Purpose
To determine the concentration of the pro-angiogenic vascular endothelial growth factor VEGF 165 (VEGF) and the anti-angiogenic VEGF 165b in vitreous samples of patients with branch retinal vein occlusion (BRVO) and central retinal vein occlusion (CRVO) in comparison to patients without retinal occlusive disease.
Design
Experimental laboratory investigation.
Methods
Vitreous samples were collected from patients undergoing surgery for arteriovenous dissection after BRVO, radial optic neurotomy after CRVO in the occlusion group, or macular pucker or macular hole in the control group. Concentrations of VEGF and VEGF 165b were determined by ELISA and an ELISA-type antibody microarray.
Results
Average vitreal concentration of VEGF was 8.6 ng/mL in the CRVO group and 2.0 ng/mL in the BRVO group as compared to 0.26 ng/mL in the control group. Average vitreal concentration of VEGF 165b was 27 pg/mL in the CRVO group, 42 pg/mL in the BRVO group, and 49 pg/mL in the control group. In patients with CRVO and BRVO, the angiogenic balance was shifted towards angiogenic stimulation.
Conclusion
The severity of RVO from BRVO to CRVO correlates with an increase of VEGF and the decrease of VEGF 165b , indicating a pro-angiogenic shift. Altering the ratio of VEGF 165b /VEGF 165 might be a feasible approach for treating retinal occlusive diseases.
Retinal vein occlusion (RVO) is the second most common retinal vascular disorder after diabetic retinopathy, and is a significant cause of visual morbidity in patients aged over 50 years. Visual impairment occurs secondary to macular edema (ME). Induction of vascular endothelial growth factor (VEGF) and pro-inflammatory cytokines by vascular occlusion is associated with a breakdown of the blood-retinal barrier as well as an increased vascular permeability, resulting in retinal and macular edema. VEGF expression correlates with the duration of branch retinal vein occlusion (BRVO). The higher the VEGF expression, the more reduction of ME was found in patients undergoing arteriovenous dissection in BRVO, whereas the visual prognosis appears to be impaired by a high vitreous level of VEGF. Lately, evidence has become available that treatment of ME by intravitreal injection of VEGF inhibitors such as bevacizumab (Avastin; Genentech, South San Francisco, California, USA) and ranibizumab (Lucentis; Novartis, Basel, Switzerland) may improve visual outcome. Studies with VEGF-Trap-Eye, a fusion protein binding VEGF and placental growth factor, in RVO patients are currently under way.
There are 2 families of VEGF isoforms formed by differential splicing, which may lead to opposite angiogenic effects. A family of splice variants with an anti-angiogenic rather than pro-angiogenic effect has been identified, named VEGF 165b . The mechanism of differential splicing of VEGF seems to be regulated by splice factors partly dependent on oxygen supply, and hypoxia has been shown to downregulate VEGF 165b in retinal pigment epithelial cells. Regarding diabetic retinopathy, it was shown that the intravitreal VEGF concentration is elevated while the VEGF 165b variant remains at control level, resulting in an angiogenic shift.
VEGF levels have been studied in RVO patients. There have been no data available about concentration of VEGF 165b yet. In our study, we examined levels of VEGF and VEGF 165b in vitreous samples of BRVO patients who underwent surgery for arteriovenous dissection and central retinal vein occlusion (CRVO) patients treated with radial optic neurotomy. Vitreous samples of patients undergoing vitrectomy with membrane peeling attributable to macular hole or macular pucker were examined in comparison.
Materials and Methods
Patients and Study Design
Patients with RVO were recruited in 2005 and 2006. A patient was considered for vitrectomy based on the clinical findings and fluorescence angiography. Regarding CRVO, patients with either ischemic occlusive disease, indicated by nonperfusion in fluorescence angiography, or clinical findings such as dark hemorrhages, a high number of cotton-wool spots, or massive leakage of the vessels and papilledema, were selected for vitrectomy and radial optic neurotomy. Regarding BRVO, patients with a good identifiable arteriovenous crossing as the location of occlusion, with significant macular edema and a visual acuity (VA) of worse than 0.2 logMAR, were eligible for vitrectomy with arteriovenous dissection. Nineteen BRVO and 17 CRVO patients were included in the study. Control specimens were collected from 32 patients undergoing vitrectomy for macular pucker and macular hole. Patient data are presented in Table 1 .
| Patient Group | Number of Patients (Male) | Time After Occlusion [Weeks] ± SD | Mean Age [Years] ± SD | Mean Visual Acuity [logMAR] ± SD |
|---|---|---|---|---|
| CRVO | 17 (9) | 9.1 ± 5.5 (range: 1–20) (median: 8) | 64.1 ± 14.0 (range: 37–90) | 1.38 ± 0.42 (range: 0.5–2.1) |
| BRVO | 19 (10) | 18.7 ± 22.8 (range: 3–96) (median: 8) | 64.7 ± 9.4 (range: 50–83) | 0.61 ± 0.32 (range: 0.2–1.3) |
| Control | 32 (11) | — | 68.3 ± 8.0 (range: 40–83) | — |
A standard 3-port vitrectomy was performed during surgery. Sample acquisition was achieved as the first step of the surgery, avoiding dilution by the infusion. Depending on clinical findings, additional procedures such as laser photocoagulation or intravitreal administration of triamcinolone or bevacizumab could be included at the end of surgery. Samples were immediately stored at −80C until further investigation.
Patients with other proliferative eye diseases and conditions with a possible reduction of VA (uveitis, diabetes, intraocular surgery within the last 6 months, lens or corneal opacities that would preclude measurement of the VA) were excluded from the study. VA was converted to logarithm of the minimal angle of resolution (logMAR) for statistical evaluation.
