Egyptian Retina Journal

: 2019  |  Volume : 6  |  Issue : 2  |  Page : 38--42

Vitreous tumor necrosis factor-alpha level in patients with wet type of age-related macular degeneration

Hegan Rabie Abdel Tawwab1, Mohamed Ahmed Attya1, Riham Assem El Essawy2, Mohamed Amin Zayed1,  
1 Department of Ophthalmology, Faculty of Medicine, Cairo University, Cairo, Egypt
2 Department of Clinical and Chemical Pathology, Faculty of Medicine, Cairo University, Cairo, Egypt

Correspondence Address:
Dr. Mohamed Amin Zayed
3 Mesaddak Street, Dokki, Giza


Aim of Study: The aim of the study was to compare the level of tumor necrosis factor (TNF)-α in vitreous samples of naïve wet age-related macular degeneration (AMD) patients and normal individuals, in an attempt to evaluate its role in the disease development. Materials and Methods: A prospective case–controlled study was conducted on 50 eyes –25 naïve wet AMD eyes as “Cases” and 25 eyes without AMD as “Controls.” Vitreous samples were collected by manual aspiration using a 25G needle attached to a 3 ml syringe directed toward the mid-vitreous cavity. The samples were collected, frozen, and stored at − 20°C. The level of TNF-α was assessed using enzyme-linked immunosorbent assay kits. Results: Vitreous TNF-α level was higher in naive wet AMD patients (4.07 ± 0.93 [2.80–6.60] pg/ml; mean ± standard deviation [minimum–maximum]) than patients without AMD (3.54 ± 0.60 [2.60–5.00] pg/ml; P = 0.019). There was no significant difference in vitreous TNF-α level between different choroidal neovascularization subtypes among studied patients (P = 0.460). Conclusion: The levels of TNF-α were shown to be significantly higher in vitreous samples of patients with naïve wet AMD as compared to normal individuals. These results may shed light on the role of TNF-α in the pathogenesis of wet AMD.

How to cite this article:
Tawwab HR, Attya MA, El Essawy RA, Zayed MA. Vitreous tumor necrosis factor-alpha level in patients with wet type of age-related macular degeneration.Egypt Retina J 2019;6:38-42

How to cite this URL:
Tawwab HR, Attya MA, El Essawy RA, Zayed MA. Vitreous tumor necrosis factor-alpha level in patients with wet type of age-related macular degeneration. Egypt Retina J [serial online] 2019 [cited 2020 Jul 5 ];6:38-42
Available from:

Full Text


Age-related macular degeneration (AMD) is a progressive vision-threatening disease, affecting the central region of the retina – the macula – and manifesting in the elderly. AMD is a degenerative disease, affecting primarily the retinal pigment epithelium (RPE) cells and secondarily the photoreceptors, leading to distortion or destruction of central vision and legal blindness.[1]

It is classified clinically as dry AMD (85%–90% of all AMD cases) and wet AMD (10%–15% of all AMD cases).[2] The pathogenesis of AMD is complex and multifactorial. It is thought to be due to oxidative stress, mitochondrial dysfunction, and inflammatory processes.[3]

In wet AMD, abnormal blood vessels start to proliferate from the choriocapillaris and penetrate the Bruch's membrane reaching either the sub-RPE or the subretinal space. The promotion and inhibition of this growth are mediated by cytokines, such as vascular endothelial growth factor (VEGF), tumor necrosis factor alpha (TNF-α), interleukin-6, pigment epithelium-derived factor, and others.[4]

TNF-α is an inflammatory cytokine secreted primarily by macrophages, natural killer cells, and lymphocytes. TNF-α receptors are expressed on many cell types in the retina and choroid, including endothelial cells of fibrovascular tissue.[5] Several lines of evidence suggest that TNF-α increases the secretion of VEGF by the human RPE cells and choroidal fibroblasts; with VEGF being the most important factor for initiating the process of pathological ocular neovascularization.[6]

Theoretically, neutralization of TNF-α activity results in deactivation of the pro-inflammatory cytokine cascade, diminished the recruitment of inflammatory cells from blood to the site of inflammation, decreased angiogenesis mediated by VEGF and alterations in chemokines and vascular permeability. Therefore, a drug that interferes with TNF-α/TNF-α receptor interactions may represent a therapeutic approach for patients with neovascular AMD.[7]

Few studies reported the improvement of best-corrected visual acuity (BCVA) and central foveal thickness as a result of intravitreal administration of infliximab, which is an anti-TNF-α monoclonal antibody, in patients with wet AMD.[8],[9]

Our aim in this study was to measure the levels of TNF-α in the vitreous samples of naïve wet AMD and normal age-matched individuals, in an attempt to evaluate its role in the pathogenesis of wet AMD.

