Spearman correlation revealed positive associations between BMI and VF IL-6 (p=0.043) and IP-10 (p=0.025); between SBP and IL-6 (p=0.021), IL-7 (p=0.022), IL-13 (p=0.031), eotaxin (p=0.039), and IP-10 (p=0.002); and between DBP and IL-6 (p=0.024), eotaxin (p=0.011), and IP-10 (p=0.035). No significant correlations between systemic factors and VF cytokines were found in controls. Hierarchical cluster analysis showed that in controls, MCP-1 and VEGF-A clustered independently with different systemic factors, but in PDR, these two cytokines clustered together, losing property similarity with systemic factors. PCA demonstrated that in PDR, the cumulative contribution rate of VF cytokines dominating PC1 and PC2 was 42.3%, while systemic factors mainly constituting PC3 contributed only 10.4% out of 52.7% total variance. In PC1, VEGF-A and IL-6 had the highest loadings (0.86 and 0.83), along with DBP (0.81) and SBP (0.70), suggesting that younger age is associated with higher VF VEGF-A and IL-6 accompanied by blood pressure elevation.
**Clinical Implications:** This study demonstrates that systemic factors, particularly blood pressure and BMI, correlate with intraocular inflammatory cytokine levels in PDR, but their overall contribution to the intraocular cytokine milieu is modest compared to local factors. The finding that VEGF-A did not correlate with systemic factors or IL-6 suggests that systemic control alone may be insufficient to reduce intraocular VEGF-A levels, supporting the need for intravitreal anti-VEGF therapy. The results provide a rationale for combined systemic management (blood pressure and BMI control) and topical ophthalmic treatments to optimize PDR management. The study's limitations include its retrospective design, small sample size, lack of serum cytokine measurements, and absence of data on potential confounders such as disease duration, hyperlipidemia, diet, exercise, and smoking. Future prospective studies with larger cohorts and comprehensive systemic and intraocular factor analysis are warranted.