Systemic Erythropoietin Concentration and its Correlation with Stage of Diabetic Retinopathy

Summary: Background: Erythropoietin (Epo) is one of systemic angiogenic factors, and its role in ocular angiogenesis and in diabetic retinopathy (DR) is not yet fully understood. Latest research data reveal possible correlation of higher EPO concentrations of erythropoietin in blood and in the eye, with more severe of stages of DR. The main aim of this work was to examine the possible influence of serum concentrations of erythropoietin on the development and stages of diabetic retinopathy in patients with diabetes mellitus type 2. Methods: The research involved 90 patients examined at University Eye Clinic in Clinical Center of Vojvodina in Novi Sad, Serbia. First group comprised of 60 patients with diabetes mellitus lasting 10 years or more, with diabetic retinopathy. Second, control group, consisted of 30 healthy individuals. In the first group of 60 diabetic patients, 30 of them had non-proliferative diabetic retinopathy (NPDR), and 30 had proliferative diabetic retinopathy (PDR). Laboratory EPO serum levels were determined, and they were correlated to the stage of DR. Concentration of EPO was assessed by ELISA method at the end of the study. Results: The highest average concentration of EPO in serum (9.95 mIU/ml) was determined in group of diabetics with PDR. The lowest average concentration of EPO in serum (6.90 mIU/ml) was found in control group. The average concentration of Epo in serum in group of diabetics with NPDR was 7.00 mIU/ml. EPO concentration in serum was elevated in group of PDR, and it was directly proportional to the level of clinical stadium of PDR, being significantly higher in moderate and severe subgroup of PDR comparing to control healthy subjects, NPDR and mild PDR (h=9.858, p=0.007). Conclusions: Significantly elevated serum concentration of EPO in advanced stages of DR, and positive correlation between EPO serum concentration and clinical stadium of PDR, suggest that erythropoietin presents one of the important growth factors from blood, which plays role in retinal ischemia and angiogenesis


Introduction
Diabetes mellitus (DM) is a multi organic disease with a high incidence in the population, whose main cause is a carbohydrate metabolism disorder. One of the most common and at the same time one of the most difficult chronic complications of DM is diabetic retinopathy (DR). Diabetic retinopathy leads to changes in the small blood vessels of the eye, and DR is considered nowadays as one of the main causes of impaired visual function and blindness, with significant socioeconomic consequences due to the big, mainly working population which it affects (1).
Changed quality of blood vessels in DM and DR, leads to increased liquid permeability which manifests as bleeding, edema and exudates in the eye. Another key point in DM and DR pathogenesis is ischemia in the tissues, which through production of vasoproliferative, angiogenic factors leads to neovascularization-growth of pathological blood vessels in the eye. These vessels have poor wall quality, and they will furthermore lead to circulus vitious of new ischemia attacs and bleeding. Increased vascular permeability and pathologic neovascularisation are considered as two major vascular pathogenic pathways for the development of diabetic retinopathy (2).
Development of DR has been studied for decades. lt has been known that local angiogenic factors in the eye are dominant, but many older as well as many recent studies indicate the importance of systemic angiogenic growth factors, too (3,4,5). One of angiogenic factors is erythropoietin, which is excreted in organism, as a response to tissue hypoxia, with a role to increase erythropoiesis.
Growth factor is the substance that stimulates cell growth, proliferation and 4 differentiation. Most frequently these factors work as cytokines and/or hormones, i.e. as signal transmitters between cells, and it is important to distinguish them from cytokines.
Erythropoietin is a glycoprotein by its structure, and a hormone with the circulatory growth factor by its function (8). It is produced mainly in the kidneys, and only in small part in the liver (10%); main stimulus for its releasing is tissue hypoxia (9). It exhibits it's influences by binding to transmembrane erythropoietin receptors (EPOR), which primarily are found on the hematopoietic cells, but also can be found on the endothelial, myocardial, neural cells, and as well on the cells of liver, uterus and retina (10).
Circulating erythropoietin primarily affects hematopoietic cells, and afterwards by stimulating angiogenesis, it is even considered as tumor-cells stimulator (11,12). There are theories about erythropoietic induction of neovascularization caused by inflammation or ischemia. This occurs by mobilizing the endothelial progenitor cells from the bone marrow, and thus increasing the number of cells in the circulation. From existing capillaries and postcapillary venules, new blood vessels are being formed by degradation of basal membrane of the blood vessel, migration of the endothelial cells and their mitosis, by forming bourgeons, lumens and vascular loops. All these steps represent angiogenesis that has to be distinguished from vasculogenesis (13). Vasculogenesis occurs by the differentiations of endothelial cells from angioblasts Angiogenesis can be physiological and pathological. Physiological angiogenesis presents the balance between proangiogenic and antiangiogenic factors in organism, and occurs in the reproductive cycle, pregnancy and wound healing (14,15). While pathological occurs in neoplastic diseases leading to acceleration of disease progression, or it is the pathological mediator like in diabetic retinopathy (16,17,18).
Numerous new studies indicate the main role of angiogenic growth factors on the development of proliferative diabetic retinopathy, and VEGF is considered to be at the first place (19), followed by erythropoietin, IGF-1, PDGF, etc. Intraocular synthesis of proangiogenic factors is in counterbalance with production of antiangiogenic ones. It is considered that erythropoietin can have direct role in pathophysiology of diabetic retinopathy, however there are still contradictory opinions about whether the role is aggravating or protective. It has been proved that erythropoietin is significantly correlated with origin of proliferative diabetic retinopathy (20). In DM, retinal cells are trying to compensate hypoglycemia and hypoxia by manifestation of increased number of erythropoietin receptors on them. It is considered by some authors this might be how they survive in unfavorable conditions like in DM, inducing increased binding of erythropoietin molecules (20,21). In experimental rats with DM, intravitreal injection of Epo caused increased number of erythropoietin receptors (EpoR) in neurosensory retina, with protective effect against retinal neovascularization and degenerative changes of photoreceptors (22,23).
In certain researches is found that the same Epo injection slows down retinal cells death and promotes the function of hematoretinal barrier, therefore Epo is being considered as one of new therapeutical options in the treatment of early DR and diabetic macular edema (24). However, there are different theories between authors about favorable influence of Epo on progression of non-proliferative and proliferative DR, some data show improvement of DR by blocking of production and effects of Epo in the eye (25).

