Influence of inflammatory plasma biomarkers on choroidal thickness in type 2 diabetes mellitus

Abstract

Purpose

To assess the influence of inflammatory plasma biomarkers on choroidal thickness (CT) in patients with type 2 diabetes (T2D).

Methods

Cross-sectional study enrolling T2D patients and age-matched healthy controls (>55 years of age, Caucasian, axial length <26 mm, no macular edema, and naïve). Patients were examined with swept-source OCT Triton, obtaining automatic measurements. CT was analyzed using the ETDRS grid and the recently proposed choroidal division. A blood analysis was commanded: general biochemical profile, liver status, T2D status, thyroid and parathyroid activity, coagulation, general immunological profile, and inflammatory biomarkers.

Results

124 eyes of 124 patients with a mean age between 66 and 68 years were examined. The new choroidal division showed differences between groups (p < 0.05) in more sectors than the ETDRS grid, and more biomarkers influenced these new sectors. T2D patients had higher levels of IL-8, TNF-α, MCP1, adiponectin and L-selectin. CT was influenced by TNF-α, IL-17, leukocytes and erythrocyte sedimentation rate, as well as by HDL cholesterol, albumin, liver function biomarkers, and TSH. HbA1c showed little influence on CT.

Conclusions

T2D patients present increased plasma inflammatory biomarkers, exhibiting an influence on CT. IL-17 is related to a thicker choroid but TNF-α is related to a thinner choroid. HbA1c has little influence on CT. The recently proposed choroidal division is more sensitive to CT changes than the ETDRS grid. Some sectors are more sensitive to plasma biomarkers.

Keywords

Choroidal thickness type 2 diabetes mellitus inflammation plasma biomarkers swept-source optical coherence tomography

Introduction

Type 2 diabetes mellitus (T2D) is characterized by peripheral insulin resistance and is caused by genetic and environmental factors such as aging, sedentary lifestyle, and obesity. As a result, enzymatic glycosylation, polyol pathways, and nonenzymatic glycosylation occur.¹ In addition, expansion of adipose tissue is associated with the accumulation of activated macrophages expressing a number of proinflammatory genes.² Thus, T2D is a disease with multisystemic affection and various pro-inflammatory and angiogenic mediators related to its pathogenesis.³ This low-grade systemic inflammation implies an increase in plasma concentrations of cytokines.⁴

Diabetic retinopathy (DR) is a microvascular affection of T2D and the role of inflammation and angiogenesis has been verified in its pathogenesis. Mediators have been detected in serum, aqueous humor, vitreous, retina, and tears.⁵ This chronic inflammation is responsible for not only the progression of the different stages of DR, but also for major complications such as diabetic macular edema (DME) and neovascularization.⁶ These mediators include: C-reactive protein (CRP), tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, IL-8, vascular endothelial growth factor (VEGF), intracellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), insulin-like growth factor-1 (IGF-1), monocyte chemotactic protein-1 (MCP-1), and chemokine ligand-5 (CCL5).

Swept-source optical coherence tomography (SS-OCT) allows better definition of choroidal boundaries and a deep analysis of choroidal thickness (CT). It can be measured manually in linear OCT slabs or automatically in the ETDRS grid. There was also a recent attempt to create a choroidal pattern using automatic measurements.⁷ Previous studies have reported decreased CT accross the entire macula in patients with DR.^7–9 A recent meta-analysis suggested that CT is already reduced prior to development of DR, and that decreased CT could be a risk factor in its development.¹⁰ Nevertheless, its pathophysiological role in DME is not yet clear, and both thickening and thinning may be observed.^11–13 The choroid also varies with other pathologies and conditions, such as age,¹⁴ glaucoma,¹⁵ high myopia,¹⁶ age-related macular degeneration¹⁷ and pathologies classified within the pachychoroid spectrum.¹⁸

Variations in CT associated with inflammatory and angiogenic mediators in T2D patients have not yet been studied extensively. They would be expected to have an influence on CT because the choroid is mainly composed of blood vessels, and vascular dysfunction has been verified in T2D. Thus, the aim of this study is to assess the possible influence of plasma inflammatory and angiogenic biomarkers on CT in patients with T2D without DME.

Methods

A cross-sectional study approved by the Ethics Committee (PI15/0102) was performed at the Miguel Servet University Hospital, Zaragoza (Spain) between December 2015 and July 2016. Patients were informed about the study and written informed consent was obtained from all of them.

Two study groups were enrolled. The first group comprised diabetic patients fulfilling the ADA (American Diabetes Association) diagnostic criteria for T2D.¹⁹ The second group comprised age-matched healthy control subjects. The inclusion criteria for both groups were: age ≥ 55 years, Caucasian, and axial length (AL) < 26 mm. Systemic exclusion criteria were the presence of autoantibodies, other types of diabetes, pregnancy or puerperium, history of cancer, any other endocrine disease, other syndromes with multisystemic affection, and systemic treatment with corticosteroids, immunosuppressive drugs, or biopharmaceuticals. Ocular exclusion criteria were: DME, glaucoma and ocular hypertension, history of anterior, intermediate or posterior uveitis, retinal inflammation and dystrophies, any other ophthalmological pathology, cataract surgery in the last three months, hazy media that could interfere with image quality, and history of previous laser photocoagulation, intravitreal injection, or any other ophthalmological treatment.

Patients underwent through a deep ophthalmological examination comprising slit-lamp examination, measurement of intraocular pressure (IOP) using Goldmann tonometry, and funduscopy. All these examinations were performed between 16:00 and 19:00 h by the same experienced ophthalmologist (FBP). AL was measured using IOLmaster 500 (Carl Zeiss, Jena, Germany). CT values were obtained using Deep Range Imaging (DRI) SS-OCT Triton (software version 1.1.4, Topcon Corp., Tokyo, Japan) through dilated pupils and under scotopic lighting conditions. Color fundus pictures were also acquired with this OCT to determine the severity of DR. Fovea-centered 6 × 6 mm macular cube analysis was performed and choroidal boundaries were automatically delimited by the internal algorithm. The retina was thus measured between the internal limiting membrane (ILM) and the outer segments/retinal pigment epithelium (OS/RPE), the retinal nerve fiber layer (RNFL) was measured between the ILM and the RNFL–ganglion cell layer (GCL), and the GCL was measured between the RNFL/GCL and the internal plexiform layer/internal nuclear layer (IPL/INL), RNFL + GCL thickness was also measured, and the choroid was delimited between the Bruch membrane (BM) and the choroid–scleral interface (CSI). Three OCT examinations were performed but only one was selected for analysis.

