Anti–vascular endothelial growth factor therapy reduces epiretinal macrophage-like cell density in BRVO-ME patients, correlating with decreased macular edema. Early responders show a significant reduction in epiretinal macrophage-like cell density, while limited early responders do not. Epiretinal macrophage-like cell may serve as biomarkers for inflammation and treatment response, aiding personalized therapy.
Key words: biomarker, epiretinal macrophage-like cells, anti-VEGF therapy, retinal vein occlusion, macular edema
Abstract
Purpose:
To investigate the impact of anti–vascular endothelial growth factor therapy on epiretinal macrophage-like cells (eMLC) in branch retinal vein occlusion with macular edema (BRVO-ME) patients.
Methods:
This retrospective study included 58 BRVO-ME patients treated with anti–vascular endothelial growth factor agents, divided into 2 groups: 28 eyes (1–2 Injection group) and 30 eyes (3 Injection group). The 3-injection group comprised 20 early responders and 10 limited early responders (LERs). Longitudinal changes in macular eMLC count and density were assessed through en face optical coherence tomography during the early response phase (baseline to 3 months postinitial injection). Linear regression and Spearman correlation analyzed eMLC density's relationship with optical coherence tomography angiography parameters and systemic inflammatory indices.
Results:
The 3-injection group exhibited greater reductions in macular edema and eMLC density than the 1 to 2 injection group. Macular eMLC density significantly decreased in early responders, whereas LERs showed no significant alteration. Posttreatment central macular thickness change positively correlated with eMLC density change (r = 0.57). Baseline systemic inflammatory indices (neutrophil/lymphocyte ratio, r = −0.46; neutrophil/platelet ratio, r = −0.36; systemic immune-inflammation index, r = −0.33; systemic inflammatory response index, r = −0.31) negatively correlated with posttreatment eMLC density change.
Conclusion:
Anti–vascular endothelial growth factor therapy reduces macular eMLC in BRVO-ME patients, especially in early responders receiving three injections. Macular eMLCs are closely associated with inflammation and may serve as a valuable treatment response biomarker.
Retinal vein occlusion (RVO) is characterized by the obstruction of the retinal venous system, encompassing branch retinal vein occlusion and central retinal vein occlusion.1 Macular edema constitutes the primary cause of vision loss in patients afflicted with RVO, exerting a substantial impact on visual function and quality of life. The present therapeutic paradigm for macular edema secondary to RVO is intravitreal injection of anti–vascular endothelial growth factor (anti-VEGF) drugs.2 However, there is considerable interindividual variability in the response to anti-VEGF therapy, with some patients showing little improvement in macular edema despite multiple injections.3 Evidence suggests that inflammation plays a key role in macular edema development in RVO,4 but identifying inflammation-driven cases remains challenging. Thus, reliable biomarkers of retinal inflammation offer valuable assistance in the diagnosis, treatment, and prognosis of RVO.
Epiretinal macrophage-like cells (eMLCs) have emerged as a promising biomarker for inflammation in retinal vascular disease.5–8 Epiretinal macrophage-like cells represent a heterogeneous group of macrophage-like cells, including vitreous cells, microglia, perivascular macrophages, and monocyte-derived macrophages.9–11 These cells play a crucial role in the pathophysiology of macular edema,5,7,8,12 with their quantity and density potentially correlating with the severity of retinal inflammation. Optical coherence tomography angiography (OCTA), a noninvasive imaging technique, allows En face OCT to visualize eMLC near the inner limiting membrane,13 providing a valuable tool for evaluating eMLC in retinal diseases.
This study aims to examine the longitudinal impact of anti-VEGF treatment on macular eMLC density, explore its correlation with macular edema prognosis, and evaluate eMLC's potential clinical value in assessing treatment response in BRVO-ME.
Materials and Methods
Study Participants
This retrospective longitudinal study was conducted on 58 patients with BRVO-ME, diagnosed, and treated with intravitreal anti-VEGF agent (ranibizumab, aflibercept, or conbercept) injections at Nanjing Medical University Eye Hospital between 2020 and 2024. Patients were divided into 2 groups: 30 eyes in the 3-injection group (receiving 3 injections during the loading phase and 28 eyes in the 1–2-injection group) (receiving one or two injections during the loading phase). Patients included in the study, all with a primary diagnosis of BRVO-ME, underwent comprehensive ophthalmic examinations and serial OCTA imaging. This imaging was conducted using identical equipment and parameters from baseline up to 3 months after the initial injection, with assessments performed after each injection. Patients with the following conditions were excluded: 1) age 18 years or younger; 2) refractive error ≥3D; 3) other inflammatory and vascular diseases; 4) ocular trauma history; 5) history of anti-VEGF, laser, intraocular surgery, or steroid treatment; 6) patients with epiretinal membrane, vitreous or preretinal hemorrhage; and 7) poor-quality OCTA images (scan quality ≤ 6/10). Patients were also classified as early responders (ERs) or limited early responders (LERs). Limited early responders were defined as patients with either a <10% reduction in central macular thickness (CMT) or <5 Early Treatment Diabetic Retinopathy Study (ETDRS) letter best-corrected visual acuity gain (or both) at 3 months after anti-VEGF initiation.3 This study was approved by the Ethics Committee of the Eye Hospital of Nanjing Medical University, and written informed consent was obtained from all participants.