6
Normal or low concentrations of erythropoietin are found in conditions of primary polycythemia, some erythropoietin-independent anemias, but also in kidney-derived anemia (26). The importance of reduced concentration of erythropoietin is clinically confirmed in early diabetic nephropathy, as it resulted in anemia which worsened diabetic retinopathy (27). Also, in curing anemia of renal origin, Epo given intravenously had positive effects on macular edema, had improved visual acuity in patients with DR, and also led to reduction of exudative maculopathy and proliferative changes of the disease (28,29).

Study group
This cross-sectional study included 90 examinees, over 50 years of age, 60 of them with The first group consisted of 60 patients suffering from DM type 2 who were treated for 10 years or more with oral, insulin or combined therapy, and who were referred to the University eye clinic for complete eye examination, and evaluation of possible diabetic ocular fundus changes The second group of 30 patients served as control group of healthy subjects who have not been diagnosed with any systemic disease or ophthalmic disease, except for the senile cataract. The study did not include patients with an asymmetric 7 finding on the eyes in terms of diabetic retinopathy, as well as patients with unregulated arterial hypertension, anemia, diabetic patients with severe kidney damage, and patients on recombinant erythropoietin therapy.
The following data were recorded: name, surname, age, gender, and duration of diabetes mellitus. The height of the intraocular pressure was measured by the applanation tonometer, where the reference values ranged from 10 to 21 mmHg. Best corrected visual acuity was determined by the Snellen chart, and expressed in decimal values (Table 1).
In the group of patients with diabetes, all patients had DM for more than 10 years. The presence of diabetes retinopathy and the progression of changes were determined.
Patients with DR were according to ocular fundus changes further classified according to the existing international classification into two main groups. As a non-proliferative or proliferative DR group, and further into one of three subgroups, depending on the progression of changes -mild, moderate and severe (Table 2.) (30). Special blood samples (5ml) were separated for the ELISA human erythropoietin immunoassay (Enzyme-Linked Immunosorbent Assay). They were frozen and stored at a temperature of -20 Celsius degrees within the hour of sampling and centrifugation, until the analysis took place. Epo was measured by the commercially available human Epo ELISA commercial kit by the manufacturer R&D Systems Quantikine IVD for the quantitative determination of human serum or plasma erythropoietin. According to the manufacturer, medicines should not affect the accuracy of the results. The sensitivity of this test is very high, less than 0.6 mIU/ml. In addition, from blood samples of examinees, there were implemented laboratory measurements of concentration of the glycated 8 hemoglobin (HgA1c) in serum.