CT measurements were given in both the ETDRS grid and in a grid of 30 × 30 macular cubes of 200 × 200 μm each. Based on this 30 × 30-cube grid, CT values were transformed into the 10 CT zones of the choroidal division introduced by Bartol-Puyal et al.⁷

Blood samples were analyzed to determine plasma biomarkers for general biochemical profile, liver status, T2D status, thyroid and parathyroid activity, general immunological profile, coagulation, and inflammatory and angiogenic biomarkers. The first six test measures: glucose, glycated hemoglobin (HbA1c), urea, creatinine, glomerular filtration rate (GFR), Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), uric acid, sodium, potassium, chlorine, cholesterol (CL), high-density lipoprotein-cholesterol (HDL-c), low density lipoprotein-cholesterol (LDL-c), triglycerides (TG), apolipoprotein A1 (ApoA1), apolipoprotein B (ApoB), lipoprotein(a) (Lp(a)), homocysteine, pro-brain natriuretic peptide (ProBNP), albumin, aspartate transaminase (AST), alanine transaminase (ALT), alkaline phosphatase (ALP), gamma-glutamyltransferase (GGT), lactate dehydrogenase (LDH), C-peptide, insulin, homeostatic model assessment of insulin resistance (HOMA-IR), homeostatic model assessment of beta-cell function (HOMA-β), thyroid-stimulating hormone (TSH), free thyroxine 4 (free T4), sex hormone-binding globulin (SHBG), prostaglandin E2 (PgE2), somatostatin, leptin, iron, ferritin, calcifediol, calcium, phosphorus, parathyroid hormone (PTH), insulin-like growth factor 1 (IGF1), erythrocyte sedimentation rate (ESR), leukocytes, lymphocytes, neutrophils, hemoglobin, hematocrit, prothrombin time (PT), prothrombin activity, fibrinogen, and microalbumin/creatinine ratio. The latter tests measure: CRP, IL-1β, IL-6, IL-8, IL-17, TNF-α, VCAM-1, ICAM-1, epidermal growth factor (EGF), monocyte chemoattractant protein 1 (MCP1), stromal-cell derived factor 1 (SDF1), VEGF-A, CCL5, adiponectin, and L-selectin. All blood analyses were performed the day following the ophthalmological examination after an eight-hour fast.

Four study groups were established: 33 healthy individuals (group 1), 34 T2D patients without DR (group 2), 26 T2D patients with mild DR (group 3), and 31 T2D patients with severer DR (group 4). This last group comprised patients with moderate, severe and proliferative diabetic retinopathy (PDR).

Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) software for Windows (software version 20, IBM Corporation, Somers, NY, USA). Mean and standard deviations were calculated for all studied parameters. Comparisons were performed using the Kruskal-Wallis test, and the Mann-Whitney U test was used for post hoc analysis in the case of statistically significant differences. Multiple regression analyses (successive steps) were also performed including all plasma biomarkers and the stage of DR. In the case of a non-statistically significant constant, regression analysis was repeated while excluding the constant. Only T2D patients were included in regression analysis. Statistical significance was considered as p < 0.05 for all analyses.

Pictures were created using Photoshop CS6 (version 13.0, Adobe Inc., San Jose, CA, USA).

Results

124 eyes of 124 patients (mean age between 66 and 68 years) were included in the study. Demographic data are displayed in Table 1. There were no differences between groups in age, IOP or AL. There were only differences in VA between groups 1 (healthy individuals) and 4 (T2D patients with moderate or severer DR).

Table 1.

Demographic data.

	Healthy (group 1)	No DR (group 2)	Mild DR (group 3)	Moderate-severe DR (group 4)	p (Kruskal-Wallis)	p between groups (if Kruskal-Wallis significant)
	Healthy (group 1)	No DR (group 2)	Mild DR (group 3)	Moderate-severe DR (group 4)	p (Kruskal-Wallis)	1-2	1-3	1-4	2-3	2-4	3-4
Eyes, number	33	34	26	31	N/A
Age, years	66.83 ± 7.31	68.47 ± 8.35	67.57 ± 7.87	67.84 ± 7.89	.862
VA, logMAR	.04 ± .05	.08 ± .12	.10 ± .20	.13 ± .13	.039*	1.00	.513	.029*	1.00	.535	1.00
IOP, mmHg	16.06 ± 2.40	16.74 ± 3.04	15.96 ± 3.33	17.32 ± 2.98	.241
AL, mm	23.95 ± 1.68	23.52 ± .80	23.04 ± .99	23.54 ± .96	.315

DR: diabetic retinopathy; N/A: not applicable; VA: visual acuity; IOP: intraocular pressure; AL: axial length.

Significant differences are marked with *.

Retinal and choroidal thicknesses of each layer are displayed in Table 2. Mean image quality was greater than 47 and there were no differences between groups. When using the ETDRS grid to compare retinal thickness, we only found differences between group 2 (T2D patients without DR) and group 4 in the outer superior sector of the RNFL. Meanwhile, we found differences between group 1 (healthy individuals) and group 4 in nearly all the sectors, with group 4 presenting thinner measurements. The only ETDRS sector not to exhibit differences was the outer nasal.

Table 2.

Retinal and choroidal thicknesses by layer.