Optical Coherence Tomography Angiography and Macular Epiretinal Macrophage-Like Cells Quantification
The OCTA system (RTVue XR Avanti, Optovue, Inc., Fremont, CA, USA) was used for macular examination, and optical coherence tomography angiography and en face OCT images were automatically generated. To minimize motion artifacts and speckle noise, 10 repeated scans of each 6-mm × 6-mm area were acquired and averaged using the built-in software. This software then calculated structural parameters around the macular region, including CMT, mean retinal thickness (MRT), vascular density of the superficial (SCP-VD) and deep (DCP-VD) capillary plexuses, and foveal avascular zone (FAZ) area and perimeter. For visualization of eMLC, En face OCT images in a 6-mm × 6-mm scan mode were used, focusing on the retinal layer 3 µm above the internal limiting membrane within the macular region. To ensure segmentation quality, all automatically segmented OCT B-scan images underwent rigorous manual review and necessary manual correction. Two certified professional graders assisted in the preliminary screening and marking of images where automatic segmentation lines deviated from the actual internal limiting membrane anatomical boundary. They then dragged, added, or deleted nodes to precisely match the retinal layer boundaries. The corrected and uncorrected results were reviewed by a team of two experienced retinal specialists (shown in Figure 1). As described in previous studies, macular eMLC quantification was performed using ImageJ software, following established protocols,14 for noise reduction, signal enhancement, and binarization. The number of eMLC was counted using the Analyze Particles function, and eMLC density was calculated as the number of cells per 36 mm2. All evaluations were performed by a single blinded evaluator, unaware of the disease severity.
Fig. 1.
Representative images illustrating the rigorous manual correction process for OCT B-scan segmentation at baseline (A–E) and posttreatment (F–J). OCT angiograms (A and F). Retinal thickness maps (B and G). Uncorrected central macular OCT cross-sections demonstrating initial internal limiting membrane (ILM) segmentation lines, with segmentation errors indicated by white arrows (C and H). Corrected central macular OCT cross-sections after manual adjustment of ILM segmentation lines (D and I). Corrected 3-mm en face OCT slabs generated from the manually refined segmentation (E and J), allowing for accurate visualization of macular eMLC.
Systemic Inflammatory Indices Calculation
BRVO-ME patients underwent hematologic evaluation before intravitreal anti-VEGF injections. White blood cells, neutrophils (NEUT), lymphocytes (LY), monocytes (MONO), platelets (PLT), and C-reactive protein (CRP) levels were recorded. The following inflammation-related indices were calculated: 1) systemic inflammatory response index (SIRI): MONO × NEUT/LY; 2) systemic immune-inflammation index (SII): PLT × NEUT/LY; 3) pan-immunoinflammatory value (PIV): NEUT × MONO × PLT/LY; 4) neutrophil/Lymphocyte ratio (NLR); 5) platelet/lymphocyte ratio (PLR); 6) lymphocyte/monocyte ratio (LMR); 7) neutrophil/platelet ratio (NPR); and 8) C-reactive protein/lymphocyte ratio (CLR).
Statistical Analysis
Data analysis was performed using R software. For two-group comparisons, normally distributed continuous variables were analyzed with the independent samples t-test and non-normally distributed variables with the Wilcoxon rank-sum test. Repeated measures were analyzed using repeated measures analysis of variance for normally distributed variables, and the Friedman test for non-normally distributed ones, with post hoc pairwise comparisons using the Wilcoxon signed-rank test with Bonferroni correction. In the univariate regression analysis, linear regression was performed for each independent variable to assess its relationship with macular eMLC density changes. Variables with a P-value less than 0.05 in the univariate analysis were included in a multivariate linear regression model. Collinearity was checked using the variance inflation factor, and model fit was assessed with adjusted R-squared. Spearman correlation coefficients evaluated relationships between macular eMLC density and other variables, including OCTA macular structural parameters and systemic inflammatory indices. Data are presented as mean ± SD, and P-value < 0.05 was considered significant.
Results
A total of 58 patients were included in this study: 30 eyes (30 patients) in the 3-injection group and 28 eyes (28 patients) in the 1- to 2-injection group. In the 1- to 2-injectio group, 5 patients received one anti-VEGF injection during the loading phase, and 23 patients received two. Baseline demographic, clinical, and retinal structural parameters are summarized in Table 1. The baseline characteristics were similar between the two groups (P > 0.05), with the only notable difference being the laterality of the affected eye (P = 0.036). Systemic inflammatory indices were calculated, revealing similar initial inflammatory conditions (P > 0.05). Macular parameters, including macular edema thickness and macular eMLC density, also showed no significant baseline differences (P > 0.05).
Table 1.
