Retinal Nerve Fiber Layer Thickness in Early-Stage Diabetic Retinopathy With Vitamin D Deficiency

Abstract

Purpose: To evaluate retinal nerve fiber layer (RNFL) thickness in early-stage diabetic retinopathy (DR) patients with and without vitamin D deficiency (VDD).

Methods: This study compared 50 early-stage DR patients with VDD (group 1) and 50 early-stage DR patients without VDD (group 2). All patients were examined by the same ophthalmologist. Mean RNFL thickness was determined by optical coherence tomography (OCT) performed by the two independent ophthalmologists for all subjects. Vitamin D levels were measured by using a radioimmunoassay. Vitamin D deficiency was defined, in accordance with the general standards, as a 25-hydroxyvitamin D (25(OH)D) level lower than 20 ng/mL.

Results: There were no significant differences between the groups in terms of age and sex distribution (P > 0.05). The mean serum 25(OH)D concentration of group 1 was significantly lower than that of group 2 (P < 0.001). The mean RNFL thickness of group 1 was significantly reduced compared to that of group 2 (P < 0.001). A significant relationship between the mean RNFL thickness and serum 25(OH)D concentrations was observed in group 1 (P < 0.001).

Conclusions: The results indicate that vitamin D functions as a neuroprotective component for optic nerves. Low serum 25(OH)D concentrations contribute to RNLF thinning in early-stage DR patients with VDD.

Diabetic complications include microvascular complications such as retinopathy, nephropathy, and neuropathy. Diabetic retinopathy (DR) is one of the most common causes of blindness worldwide.¹ In addition to vascular changes, the earliest stages of diabetic retinopathy feature neurodegenerative changes such as loss of ganglion cells, glial reactivity, and thinning of the retinal layers.² Moreover, in recent clinical and experimental studies^3–5 it has been observed that these neurodegenerative changes cause abnormalities in electroretinogram, contrast sensitivity, dark adaptation, and microperimetry. It has also been reported that DR is associated with thinning of the retinal nerve fiber layer (RNFL).^6,7 Similarly, histologic and immunohistochemical studies have demonstrated that DR influences retinal ganglion cells, horizontal cells, amacrine cells, and photoreceptors in the neural retina and results in a significant decrease in RNFL thickness.^8,9 Spectral-domain optical coherence tomography (SD-OCT) has been used to show that RNFL thinning in DR is due to ganglion cell loss.¹⁰ 

Many studies have shown the nonskeletal effects of low serum concentrations of 25-hydroxyvitamin D (25(OH)D); for example, low concentrations of 25(OH)D are associated with poor visual acuity in older adults.¹¹ The most common cause of poor visual acuity related to low serum 25(OH)D concentrations is age-related macular degeneration (AMD) in older adults.^11,12 Several studies^13,14 have found an association between vitamin D deficiency (VDD) and the pathogenesis of AMD. It has also been hypothesized that low 25(OH)D concentrations may result in optic neuropathy through a decreased neuroprotective effect, thus leading to poorer visual acuity in older adults.¹⁵ Furthermore, it has been shown that vitamin D can regulate the renin–angiotensin system as well as retinal microvessel circulation by improving endothelial cell–dependent vessel vasodilatation.¹¹ 

The evaluation of RNFL thickness is a sensitive indicator for determining optic nerve health and nerve injury.¹⁶ Optical coherence tomography, a noninvasive technique for cross-sectional tomography imaging of the retina and optic nerve, is used for measuring RNFL thickness.¹⁷ The RNFL thickness measurement is a valuable tool for the early diagnosis of optic neuropathies and as a method for monitoring the progression of glaucoma.^18,19 

The purpose of this study was to evaluate differences in mean RNFL thickness between early-stage DR patients with and without VDD, by using OCT. 

Methods

Fifty early-stage DR patients with VDD (group 1) and 50 early-stage DR patients without VDD (group 2) were included in this study. All participants attending the Atatürk University Medical Centre of Diabetes between November 2014 and February 2015 were included in our study. Approval of this study was granted by the Atatürk University Ethics Committee. All participants were given detailed information concerning the purpose and procedures of the study in accordance with the Declaration of Helsinki and all participants provided informed consent before their participation in this prospective study. 