	Healthy (group 1)	No DR (group 2)	Mild DR (group 3)	Moderate-severe DR (group 4)	p (Kruskal-Wallis)	p between groups (if Kruskal-Wallis significant)
	Healthy (group 1)	No DR (group 2)	Mild DR (group 3)	Moderate-severe DR (group 4)	p (Kruskal-Wallis)	1-2	1-3	1-4	2-3	2-4	3-4
Retinal thickness in the ETDRS grid, μm
Center	239.48 ± 25.47	235.30 ± 29.42	242.81 ± 22.52	244.32 ± 29.97	.560
Inner temporal	293.31 ± 17.41	288.66 ± 17.54	293.31 ± 18.45	294.78 ± 19.41	.551
Inner superior	303.34 ± 17.33	298.72 ± 13.50	304.51 ± 18.54	303.48 ± 24.12	.606
Inner nasal	304.02 ± 19.48	300.27 ± 16.95	304.62 ± 17.09	304.82 ± 22.49	.750
Inner inferior	300.91 ± 16.65	298.39 ± 20.54	301.02 ± 17.94	302.89 ± 20.32	.819
Outer temporal	253.13 ± 16.69	246.74 ± 12.73	251.28 ± 16.69	254.45 ± 16.56	.211
Outer superior	268.79 ± 17.39	259.69 ± 13.28	263.23 ± 15.77	265.18 ± 24.34	.230
Outer nasal	282.11 ± 18.63	273.70 ± 13.69	276.13 ± 18.64	277.97 ± 22.96	.314
Outer inferior	258.18 ± 14.88	250.94 ± 12.36	253.12 ± 17.96	256.14 ± 20.36	.299
RNFL thickness in the ETDRS grid, μm
Center	6.51 ± 10.08	4.63 ± 5.20	5.03 ± 3.80	5.30 ± 5.72	.702
Inner temporal	20.51 ± 5.10	19.98 ± 6.18	19.55 ± 4.47	19.59 ± 4.70	.875
Inner superior	28.54 ± 5.99	27.39 ± 2.91	28.25 ± 4.40	30.56 ± 6.07	.083
Inner nasal	26.15 ± 16.64	23.75 ± 5.94	24.11 ± 3.98	25.14 ± 6.22	.758
Inner inferior	29.95 ± 10.04	28.71 ± 8.20	28.91 ± 5.29	30.15 ± 7.69	.861
Outer temporal	21.80 ± 3.24	21.17 ± 3.75	21.43 ± 3.27	21.53 ± 5.11	.930
Outer superior	40.71 ± 8.56	36.50 ± 6.24	38.92 ± 5.25	41.81 ± 9.37	.031*	.152	1.00	1.00	1.00	.034*	.928
Outer nasal	50.26 ± 16.83	45.96 ± 6.81	48.08 ± 7.52	50.86 ± 10.28	.281
Outer inferior	42.24 ± 9.26	38.37 ± 6.46	41.26 ± 7.58	41.41 ± 8.01	.210
GCL thickness in the ETDRS grid, μm
Center	44.79 ± 9.29	42.94 ± 9.55	44.85 ± 7.36	44.69 ± 9.78	.802
Inner temporal	85.96 ± 8.07	82.89 ± 8.21	84.62 ± 10.24	82.29 ± 11.62	.406
Inner superior	86.88 ± 7.99	84.94 ± 6.90	84.85 ± 10.98	83.08 ± 13.68	.404
Inner nasal	85.46 ± 11.27	83.44 ± 7.91	86.61 ± 8.01	85.49 ± 12.34	.665
Inner inferior	85.60 ± 8.66	83.65 ± 8.71	85.28 ± 11.23	84.12 ± 12.33	.853
Outer temporal	67.83 ± 6.53	64.78 ± 6.59	68.27 ± 7.40	65.96 ± 6.31	.142
Outer superior	62.28 ± 4.85	59.91 ± 6.25	61.99 ± 5.90	59.05 ± 7.62	.121
Outer nasal	66.04 ± 6.11	64.95 ± 6.19	66.19 ± 6.55	64.62 ± 9.12	.776
Outer inferior	59.48 ± 5.04	58.31 ± 7.36	59.62 ± 6.07	58.25 ± 7.19	.760
RNFL + GCL thickness in the ETDRS grid, μm
Center	51.31 ± 17.09	47.58 ± 14.13	49.88 ± 10.61	49.90 ± 14.52	.766
Inner temporal	106.48 ± 10.42	102.87 ± 9.24	104.17 ± 13.28	101.30 ± 14.34	.354
Inner superior	115.42 ± 10.31	112.33 ± 8.63	115.10 ± 13.55	112.60 ± 17.86	.686
Inner nasal	111.62 ± 14.91	107.20 ± 9.12	110.73 ± 9.75	109.81 ± 16.16	.541
Inner inferior	115.55 ± 11.38	112.36 ± 10.05	114.19 ± 13.51	113.26 ± 18.36	.801
Outer temporal	89.64 ± 9.02	85.96 ± 8.89	89.71 ± 9.49	86.88 ± 10.29	.277
Outer superior	102.99 ± 11.88	96.41 ± 11.51	100.91 ± 8.88	99.58 ± 16.15	.188
Outer nasal	116.30 ± 15.87	110.91 ± 10.80	114.28 ± 10.96	113.95 ± 17.63	.485
Outer inferior	101.71 ± 10.56	96.68 ± 11.59	100.88 ± 11.50	98.37 ± 14.23	.310
CT in the ETDRS grid, μm
Center	236.41 ± 67.29	192.97 ± 74.84	203.96 ± 74.13	174.76 ± 60.64	.005*	.070	.462	.003*	1.00	1.00	.696
Inner temporal	230.57 ± 60.65	185.91 ± 80.18	190.16 ± 65.78	164.08 ± 62.83	.002*	.050	.153	.001*	1.00	1.00	.914
Inner superior	238.30 ± 59.51	197.90 ± 69.60	211.24 ± 74.55	184.55 ± 74.01	.017*	.112	.835	.014*	1.00	1.00	.901
Inner nasal	218.69 ± 69.89	179.17 ± 73.47	192.59 ± 85.32	165.53 ± 67.81	.032*	.183	1.00	.028*	1.00	1.00	1.00
Inner inferior	225.99 ± 65.56	185.29 ± 83.49	185.45 ± 73.14	166.54 ± 68.95	.012*	.149	.221	.009*	1.00	1.00	1.00
Outer temporal	213.85 ± 55.32	169.39 ± 72.38	172.63 ± 55.16	150.48 ± 57.94	.001*	.021*	.068	<.001*	1.00	1.00	1.00
Outer superior	232.80 ± 66.39	192.59 ± 64.39	202.63 ± 72.90	181.23 ± 68.46	.019*	.100	.553	.017*	1.00	1.00	1.00
Outer nasal	170.07 ± 73.00	144.70 ± 68.82	151.10 ± 85.60	134.66 ± 70.97	.276
Outer inferior	204.06 ± 63.95	175.51 ± 72.31	169.83 ± 68.82	155.44 ± 62.33	.035*	.502	.322	.026*	1.00	1.00	1.00
CT in the new division, μm
1 nasal	170.87 ± 76.06	139.86 ± 53.53	148.13 ± 80.21	132.67 ± 64.56	.134
1 temporal	173.44 ± 65.61	139.96 ± 76.43	156.76 ± 69.98	138.52 ± 55.80	.129
2 nasal	182.83 ± 71.34	150.02 ± 53.80	150.87 ± 69.77	139.51 ± 58.88	.042*	.220	.344	.044*	1.00	1.00	1.00
2 temporal	185.84 ± 63.66	147.97 ± 73.41	153.89 ± 60.83	143.64 ± 54.13	.036*	.099	.351	.055	1.00	1.00	1.00
3 nasal	199.96 ± 67.67	164.03 ± 55.72	168.23 ± 71.14	152.75 ± 58.84	.022*	.130	.347	.021*	1.00	1.00	1.00
3 temporal	204.35 ± 63.76	162.02 ± 72.09	169.01 ± 62.63	157.29 ± 55.01	.015*	.046*	.223	.023*	1.00	1.00	1.00
4 nasal	221.32 ± 64.37	180.29 ± 58.45	188.38 ± 68.77	170.04 ± 57.71	.008*	.047*	.273	.008*	1.00	1.00	1.00
4 temporal	224.51 ± 62.17	180.79 ± 67.37	187.85 ± 62.06	174.22 ± 59.78	.007*	.029*	.161	.010*	1.00	1.00	1.00
5 nasal	238.64 ± 65.12	192.63 ± 60.00	202.92 ± 70.40	187.48 ± 61.08	.007*	.023*	.211	.011*	1.00	1.00	1000
5 temporal	243.48 ± 63.96	194.91 ± 64.80	206.18 ± 70.30	191.13 ± 60.89	.005*	.016*	.181	.010*	1.00	1.00	1.00

DR: diabetic retinopathy; N/A: not applicable; RNFL: retinal nerve fiber layer; GCL: ganglion cell layer; CT: choroidal thickness.

Significant differences are marked with *.

Analysis using the new choroidal division revealed several differences in comparison with the previous analysis. We found differences between groups 1 and 4 in nearly all sectors, with group 4 presenting thinner measurements. There were also differences between groups 1 and 2 in the central sectors, with group 2 presenting thinner choroid measurements. There were no differences between groups in sectors 1 nasal and 1 temporal, i.e. the choroid located most nasally and temporally.

Table 3 shows plasma biomarkers for general biochemical profile, liver status, T2D status, thyroid and parathyroid activity, general immunological profile, and coagulation. As expected, glucose and HbA1c levels were significantly lower in healthy individuals, as were HOMA-IR and HOMA-β. Renal biomarkers (urea, creatinine, GFR, and CKD-EPI) indicated poorer kidney function in group 4 than in group 1. All CL values were higher in healthy individuals and there were no clinically significant differences regarding liver status, coagulation, or thyroid and parathyroid activity, despite calcium being slightly higher in group 1 than in group 2. Neutrophils were significantly lower in group 1.

Table 3.

Plasma biomarkers for general biochemical profile, liver status, T2D status, thyroid and parathyroid activity, general immunological profile, and coagulation.

	Healthy (group 1)	No DR (group 2)	Mild DR (group 3)	Moderate-severe DR (group 4)	p*	p between groups**
	Healthy (group 1)	No DR (group 2)	Mild DR (group 3)	Moderate-severe DR (group 4)	p*	1-2	1-3	1-4	2-3	2-4	3-4
Glucose, mg/dl	94.00 ± 7.75	138.79 ± 33.83	153.35 ± 50.15	147.24 ± 44.76	<.001*	<.001*	<.001*	<.001*	.798	1.000	1.000
Urea, mg/dl	39.09 ± 8.51	49.12 ± 18.56	49.12 ± 26.59	55.00 ± 24.59	.022*	.279	.369	.015*	1.00	1.00	1.00
Creatinine, mg/dl	0.74 ± 0.16	0.99 ± 0.54	0.97 ± 0.31	1.12 ± 0.54	.005*	.096	.205	.003*	1.00	1.00	1.00
GFR, ml/min·1.73m²	88.68 ± 15.47	77.92 ± 26.21	76.76 ± 20.55	70.46 ± 24.35	.014*	.297	.248	.009*	1.00	1.00	1.00
CKD-EPI, ml/min·1.73m²	87.58 ± 10.76	76.69 ± 22.44	77.30 ± 19.49	71.01 ± 23.67	.011*	.155	.288	.007*	1.00	1.00	1.00
Uric acid, mg/dl	5.17 ± 1.03	5.85 ± 1.79	5.88 ± 1.77	6.27 ± 1.80	.063
Na, mEq/l	140.33 ± 1.67	139.76 ± 1.71	139.23 ± 2.21	139.21 ± 1.76	.057
K, mEq/l	4.39 ± 0.31	4.39 ± 0.48	4.38 ± 0.38	4.55 ± 0.45	.316
Cl, mEq/l	105.12 ± 1.73	102.58 ± 2.15	102.35 ± 2.13	102.90 ± 3.21	<.001*	<.001*	<.001*	.002*	1.00	1.00	1.00
CL, mg/dl	221.09 ± 24.56	178.91 ± 33.83	161.96 ± 32.36	161.72 ± 40.18	<.001*	<.001*	<.001*	<.001*	.315	.258	1.00
HDL-c, mg/dl	64.03 ± 15.76	48.45 ± 11.14	44.69 ± 8.93	43.67 ± 11.77	<.001*	<.001*	<.001*	<.001*	1.00	.932	1.00
LDL-c, mg/dl	139.18 ± 23.99	101.42 ± 28.19	90.00 ± 31.69	85.22 ± 27.36	<.001*	<.001*	<.001*	<.001*	.758	.155	1.00
Non-HDL-c, mg/dl	157.06 ± 25.08	130.45 ± 31.42	113.42 ± 25.30	117.76 ± 38.62	<.001*	.004*	<.001*	<.001*	.215	.634	1.00
TG, mg/dl	89.85 ± 40.24	144.94 ± 52.90	153.69 ± 105.40	161.07 ± 140.69	.010*	.094	.052	.016*	.094	1.00	1.00
ApoA1, mg/dl	187.45 ± 27.23	158.88 ± 29.14	154.12 ± 23.78	146.37 ± 29.65	<.001*	<.001*	<.001*	<.001*	1.00	.470	1.00
ApoB, mg/dl	103.30 ± 20.64	95.09 ± 22.56	87.15 ± 22.52	88.53 ± 26.45	.029*	.903	.052	.079	1.00	1.00	1.00
Lp(a), mg/dl	33.92 ± 39.94	40.77 ± 41.87	51.96 ± 68.83	55.24 ± 56.86	.351
Homocysteine, μmol/l	11.42 ± 6.44	12.54 ± 3.91	12.03 ± 4.20	12.94 ± 6.21	.701
ProBNP, pg/ml	103.81 ± 95.82	150.20 ± 226.56	367.40 ± 928.84	669.40 ± 1259.19	.018*	1.00	1.00	.025*	1.00	.051	.869
Albumin, g/dl	4.30 ± 0.244	4.33 ± 0.24	4.33 ± 0.23	4.20 ± 0.29	.164
AST, U/l	21.82 ± 6.60	24.31 ± 7.50	19.31 ± 3.52	25.41 ± 15.15	.069
ALT, U/l	17.15 ± 7.65	26.33 ± 18.15	19.46 ± 7.09	21.90 ± 10.06	.017*	0.013*	1.00	.715	.176	.869	1.00
ALP, U/l	72.03 ± 21.23	79.97 ± 23.14	73.27 ± 16.29	94.83 ± 73.64	.121
GGT, U/l	23.52 ± 15.33	45.18 ± 71.24	30.69 ± 16.71	58.24 ± 82.31	.080
LDH, U/l	182.30 ± 29.15	176.97 ± 27.60	168.69 ± 29.18	198.62 ± 36.88	.004*	1.00	.569	.238	1.00	.043*	.003*
HbA1c, %	5.67 ± 0.35	7.32 ± 1.64	7.61 ± 1.37	7.94 ± 1.31	<.001*	<.001*	<.001*	<.001*	1.00	.332	1.00
C-peptide, mmol/mol	8 521.97 ± 5 542.77	14 417.99 ± 13 972.05	8 392.14 ± 18 218.26	9 743.42 ± 7 840.06	.014*	.365	.022*	1.00	1.00	.882	.074
Insulin, ng/ml	7.43 ± 4.52	8.79 ± 4.46	8.40 ± 3.30	7.88 ± 4.06	.695
HOMA-IR	1.77 ± 1.17	3.00 ± 1.90	3.27 ± 2.00	2.99 ± 1.98	.012*	.042*	.058	.565	1.00	1.00	1.00
HOMA-β	84.74 ± 42.10	49.94 ± 30.72	55.63 ± 67.82	36.47 ± 23.29	.010*	.030*	.375	.098	1.00	1.00	1.00
TSH, μU/ml	1.98 ± 1.05	2.01 ± 0.88	1.84 ± 0.86	2.26 ± 1.72	.606
Free T4, ng/dl	0.91 ± 0.13	0.96 ± 0.14	0.96 ± 0.14	0.94 ± 0.14	.375
SHBG, nmol/l	68.90 ± 32.33	51.08 ± 25.32	47.88 ± 20.97	53.10 ± 28.58	.017*	.056	.025*	.414	1.00	1.00	1.00
PgE2, pg/ml	1 226.12 ± 806.18	1 086.15 ± 685.81	1 127.38 ± 704.91	955.00 ± 687.53	.533
Somatostatin, pmol/l	5.34 ± 4.23	10.27 ± 5.96	8.54 ± 4.32	9.48 ± 6.56	.002*	.002*	.174	.019*	1.00	1.00	1.00
Leptin, ng/ml	15.27 ± 13.42	16.08 ± 15.08	14.86 ± 12.70	23.86 ± 21.64	.134
Fe, μg/dl	84.58 ± 31.19	86.58 ± 34.10	79.23 ± 31.27	75.44 ± 33.28	.530
Ferritin, ng/ml	140.98 ± 109.13	178.52 ± 196.03	121.42 ± 108.52	201.54 ± 258.75	.331
Calcifediol, nmol/l	46.88 ± 22.69	47.58 ± 20.55	48.33 ± 20.52	44.39 ± 19.05	.902
Ca, mg/dl	9.26 ± 0.36	9.52 ± 0.38	9.36 ± 0.31	9.31 ± 0.39	.024*	.023*	1.00	1.00	.472	.132	1.00
P, mg/dl	3.28 ± 0.44	3.35 ± 0.52	3.60 ± 0.50	3.64 ± 0.51	.009*	1.00	.084	.029*	.335	.143	1.00
PTH, pg/ml	57.73 ± 27.63	59.56 ± 29.46	57.56 ± 28.31	72.04 ± 68.37	.491
IGF1, ng/ml	108.92 ± 39.16	127.28 ± 66.76	117.51 ± 58.45	103.41 ± 50.09	.333
ESR, mm/h	16.82 ± 12.63	17.56 ± 14.71	18.27 ± 16.63	23.76 ± 17.42	.288
Leukocytes, μl⁻¹	5 706.06 ± 1 300.71	7 987.88 ± 2 021.10	8 153.85 ± 2 389.01	8 000.00 ± 2 197.08	<.001*	<.001*	<.001*	<.001*	1.00	1.00	1.00
Lymphocytes, μl⁻¹	1 969.70 ± 545.14	2 436.36 ± 839.21	2 403.85 ± 812.64	2 144.83 ± 830.91	.051
Neutrophils, μl⁻¹	3 054.55 ± 892.36	4 678.79 ± 1 457.47	4 667.62 ± 2 121.50	4 879.31 ± 1 552.96	<.001*	<.001*	.001*	<.001*	1.00	1.00	1.00
Hemoglobin, g/dl	13.88 ± 1.14	14.37 ± 1.46	13.98 ± 1.44	13.33 ± 1.67	.046*	.978	1.00	.822	1.00	.031*	.582
Hematocrit, %	42.35 ± 3.54	44.16 ± 3.91	42.74 ± 4.41	41.16 ± 4.75	.044*	.476	1.00	1.00	1.00	.031*	.959
PT, s	10.70 ± 0.53	10.84 ± 0.77	11.33 ± 1.85	11.27 ± 0.73	.070
Prothrombin activity, %	105.52 ± 8.56	102.52 ± 15.66	99.21 ± 16.23	97.35 ± 9.94	.097
Fibrinogen, g/l	3.92 ± 0.49	4.35 ± 0.82	4.29 ± 0.62	4.48 ± 0.81	.012*	.079	.269	.012*	1.00	1.00	1.00
Microalbumin/ creatinine, mg/g	11.42 ± 10.96	174.16 ± 662.56	43.23 ± 105.98	279.11 ± 657.55	.115