Comparing Baseline and Postinjection Characteristics Across 3-Injection Group and 1- to 2-Injection Group
| 3 Injection Group | 1–2 Injection Group | P | |
| Baseline | |||
| Sample size | 30 | 28 | N/A |
| Age, year | 59.13 ± 8.65 | 58.43 ± 9.82 | 0.773* |
| Sex (M/F) | 12/18 | 15/13 | 0.301† |
| Laterality (R/L) | 11/19 | 18/10 | 0.036† |
| DM (No/Yes) | 29/1 | 27/1 | 0.960† |
| HP (No/Yes) | 15/15 | 18/10 | 0.272† |
| BCVA (logMAR) | 0.77 ± 0.50 | 0.76 ± 0.43 | 0.821‡ |
| SCP-VD | 44.58 ± 5.12 | 43.59 ± 4.72 | 0.447* |
| DCP-VD | 47.25 ± 5.43 | 46.48 ± 4.84 | 0.570* |
| FAZ area (mm2) | 0.30 ± 0.11 | 0.34 ± 0.18 | 0.469‡ |
| PERIM (mm) | 2.44 ± 0.53 | 2.47 ± 0.75 | 0.883‡ |
| CMT (μm) | 481.23 ± 81.70 | 489.32 ± 125.07 | 0.809‡ |
| MRT (μm) | 387.63 ± 54.82 | 388.57 ± 74.59 | 0.518‡ |
| CRP | 1.48 ± 1.17 | 1.81 ± 1.96 | 0.927‡ |
| SIRI | 1.37 ± 0.75 | 1.27 ± 0.93 | 0.326‡ |
| SII | 684.52 ± 403.18 | 545.52 ± 314.14 | 0.151‡ |
| PIV | 328.01 ± 215.67 | 286.76 ± 231.82 | 0.236‡ |
| NLR | 2.83 ± 1.21 | 2.54 ± 1.61 | 0.088‡ |
| PLR | 132.73 ± 61.36 | 109.05 ± 41.55 | 0.225‡ |
| LMR | 4.82 ± 2.54 | 5.19 ± 2.56 | 0.531‡ |
| NPR | 0.03 ± 0.01 | 0.03 ± 0.02 | 0.932‡ |
| CLR | 0.72 ± 0.49 | 0.93 ± 1.20 | 0.331‡ |
| Macular eMLC count | 254.33 ± 132.00 | 259.21 ± 154.31 | 0.876‡ |
| Macular eMLC density (cells/mm2) | 7.06 ± 3.67 | 7.20 ± 4.29 | 0.876‡ |
| Postinjection | |||
| BCVA (logMAR) | 0.41 ± 0.34 | 0.60 ± 0.43 | 0.038‡ |
| SCP-VD | 44.57 ± 5.28 | 43.57 ± 4.87 | 0.457* |
| DCP-VD | 44.60 ± 4.92 | 44.48 ± 5.40 | 0.932* |
| FAZ area (mm2) | 0.42 ± 0.18 | 0.45 ± 0.21 | 0.608‡ |
| PERIM (mm) | 2.84 ± 0.90 | 2.76 ± 0.69 | 0.931‡ |
| CMT (μm) | 321.5 ± 115.85 | 373.32 ± 130.11 | 0.037‡ |
| MRT (μm) | 320.83 ± 39.95 | 341.86 ± 54.37 | 0.210‡ |
| Macular eMLC count | 132.63 ± 73.76 | 216.79 ± 103.05 | <0.001‡ |
| Macular eMLC density (cells/mm2) | 3.68 ± 2.05 | 6.02 ± 2.86 | <0.001‡ |
Two-sample t-test.
Chi-square test.
Wilcoxon rank-sum test.
BCVA, best-corrected visual acuity; CLR, C-reactive protein/lymphocyte ratio; DM, diabetes mellitus; DCP, deep capillary plexus; F, female; HP, hypertension; LMR, lymphocyte/monocyte ratio; L, left; M, male; N/A, not applicable; PERIM, perimeter circumference of the FAZ; PIV, pan-immunoinflammatory value; R, right; SCP, superficial capillary plexus.
A comprehensive eye examination was performed three months after the initial injection. Significant posttreatment changes were observed in best-corrected visual acuity, CMT, MRT, and macular eMLC density. The 3-injection group demonstrated superior improvements in visual acuity and macular edema compared with the 1- to 2-injection group, with a significantly greater reduction in macular eMLC density (Table 1).
Linear regression analysis was conducted to assess the impact of anti-VEGF injections on macular eMLC density changes, with details in Table 2. First, a univariate analysis was performed, with data on the number of anti-VEGF injections, baseline values, posttreatment values, and the changes between baseline and posttreatment values included. Univariate analysis revealed that the number of anti-VEGF injections and (P = 0.027), posttreatment CMT (P < 0.001), posttreatment MRT (P < 0.001), changes in CMT (ΔCMT, P < 0.001), and MRT (ΔMRT, P = 0.017) were significantly correlated with changes in macular eMLC density. Next, a multivariate regression model was constructed, including the significant variables from the univariate analysis. The variance inflation factor values were all less than five, indicating that there was no severe multicollinearity. This model showed a significant association between the ΔCMT and macular eMLC density changes (P = 0.017) after adjusting for confounders, with R2 = 0.44. Furthermore, Spearman correlation analysis revealed a positive correlation between changes in macular eMLC density and changes in macular edema thickness (e.g., ΔCMT (r = 0.57, P < 0.001) and ΔMRT (r = 0.31, P = 0.017)), as illustrated in Figure 2.
Table 2.
Univariate and Multivariate Linear Regression Analysis of OCTA Factors Associated With Changes in Macular eMLCs Density Following Anti-VEGF Therapy
| Variable | Univariable Analysis | Multivariable Analysis | |||
| Beta (95% CI) | P | Beta (95% CI) | P | VIF | |
| Age | −0.12 (−2.74 to 2.51) | 0.929 | — | — | |
| No. injections 3 vs. 1–2 | 51.63 (5.94 to 97.32) | 0.027* | 27.6 (−9.03 to 64.20) | 0.137 | 1.065 |
| Baseline | |||||
| SCP-VD | −1.05 (−5.93 to 3.83) | 0.669 | — | — | |
| DCP-VD | −1.39 (−6.08 to 3.29) | 0.554 | — | — | |
| FAZ (mm2) | 3.33 (−154.76 to 161.42) | 0.966 | — | — | |
| PERIM (mm) | 10.19 (−27.47 to 47.87) | 0.589 | — | — | |
| CMT (μm) | −0.03 (−0.26 to 0.20) | 0.810 | — | — | |
| MRT (μm) | −0.19 (−0.56 to 0.17) | 0.299 | — | — | |
| Posttreatment | |||||
| SCP-VD | −1.95 (−6.67 to 2.77) | 0.411 | — | — | |
| DCP-VD | −1.98 (−6.65 to 2.70) | 0.401 | — | — | |
| FAZ (mm2) | 3.33 (−154.76 to 161.42) | 0.966 | — | — | |
| PERIM (mm) | 11.26 (−18.57 to 41.09) | 0.452 | — | — | |
| CMT (μm) | −0.45 (−0.60 to −0.30) | *** | −0.12 (−0.40 to 0.17) | 0.421 | 3.923 |
| MRT (μm) | −1.07 (−1.48 to −0.66) | *** | −0.47 (−1.09 to 0.14) | 0.129 | 2.760 |
| Δ change | |||||
| ΔSCP-VD | 1.96 (−4.80 to 8.72) | 0.563 | — | — | |
| ΔDCP-VD | 0.73 (−4.58 to 6.05) | 0.783 | — | — | |
| ΔFAZ (mm2) | 0.76 (−181.27 to 182.78) | 0.993 | — | — | |
| ΔPERIM (mm) | −7.24 (−43.86 to 29.38) | 0.693 | — | — | |
| ΔCMT (μm) | 0.40 (0.24 to −0.55) | *** | 0.31 (0.06 to 0.57) | 0.017* | 3.393 |
| ΔMRT (μm) | 0.47 (0.09 to 0.86) | 0.017* | −0.28 (−0.72 to 0.17) | 0.222 | 2.232 |
*P < 0.05, **P < 0.01, and ***P < 0.001.
Δ, Change Between Baseline and Posttreatment; BCVA, best-corrected visual acuity; DCP, deep capillary plexus; PERIM, perimeter circumference of the FAZ; SCP, superficial capillary plexus; VIF, variance inflation factor.
Fig. 2.
Correlation between macular eMLC density and OCTA macular structural parameters. The left panel shows heatmaps illustrating the correlation between these parameters. The right panel displays bar graphs showing the correlation coefficients (ranked from highest to lowest) with macular eMLC density. Δ indicates the change between baseline and posttreatment measurements.
Patients in the 3-injection group were further stratified into ERs and LERs based on their treatment response. Comparing baseline data between these two subgroups (Table 3), we found that, aside from differences in the treated eye, LERs had significantly higher NLR score (P = 0.011). Table 4 presents a longitudinal assessment of visual acuity, macular structural parameters, and eMLC parameters, documenting changes from baseline through successive anti-VEGF injections in both ERs and LERs. Compared with baseline, macular eMLC count and density progressively decreased after each injection, particularly in ERs (Figure 3). After treatment, ERs demonstrated decreased CMT and MRT and increased visual acuity, while LERs showed no significant improvement. Both ERs and LERs exhibited decreased DCP-VD and increased FAZ area (Table 4). Early responders showed a pronounced reduction, reaching 72%, 60%, and 45% of baseline values after the first, second, and third injections, respectively (Table 4), with representative images in Figure 4 (left panel). Conversely, LERs showed no significant change in macular eMLC count and density between injections (Table 4), with representative images in Figure 4 (right panel).
Table 3.
Baseline and Characteristics of the ER Group and LER Group
| ERs Group | LERs Group | P | |
| Sample size | 20 | 10 | N/A |
| Age, year | 61.25 ± 8.47 | 54.90 ± 7.75 | 0.057* |
| Sex (M/F) | 8/12 | 4/6 | 1.000† |
| Laterality (R/L) | 10/10 | 1/9 | 0.032† |
| DM (No/Yes) | 19/1 | 10/0 | 0.517† |
| HP (No/Yes) | 10/10 | 5/5 | 1.000† |
| BCVA (logMAR) | 0.85 ± 0.57 | 0.60 ± 0.21 | 0.376‡ |
| SCP-VD | 44.94 ± 4.33 | 43.86 ± 6.64 | 0.595* |
| DCP-VD | 47.18 ± 6.12 | 47.41 ± 3.98 | 0.913* |
| FAZ area (mm2) | 0.29 ± 0.11 | 0.32 ± 0.14 | 0.895‡ |
| PERIM (mm) | 2.47 ± 0.51 | 2.38 ± 0.58 | 0.588‡ |
| CMT (μm) | 398.75 ± 60.39 | 365.40 ± 34.07 | 0.355‡ |
| MRT (μm) | 387.63 ± 54.82 | 388.57 ± 74.59 | 0.243‡ |
| CRP | 1.39 ± 1.27 | 1.65 ± 0.97 | 0.307‡ |
| SIRI | 1.33 ± 0.84 | 1.44 ± 0.55 | 0.530‡ |
| SII | 664.70 ± 456.46 | 724.19 ± 285.13 | 0.214‡ |
| PIV | 337.09 ± 246.25 | 309.85 ± 145.99 | 0.713‡ |
| NLR | 2.57 ± 1.28 | 3.36 ± 0.87 | 0.011‡ |
| PLR | 136.25 ± 66.66 | 125.69 ± 51.67 | 0.397‡ |
| LMR | 4.77 ± 2.48 | 4.91 ± 2.81 | 0.983‡ |
| NPR | 0.02 ± 0.02 | 0.03 ± 0.02 | 0.131‡ |
| CLR | 0.64 ± 0.48 | 0.90 ± 0.51 | 0.129‡ |
| Macular eMLC count | 212.65 ± 108.18 | 201.20 ± 45.67 | 0.355‡ |
| Macular eMLC density (cells/mm2) | 5.91 ± 3.01 | 5.59 ± 1.27 | 0.355‡ |
Two-sample t-test.
Chi-square test.
Wilcoxon rank-sum test.
BCVA, best-corrected visual acuity; CRP, C-reactive protein; CLR, C-reactive protein/lymphocyte ratio; DM, diabetes mellitus; F, female; HP, hypertension; L, left; LMR, lymphocyte/monocyte ratio; M, male; N/A, not applicable; NPR, neutrophil/platelet ratio; PERIM, perimeter circumference of the FAZ; PIV, pan-immunoinflammatory value; PLR, platelet/lymphocyte ratio; R, right; VD, vessel density.
Table 4.
Longitudinal Assessment of Visual Acuity, Macular Structure, and eMLC Parameters in ERs and LERs
| Baseline | Post 1st Injection | Post 2nd Injection | Post 3rd Injection | P | |
| ERs (n = 20) | |||||
| BCVA (logMAR) | 0.85 ± 0.57 | 0.68 ± 0.45 | 0.46 ± 0.39 | 0.35 ± 0.37 | <0.001* |
| CMT (μm) | 488.50 ± 93.02 | 344.40 ± 74.03 | 298.75 ± 63.59 | 250.55 ± 44.89 | <0.001* |
| MRT (μm) | 389.75 ± 60.39 | 334.60 ± 42.87 | 316.40 ± 36.67 | 301.00 ± 23.78 | <0.001* |
| SCP-VD | 44.94 ± 4.33 | 44.66 ± 4.18 | 44.82 ± 4.25 | 45.19 ± 4.04 | 0.870† |
| DCP-VD | 47.18 ± 6.12 | 45.19 ± 4.76 | 44.21 ± 4.79 | 45.13 ± 5.28 | 0.033† |
| FAZ area (mm2) | 0.29 ± 0.10 | 0.35 ± 0.11 | 0.37 ± 0.11 | 0.39 ± 0.11 | <0.001* |
| PERIM (mm) | 2.47 ± 0.51 | 2.59 ± 0.57 | 2.65 ± 0.69 | 2.81 ± 0.73 | 0.247* |
| eMLC count | 212.65 ± 108.18 | 154.15 ± 50.72 | 127.00 ± 45.29 | 95.90 ± 45.49 | <0.001* |
| eMLC density (cells/mm2) | 5.91 ± 3.01 | 4.28 ± 1.41 | 3.53 ± 1.26 | 2.66 ± 1.26 | <0.001* |
| LERs (n = 10) | |||||
| BCVA (logMAR) | 0.60 ± 0.21 | 0.60 ± 0.20 | 0.57 ± 0.20 | 0.53 ± 0.23 | 0.150* |
| CMT (μm) | 466.70 ± 53.74 | 448.40 ± 26.10 | 470.80 ± 57.71 | 463.40 ± 73.68 | 0.160* |
| MRT (μm) | 365.40 ± 34.07 | 351.50 ± 39.40 | 360.40 ± 31.59 | 360.50 ± 36.43 | 0.106* |
| SCP-VD | 43.86 ± 6.64 | 42.86 ± 5.62 | 42.31 ± 5.46 | 43.34 ± 7.26 | 0.473† |
| DCP-VD | 47.41 ± 3.97 | 47.16 ± 3.71 | 45.63 ± 3.74 | 43.53 ± 4.14 | 0.001† |
| FAZ area (mm2) | 0.32 ± 0.14 | 0.41 ± 0.19 | 0.47 ± 0.26 | 0.48 ± 0.27 | 0.017* |
| PERIM (mm) | 2.38 ± 0.58 | 2.80 ± 1.25 | 2.92 ± 1.16 | 2.92 ± 1.23 | 0.247* |
| eMLC count | 201.20 ± 45.67 | 207.2 ± 42.80 | 191.90 ± 64.13 | 206.10 ± 64.54 | 0.441* |
| eMLC density (cells/mm2) | 5.59 ± 1.27 | 5.76 ± 1.19 | 5.33 ± 1.78 | 5.73 ± 1.79 | 0.441* |
Friedman test.
Repeated measures ANOVA
ANOVA, analysis of variance; BCVA, best-corrected visual acuity; N/A, not applicable; PERIM, perimeter circumference of the FAZ; SCP, superficial capillary plexus; VD, vessel density.
Fig. 3.
Paired scatter bar plot showing longitudinal changes in macular eMLC density at baseline and after the first, second, and third anti-VEGF injections in the 3-injection group. LERs are represented by the red line, and ERs by the gray line.
Fig. 4.
Representative macular eMLC images in ERs (A–L). Or LERs (M–X). Across successive three injections. The columns show OCT angiograms (A, D, G, J, M, P, S, and V), central macular OCT cross-sections (A1, D1, G1, J1, M1, P1, S1, and V1), retinal thickness maps (B, E, H, K, N, Q, T, and W), 3-mm en face OCT slabs (C, F, I, L, O, R, U, and X), and extracted eMLC from the 3-mm en face OCT slabs (C1, F1, I1, L1, O1, R1, U1, and X1) in the 6-mm × 6-mm macular region.