Statistical analysis
All the data were statistically analyzed using statistical software Statistical Package for Social Sciences -SPSS 17 and were presented graphically and in tabular presentation.
Descriptive statistics, including median, arithmetic mean, and standard deviation (SD) were used to describe the studied parameters. Differences in distributions of individual parameters between study groups were analyzed using the parametric Student's t-test, or the nonparametric Mann-Whitney test in case a distribution showed a significant deviation, while Chi-square test was used for categorical data. Value of p<0.05 was considered statistically significant. Multiple regression analysis was used to form a model for predicting disease indicators. Table 1 shows basic characteristics of the examinees, and ophthalmic clinical and laboratory data of the parameters that were analyzed in the study. Out of total number of examinees, there were 48 women and 41 men, the average age was 62.65 years (in the range from 50 to 81 years).

Results
In all patients in the non-proliferative and proliferative DR group, existence of diabetes mellitus type 2 was proven, as well as the presence and type of diabetic retinopathy, according to the International Clinical Diabetic Retinopathy Disease Severity Scale -ICDRDSS ( Table 2).
The duration of diabetes mellitus in the NPDR group was 17.1 years (16±7.36 years), and in the PDR group it was 18.13 (16± 7.76 years). There was no statistically significant difference in the duration of diabetes mellitus between the observed groups (p=0.589).
The maximum duration of DM in the NPDR group was found in subgroups with an advanced form of illness18±2.83 years, followed by subgroups of moderate and mild form of NPDR.
In the PDR group, the longest duration of DM was recorded in a subgroup of patients with a mild form of disease of 20.5±6.63, followed by subgroups of moderate and severe PDR. In the group of subjects with NPDR concentration, HgA1c was 7.96%, and in the control group 5.43%. A statistically significant difference was found between the average value of HgA1c in the control group and NPDR, p=0.001, and between the control and PDR, p=0.001. While a statistically significant difference was not found between the average values of HgA1c in the blood of patients with NPDR and PDR, p=0.599.
By examination of the fundus of the eye by direct or indirect ophthalmoscopy and biomicroscopy, the control group of the subjects excluded the existence of pathological changes in the retina, except changes in arterial hypertension, which were found in 67% of persons. The study did not include patients who were previously treated with laser photocoagulation, or who had any other eye treatment. The mean Epo concentration in the NPDR group was 7.0mIU/ml, the highest concentration was recorded in the moderate NPDR group, while the lowest was found in the subgroup of patients with advanced NPDR. The Kruskal-Wallis analysis did not show significant differences in the erythropoietin level in the observed patient subgroups, p=0.781.
Further, with the Mann-Whitney test, a statistically significant difference was not found in the mean value of erythropoietin by comparison with the observed NPDR: mild and moderate p=0.519, mild and advanced p=0.933, moderate and advanced p=0.686.
The average Epo concentration in the PDR patients group was 9.95 mUI/l. Using the Kruskal-Wallis test, a statistically significant difference in Epo values in PDR patients was found in relation to the stage of clinical picture, p=0.007. By using the Mann-Whitney test, statistically significantly higher concentrations of Epo in the blood of patients in subgroups were moderate and advanced in PDR, relative to the mild, p=0.001. Whereas, there is no statistically significant difference in the serum Epo concentration between the two most severe forms of PDR, p=1,000.
Statistically significantly higher concentrations of Epo in the blood were found in the subgroup of advanced PDR compared to: control group (p=0.004), but also each subgroup of NPDR, compared to mild NPDR (p=0.035), moderate NPDR (p=0.039), and advanced NPDR (p=0.022).