DR: diabetic retinopathy; GFR: glomerular filtration rate; CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration; CL: cholesterol; HDL-c: high density lipoprotein-cholesterol; LDL-c: low density lipoprotein-cholesterol; TG: triglycerides; ApoA1: apolipoprotein A1; ApoB: apolipoprotein B; Lp(a): lipoprotein(a); ProBNP: pro-brain natriuretic peptide; AST: aspartate transaminase; ALT: alanine transaminase; ALP: alkaline phosphatase; GGT: gamma-glutamyltransferase; LDH: lactate dehydrogenase; HbA1c: glycated hemoglobin; HOMA-IR: homeostatic model assessment of insulin resistance; HOMA-β: homeostatic model assessment of beta-cell function; THS: thyroid-stimulating hormone; Free T4: free thyroxine 4; SHBG: sex hormone-binding globulin; PgE2: prostaglandin E2; CRP: C-reactive protein; PTH: parathyroid hormone; IGF1: insulin-like growth factor 1; ESR: erythrocyte sedimentation rate; PT: prothrombin time.

*Kruskal-Wallis analysis.

**Calculated if Kruskal-Wallis significant.

Significant differences are marked with *.

Plasma inflammatory and angiogenic biomarkers are displayed in Table 4. We only found differences in IL 8, TNF-α, MCP1, adiponectin and L-selectin.

Table 4.

Inflammatory and angiogenic biomarkers in the four study groups.

	Healthy (group 1)	No DR (group 2)	Mild DR (group 3)	Moderate-severe DR (group 4)	p	p between groups
	Healthy (group 1)	No DR (group 2)	Mild DR (group 3)	Moderate-severe DR (group 4)	p	1-2	1-3	1-4	2-3	2-4	3-4
CRP, mg/dl	0.28 ± 0.25	0.39 ± 0.58	0.47 ± 0.80	0.40 ± 0.35	.544
IL 1β, pg/ml	0.61 ± 0.51	1.36 ± 3.83	0.70 ± 0.63	0.52 ± 0.62	.428
IL 6, pg/ml	0.63 ± 0.39	0.99 ± 0.81	0.97 ± 0.98	1.89 ± 5.09	.337
IL 8, pg/ml	1.37 ± 1.03	2.99 ± 2.76	2.10 ± 1.13	3.07 ± 3.26	.015*	.038*	1.00	.037*	.886	1.00	.774
IL 17, pg/ml	2.44 ± 3.14	3.99 ± 7.15	1.85 ± 1.92	3.61 ± 9.94	.645
TNF-α, pg/ml	1.01 ± 0.55	2.16 ± 1.87	2.03 ± 1.24	2.22 ± 1.55	.004*	.011*	.056	.013*	1.00	1.00	1.00
VCAM-1, ng/ml	1 149.87 ± 331.50	1 177.47 ± 191.18	1 208.48 ± 232.35	1 208.33 ± 268.73	.804
ICAM-1, ng/ml	113.35 ± 44.61	142.76 ± 93.56	183.15 ± 51.09	139.57 ± 65.42	.173
EGF, pg/ml	8.24 ± 9.11	20.05 ± 36.13	35.98 ± 76.85	19.46 ± 46.32	.218
MCP1, pg/ml	331.90 ± 126.60	443.25 ± 193.57	503.41 ± 182.32	436.03 ± 175.07	.003*	.071	.002*	.142	1.00	1.00	.953
SDF1, pg/ml	6 635.54 ± 2 330.66	7 117.73 ± 2 829.32	6 947.58 ± 4 096.82	7 471.16 ± 4 210.75	.826
VEGF-A, pg/ml	161.66 ± 140.21	355.19 ± 898.57	150.66 ± 226.03	233.94 ± 466.14	.534
CCL5, pg/ml	23 955.24 ± 17 577.47	26 129.72 ± 21 710.76	34 832.20 ± 34 806.98	25 605.47 ± 27 668.61	.420
Adiponectin, ng/ml	16 505.25 ± 20 685.00	8 593.92 ± 8 076.23	6 006.65 ± 5 706.47	10 965.51 ± 11 933.98	.023*	.116	.023*	.685	1.00	1.00	1.00
L-selectin, ng/ml	994.26 ± 320.23	837.96 ± 387.43	741.20 ± 232.88	821.17 ± 339.23	.041*	.401	.035*	.311	1.00	1.00	1.00

DR: diabetic retinopathy; IL: interleukin; TNF-α: tumoral necrosis factor-α; VCAM1: vascular cell adhesion protein 1; ICAM1: intercellular adhesion molecule 1; EGF: epidermal growth factor; MCP1: monocyte chemoattractant protein 1; SDF1: stromal-cell derived factor 1; VEGF-A: vascular endothelial growth factor A; CCL5: chemokine ligand 5.