 Materials and Methods

This prospective case–controlled study was conducted on 50 eyes of 50 patients that were divided into two groups: 25 eyes of 25 patients with wet AMD as cases and 25 eyes of 25 patients without AMD as controls.

All cases were recruited from the Kasr Al Ainy Hospital in the interval between June 2018 and March 2019. This study was conducted after approval of the Cairo University, Ophthalmology Department of Ethical Committee. All patients received a thorough explanation of the study design and aims and have signed an informed written consent.

Study cases were selected from naïve wet AMD patients who were candidates for intravitreal anti-VEGF injection, while controls were selected from patients without AMD or any concurrent ocular pathology admitted for cataract surgery. Inclusion criteria include patients older than 55 years diagnosed with naïve wet AMD clinically and confirmed with fundus fluorescein angiography (“FFA”) and optical coherence tomography (“OCT”) in the absence of concurrent ocular disease in the study eye. Exclusion criteria include patients with other causes of choroidal neovascular membranes (CNVMs) (e.g., pathologic myopia, trauma, and hereditary disease); coexisting ocular pathologies (diabetic retinopathy, retinal vein occlusion, uveitis, retinal detachment, and optic neuropathy); eyes previously injected with anti-VEGF agents; previous intraocular surgery except uneventful cataract surgery performed 6 months before the study; and dense media opacities impeding visualization.

All patients meeting inclusion criteria were subjected to full ophthalmic examination that includes BCVA measurement at baseline using the Snellen's chart with conversion to logarithm of the minimum angle of resolution (logMAR) units for statistical analysis, slit-lamp biomicroscopy using +90 Diopter Volk lens, intraocular pressure measurement using Goldmann applanation tonometry, FFA (Visucam 500, Carl Zeiss Meditec, Jena, Germany) for proper classification of the type of wet AMD, and spectral domain-OCT (RTVue-100; Optovue Inc., Fremont, CA, USA) to confirm the membrane activity.

Based on FFA, wet AMD was classified into four subtypes: Classic choroidal neovascularization (CNV): a well-demarcated area of uniform hyperfluorescence with a hypofluorescent margin in the early phase, and dye leakage obscuring the boundaries during the mid and late phases; Predominantly classic CNV: the classic component occupying 50% or more of the entire neovascular complex (including occult CNV and all the fluorescence-blocking constituents); Minimally classic CNV : the classic component occupying Occult CNV with no classic component: which is further divided into two subtypes: “fibrovascular RPE detachment” – appearing as stippled hyperfluorescence with irregular RPE elevation and an “undefined area with a late-phase dye leakage from an undetermined source” – not corresponding to classic CNV and/or fibrovascular RPE detachment in FFA early or mid-phases.

Vitreous sampling procedure

The procedure was performed under topical anesthesia using benoxinate hydrochloride 0.4% drops that were instilled into the eye for 2–3 min. Povidone-iodine (betadine) 5% was applied for sterilization of the eye for 5 min prior to the procedure. A wire speculum was applied to separate the eyelids, and the patient was instructed to look away from the injection site, which was most commonly inferotemporal. Vitreous samples (0.2 ml) were takenfirst before intravitreal anti-VEGF injections. The samples were manually aspirated using a 25G needle attached to a 3 ml syringe that was advanced perpendicularly through the sclera and directed toward to mid-vitreous cavity; 4 and 3.5 mm posterior to the limbus for phakic and pseudophakic eyes, respectively. The samples were kept in Eppendorf tubes that were immediately frozen and stored at −20°C to prevent protein denaturation. The anti-VEGF was injected at another quadrant away from the site of vitreous aspiration. Control samples were similarly collected from age-matched patients undergoing cataract surgery not associated with any known retinal pathology; samples were obtained just before the surgery was performed. The level of TNF-α was assessed usingInvitrogen (TNF-α Human Instant enzyme-linked immunosorbent assay [ELISA]™ kits by “Thermo Fisher Scientific,” USA). The unit of TNF-α is picogram (pg)/ml.