Discussion
A common complication of diabetes mellitus is diabetic retinopathy which, in addition to cataract, glaucoma and senile macular degeneration, is one of the leading causes of visual impairment and blindness in the world (31,32). Although investigated in numerous studies for decades, pathophysiological mechanism of the development of diabetic retinopathy has not been completely clarified. The eye, although a small organ, is by its metabolic properties most similar to the brain, and from the general circulation it is separated by blood-ocular barrier. In the vitreous body, an increased concentration of vasoproliferative, angiogenic substances was found in patients with diabetes mellitus. However, it is not entirely clear whether this is a consequence of local production due to the presence of eye ischemia and/or the systemic impact of these factors on the eye and its metabolic processes. In explanation of the onset of diabetic retinopathy nowadays, increasing attention is also attributed to systemic angiogenic factors, and their possible ability to promote and increase diabetic vascular changes and occurring of neovascularization of the retina (33,34).
Apart from vascular endothelial factor, as a main factor, erythropoietin is one of the most prominent other representatives of systemic angiogenic factors, and its role in the occurrence of neovascularization in the eye has not yet been fully determined and understood. It plays a role as a growth factor in normal processes of erythropoiesis, by stimulating erythrocytes, their proliferation and differentiation, and the prevention of the apoptotic death of erythroid precursors with erythropoietin receptors (9,13). However, in numerous studies, the presence of erythropoietin receptors has been confirmed in various tissues such as kidney, liver, uterus, retina (10,35). In some studies, in diabetic patients was found significantly higher erythropoietin concentration in the vitreous body, along with VEGF receptors, which is explained by local production of erythropoietin in tissue hypoxia in the eye, in conditions of elevated blood sugar levels, in the presence of other cytokines in DR as well (19,36).
In various studies has been shown that erythropoietin affects endothelial cells the same as vascular endothelial growth factor (VEGF), and for instance, in renal anemia diabetic patients treated with human recombinant erythropoietin (rhEpo), rhEpo exibits the same effect promoting increased intraocular angiogenesis and consequently impairment of DR (37,38). A statistically significant worsening of DR in the group of patients using rhEpo compared to the non-receiving group was also noted, and a direct proportionality of the serum erythropoietin concentration and the deterioration of diabetic eye disease was determined (39,40). Certain number of authors claim that the recombinant therapeutic Epo application allows better delivery of oxygen to ocular tissues, and reduction of diabetic changes after its local or systemic administration (41,42).
The role of Epo is still interesting for many researches nowadays. It is still unclear whether erythropoietin in DR has a protective or aggravating role, comparing results of various studies. Certainly, an increase in the number of erythropoietin receptors on retina cells in DR has been confirmed, which is considered as a compensatory response to tissue 13 hypoxia and hyperglycemia during DM. In these conditions, increased production and binding of Epo to erythropoietin receptors is the mechanism of survival of retinal nerve cells. It has been shown that erythropoietin inserted exogenously contributes to the preservation of the external blood-retinal barrier, acting on processes at the level of the retinal pigment epithelium. In addition, intravitreal injections of rhEpo have led to the inhibition of VEGF and stabilization of the function of the blood-retinal barrier, and also to a short-term positive effect on the chronic macular edema that has until then been practically refractory to any potential therapy (43).
During our study, we found that serum erythropoietin concentrations were increased directly proportional to the severity of the clinical stage of proliferative diabetic retinopathy. The highest average value of erythropoietin in serum was found in the group of subjects with the most severe forms of proliferative diabetic retinopathy (9.95 mIU/ml).
The lowest average concentration of EPO in serum (6.90 mIU/ml) was found in control group. The average concentration of Epo in serum in group of diabetics with NPDR was 7.00 mIU/ml. There was no statistically significant difference in average Epo concentration within main groups of healthy control subjects, and groups of patients with NPDR and PDR.

Consent for publication
All authors have read through the paper, and agreed upon consent for publication of this paper.

Availability of data and material
The datasets used and analyzed during the current study are available from the corresponding author, and available in supplement, as excel sheet file.

Competing interests
All authors confirm there is no competing interests, and no financial disclosure.

Funding
The data and material used in this study was obtained, analyzed and financed from resources of Clinical Center Vojvodina, Novi Sad, Serbia, with the approval of authorities.

Authors' contributions
SD conceived the idea for the study. SD, SJ, AM, and NB contributed to the design of the research. All authors were involved in data collection. DG, SB and AM analyzed the data.