Patients who had type 2 diabetes for more than 5 years, early-stage diabetic retinopathy, and visual acuity of 6/10 or better were included in the study. The exclusion criteria for both study groups included the following: refractive error of more than spherical equivalent (SE) +5 or SE −8 diopters in at least one eye; presence of cataract; a cup-to-disc (C/D) ratio of more than 0.3 and a C/D ratio between both eyes exceeding 0.2; glaucoma or ocular hypertension in either eye or in a family member's eyes; presence of secondary causes of glaucoma; severe corneal opacity; pathology in the anterior chamber angle and high myopia; previous refractive surgeries or intraocular surgery; macular degeneration; epiretinal membrane; vitreous/preretinal hemorrhage; moderate/severe nondiabetic retinopathy and proliferative diabetic retinopathy; diabetic macular edema; optic neuropathy; retinal vein occlusion; or hypertensive retinopathy. Thirty patients who had these exclusion criteria were excluded from this study. 

The diagnosis of diabetes mellitus (DM) was verified by clinical examinations and laboratory analyses. All patients underwent ophthalmologic examination including a best-corrected visual acuity (commonly refers to the clarity of vision), measurement with the Snellen visual acuity chart, slit-lamp biomicroscopy, intraocular pressure (the fluid pressure inside the eye) measurement, dilated fundoscopy with 90-D fundus lens, OCT, fundus photography, and fundus fluorescein angiography. After mydriasis induction with tropicamide eye drops, the ophthalmologic examinations were performed by the same retina specialists. Diabetic retinopathy was graded by the same ophthalmologists blinded to 25(OH)D status, as based on the Early Treatment Diabetic Retinopathy Study (ETDRS) grading system.¹⁴ 

An OCT examination (RTVue SD-OCT; Optovue, Inc., Fremont, CA, USA) with macular thickness mapping and peripapillary RNFL thickness measurements was conducted in all eyes by the two independent ophthalmologists blinded to 25(OH)D status. The ONH (optic nerve head map) protocol was used in this study. This instrument has a light source of 840-nm wavelength. Each eye was examined after pupillary dilation. The circumpapillary RNFL thickness in four quadrants (temporal, nasal, superior, and inferior) was measured for all patients. The global RNFL thickness was obtained by calculating the mean of the total 360° RNFL thicknesses. 

Blood samples were obtained after a 12-hour fast. Vitamin D levels were measured by radioimmunoassay. Vitamin D deficiency was defined, in accordance with the general standards, as 25(OH)D levels lower than 20 ng/mL.²⁰ 

Statistical analyses were performed using Windows SPSS (version 11.5; SPSS, Inc., Chicago, IL, USA). Statistical significance was calculated by using the independent samples t-test. A P value of <0.05 was regarded as statistically significant. All results were stated as the mean and standard deviation (mean ± SD). 

Results

All 100 patients had early-stage DR; 61% of patients in group 1 and 50% in group 2 were male. The mean ages of the participants in groups 1 and 2 were 60.05 and 59.00 years, respectively. Clinical and demographic characteristics of both groups are listed in Table 1. There were no significant differences between the groups in terms of age and sex distribution, visual acuity, or intraocular pressure (P > 0.05; Table 1). 

Table 1

 

Clinical and Demographic Characteristics of All Groups

 

The mean serum 25(OH)D concentration of group 1 was 18.05 ± 8.90 ng/mL, while the mean serum 25(OH)D concentration of group 2 was 33.38 ± 7.55 ng/mL. A significant difference was observed in the mean serum 25(OH)D concentration in group 1 compared to group 2 (P < 0.001; Table 1). In group 1, the mean RNFL thickness in the right and the left eye was 94.568 ± 8.95 μm and 93.989 ± 8.69 μm, respectively. In group 2, the mean RNFL thickness in the right and the left eye was 112.703 ± 11.94 μm and 112.281 ± 12.55 μm, respectively. There were significant differences between the groups in terms of the mean RNFL thickness (P < 0.001; Figs. 1, 2; Table 2). 

Figure 1

 

Retinal nerve fiber layer loss in the right eye of a patient with vitamin D deficiency by using OCT.

Figure 2

 

Normal RNFL thickness in the right eye of a patient without vitamin D deficiency by using OCT.

Table 2

 

Comparison of RNLF Parameters Between Group 1 and Group 2

 

We also confirmed that there was a significant correlation between the mean RNFL thickness of the right eye and serum 25(OH)D concentrations (R = 0.74, P < 0.001; Fig. 3), and there was a significant correlation between the mean RNFL thickness of the left eye and serum 25(OH)D concentrations in group 1 (R = 0.88, P < 0.001; Fig. 4). 