Systemic inflammation is a key factor in RVO development,15 necessitating the exploration of its correlation with macular eMLC. Spearman correlation analysis showed a negative association between changes in macular eMLC density and several baseline systemic inflammatory indices: NLR (r = −0.46, P < 0.001), NPR (r = −0.36, P = 0.006), SII (r = −0.33, P = 0.013), and SIRI (r = −0.31, P = 0.019) (Figure 5).
Fig. 5.
Correlation between macular eMLC density and systemic inflammatory indices. The left panel shows heatmaps illustrating the correlation between these parameters. The right panel displays bar graphs showing the correlation coefficients (ranked from highest to lowest) with macular eMLC density. Δ indicates the change between baseline and posttreatment measurements.
Discussion
Epiretinal macrophage-like cells, as potential inflammatory biomarkers,5–8 play a crucial role in retinal vascular diseases. Research indicates that macular eMLC density increases and their morphology alters in RVO-ME patients,12,14,16 suggesting their involvement in the pathogenesis and progression of the condition. Given that anti-VEGF therapy is the primary treatment for BRVO-ME, understanding its effects on eMLC is crucial for optimizing treatment strategies.
Previous studies have largely relied on cross-sectional observations. Our study provides a longitudinal assessment within the same patient cohort, effectively addressing the issue of considerable interindividual variability in eMLC density.17 The results of this study are highly consistent with existing literature, further confirming the role of eMLCs in BRVO-ME. Previous cross-sectional studies have indicated that macular eMLC density is increased and their morphology altered in BRVO-ME patients, and this correlates with the severity of macular edema.12,14,16 In line with these findings, our study also observed a positive correlation between the change in macular eMLC density and the change in macular edema thickness (ΔCMT and ΔMRT). This consistency with previous research, which reported a positive correlation between macular eMLC density and macular thickness, suggests that eMLCs may be involved in the formation of macular edema.
The number and density of macular eMLCs significantly decreased after anti-VEGF treatment. This decline was particularly pronounced in patients who received more injections. Three-injection group also demonstrated superior outcomes for visual acuity improvement and macular edema resolution. On one hand, anti-VEGF drugs may directly affect eMLC function and activity. Although eMLCs are not the primary target of these drugs, studies have shown that retinal microglia and macrophages express the VEGF receptor Flt-1 under inflammatory conditions.18 As RVO is an inflammatory disease, its microenvironment may induce VEGF receptor expression on eMLC, making them susceptible to anti-VEGF treatment. On the other hand, the reduction of macular eMLC by anti-VEGF drugs may be linked to reduced inflammation. Anti-VEGF drugs can modulate inflammatory responses through several mechanisms,19 including inhibiting vascular permeability, reducing inflammatory cell infiltration, modulating retinal neurovascular unit, and regulating inflammatory mediators. Studies have confirmed that VEGF can activate, proliferate, and promote microglia migration, contributing to the feedback loop of retinal immune cell VEGF expression.20 Furthermore, the restoration of blood–retinal barrier integrity and the resolution of macular edema after anti-VEGF treatment inherently contribute to a reduction in the intraretinal inflammatory environment. The reduction in hyperreflective foci,19 a retinal inflammatory marker, after anti-VEGF treatment in BRVO-ME further supports its anti-inflammatory effect.
Foveal avascular zone area and DCP-VD abnormalities are commonly linked to macular ischemia. We observed decreased DCP-VD and increased FAZ area after continuous anti-VEGF therapy, a finding consistent with some studies in RVO.21–23 However, other research has not corroborated this.24–26 Although anti-VEGF effectively resolves fluid accumulation, its ability to induce reperfusion or reverse established retinal ischemia remains debated and often limited.27 If underlying ischemia persists or worsens, FAZ enlargement might continue even with edema resolution, suggesting ongoing retinal damage. Conversely, macular edema can physically distort the macula and obscures the true FAZ boundaries.28 When anti-VEGF treatment resolves the edema, the retina flattens, revealing the FAZ's true extent, including preexisting or progressive nonperfusion areas. Thus, FAZ enlargement is not always a consistent sign of persistent retinal ischemia.
This study highlights the close relationship between macular eMLC density and the response to anti-VEGF treatment. Other researchers have observed that RVO patients with higher macular eMLC density require more injections,16 suggesting that macular eMLCs could serve as an indicator of sensitivity to anti-VEGF therapy. Our study observed differential macular eMLC responses between ERs and LERs, supporting eMLCs as dynamic biomarkers for real-time treatment efficacy assessment. ERs exhibited a significant and progressive decrease in macular eMLC density, which correlated with their substantial improvements in visual acuity and anatomical parameters; conversely, LERs showed no significant change in macular eMLC. Longitudinal monitoring of macular eMLC density can enable clinicians to identify LERs early in the course of treatment. Patients who continue to exhibit persistent macular edema and high macular eMLC density after one or two anti-VEGF injections may be identified as potential LERs, possibly necessitating more aggressive or alternative treatment approaches. These approaches might include intensifying the anti-VEGF dosing regimen or an earlier transition to, or combination with, anti-inflammatory therapies such as intravitreal corticosteroids.29–31
Systemic inflammation plays a key role in RVO progression, potentially through mechanisms such as atherosclerosis and a hypercoagulable state.15 Elevated systemic inflammatory markers (such as SIRI, NLR, and CRP) are associated with various ocular inflammatory diseases and can reflect the body's overall inflammatory burden.32–34 While the impact of systemic inflammation on eMLCs is not fully understood, systemic inflammatory indices like NLR, PLR, SIRI and SII are linked to RVO progression, with NLR showing the highest risk.35–37 Our study also found that LERs had a higher NLR and showed worse improvement in both visual acuity and macular edema. This study provides evidence of a connection between systemic inflammatory status and the response of local retinal inflammatory cells to treatment. We found that elevated baseline systemic inflammatory levels, include NLR, NPR, SII, and SIRI, appear to hinder the expected reduction of local retinal eMLCs by anti-VEGF treatment. A high systemic inflammatory burden may create an unfavorable microenvironment, potentially limiting the anti-inflammatory efficacy of anti-VEGF drugs at the retinal level.
