Prospective Cytokine Profiling in Aqueous Humor Reveals a Proinflammatory Microenvironment in Highly Dense Nuclear Cataracts

Abstract

Purpose: To investigate if the cytokine profile in the aqueous humor (AH) of cataract patients varies according to cataract type and severity.

Methods: This prospective study included 397 eyes of 397 patients (median age, 76 years; range, 30-94 years) who underwent standard small-incision phacoemulsification surgery. Cataracts were graded using the LOCS III system: mild (≤3), moderate (3.5-5), and severe (≥5). Biometric measurements from the IOL master 700 (Zeiss, Oberkochen, Germany) were used to differentiate between thick (>4.5 mm) and thin lenses. Information about age, gender, and self-reported diseases was obtained from patient records. Eleven different proteins were measured in AH using a multiplex cytokine assay (AYOXXA Biosystems, Cologne, Germany), including IL- 6, IL-8, angiopoietin 2, C-reactive protein (CRP), vascular endothelial growth factor A (VEGF-A), platelet-derived growth factor BB, placental growth factor, CXCL12, CXCL13, and CXCL10. Statistical analysis was performed using R and included nonparametrical testing, linear regression, and k-means clustering.

Results: Higher nuclear LOCS grades correlated with increased levels of CCL2 (360 vs. 387 vs. 517 pg/mL, P < 0.001), VEGF-A (270 vs. 292 vs. 390 pg/mL, P = 0.012), IL-8 (3.1 vs. 4.2 vs. 5.7 pg/mL, P = 0.018), and CXCL10 (52 vs. 61 vs. 90 pg/mL, P = 0.003). No associations were observed for cortical and subcapsular cataracts. Thicker lenses were associated with significantly increased levels of CRP, CXCL10, CXCL12, IL-6, IL-8, and VEGF-A.

Conclusions: The cytokine profile of AH varies based on cataract grading and lens thickness. In highly dense nuclear cataracts, CCL2, VEGF-A, IL-8, and CXCL10 were elevated.

Figure 1.

Association of cytokines to LOCS III grading. (A) Histograms displaying the distribution of the LOCS III grading categories in our cohort. (B) Boxplot showing the significantly dysregulated cytokines and their respective levels for the LOCS III category nuclear opalescence. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant. (C) Correlation matrix displaying the level of positive (shades of blue) or negative (shades of red) correlations for the indicated parameters. Size of the circle indicates higher levels of correlation.

Figure 2.

Association of aqueous humor cytokine levels to baseline characteristics. (A) Boxplot showing the significantly dysregulated cytokines and their respective levels stratified by sex. (B) Boxplot showing the significantly dysregulated cytokines and their respective levels stratified by self-reported T2D. (C) Boxplot showing the significantly dysregulated cytokines and their respective levels stratified by the presence of arterial hypertension. (D) Boxplot showing the significantly dysregulated cytokines and their respective levels stratified by the presence of glaucoma. (E) Boxplot showing the significantly dysregulated cytokines and their respective levels stratified by lens thickness with a cutoff of 4.5 mm. (F) Boxplot showing the significantly dysregulated cytokines and their respective levels stratified by cardiovascular risk factors. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant.

Figure 3.

K-means cluster analysis. (A) Visual representation of k-means cluster analysis dividing the cohort into two cohorts. Each dot represents one individual, and lines connect dots to the cluster centroid. (B) Characteristics of cohort stratified by clusters as identified by k-means clustering. (C) Boxplots of cytokines stratified by clusters. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant.