Inflammatory citokines in tears and plasma of patients with type 2 diabetes and healthy controls

Patients and Eligibility Criteria

This study includes 108 participants recruited from health centers in Málaga province, Spain, between May and June 2023. After applying inclusion and exclusion criteria, the final sample comprised 81 patients, categorized into the T2DM group (40 patients diagnosed with type 2 diabetes) and the control group (41 non-diabetic individuals).

Inclusion criteria required men and women aged 18 years or older who provided voluntary, informed consent. For the T2DM group, participants were selected from a patient list at a family medicine clinic using systematic, non-random sampling (every fourth patient with type 2 diabetes was selected; if a patient did not meet the inclusion criteria, the next individual on the list was chosen) to minimize selection bias. Patients in this group had a diagnosis of type 2 diabetes for over two years and were receiving medical treatment. The control group included non-diabetic participants selected from the same list, following identical sampling criteria.

Exclusion criteria included individuals unable to attend the health center, those with a confirmed or suspected active infectious disease at recruitment, a prior diagnosis of diabetic retinopathy, chronic inflammatory diseases (including autoimmune conditions), pregnant or breastfeeding women, individuals undergoing hormonal therapy, and those with cognitive impairments that could impede understanding of the study’s purpose and procedures.

Clinical Evaluations

All participants in the study were scheduled for a primary care consultation and evaluations were conducted by a healthcare team comprising two physicians and one nurse, who attended to all participants.

During the clinical interview, sociodemographic data were collected and confirmed, along with cardiovascular risk factors, including any history of myocardial infarction and stroke. Additionally, eligibility criteria were assessed, and after participants signed the informed consent form, biological samples were collected.

Sample Collection and Processing

Plasma

Blood samples were collected in the morning after participants had fasted for 8–12 h. Venous blood was drawn into 10 mL K2 EDTA tubes (BD, Franklin Lakes, NJ, USA) and immediately processed to obtain plasma. Specifically, the blood samples were centrifuged at 2200 ×g for 15 minutes at 4°C, after which the supernatant (plasma) was collected. Plasma samples were then individually characterized, logged, and stored at −80°C until further analyses. A small aliquot of plasma from each sample was tested for infectious diseases using commercial rapid tests for HIV, hepatitis B, hepatitis C (Strasbourg, Cedex, France), and SARS-CoV-2 (Bio-Connect, Huissen, The Netherlands).

Tears

Tear samples were collected from each eye of each patient during the same consultation session as the clinical evaluation and blood collection, using the Schirmer test with Schirmer-Plus® strips (GECIS, Neung sur Beuvron, France). The paper strips were placed in the inferior fornix of each eye without the prior application of topical anesthetic. Samples with less than 6 mm of moisture on the strip after 5 minutes were excluded. Tear samples were immediately frozen at -80°C until analysis, as previously reported [12].

Briefly, for protein elution, each trip containing tear fluid was cut into small pieces and immersed in 100 μL of phosphate-buffered saline (PBS) with 0.3% Tween® 20, 0.5% bovine serum albumin (BSA), and a protease inhibitor. The mixture was incubated overnight at 4°C, after which the supernatant was collected. Total protein content in each sample was quantified using a NanoDropTM One spectrophotometer (Thermo Fischer Scientific, Waltham, MA, USA) by measuring the absorbance at 280 nm.

Cytokine Determinations

The concentrations of inflammatory mediators in plasma and tear samples were quantified according to the protocol provided by the Bio-Plex Pro™ Human Cytokine 27-plex Assay kit (#M500KCAF0Y; Bio-Rad Laboratories, Hercules, CA, USA). This immunoassay is based on Luminex® MAGPIX® technology and was conducted at the IBIMA-BIONAND Platform Laboratory at the Málaga Technology Park in collaboration with the University of Málaga. A total of 27 inflammatory mediators were analyzed: interleukin (IL)-1β, IL-1ra, IL-2, IL-4, IL-5, IL-6, IL-7, CXCL8 (IL-8), IL-9, IL-10, IL-12p70, IL-13, IL-15, IL-17A, CCL11 (eotaxin-1), fibroblast growth factor basic (FGF basic), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), interferon (IFN)-γ, CXCL10 (IFN-γ-induced protein 10, IP-10), CCL2 (monocyte chemoattractant protein, MCP-1), CCL3 (macrophage inflammatory protein-1α, MIP-1α), platelet-derived growth factor (PDGF-BB), CCL4 (MIP-1β), CCL5 (regulated on activation normal T cell expressed and secreted, RANTES), tumor necrosis factor (TNF-α), and vascular endothelial growth factor (VEGF). The 96-well plates were measured using a Bio-Plex MAGPIX™ reader and Bio-Plex Manager™ MP software (Luminex, Austin, TX, USA) in the Proteomics Unit at the Central Research Support Services of the University of Málaga.

All samples were run in duplicate to enhance the reliability and accuracy of measurements and to minimize measurement bias based on previous determinations. For samples with an optical density (OD) lower than the limit of detection in the multiplex assay but higher than the background (zero values), the assignment of concentrations was arranged from the sample with the lowest OD, which was assigned with half of the minimum concentration that could be interpolated in the standard curves [12, 14]. The intra-assay coefficient of variability (CV) was less than 7% and the inter-assay CV was less than 8%. Concentrations of these inflammatory mediators were measured in pg/mL or ng/mL.

Statistical Analysis

Data were presented as the number and percentage of events [n (%)], mean and standard deviation (mean ± SD), or median and interquartile range [median (IQR, 25%-75%)], according to variable type and distribution.

The statistical significance of differences in categorical variables was assessed with Fisher's exact test, while differences in continuous variables were evaluated using either the Mann-Whitney U test for non-normally distributed variables or the Student’s t-test for normally distributed variables. To control the false discovery rate (FDR) arising from multiple comparisons between the T2DM and control groups, the Benjamini-Hochberg procedure was applied to calculate adjusted p-values (q-values).

Multiple correlation analyses between plasma and tear cytokine concentrations were conducted using the Spearman correlation coefficient (rho). Tear cytokine concentrations for each patient were calculated as the mean of both eyes, with a mean coefficient of variation <6%. Analysis of covariance (ANCOVA) (F statistic) was used to assess plasma and tear cytokines based on type 2 diabetes diagnosis, adjusting for independent variables and covariates to minimize bias. Raw cytokine concentration data were inverse-transformed to approximate a normal distribution and meet ANCOVA assumptions. Estimated marginal means and 95% confidence intervals (95% CIs) of the inverse-transformed cytokine concentrations were reported.

A post-hoc power analysis indicated that the final sample size (n=81) achieved approximately 80% power to detect a moderate-to-large effect (d≈0.63) at a 5% significance level, which is considered sufficient for identifying differences in cytokine profiles between the T2DM and control groups [15].

All statistical analyses were conducted using GraphPad Prism version 5.04 (GraphPad Software, San Diego, CA, USA) and IBM SPSS Statistics version 22 (IBM, Armonk, NY, USA). P-values less than 0.05 were considered statistically significant.