*Kruskal-Wallis.

**Calculated if Kruskal-Wallis significant.

Significant differences are marked with *.

Multiple regression analyses showed that not all plasma biomarkers had an influence on CT, and not all choroidal regions were influenced by the same plasma biomarkers. TNF-α was the biomarker that influenced the highest number of choroidal sectors in the ETDRS grid (central, inner temporal, inner inferior and outer inferior). Meanwhile, the outer nasal ETDRS sector was the one influenced by the highest number of biomarkers (ApoA1, ApoB, ALP, C-peptide, leukocytes, prothrombin time, IL1 7, ICAM1 and VEGFA). These findings are displayed in Table 5.

Table 5.

Multiple regression analyses between plasma biomarkers and choroidal thickness in the ETDRS grid.

CT, μm	Predictive variables (coefficient, p value)																		Constant (p)	R (R²)
CT, μm	Urea	UA	Cl	ApoA1	ApoB	ALP	LDH	C-P	Leu	Lym	PT	TNF-α	IL-6	IL-8	IL-17	ICAM1	SDF1	VEGFA	Constant (p)	R (R²)
Center												−25.198 (.008)							284.396 (<.001)	.575 (.331)
Inner temp												−32.185 (.001)	39.256 (.032)						244.399 (<.001)	.698 (.487)
Inner sup	−1.790 (.014)																−.022 (.005)		463.079 (<.001)	.700 (.491)
Inner nasal							−2.352 (.005)												604.914 (<.001)	.602 (.362)
Inner inf												−24.765 (.021)							267.053 (<.001)	.513 (.263)
Outer temp		−37.293 (<.001)	5.020 (<.001)											−22.208 (<.001)	26.573 (.003)		−017 (.001)			.988 (.977)
Outer sup
Outer nasal				.917 (<.001)	−1.042 (<.001)	−.369 (.038)		−26.752 (<.001)	.041 (<.001)		−18.062 (<.001)				23.367 (<.001)	−.001 (<.001)		.206 (<.001)	152.420 (.018)	.995 (.991)
Outer inf										.039 (.013)		−18.805 (.015)							142.294 (.003)	.723 .523

UA: uric acid; Cl: chlorine; ApoA1: apolipoprotein A1; ApoB: apolipoprotein B; ALP: alkaline phosphatase; C-P: C-peptide; Leu: leukocytes; Lym: lymphocytes; PT: prothrombin time; TNF-α: tumor necrosis factor α; IL: interleukin; ICAM1: intracellular adhesion molecule 1; SDF1: stromal-cell derived factor 1; VEGFA: vascular endothelial growth factor A.

Only significant outcomes are displayed.

Tables 6 and 7 display the findings of the multiple regression analyses between plasma biomarkers and the new choroidal division. As found in the ETDRS grid, not all plasma biomarkers had an influence on CT, nor were all choroidal regions influenced by the same plasma biomarkers. Additionally, the number of biomarkers influencing CT was considerably higher. The central choroidal sectors (5N and 5T), together with T1, were influenced by the fewest biomarkers. Based on this choroidal division, the biomarkers that influenced the highest number of choroidal sectors were HDL, albumin and IL 17.

Table 6.

Multiple regression analyses between plasma biomarkers and choroidal thickness in the new choroidal division (part 1).

	Predictive variables coefficient, p value																						C p	R R²
	Glu	UA	Na	K	Cl	HDL	Hom	Alb	ALP	LDH	HbA1c	C-P	TSH	T4	Ca	P	ESR	Neu	PT	M/C	IL17	CCL5
1N			−13.952 <.001			1.752 <.001	−1.737 <.001	21.009 .002	−974 <.001		13.406 <.001							.004 .001	−36.091 <.001		35.713 <.001		24.129 <.001	.999 .998
1T						1.758 .022																	444.584 .001	.509 .259
2N					−4.271 <.001	4.954 <.001	1.770 <.001	188.795 <.001	-.72 .017		.762 <.001	27.341 <.001	9.415 <.001	−224.149 <.001		9.830 <.001	1.271 <.001			-.006 .006			−448.462 <.001	1.00 1.00
2T	.506 .001	−41.262 <.001		100.831 <.001						−2.506 <.001					−66.051 <.001						11.502 .015	.001 .001	6.988 <.001	.980 .960

N: nasal; T: temporal; Glu: glucose; UA: uric acid; Na: sodium; K: potassium; Cl: chlorine; HDL: high density lipoprotein-cholesterol; Hom: homocysteine; Alb: albumin; ALP: alkaline phosphatase; LDH: lactate dehydrogenase; HbA1c: glycated hemoglobin; C-P: C-peptide; TSH: thyroid stimulating hormone; T4: free thyroxine 4; Ca: calcium; P: phosphorus; ESR: erythrocyte sedimentation rate; Neu: neutrophils; PT: prothrombin time; M/C: microalbumin/creatinine ratio; IL: interleukin; CCL5: chemokine-ligand 5; C: constant.

Only significant outcomes are displayed.

Table 7.

Multiple regression analyses between plasma biomarkers and choroidal thickness in the new choroidal division (part 2).

	Predictive variable coefficient, p value															C p	R R²
	Urea	GFR	Uric acid	HDL	Lpa	Alb	ALT	TSH	ESR	CRP	Leu	IL17	TNFα	SDF1	Adi	C p	R R²
3N				2.851 .004		132.070 .001		21.333 .031								−4.501 .001	.843 .711
3T	-.948 .019		−25.751 <.001				−4.069 <.001	30.451 <.001	−2.240 .006	40.702 .016						11.291 <.001	.944 .890
4N		−1.135 .002		3.719 <.001	−.780 <.001						.021 <.001	26.838 <.001	−13.403 .003		−.003 .014		.997 .993
4T											.014 .003	24.554 .007	−19.048 .001			93.994 .023	.825 .680
5N						68.469 <.001								−.012 .043			.976 .952
5T							−2.811 .037						−16.634 .023			306.906 <.001	.624 .690

N: nasal; T: temporal; GFR: glomerular filtration rate; HDL: high density lipoprotein-cholesterol; Lpa: lipoprotein a; Alb: albumin; ALT: alanine transaminase; TSH: thyroid stimulating hormone; ESR: erythrocyte sedimentation rate; CRP: C-reactive protein; Leu: leukocytes; IL: interleukin; TNF-α: tumor necrosis factor α; SDF1: stromal-derived factor 1; Adi: adiponectin; C: constant.

Only significant outcomes are displayed.

Figure 1 displays the findings of several regression analyses in both the ETDRS grid and the recently proposed choroid-specific grid. Choroidal sectors with a significant association with a plasma biomarker are colored in gray.

Figure 1.