Statistical analysis

Computer-based IBM Statistical Package for the Social Sciences version 21 was used. Data for qualitative variables were summarized using number and percent, whereas mean and standard deviation for normally distributed quantitative variables and median and interquartile range for quantitative variables were not normally distributed. Chi-square test was used to compare qualitative variables, t-test for normally distributed quantitative variables, and Kruskal–Wallis test for quantitative variables, which were not normally distributed. Correlations testing linear relation between variables were done. The statistical significance was defined as P < 0.05.


In the current study, 50 eyes of 50 patients meeting our inclusion criteria were included and classified into two groups. Thefirst group (cases) includes 25 eyes of 25 patients of naïve wet AMD, whereas the second group (controls) includes 25 eyes of 25 normal individuals. In all studied patients, the vitreous TNF-α level was measured in pg/ml.

The mean age of cases was 63.64 ± 6.53 years, whereas the mean age of controls was 66.44 ± 6.31 years. There was no statistically significant difference in the age of both groups (P = 0.129). The percentage of females was 65% among cases and 68% among controls, showing no statistically significant difference between both groups (P = 0.832).

The clinical data of the studied cases regarding the BCVA and central subfield macular thickness (CSMT) are presented in [Table 1]. The BCVA was measured at baseline using the Snellen's chart and converted to LogMAR units for statistical analysis. The CSMT, defined as the distance from the inner limiting membrane “ILM” to the “RPE” within the central subfield, was evaluated using retinal cross-sectional image and three-dimensional macular scan pattern. According to the FFA, the four subtypes of CNV: classic, predominantly classic, minimally classic, and occult represented 36%, 20%, 12%, and 32% of the studied patients, respectively.{Table 1}

The mean vitreous TNF-α level among the studied cases was 4.07 ± 0.93 pg/ml, ranging from 2.80 to 6.60 pg/ml. However, among controls, it was 3.54 ± 0.60 pg/ml, ranging from 2.60 to 5.00 pg/ml. On comparing vitreous TNF-α levels between cases and controls, it showed a statistically significant difference between both groups (P = 0.019).

There was no statistically significant relation between vitreous TNF-α level and the patients' age, BCVA, and CSMT as seen in [Table 2], or CNV subtypes as seen in [Table 3]. Furthermore, no statistically significant relation was found between vitreous TNF-α level and the patients' gender (P = 0.480).{Table 2}{Table 3}


AMD is one of the leading causes of visual impairment worldwide,[10] and its wet subtype is considered to be a devastating form of this disease. Tremendous efforts are ongoing worldwide to shed light on the most important factors in its pathogenesis of CNVM. Although VEGF is considered to be one of the main contributing factors, however; other factors including TNF-α, fibroblast growth factor, transforming growth factor, ILs, and complement are suggested to play an important role in the disease development.

In CNV, TNF-α is thought to stimulate the synthesis of IL-8, monocyte colonization protein, and RPE-secreted VEGF.[6] Furthermore, it is a major regulator of RPE activation responses, including cell attachment, spreading, chemotaxis, migration, and proliferation.[11],[12] Moreover, it upregulates the expression of various apoptotic factors in RPE cells.

The aim of our study was to compare the level of TNF-α in vitreous samples of naïve wet AMD patients and normal individuals, in an attempt to evaluate its role in disease development. To the best of our knowledge, this is thefirst study, in which TNF-α has been isolated and measured in vitreous of patients with naïve wet AMD.

In the current study, patients with naïve wet AMD had statistically significant higher vitreous levels of TNF-α (4.07 ± 0.93 pg/ml) compared to patients without AMD (3.54 ± 0.60 pg/ml). This is not consistent with the results of Ioanna et al., who analyzed the plasma levels of six hypoxia-related mediators, including TNF-α, using ELISA in patients with AMD. They include patients affected by dry AMD (n = 5), wet AMD (n = 11), and there was no elevated concentration of TNF-α in patients with wet AMD.[13] Possibly, the ocular concentrations of certain factors in the diseased eye do not necessarily correlate with their plasma concentrations. Moreover, the response to hypoxic insult in retinal diseases remains mostly local although many factors produced by such conditions are secreted proteins.[13] Our study is considered superior as such possibility was taken into our consideration when measuring the levels of TNF-α in vitreous and not in plasma.

Our study agrees with Suzuki et al. who studied the expression profile of cytokines in the vitreous fluid of 86 patients with diabetic retinopathy (DR) (n = 76) and central retinal vein occlusion (CRVO) (n = 10); as “cases group compared to 23 eyes with an epiretinal membranes and macular holes as “control group,” noting a significantly higher concentration of TNF-α in the vitreous of CRVO patients compared to the control group. They conclude that TNF-α, as an inflammatory cytokine, appears likely to be involved in the development of macular edema and neovascularization.[14]

Our results are consistent with the conclusion of Spranger et al. who studied TNF-α level in the vitreous samples of three groups with different retinal stages of ischemia: a “control group” without angiogenesis, a “diabetes group” with less severe to moderate angiogenesis of patients with PDR, and a “rubeosis group” with massive angiogenesis (rubeosis iridis) resulting from different causes. They found that TNF-α level of vitreous in rubeosis (25.8 pg/ml; n = 6) was 2-fold elevated compared to controls (13.1 pg/ml; n = 10; P < 0.05) and 1.4-fold elevated compared to the diabetes group (18.2 pg/ml; n = 17). They suggested that TNF-α seems to be a mediator of angiogenesis in the eye.[15]

Our study gives an explanation for improvement observed in neovascular AMD patients involved in Giganti et al. and Theodossiadis et al. studies who were treated with intravitreal infliximab.[8],[16] It also explains the reported improvement of ocular inflammatory signs and visual acuity in patients with noninfectious uveitis receiving infliximab[17],[18],[19] and regression of CNV in patients with AMD who received intravenous administration of infliximab for the treatment of rheumatoid arthritis.[20]

There was a concern regarding our sampling technique using 25G needle connected to syringe instead of vitrector, but this issue was evaluated in the previous study by Pfahler et al. who conducted a prospective, consecutive case series of 578 diagnostic vitreous sample attempts. A total of 550-positive diagnostic vitreous samples were collected. The rate of dry sample (<0.05 mL of vitreous was obtained) was 4.8%. Complications include transient vitreous hemorrhage (n = 3), transient intraocular inflammation (n = 3), localized, and chronic retinal detachment (n = 1). None of the complications resulted in permanent visual sequelae. These complications were no different than those expected from intravitreal injection alone. They concluded that in-office vitreous sampling is a safe and reproducible procedure.[21]

We preferred to use a vitreous sample rather than aqueous sample in our study to get more accurate results. Although previous studies on a small number of proteins have shown that the vitreous and aqueous protein levels do correlate. Ecker et al. conducted a study that analyzed the correlation between aqueous and vitreous by studying 34 proteins (i.e., cytokines, growth factors, and TNF-α) in patients with retinal diseases. The quantitative native protein analysis method named reverse phase protein microarray technology was the method used to examine aqueous and vitreous samples. The results revealed a significant correlation of 8 proteins, correlation of 3 proteins, but fell short of significance and no correlation of 23 proteins.[22]

Ecker et al. postulated that several factors determine the levels of proteins in aqueous and vitreous, including lens status, vitreous structure, disease state, and protein size. Furthermore, they elucidated that the difference in rates of protein synthesis and metabolism between the aqueous and vitreous and the histological differences between the eye segments may also contribute to the difference in levels of various proteins.[22]

These findings adopted by Ecker et al. put a strong emphasis on the importance of measuring TNF-α in vitreous samples of patients with naïve wet AMD in our study rather than in aqueous humor. It is postulated that in eye diseases localized to the retina and choroid, such as macular degeneration, the vitreous protein levels might be a better indicator than the aqueous protein levels. Moreover, studying the pathophysiology of vitreoretinal diseases would be the best by sampling the vitreous because of its near vicinity to the retina.


The levels of TNF-α were shown to be significantly higher in vitreous samples of patients with naïve wet AMD as compared to normal individuals. This may put some emphasis on the possible role of TNF-α in the pathogenesis of wet AMD and may shed light on the therapeutic role of anti-TNF-α (infliximab) in treating patients with wet AMD.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interes


1Fine SL, Berger JW, Maguire MG, Ho AC. Age-related macular degeneration. N Engl J Med 2000;342:483-92.
2Ferris FL 3rd, Wilkinson CP, Bird A, Chakravarthy U, Chew E, Csaky K, et al. Clinical classification of age-related macular degeneration. Ophthalmology 2013;120:844-51.
3Nowak JZ. Age-related macular degeneration (AMD): Pathogenesis and therapy. Pharmacol Rep 2006;58:353-63.
4Ambati J, Ambati BK, Yoo SH, Ianchulev S, Adamis AP. Age-related macular degeneration: Etiology, pathogenesis, and therapeutic strategies. Surv Ophthalmol 2003;48:257-93.
5Majka S, McGuire PG, Das A. Regulation of matrix metalloproteinase expression by tumor necrosis factor in a murine model of retinal neovascularization. Invest Ophthalmol Vis Sci 2002;43:260-6.
6Grossniklaus HE, Ling JX, Wallace TM, Dithmar S, Lawson DH, Cohen C, et al. Macrophage and retinal pigment epithelium expression of angiogenic cytokines in choroidal neovascularization. Mol Vis 2002;8:119-26.
7Ni Z, Hui P. Emerging pharmacologic therapies for wet age-related macular degeneration. Ophthalmologica 2009;223:401-10.
8Theodossiadis PG, Liarakos VS, Sfikakis PP, Vergados IA, Theodossiadis GP. Intravitreal administration of the anti-tumor necrosis factor agent infliximab for neovascular age-related macular degeneration. Am J Ophthalmol 2009;147:825-30, 830.e1.
9Theodossiadis PG, Liarakos VS, Sfikakis PP, Charonis A, Agrogiannis G, Kavantzas N, et al. Intravitreal administration of the anti-TNF monoclonal antibody infliximab in the rabbit. Graefes Arch Clin Exp Ophthalmol 2009;247:273-81.
10Flaxman SR, Bourne RR, Resnikoff S, Ackland P, Braithwaite T, Cicinelli MV, et al. Global causes of blindness and distance vision impairment 1990-2020: A systematic review and meta-analysis. Lancet Glob Health 2017;5:e1221-34.
11Jin M, He S, Wörpel V, Ryan SJ, Hinton DR. Promotion of adhesion and migration of RPE cells to provisional extracellular matrices by TNF-alpha. Invest Ophthalmol Vis Sci 2000;41:4324-32.
12Burke JM. Stimulation of DNA synthesis in human and bovine RPE by peptide growth factors: The response to TNF-alpha and EGF is dependent upon culture density. Curr Eye Res 1989;8:1279-86.
13Ioanna Z, Christian S, Christian G, Daniel B. Plasma levels of hypoxia-regulated factors in patients with age-related macular degeneration. Graefes Arch Clin Exp Ophthalmol 2018;256:325-32.
14Suzuki Y, Nakazawa M, Suzuki K, Yamazaki H, Miyagawa Y. Expression profiles of cytokines and chemokines in vitreous fluid in diabetic retinopathy and central retinal vein occlusion. Jpn J Ophthalmol 2011;55:256-63.
15Spranger J, Meyer-Schwickerath R, Klein M, Schatz H, Pfeiffer A. [TNF-alpha level in the vitreous body. Increase in neovascular eye diseases and proliferative diabetic retinopathy]. Med Klin (Munich) 1995;90:134-7.
16Giganti M, Beer PM, Lemanski N, Hartman C, Schartman J, Falk N. Adverse events after intravitreal infliximab (Remicade). Retina 2010;30:71-80.
17Petropoulos IK, Vaudaux JD, Guex-Crosier Y. Anti-TNF-alpha therapy in patients with chronic non-infectious uveitis: The experience of Jules Gonin Eye Hospital. Klin Monbl Augenheilkd 2008;225:457-61.
18Tabbara KF, Al-Hemidan AI. Infliximab effects compared to conventional therapy in the management of retinal vasculitis in Behçet disease. Am J Ophthalmol 2008;146:845-500.
19Diaz-Llopis M, García-Delpech S, Salom D, Udaondo P, Hernández-Garfella M, Bosch-Morell F, et al. Adalimumab therapy for refractory uveitis: A pilot study. J Ocul Pharmacol Ther 2008;24:351-61.
20Markomichelakis NN, Theodossiadis PG, Sfikakis PP. Regression of neovascular age-related macular degeneration following infliximab therapy. Am J Ophthalmol 2005;139:537-40.
21Pfahler SM, Brandford AN, Glaser BM. A prospective study of in-office diagnostic vitreous sampling in patients with vitreoretinal pathology. Retina 2009;29:1032-5.
22Ecker SM, Hines JC, Pfahler SM, Glaser BM. Aqueous cytokine and growth factor levels do not reliably reflect those levels found in the vitreous. Mol Vis 2011;17:2856-63.