Diabetic retinopathy (DR) is a common microvascular complication of diabetes mellitus and an important cause of preventable blindness in working-age adults.1–3 This diabetes-related complication leads to poor quality of life in patients. As a serious ophthalmological problem, DR also presents a heavy burden to health care systems and society.
Previous studies have confirmed that myopia has a protective association with DR,4,5 and axial length may contribute most to this relationship.6,7 In diabetic patients, eyes with longer axial length are less likely to have DR and diabetic macular edema (DME).6 Several studies have reported that the intraocular concentration of vascular endothelial growth factor-A (VEGFA) was negatively correlated with axial length of patients without retinal diseases.8–13 VEGFA is the primary mediator of retinal angiogenesis and plays an important role in the progression of proliferative diabetic retinopathy (PDR).14,15 High myopia may lead to the degeneration of retinal pigment epithelial cells (RPE), which may reduce the secretion of VEGFA.16,17 Decreased levels of VEGFA can prevent vascular leakage, reduce angiogenesis, and ultimately slow down the development of DR.18 Meanwhile, the excessive lengthening of the eyeball in high myopia results in the mechanical dilation and thinning of the retina, which may lead to the straightening and diameter reduction of retinal vessels and decrease of retinal blood flow.19,20 Longer axial length of the eye may decrease the risks of DR by reducing the retina's oxygen consumption and alleviating retinal hypoxia in diabetic patients.7 In conclusion, the decrease of VEGFA concentrations with increasing axial length might partially explain why the myopic eyes have a lower prevalence of DR.
The pathogenesis of DR is complex, involving various vascular, inflammatory, and neurological mechanisms.21 It has been proven that other angiogenesis factors and inflammatory cytokines are also associated with the development of DR.22–24 A negative relationship between intraocular VEGFA and axial length was found in non-neovascular ocular disease, but the relationship between other intraocular cytokines and axial length in patients with neovascular ocular disease such as DR has not been studied. If the aqueous levels of cytokines in DR patients are correlated with the axial length, it may provide more evidence for myopia as a protective factor of DR.
The aim of this study is to further assess the relationship between the aqueous humor levels of interleukin-8 (IL-8), IL-10, VEGFA, vascular adhesion molecule-1 (VCAM-1), basic fibroblast growth factor (bFGF), vascular endothelial growth factor-B (VEGFB), and placental growth factor (PLGF) and axial length in eyes with DR. At the same time, we measured the concentration of the above cytokines in serum to help analyze the meaning of cytokine concentration in aqueous humor.
This study included 65 patients who visited the Department of Ophthalmology at the Affiliated Hospital of Inner Mongolia Medical University. The study group consisted of 30 DR patients who underwent cataract surgery with or without vitrectomy between October 2017 and February 2018, and the control group consisted of 35 age-related cataract patients who underwent conventional cataract surgery in October 2017. All patients were treated at the same institution by the same surgeon. We excluded patients with previous intraocular surgery, keratitis, uveitis, glaucoma, retinopathy such as retinal detachment, retinal vein occlusion and age-related macular degeneration, systemic diseases, and malignant tumors. The study included only 1 eye of each patient. If both eyes were eligible, only the first eye that underwent surgery was included in the study. This study was performed in accordance with the Declaration of Helsinki. The Ethics Committee of the Affiliated Hospital of Inner Mongolia Medical University approved this study, and all participants gave written informed consent.
All patients underwent comprehensive ophthalmological examinations using noncontact tonometer, preoperative best corrected visual acuity (BCVA) by standardized logMAR visual acuity chart, slit lamp biomicroscopy of the anterior segment, a dilated fundus examination with both slit lamp biomicroscopy and indirect ophthalmoscopy, fundus fluorescein angiography (when necessary), fundus color photography (CF-60UV, Canon, Tokyo, Japan), and optical coherence tomography (3DOCT-2000, Topcon, Tokyo, Japan). Biometry (IOL Master Version 3.01, Carl Zeiss, Jena, Germany) was also performed to measure axial length. DR was diagnosed and classified according to the results of indirect ophthalmoscopy and fundus color photography.