Plasma biomarkers influencing choroidal thickness in the ETDRS grid and the recently proposed choroidal division in a right-eye model. Choroidal sectors in which regression analysis revealed significant influence (p < 0.05) are highlighted in gray.

Discussion

Inflammation plays an important role in the pathophysiology of T2D and this low-grade inflammation contributes to microvascular damage, which encompasses DR. As the choroid is mainly composed of blood vessels, diabetic choroidopathy could be a risk factor in development of DR,¹⁰ although no direct correlation between the two entities has been established yet. This may be attributed to the different characteristics of retinal and choroidal blood vessels.

CT is difficult to measure with spectral-domain (SD)-OCT. However, the development of enhanced-depth-imaging (EDI) technology and SS-OCT has made CT measurement much easier and more accurate. The repeatability and reliability of the Triton DRI OCT device have been demonstrated both in healthy individuals²⁰ and in T2D patients.²¹ Automatic measurements also reduce variability due to their predictability and minimal scan artifacts.²²

Some authors have found some level of retinal neurodegeneration,²³ although this may not always be detectable in the early stages of the disease.²⁴ Our study did not find any statistical differences between groups.

In our study, comparison using the ETDRS grid found differences in CT between healthy individuals and patients with moderate or severer DR. Previous studies had already confirmed decreased CT in diabetic patients, even prior to the development of DR.^7–9 It is possible that we did not find differences in the other study groups because of the reduced sample size. Meanwhile, comparison using the choroidal division introduced by Bartol-Puyal et al.⁷ showed differences in CT between healthy and T2D subjetctswithout DR in more choroidal sectors than the ETDRS grid. The most nasal and temporal zones showed no differences, while the central zones were the most sensitive to CT reduction. The distribution of choroidal sectors along the vertical axis instead of in a circular pattern may be the underlying reason why some differences could not be found using the ETDRS grid.

As expected, healthy individuals presented lower levels of glucose, HbA1c, HOMA-IR, and HOMA-β in plasma, and as well as presenting better kidney function. Kidney function is well known to be related to the degree of DR.^25,26 Nevertheless, we found lower CL, ApoA1 and ApoB levels in T2D patients, maybe due to the effect of the lipid-lowering medication that T2D patients are recommended to take in order to reduce the risk of cardio-vascular disease (CVD). In fact, pro-BNP is a biomarker for heart failure, whose levels were higher in those patients with severer DR. C-peptide has been proposed to be a biomarker of CVD, too.²⁷ Healthy individuals in our study presented lower plasma levels of C-peptide than patients with mild DR. There were no clinical differences regarding liver function, thyroid function, serum iron and ferritin, serum calcifediol, parathyroid function, or coagulation activity.

As part of the inflammatory pathogenesis, leukocytes²⁸ and neutrophils²⁹ were significantly higher in diabetic patients than in healthy individuals. As a result of dysregulated innate immunity, these produce higher levels of proinflammatory cytokines and adhesion molecules, despite reduced phagocytosis and bactericidal activity.²⁹ Hence, we found high serum levels of IL-8, TNF-α, and MCP1 in T2D patients. Some authors have found high levels of these cytokines both in serum and in aqueous and vitreous humors,⁵ while others have not.³⁰ In our study we only found higher serum levels of proinflammatory cytokines, but not of others such as CRP, IL-1β, IL-6, IL-17, VCAM-1, ICAM-1, EGF, SDF1, VEGFA, or CCL5. It may be that some cytokines only increase locally in retina or choroid while others do so in the entire body.

Adiponectin has been proposed as a protection factor against the development of DR, with anti-inflammatory effects.³¹ Although we did not find differences between groups, healthy controls presented higher plasma levels.

Multiple regression analysis produced different outcomes depending on the choroidal division. In the case of the ETDRS grid, the outer superior sector was the only one that was not influenced by any biomarker. TNF-α was the biomarker that influenced most choroidal sectors, followed by IL-17 and SDF1. Several others had some influence as well, but only in one sector. TNF-α and SDF1 were negatively related to CT, whereas higher levels of IL-17 implied a thicker choroid.

Regression analysis performed using the new choroidal division revealed a higher number of plasma biomarkers with some influence on CT. We could therefore posit that this division is more sensitive to variations in CT than the ETDRS grid. HDL cholesterol had a direct influence on nasal choroid (sectors 1N, 1T, 2N, 3N and 4N), as did homocysteine (sectors 1N and 2N) and albumin (1N, 2N, 3N and 5N). Poor liver function –higher levels of ALT and ALP– would imply a thinner choroid. High levels of TSH were associated with a thicker choroid, but T4 had little influence and could even imply thinning. Given that the quantity of leukocytes, neutrophils and IL-17 was associated with greater CT, inflammation would cause thickening. These findings remain in line with current knowledge about inflammation and the choroid. Pathologies with an inflammatory pathogenesis such as uveitis present a thicker CT.³² However, high serum levels of IL-17 have also been reported in patients with choroidal neovascularization (CNV),³³ in which choroidal thinning occurs. Conversely, a thinner choroid is associated with higher plasma levels of TNF-α. This is an inflammatory cytokine that contributes to angiogenesis and also stimulates the expression of VEGF and ICAM1 on the vascular endothelium.³⁴ However, to date, its role in CT has not been studied.

In our study we found that HbA1c has little influence on CT. Some previous authors did not detect any influence,³⁵ while others did.³⁶

The strength of this study is that including only naïve T2D patients eliminates possible bias due to retinal and choroidal changes secondary to previous treatment, such as photocoagulation³⁷ or antiVEGF.³⁸ Patients with DME were also excluded because of the potential interference with choroidal measurements. Excluding other pathologies associated with choroidal thinning or thickening also makes our results more reliable. Possible individual variations were minimized because all measurements were taken at the same day time and by the same operator.³⁹ As far as we know, this is the first study to analyze the influence of serum inflammatory biomarkers on CT in T2D patients.

A possible wWeakness is that it is possible that some differences could not be found due to the reduced sample size. Further studies with larger sample sizes should be performed in order to confirm these results.

T2D is characterized by an increase in plasma pro-inflammatory cytokines (IL-8, TNF-α, MCP1, adiponectin and L-selectin), and some of them have an influence on CT. IL-17 is related to a thicker choroid while TNF-α is related to a thinner choroid. HbA1c has little influence on CT. The recently proposed choroidal division is more sensitive to CT changes than the ETDRS grid when applied to the choroid. Some choroidal sectors are more sensitive to serum inflammatory biomarkers than others.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Francisco de Asís Bartol-Puyal

References

Guthrie

. Pathophysiology of diabetes Mellitus. Crit Care Nurs Q. 2004; 27: 113–125.

Kahn

Cooper

Del Prato

. Pathophysiology and treatment of type 2 diabetes: perspectives on the past, present, and future. Lancet. 2014; 383: 1068–1083.

Kolluru

Bir

Kevil

. Endothelial dysfunction and diabetes: effects on angiogenesis, vascular remodeling, and wound healing. Int J Vasc Med. 2012; 2012: 1–30.

Calle

Fernandez

. Inflammation and type 2 diabetes. Diabetes Metab. 2012; 38: 183–191.