This study has several limitations. First, as a retrospective analysis, we could not examine factors such as blood lipid levels, posterior vitreous detachment, and refractive errors related to macular eMLC changes. Second, despite including 58 patients, the sample size remains moderate for subgroup analyses; a larger cohort would help enhance the statistical power and generalizability of the findings. Additionally, the study included different types of anti-VEGF drugs; although they all belong to the anti-VEGF class, their molecular characteristics or efficacy might subtly differ, and this study did not differentiate between them.
The results of this study directly translate into potential, actionable clinical tools. Monitoring macular eMLC density through en face OCT can provide information for identifying patients at risk of poor anti-VEGF treatment response. Anti-VEGF treatment effectively reduces eMLC density, and this reduction is closely related to the resolution of macular edema. LERs exhibited limited changes in eMLC density. Patients' baseline systemic inflammatory levels had a negative impact on the posttreatment change in macular eMLC density. In summary, this study provides important insights into the dynamic changes of eMLCs in BRVO-ME patients treated with anti-VEGF therapy.
Footnotes
Supported by grants from the National Natural Science Foundation of China (grant numbers 82271107 to J.Y.), Natural Science Foundation of Jiangsu Province (BK20221186 to J.Y.), Jiangsu Provincial Health Commission Medical Research Key Project (K2023060 to J.Y.), Postgraduate Research and Practice Innovation Program of Jiangsu Province (JX10414318 to J.Y.).
None of the authors have any conflicting interests to disclose.
W. Bai, Y. Zhao, Y. -Y. Zhang, F. Wen authors have contributed equally to this work.
Contributor Information
Wen Bai, Email: mubaiwen@163.com.
Yue Zhao, Email: 517709972@qq.com.
Yi-Yang Zhang, Email: zyynjmu@stu.njmu.edu.cn.
Jia-Yi Li, Email: lijiayi0719hf@163.com.
Shan-Shan Zhu, Email: zss@stu.njmu.edu.cn.
Cheng Sun, Email: suncheng22260105@163.com.
Feng Wen, Email: wenfeng208@foxmail.com.
References
- 1.Scott IU, Campochiaro PA, Newman NJ, Biousse V. Retinal vascular occlusions. Lancet 2020;396:1927–1940. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Shalchi Z, Mahroo O, Bunce C, Mitry D. Anti-vascular endothelial growth factor for macular oedema secondary to branch retinal vein occlusion. Cochrane Database Syst Rev 2020;7:CD009510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kailar RS, Kuo BL, Perkins SW, Singh RP. Long-term outcomes in early versus limited response to Anti-VEGF treatment for retinal vein occlusion. Ophthalmol Retina 2024;8:55–61. [DOI] [PubMed] [Google Scholar]
- 4.Daruich A, Matet A, Moulin A, et al. Mechanisms of macular edema: beyond the surface. Prog Retin Eye Res 2018;63:20–68. [DOI] [PubMed] [Google Scholar]
- 5.Ong JX, Nesper PL, Fawzi AA, et al. Macrophage-like cell density is increased in proliferative diabetic retinopathy characterized by optical coherence tomography angiography. Invest Ophthalmol Vis Sci 2021;62:2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Pichi F, Neri P, Aljeneibi S, et al. In vivo visualization of macrophage-like cells in patients with uveitis by use of en face swept source optical coherence tomography. Ocul Immunol Inflamm 2024;32:1532–1538. [DOI] [PubMed] [Google Scholar]
- 7.Wang Z, An H, Tang J, et al. Elevated number and density of macrophage-like cell as a novel inflammation biomarker in diabetic macular edema. Sci Rep 2023;13:5320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zeng Y, Wen F, Zhuang X, et al. Epiretinal macrophage-like cells on optical coherence tomography: potential inflammatory imaging biomarker of severity in diabetic retinopathy. Retina 2024;44:1314–1322. [DOI] [PubMed] [Google Scholar]
- 9.Hammer DX, Agrawal A, Villanueva R, et al. Label-free adaptive optics imaging of human retinal macrophage distribution and dynamics. Proc Natl Acad Sci U S A 2020;117:30661–30669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rajesh A, Droho S, Lavine JA. Macrophages in close proximity to the vitreoretinal interface are potential biomarkers of inflammation during retinal vascular disease. J Neuroinflammation 2022;19:203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Rui Y, Zhang M, Lee DMW, et al. Label-free imaging of inflammation at the level of single cells in the living human eye. Ophthalmol Sci 2024;4:100475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wang W, Sun G, He L, Chen C. Increased macrophage-like cell density in retinal vein occlusion as characterized by en face optical coherence tomography. J Clin Med 2022;11:5636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Castanos MV, Zhou DB, Linderman RE, et al. Imaging of macrophage-like cells in living human retina using clinical OCT. Invest Ophthalmol Vis Sci 2020;61:48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zeng Y, Zhang X, Mi L, et al. Characterization of Macrophage-Like Cells in Retinal Vein Occlusion Using En Face Optical Coherence Tomography. Front Immunol 2022;13:855466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kumar MJ, Jr., Kotak PS, Acharya S, et al. A comprehensive review of ocular manifestations in systemic diseases. Cureus 2024;16:e65693. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Rangwani SM, Hawn S, Sklar NC, et al. Macrophage-like cells are increased in retinal vein occlusion and correlate with more intravitreal injections and worse visual acuity outcomes. J Personalized Med 2022;13:45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wang W, Chen C, Yi Z, et al. Characteristics of macrophage-like cells in acute nonarteritic anterior ischemic optic neuropathy and the normal fellow eyes on en face optical coherence tomography. Front Immunol 2022;13:1095296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Couturier A, Bousquet E, Zhao M, et al. Anti-vascular endothelial growth factor acts on retinal microglia/macrophage activation in a rat model of ocular inflammation. Mol Vis 2014;20:908–920. [PMC free article] [PubMed] [Google Scholar]