VEGF and VEGF 165b Determination
Vitreal concentrations of VEGF were measured with an ELISA-type antibody microarray (Quantibody, Raybiotech Inc, Norcross, Georgia, USA) following the manufacturer’s instructions. A total of 15 μL of vitreous was used for each patient. The detection antibody was labeled with biotin, which was detected with Alexa Fluor 555–conjugated streptavidin. The signals were read with a G2565 microarray reader (Agilent Technologies, Santa Clara, California, USA). TM4 Spotfinder ( http://www.tm4.org ) was used for quantification of the spots. The concentrations of the factors were calculated from the median intensities of the spots using standard curves obtained with a mix of peptide standards. VEGF 165b was determined by ELISA using a pan-VEGF capture antibody and a mouse monoclonal detection antibody raised against the terminal 9 amino acids of VEGF 165b , and a standard curve using recombinant human VEGF 165b . A biotinylated secondary goat anti-mouse IgG was used to detect the VEGF 165b antibody. Protein concentrations were determined by the bicinchoninic acid protein assay.
Mean concentrations for CRVO, BRVO, and control groups were compared by t test. Concentrations of VEGF, VEGF 165b , and the relation of VEGF 165b to VEGF were tested for correlation to VA and change of vision after surgery by assessment of the Pearson correlation coefficient. For this calculation, patients receiving further surgery (such as cataract surgery or repeated injections of bevacizumab) were excluded. P < .05 was considered significant.
Results
Concentration of VEGF and VEGF 165b in Ocular Fluid
The total amount of vitreal protein was elevated by a factor of 1.5 in the CRVO group compared to the controls. The results in the BRVO group were comparable to the control level ( Table 2 ).
| Patient Group | Number of Patients | VEGF | VEGF 165b | VEGF 165b /VEGF | Protein Conc. [mg/mL] ± SEM | |||
|---|---|---|---|---|---|---|---|---|
| Conc. [pg/mL] ± SEM | Fold Relative to Control ( P Value) | Conc. [pg/mL] ± SEM | Fold Relative to Control ( P Value) | Value b ± SEM | Fold Relative to Control ( P Value) | |||
| CRVO | 17 | 8653 ± 3020 | 33.8 (.013) a | 27 ± 7.5 | 0.54 (.046) a | 0.0073 ± 0.0026 | 42.4 (.007) a | 7.7 ± 0.60 a |
| BRVO | 19 | 1994 ± 582 | 7.8 (.008) a | 42 ± 10 | 0.85 (.585) | 0.046 ± 0.020 | 6.4 (.046) a | 5.2 ± 0.29 |
| Control | 32 | 256 ± 43 | — | 49 ± 8.1 | — | 0.31 ± 0.15 | — | 5.1 ± 0.25 |
a Significant as determined by t test ( P < .05).
b Mean value is based on the individual ratios of VEGF 165b /VEGF.
Average vitreal VEGF concentration in the CRVO group was 8.6 ng/mL (977 pg/mg protein) compared to 0.26 ng/mL (49 pg/mg protein) in the control group, indicating a significant 33-fold higher concentration ( Table 2 ). Similarly, the average vitreal VEGF concentration in the BRVO group was 2.0 ng/mL (358 pg/mg protein) compared to 0.26 ng/mL (49 pg/mg protein) in the control group, which is a significant 8-fold higher concentration. In contrast, the vitreal VEGF 165b concentration was reduced from 49 pg/mL (11 pg/mg protein) in the control group to 27 pg/mL (3.3 pg/mg protein) in the CRVO group (2-fold, significant) and to 42 pg/mL (8.2 pg/mg protein) in the BRVO group (1.2-fold, not significant). In conclusion, in both CRVO and BRVO the angiogenic balance, indicated by the ratio of VEGF 165b to VEGF, is shifted towards angiogenic stimulation, from 0.3 in the control group to 0.007 (42-fold) in CRVO or to 0.05 (6-fold) in BRVO. Please note that the values related to protein were calculated for each patient before averaging. That implicates that the values are slightly different from those obtained by dividing the means from Table 2 .
Neither concentrations of VEGF or VEGF 165b nor the relation of VEGF 165b to VEGF showed a correlation to VA and change of vision after surgery.
Results
Concentration of VEGF and VEGF 165b in Ocular Fluid
The total amount of vitreal protein was elevated by a factor of 1.5 in the CRVO group compared to the controls. The results in the BRVO group were comparable to the control level ( Table 2 ).
| Patient Group | Number of Patients | VEGF | VEGF 165b | VEGF 165b /VEGF | Protein Conc. [mg/mL] ± SEM | |||
|---|---|---|---|---|---|---|---|---|
| Conc. [pg/mL] ± SEM | Fold Relative to Control ( P Value) | Conc. [pg/mL] ± SEM | Fold Relative to Control ( P Value) | Value b ± SEM | Fold Relative to Control ( P Value) | |||
| CRVO | 17 | 8653 ± 3020 | 33.8 (.013) a | 27 ± 7.5 | 0.54 (.046) a | 0.0073 ± 0.0026 | 42.4 (.007) a | 7.7 ± 0.60 a |
| BRVO | 19 | 1994 ± 582 | 7.8 (.008) a | 42 ± 10 | 0.85 (.585) | 0.046 ± 0.020 | 6.4 (.046) a | 5.2 ± 0.29 |
| Control | 32 | 256 ± 43 | — | 49 ± 8.1 | — | 0.31 ± 0.15 | — | 5.1 ± 0.25 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