Figure 3

 

Correlation between the mean RNFL thickness in right eye and serum 25(OH)D concentrations in early-stage DR patients with vitamin D deficiency (R = 0.74, P < 0.001).

Figure 4

 

Correlation between the mean RNFL thickness in left eye and serum 25(OH)D concentrations in early-stage DR patients with vitamin D deficiency (R = 0.88, P < 0.001).

Discussion

Diabetic retinopathy causes not only capillary degeneration but also neurodegenerative changes and reduction in thickness of the inner retinal layers.^21,22 Many clinical and experiential studies^23,24 have provided evidence of neuronal abnormalities and different types of apoptotic neuronal cells in all retinal layers in the early stages of DR. Furthermore, an in vitro study²⁵ has demonstrated that diabetes affects both retinal neurons and glial cells. Retinal ganglion cell apoptosis is observed even before the onset of retinopathy, and the location of apoptotic cells in the retina during the early stage of diabetes is independent of the location of the blood vessels. 

Recently, thinning of the RNFL, inner retinal layers, has been associated with neural apoptosis and loss of ganglion cell in DR as compared with control subjects.^1,4 A more recent study⁹ has demonstrated that neurodegenerative changes, including loss of ganglion cell bodies and glial reactivity, cause thinning of the inner retinal layers in type 2 diabetic patients. Similarly, another study²⁶ reports that DR is associated with thinning of RNFL and that the average peripapillary RNFL layer at the superior region, as measured by OCT, is slightly thinner in diabetics before abnormal vascular manifestations occur, as compared with a nondiabetic control group. Our study measured the RNFL thickness, inner retinal layer, as an indicator of early sign of neurodegeneration in DR. 

Recent studies have suggested that 1α,25-dihydroxyvitamin D3 (1,25-(OH)2D3) is necessary for normal insulin secretion and that 1,25-(OH)2D3 directly affects the increase in intracellular channels and insulin secretion in beta cells. Furthermore, 1,25-(OH)2D3 increases insulin secretion and improves glucose tolerance.²⁷ Plasma 1,25-(OH)2D3 concentration is decreased in diabetic rats as well as in animals with VDD.²⁸ Recent studies²⁹ have also shown a relationship between 1.25-(OH)2D3 levels and retinopathy stages and a relationship between low levels of 25(OH)D and proliferative retinopathy. These recent findings indicate that VDD is an important factor for the emergence and progression of various DR stages.^30,31 Several experiments have confirmed that vitamin D regulates the neuroprotective effect of the central nervous system, which contains the optic nerves.³² 

Furthermore, these studies^32,33 suggest that vitamin D levels affect intraocular pressure and ocular blood flow during the development of chronic metabolic diseases such as diabetes, hypertension, and dyslipidemia. Therefore, it is believed that reduced vitamin D levels may be related to chronic optic neuropathy. Many studies have shown that vitamin D has an influence on peripheral and microvessel circulation by regulating the renin–angiotensin system and improving endothelial cell–dependent vessel vasodilatation. These studies^33,34 have suggested that vitamin D can decrease the risk of developing open-angle glaucoma by improving ocular blood flow, which is regulated by both endothelial function and the renin–angiotensin system. Annweiler et al.¹¹ suggest that vitamin D is a neuroprotective neurosteroid hormone with anti-inflammatory and anti-ischemic properties. Since both inflammation and ischemia are potent mechanisms promoting optic neuropathy, vitamin D may assist in preventing this condition. Age-related hypovitaminosis D may contribute to nerve degeneration and could play a role in the atrophy of retinal ganglion cells in the optic chiasm.¹¹ 

We found that the mean RNFL layer was thinner in early-stage DR patients with VDD than in early-stage DR patients without VDD, and a significant correlation between the mean RNFL thickness and serum 25(OH)D concentrations was observed in early-stage DR patients with VDD. We demonstrated that vitamin D functions as a neuroprotective component for the central nervous system, including the optic nerves. Low serum 25(OH)D concentrations contribute to RNLF thinning. However, this study was limited in its ability to conclude that VDD is directly related to the development of RNLF thickness in early-stage DRP patients with VDD. Therefore, the direct role of vitamin D should be examined by further prospective clinical studies that use treatment with vitamin D. 

Acknowledgments

The authors had no financial or material support for this manuscript. The authors alone are responsible for the content and writing of the paper. 

Disclosure: A. Gungor, None; O. Ates, None; H. Bilen, None; I. Kocer, None