The patients without diabetes served as the control group. According to the fundus examination, the patients with diabetes were classified into 2 groups. The patients without any retinal neovascularization were placed in the nonproliferative diabetic retinopathy (NPDR) group, and the patients with retinal neovascularization were defined as the PDR group.
Aqueous humor was collected during routine cataract surgery and sampled using a 26-gauge needle. The volume of aqueous humor collected was at least 100 μL. Venous blood sample was collected in the morning of operation, and serum was separated by centrifugation. All samples were sealed in Eppendorf tubes and stored at −80°C until analysis. The levels of IL-8, IL-10, VEGFA, VCAM-1, and bFGF were measured by Becton, Dickinson and company (BD) cytometric bead array (CBA) analysis software following standard protocol using the BD CBA Flex Set (BD Biosciences, San Jose, CA). VEGFB and PLGF were measured by enzyme-linked immunosorbent assay (ELISA) using Quantikine Human PLGF Immunoassay kit (R&D Systems, Minneapolis, MN) and human VEGFB ELISA kit (YB Science, Shanghai, SHH, China), according to the manufacturer's instructions. Both CBA and ELISA technology have been described in detail in previous studies.22,25
Statistical analysis was carried out using SPSS for Windows version 20.0 (IBM, Armonk, NY). The normal distribution of data was tested by the Kolmogorov-Smirnov test. The differences among the control, NPDR, and PDR groups were tested by 1-way analysis of variance or Kruskal-Wallis tests, according to normality assumptions and homogeneity of variances. The correlations between axial length and cytokines were analyzed by the Pearson or Spearman correlation coefficient, depending on normality assumptions. According to the results of the normality test, the Pearson correlation coefficient was adopted to analyze bivariate correlation between the axial lengths and aqueous levels of IL-10, VEGFA, VEGFB, and VCAM-1, and the Spearman correlation coefficient was used to evaluate the bivariate correlation between the axial lengths and aqueous levels of IL-8, bFGF, and PLGF. Similarly, the levels of VEGFA and VEGFB were compared among the three groups by 1-way analysis of variance test, and the other cytokines were analyzed by Kruskal–Wallis test. Multivariate stepwise regression analysis was used to detect the significance of influencing factors. Two-tailed P < 0.05 was considered statistically significant.
The study included 65 eyes of 65 patients (27 men, 38 women). Among these patients, 14, 16, and 35 patients were included in the NPDR, PDR, and control groups, respectively. Detailed demographic and baseline characteristics of patients are given in Table 1. No statistical differences were found among age, sex, intraocular pressure, and axial length of patients in the 3 groups.
There were significant differences in the aqueous concentrations of IL-8, IL-10, VEGFA, VCAM-1, and VEGFB among the 3 groups (all P values were <0.005). There were no significant differences in the aqueous concentrations of bFGF (P = 0.46) and PLGF (P = 0.82) among the 3 groups. The concentrations of IL-8 and VEGFA in aqueous humor were higher than those in serum, whereas the concentrations of IL-10, VCAM-1, and VEGFB in aqueous humor were lower than those in serum (all P values were <0.005; Table 2). There was no difference in bFGF between aqueous humor and serum.
Figure 1 shows the concentration of various cytokines in the aqueous humor and serum of the 3 groups of patients. In the NPDR group, as compared with the control group, significantly higher aqueous humor of concentrations were measured for IL-8 (P = 0.005). In the PDR group as compared with the control group, significantly higher aqueous humor of concentrations were measured for VEGFA (P < 0.001), IL-8 (P = 0.005), IL-10 (P = 0.001),VCAM-1 (P = 0.001), and VEGFB (P = 0.005). In the PDR group as compared with the NPDR group, significantly higher aqueous humor of concentrations were measured for VEGFA (P = 0.001) and VEGFB (P = 0.001). In the serum samples, only the concentration of VCAM-1 in PDR group was higher than that in control group (P = 0.045), and there was no statistical difference in the concentration of other cytokines among groups.
The correlation between cytokines and levels of aqueous humor and axial length was analyzed in the 3 groups. In the control group, there was a negative correlation between VEGFA and axial length [Pearson correlation coefficient, r = −0.440, P = 0.008; equation of the regression line: VEGF concentration: −3.537 × axial length (mm) + 120.625; Fig. 2], and a positive correlation between VCAM-1 and axial length [Pearson correlation coefficient, r = 0.395, P = 0.019; equation of the regression line: VCAM-1 concentration: 43.689 × axial length (mm) −707.147; see Fig. 2].
In the NPDR group, PLGF was negatively correlated with axial length [Spearman correlation coefficient, r = −0.576, P = 0.031; equation of the regression line: PLGF concentration: −3.206 × axial length (mm) + 94.21], and IL-10 [Pearson correlation coefficient, r = 0.533, P = 0.049; equation of the regression line: IL-10 concentration: 0.19 × axial length (mm) + 0.629] and VCAM-1 [Pearson correlation coefficient, r = 0.566, P = 0.035; equation of the regression line: VCAM-1 concentration: 152.435 × axial length (mm) −3214.291] were positively correlated with axial length (Fig. 3). In the PDR group, the aqueous levels of all cytokines did not significantly correlate with axial length.
Multivariate stepwise regression analysis was used to detect the related factors of DR, such as axial length and cytokine concentration in aqueous humor, but there was no statistical significance. Then multivariate analysis was used in NPDR group: the concentration of PLGF IL-10 and VCAM-1 in aqueous humor was taken as a dependent variable, and the age, sex, intraocular pressure, duration of diabetes, axial length, and anterior chamber depth were independent variables. It indicates that PLGF and IL-10 in aqueous humor is only related to the axial length [P
= 0.041; standardized coefficients beta:−0.552; regression coefficient B:−3.206; 95% confidence interval (CI):−6.252 to −0.160 and p
= 0.049; β: 0.533; B:0.190; 95% CI:0–0.379; respectively] and is not related to other factors, and VCAM-1 in aqueous humor was related to age (P
= 0.016; β:−0.546; B:−15.840; 95% CI:−28.175 to −3.506) and axial length (P
= 0.038; β: 0.455; B: 122.547; 95% CI: 8.156–236.937).
The pathogenesis of DR was associated with 4 biochemical pathways and led to the exhibition of numerous pathological events, including inflammation, oxidative stress, and angiogenesis.26 Each pathway could cause the upregulation of cytokines.21 Normally, the dynamic balance between proangiogenic factors and antiangiogenic factors determines the development of PDR.27 Also, it is suggested that inflammation plays a major role in the development of early and late-stage of DR.21,24,28 In this study, increasing aqueous levels of IL-8 found in the NPDR group might be attributed to the activation of the inflammatory process. Moreover, the levels of IL-8, IL-10, VCAM-1, VEGFB, and VEGFA in the PDR group were significantly higher than in the control group. Meanwhile, the concentrations of IL-8 and VEGFA in aqueous humor were higher than those in serum in the NPDR and PDR groups. Those findings indicate that VEGFA, and other angiogenesis factors and inflammatory cytokines, may be involved in the pathogenesis of DR. Therefore, it is necessary to study the relationship between cytokines and axial length in DR patients.
In our study, the aqueous level of VEGFA decreased significantly with increasing axial length in controls. This finding is in agreement with previous investigations.8,9,11–13 Similar to VEGFA, the aqueous level of PLGF was negatively correlated with axial length in the NPDR group. This correlation has never been reported. The role of PLGF, a homologue factor of VEGFA, is to stimulate the proliferation and migration of vascular endothelial cells and induce angiogenesis.29 In recent years, evidence has emerged showing that PLGF is not essential in physiological angiogenesis but does play an important role in pathological angiogenesis in animal models.30,31 It is reported that although the level of PLGF in the vitreous of DR patients significantly increases and the level of PLGF in the aqueous humor is significantly correlated with the level of VEGFA,32 the level of PLGF in aqueous humor is low or undetectable.33,34 The concentration of PLGF in aqueous humor may be lower than the detection limit.33 In some studies, the concentration of PLGF in aqueous humor of cataract patients was very low,32,33,35 but in our study, we could not find any difference in PLGF concentrations in aqueous humor among the 3 groups. This is related to several extreme values in the control group, but they meet the criteria for inclusion. We did not exclude these values in order not to create selection bias. At the same time, there was no correlation between PLGF and VEGFA (P > 0.05). The difference in level of PLGF may be attributed to the different kinds of samples and detection methods. Due to the characteristics of PLGF, the level of PLGF was low or absent in the normal retina and low in DR without neovascularization.36 This is why the correlation between PLGF and axial length only appears in the NPDR group and not in the control group. The results of our study fit with the observation that myopia is associated with the lower prevalence of DR. It is suggested that PLGF plays an important role in the early development of DR, and longer axial length is a protective factor of DR. In addition, unlike VEGFA, PLGF only upregulated in the pathological process; thus, PLGF blockade might inhibit disease progression more selectively and cause fewer side effects. Anti-PLGF therapy may provide new therapeutic prospects for ocular angiogenesis diseases.
The negative correlation between axial length and VEGFA and PLGF may explain why myopia eyes have a lower prevalence of DR, and here are several possible explanations for the negative correlation. One reason is that VEGFA and PLGF might be more diluted in the large vitreous cavity with longer axial length, and a higher vitreous liquefaction might shorten the turnover time of substances in the liquefied vitreous body. Another possible explanation is that the production of VEGFA and PLGF might decrease because of the thinning of retina with axial elongation.
VCAM-1, an active member of the adhesion molecule immunoglobulin superfamily, is involved in the recruitment of leukocytes, vascular endothelium, and migration.37 A previous study showed that the level of VCAM-1 was significantly higher in the serum and vitreous of DR patients.38 In our study, the VCAM-1 levels in the aqueous humor of the PDR group were significantly higher than in the control group. Meanwhile, the VCAM-1 level was positively correlated with axial length in both the control and NPDR groups. The multivariate analysis indicated that VCAM-1 in aqueous humor was related to age and axial length. The positive correlation between them may be influenced by other variables such as patients with various degrees of cataract and patients of different ages. Previous studies have shown that elevated VCAM-1 is associated with the degree of macular edema21 and longer axial length is a protective factor for DME, which is inconsistent with our conclusion. Future studies should examine the relationship among the axial length, DME, and VCAM-1.
IL-10 is a protective cytokine with anti-inflammatory effect.39,40 Mao et al41 suggested that the proinflammatory cytokines elevated in PDR, and high expression of IL-10 played a protective role against the development of PDR. Our study found the aqueous humor levels of IL-10 significantly higher in the PDR group. Meanwhile, the concentration of IL-10 in aqueous humor was positively correlated with axial length in the NPDR group. The increased IL-10 concentrations suggested a balanced change of the intraocular environment in the DR patients. Our study showed that the elevation of IL-10 may be a protective factor for PDR.
In our study, we did not find significant correlation between the levels of bFGF, VEGFB, and IL-8 and axial length in any group, which suggested that the influence of axial length on these cytokines may be not as pivotal as that of VEGFA, PLGF, VCAM-1, and IL-10. Our study showed that there was no significant difference in bFGF concentration in aqueous humor and serum among the 3 groups. Our results also suggest this conclusion. It has remained unclear why these cytokine concentrations were not correlated with axial length. One explanation could be that these cytokines were not mainly produced in the eyes. These cytokines may instead leak into the eyes from the serum via the inner ocular surface, and the dilution effect of increased axial length may be compensated by the increase of inner ocular surface. Another explanation is that PRE, Muller, and retinal endothelial cells may have secreted cytokines in DR patients, affecting the correlation between cytokines and axial length. This deserves our further study.
The limitations of our research should be discussed. First, the number of patients enrolled in the study was relatively small. However, our results were statistically significant, so small samples may strengthen our conclusions. Second, this study is a retrospective study and may have a selective bias. There was a large sex gap between the NPDR and PDR groups because there were more female patients in the hospital at that time. To avoid selection bias, we included all eligible patients. Third, the samples we collected were aqueous humor rather than vitreous humor. Although vitreous humor can better reflect the condition of the retina, it may lead to unnecessarily complicated surgical treatment. Fourth, our study did not contain HbA1c, but our patients’ glucose concentrations were well controlled during the study.
In conclusion, we observed negative or positive correlations with PLGF, IL-10, and VCAM-1 and axial length, suggesting that these cytokines may have protective or promotive effects in the development of DR and further explain the protective effect of axial length on DR. To our knowledge, the present report is the first to show a negative correlation between aqueous PLGF levels and axial length. Further studies will be required to confirm the benefits of longer axial length for DR.
1. Cheung N, Mitchell P, Wong TY. Diabetic retionpathy. Lancet
2. Mohamed Q, Gillies MC, Wong TY. Management of diabetic retinopathy: a systematic review. JAMA
3. Zheng Y, He M, Congdon N. The worldwide epidemic of diabetic retinopathy. Indian J Ophthalmol
4. Bazzazi N, Akbarzadeh S, Yavarikia M, et al. High myopia and diabetic retinopathy: a Contralateral Eye Study in diabetic patients with high myopic anisometropia. Retina
5. Chao DL, Lin SC, Chen R, et al. Myopia is inversely associated with the prevalence of diabetic retinopathy in the South Korean Population. Am J Ophthalmol
6. Man RE, Sasongko MB, Sanmugasundram S, et al. Longer Axial Length Is Protective of Diabetic Retinopathy and Macular Edema. Ophthalmology
7. Man RE, Lamoureux EL, Taouk Y, et al. Axial length, retinal function, and oxygen consumption: a potential mechanism for a lower risk of diabetic retinopathy in longer eyes. Invest Ophthalmol Vis Sci
8. Zhu D, Yang DY, Guo YY, et al. Intracameral interleukin 1β, 6, 8, 10, 12p, tumor necrosis factor α and vascular endothelial growth factor and axial length in patients with cataract. PLoS One
9. Jonas JB, Tao Y, Neumaier M, et al. VEGF and refractive error. Ophthalmology
2010; 117: 2234.
10. Sawada O, Kawamura H, Kakinoki M, et al. Vascular endothelial growth factor in the aqueous humour in eyes with myopic choroidal neovascularization. Acta Ophthalmol
11. Sawada O, Miyake T, Kakinoki M, et al. Negative correlation between aqueous vascular endothelial growth factor levels and axial length. Jpn J Ophthalmol
12. Wakabayashi T, Ikuno Y, Oshima Y, et al. Aqueous concentrations of vascular endothelial growth factor in eyes with high myopia with and without choroidal neovascularization. J Ophthalmol
13. Hu Q, Liu G, Deng Q, et al. Intravitreal vascular endothelial growth factor concentration and axial length. Retina
14. Qian J, Lu Q, Tao Y, et al. Vitreous and plasma concentrations of apelin and vascular endothelial growth factor after intravitreal bevacizumab in eyes with proliferative diabetic retinopathy. Retina
15. Fisher C, Jonckx B, Mazzone M, et al. Anti-PlGF
inhibits growth of VEGF(R)-inhibitor-resistant tumors without affecting healthy vessels. Cell
16. Saint-Geniez M, Kurihara T, Sekiyama E, et al. An essential role for RPE-derived soluble VEGF in the maintenance of the choriocapillaris. Proc Natl Acad Sci U S A
17. Tolmachova T, Wavre-Shapton ST, Barnard AR, et al. Retinal pigment epithelium defects accelerate photoreceptor degeneration in cell type-specific knockout mouse models of choroideremia. Invest Ophthalmol Vis Sci
18. Ferrone PJ, Jonisch J. Comparison of ranibizumab 0.5 mg versus 1.0 mg for the treatment of patients with clinically significant diabetic macular edema: a randomized, clinical trial. Ophthalmic Surg Lasers Imaging Retina
19. Shimada N, Ohno-Matsui K, Harino S. Reduction of retinal blood flow in high myopia. Graefes Arch Clin Exp Ophthalmol
20. Benavente-Pérez A, Hosking SL, Logan NS, Broadway DC. Ocular blood flow measurements in healthy human myopic eyes. Graefes Arch Clin Exp Ophthalmol
21. Semeraro F, Cancarini A, dell’Omo R, et al. Diabetic retinopathy: vascular and inflammatory disease. J Diabetes Res
2015; 2015: 582060.