Kaštelan

Orešković

Bišćan

, et al. Inflammatory and angiogenic biomarkers in diabetic retinopathy. Biochem Medica. 2020; 30: 1–15.

Gouliopoulos

Kalogeropoulos

Lavaris

, et al. Association of serum inflammatory markers and diabetic retinopathy: a review of literature. Eur Rev Med Pharmacol Sci. 2018; 22: 7113–7128.

Bartol-Puyal

FDA

Isanta

Ruiz-Moreno

, et al. Distribution of choroidal thinning in high myopia, diabetes mellitus, and aging: a swept-source OCT study. J Ophthalmol. 2019; 2019: 1–9.

Querques

Lattanzio

Querques

, et al. Enhanced depth imaging optical coherence tomography in type 2 diabetes. Investig Opthalmol Vis Sci. 2012; 53: 6017.

Esmaeelpour

Považay

Hermann

, et al. Mapping choroidal and retinal thickness variation in type 2 diabetes using three-dimensional 1060-nm optical coherence tomography. Invest Ophthalmol Vis Sci. 2011; 52: 5311–5316.

10.

Endo

Kase

Saito

, et al. Choroidal thickness in diabetic patients without diabetic retinopathy: a meta-analysis. Am J Ophthalmol. 2020; 218: 68–77.

11.

Kim

Lee

Joe

, et al. Changes in choroidal thickness in relation to the severity of retinopathy and macular edema in type 2 diabetic patients. Invest Ophthalmol Vis Sci. 2013; 54: 3378–3384.

12.

Eliwa

Hegazy

Mahmoud

, et al. Choroidal thickness change in patients with diabetic macular edema. Ophthalmic Surg Lasers Imaging Retina. 2017; 48: 970–977.

13.

, et al. Subfoveal choroidal thickness in diabetes and diabetic retinopathy. Ophthalmology. 2013; 120: 2023–2028.

14.

Ruiz-Medrano

Flores-Moreno

Peña-García

, et al. Macular choroidal thickness profile in a healthy population measured by swept-source optical coherence tomography. Investig Ophthalmol Vis Sci. 2014; 55: 3532–3542.

15.

Pablo

Cameo

Bambo

, et al. Peripapillary choroidal thickness analysis using swept-source optical coherence tomography in glaucoma patients: a broader approach. Ophthalmic Res. 2018; 59: 7–13.

16.

Zhang

Neitz

, et al. Geographic mapping of choroidal thickness in myopic eyes using 1050-nm spectral domain optical coherence tomography. J Innov Opt Health Sci. 2015; 8: 1550012.

17.

Park

Kang

M-J

Kim

, et al. Topographic changes in choroidal thickness in age-related macular degeneration during the development of active choroidal neovascularization. Retina. 2021; 41: 409–422.

18.

Cheung

CMG

Lee

Koizumi

, et al. Pachychoroid disease. Eye (Lond). 2019; 33: 14–33.

19.

American Diabetes Association. Standards of medical care in diabetes - 2013. Diabetes Care. 2013; 36: S11–S66.

20.

Matsuo

Sakamoto

Yamashita

, et al. Comparisons of choroidal thickness of normal eyes obtained by two different spectral-domain OCT instruments and one swept-source OCT instrument. Investig Ophthalmol Vis Sci. 2013; 54: 7630–7636.

21.

Abadía

Calvo

Bartol-Puyal

, et al. Repeatability of choroidal thickness measurements assessed with swept-source optical coherence tomography in healthy and diabetic individuals. Retina. 2019; 39: 786–793.

22.

Mansouri

Medeiros

Tatham

, et al. Evaluation of retinal and choroidal thickness by swept-source optical coherence tomography: repeatability and assessment of artifacts. Am J Ophthalmol. 2014; 157: 1022–1032.e3.

23.

Vujosevic

Midena

. Retinal layers changes in human preclinical and early clinical diabetic retinopathy support early retinal neuronal and Müller cells alterations. J Diabetes Res. 2013; 2013: 1–8.

24.

Schröder

Szendroedi

Benthin

, et al. German diabetes study - baseline data of retinal layer thickness measured by SD-OCT in early diabetes mellitus. Acta Ophthalmol. 2019; 97: e303–e307.

25.

Bartol-Puyal

FDA

Isanta

Calvo

, et al. Macro and microangiopathy related to retinopathy and choroidopathy in type 2 diabetes. Eur J Ophthalmol. 2022; 32: 2412–2418.

26.

Sabanayagam

Chee

Banu

, et al. Association of diabetic retinopathy and diabetic kidney disease with all-cause and cardiovascular mortality in a multiethnic Asian population. JAMA Netw Open. 2019; 2: e191540.

27.

Wang

Wan

Chen

, et al. Association of C-peptide with diabetic vascular complications in type 2 diabetes. Diabetes Metab. 2020; 46: 33–40.

28.

Forrester

Kuffova

Delibegovic

. The role of inflammation in diabetic retinopathy. Front Immunol. 2020; 11: 583687.

29.

Chen

. Diabetic retinopathy and dysregulated innate immunity. Vision Res. 2017; 139: 39–46.

30.

X-A

Zhou

X-F

, et al. Angiogenesis-related cytokines in serum of proliferative diabetic retinopathy patients before and after vitrectomy. Int J Ophthalmol. 2012; 5: 726–730.

31.

Gong

Löfqvist

, et al. Review: adiponectin in retinopathy. Biochim Biophys Acta. 2016; 1862: 1392–1400.

32.

Kim

Park

Y-H

. Choroidal vascularity index and choroidal thickness in human leukocyte antigen-B27-associated uveitis. Ocul Immunol Inflamm. 2019; 27: 1280–1287.

33.

Guo

Yin

Zheng

, et al. Cytokine profiling reveals increased serum inflammatory cytokines in idiopathic choroidal neovascularization. BMC Ophthalmol. 2019; 19. DOI: 10.1186/S12886-019-1101-6.

34.

Jasielska

Semkova

Shi

, et al. Differential role of tumor necrosis factor (TNF)-alpha receptors in the development of choroidal neovascularization. Invest Ophthalmol Vis Sci. 2010; 51: 3874–3883.

35.

Sahinoglu-Keskek

Altan-Yaycioglu

Canan

, et al. Influence of glycosylated hemoglobin on the choroidal thickness. Int Ophthalmol. 2018; 38: 1863–1869.

36.

Unsal

Eltutar

Zirtiloğlu

, et al. Choroidal thickness in patients with diabetic retinopathy. Clin Ophthalmol. 2014; 8: 637–642.

37.

Okamoto

Matsuura

Ogata

. Effects of panretinal photocoagulation on choroidal thickness and choroidal blood flow in patients with severe nonproliferative diabetic retinopathy. Retina. 2016; 36: 805–811.

38.

Kim

Lee

Chang

, et al. Short-term choroidal thickness changes in patients treated with either ranibizumab or aflibercept: a comparative study. Br J Ophthalmol. 2016; 100: 1634–1639.

39.

Gabriel

Esmaeelpour

Shams-Mafi

, et al. Mapping diurnal changes in choroidal, Haller’s and Sattler’s layer thickness using 3–dimensional 1060-nm optical coherence tomography. Graefe’s Arch Clin Exp Ophthalmol. 2017; 255: 1957–1963.