- 19.Qin HF, Shi FJ, Zhang CY, et al. Anti-VEGF reduces inflammatory features in macular edema secondary to retinal vein occlusion. Int J Ophthalmol 2022;15:1296–1304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Mitra S, Devi S, Lee MS, et al. Vegf signaling between muller glia and vascular endothelial cells is regulated by immune cells and stimulates retina regeneration. Proc Natl Acad Sci U S A 2022;119:e2211690119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Feucht N, Schonbach EM, Lanzl I, et al. Changes in the foveal microstructure after intravitreal bevacizumab application in patients with retinal vascular disease. Clin Ophthalmol 2013;7:173–178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Tripathy S, Le HG, Cicinelli MV, Gill MK. Longitudinal changes on optical coherence tomography angiography in retinal vein occlusion. J Clin Med 2021;10:1423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Cheng BT, Mishra S, Bryan JM, et al. Retinal vessel density and treatment intensity among adults with retinal vein occlusion: a swept-source optical coherence tomography angiography study. J Clin Med 2022;11:2892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Spiewak D, Witek K, Drzyzga L, Mrukwa-Kominek E. An analysis of optical coherence tomography angiography (OCT-A) perfusion density maps in patients treated for retinal vein occlusion with intravitreal aflibercept. Diagnostics (Basel) 2023;13:3100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ghasemi Falavarjani K, Iafe NA, Hubschman JP, et al. Optical coherence tomography angiography analysis of the foveal avascular zone and macular vessel density after Anti-VEGF therapy in eyes with diabetic macular edema and retinal vein occlusion. Invest Ophthalmol Vis Sci 2017;58:30–34. [DOI] [PubMed] [Google Scholar]
- 26.Cheong KX, Lee SY, Ang M, Teo KYC. Vessel density changes on optical coherence tomography angiography after vascular endothelial growth factor inhibitor treatment for diabetic macular edema. Turk J Ophthalmol 2020;50:343–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Elnahry AG, Abdel-Kader AA, Habib AE, et al. Review on recent trials evaluating the effect of intravitreal injections of Anti-VEGF agents on the macular perfusion of diabetic patients with diabetic macular edema. Rev Recent Clin Trials 2020;15:188–198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Winegarner A, Wakabayashi T, Fukushima Y, et al. Changes in retinal microvasculature and visual acuity after antivascular endothelial growth factor therapy in retinal vein occlusion. Invest Ophthalmol Vis Sci 2018;59:2708–2716. [DOI] [PubMed] [Google Scholar]
- 29.Yuan Q, Gao Y, Liu Y, et al. Efficacy of single-dose intravitreal dexamethasone implantation for retinal vein occlusion patients with refractory macular edema: a systematic review and meta-analysis. Front Pharmacol 2022;13:951666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Darabus DM, Darabus RG, Munteanu M. The diagnosis and treatment of branch retinal vein occlusions: an update. Biomedicines 2025;13:105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zhu Q, Pan X, Du Z, et al. From monotherapy to combination strategies: redefining treatment approaches for multiple-cause macular edema. Clin Ophthalmol 2025;19:887–897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Alhalwani AY, Hafez SY, Alsubaie N, et al. Assessment of leukocyte and systemic inflammation index ratios in dyslipidemia patients with dry eye disease: a retrospective case‒control study. Lipids Health Dis 2024;23:179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kurtul BE, Ozer PA. Neutrophil-to-lymphocyte ratio in ocular diseases: a systematic review. Int J Ophthalmol 2019;12:1951–1958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Madanagopalan VG. Commentary: systemic immune-inflammatory indices and their association with ocular disorders-Do we have economical and reliable biomarkers? Indian J Ophthalmol 2020;68:859–860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Dogan E, Gundogdu KO, Bursali O, et al. Systemic inflammatory marker levels in serous macular detachment secondary to retinal vein occlusion. J Curr Ophthalmol 2023;35:177–181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ucer MB, Cevher S. Assessment of systemic inflammatory response index and other inflammatory indicators in retinal vein occlusion. Medicine (Baltimore) 2023;102:e36512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Zuo W, Chen T, Song J, Ai M. Assessment of systemic immune-inflammation index levels in patients with retinal vein occlusion. Ocul Immunol Inflamm 2023;31:491–495. [DOI] [PubMed] [Google Scholar]