22. Wu H, Hwang DK, Song X, Tao Y. Association between aqueous cytokines and diabetic retinopathy stage. J Ophthalmol
2017; 2017: 9402198.
23. Cheung CM, Vania M, Ang M, et al. Comparison of aqueous humor cytokine and chemokine levels in diabetic patients with and without retinopathy. Mol Vis
24. Zhou J, Wang S, Xia X. Role of intravitreal inflammatory cytokines and angiogenic factors in proliferative diabetic retinopathy. Curr Eye Res
25. Chen R, Lowe L, Wilson JD, et al. Simultaneous quantification of six human cytokines in a single sample using microparticle-based flow cytometric technology. Clin Chem
26. Behl T, Kaur I, Kotwani A. Role of leukotrienes in diabetic retinopathy. Prostaglandins Other Lipid Mediat
27. Qazi Y, Maddula S, Ambati BK. Mediators of ocular angiogenesis. J Genet
28. Feng S, Yu H, Yu Y, et al. Levels of inflammatory cytokines IL-1β, IL-6, IL-8, IL-17A, and TNF-α in aqueous humour of patients with diabetic retinopathy. J Diabetes Res
2018; 2018: 8546423.
29. Nguyen QD, De Falco S, Behar-Cohen F, et al. Placental growth factor and its potential role in diabetic retinopathy and other ocular neovascular diseases. Acta Ophthalmol
30. De Falco S. The discovery of placenta growth factor and its biological activity. Exp Mol Med
31. Gigante B, Morlino G, Gentile MT, et al. Plgf
-/-eNos-/- mice show defective angiogenesis associated with increased oxidative stress in response to tissue ischemia. FASEB J
32. Al Kahtani E, Xu Z, Al Rashaed S, et al. Vitreous levels of placental growth factor correlate with activity of proliferative diabetic retinopathy and are not influenced by bevacizumab treatment. Eye (Lond)
33. Yamashita H, Eguchi S, Watanabe K, et al. Expression of placenta growth factor (PIGF) in ischaemic retinal diseases. Eye (lond)
34. Khuu LA, Tayyari F, Sivak JM, et al. Aqueous humour concentrations of TGF-β, PLGF
and FGF-1 and total retinal blood flow in patients with early non-proliferative diabetic retinopathy. Acta Ophthalmol
35. Ando R, Noda K, Namba S, et al. Aqueous humour levels of placental growth factor in diabetic retinopathy aqueous humour levels of placental growth factor in diabetic retinopathy. Acta Ophthalmol
36. Khaliq A, Foreman D, Ahmed A, et al. Increased expression of placenta growth factor in proliferative diabetic retinopathy. Lab Invest
37. Hernández C, Burgos R, Cantón A, et al. Vitreous levels of vascular cell adhesion molecule and vascular endothelial growth factor in patients with proliferative diabetic retinopathy. Diabetes Care
38. Adamiec-Mroczek J. Assessment of selected adhesion molecule and proinflammatory cytokine levels in the vitreous body of patients with type 2 diabetes—role of the inflammatory-immune process in the pathogenesis of proliferative diabetic retinopathy. Graefes Arch Clin Exp Ophthalmol
39. Jagannathan M, McDonnell M, Liang Y, et al. Toll-like receptors regulate B cell cytokine production in patients with diabetes. Diabetologia
40. Lee JH, Lee W, Kwon OH, et al. Cytokine profile of peripheral blood in type 2 diabetes mellitus patients with diabetic retinopathy. Ann Clin Lab Sci
41. Mao C, Yan H. Roles of elevated intravitreal IL-1β and IL-10 levels in proliferative diabetic retinopathy. Indian J Ophthalmol