Peripheral white blood cell subtypes and the development/progression of diabetic macular edema in type 2 diabetic patients: a comparative study

Introduction

Diabetic macular edema (DME) is one of the characteristic manifestations of diabetes retinopathy (DR) and an important cause of vision loss in patients with diabetes. Anti-vascular endothelial growth factor (anti-VEGF) injections are currently the first-line treatment option for improving vision in DME; however, 40% of DME patients do not respond or respond poorly to anti-VEGF therapy (1). The macula is a highly specialized area of the central retina and responsible for high-acuity vision. Chronic macular edema can lead to irreversible damage of the retinal structure, so a simple and convenient marker is urgently needed to predict the occurrence and development of DME. Early detection of the relevant risk factors of DME, close follow-up and strict blood glucose control, and timely symptomatic treatment are wise choices.

With the progress of DR, the progressive increase of systemic neutrophils can lead to an increase in the number of neutrophils in the retina, leading to sustained damage to the retinal vascular system and destruction of the blood retinal barrier (BRB), and DME is the direct consequence of BRB destruction (2). Lymphocytes play a key role in host defense against non-specific injury. A study has shown that lymphocytes play a protective role in vascular lesions caused by type 2 diabetes (3). Our previous study of patients with severe DR showed that lymphocyte percentage (%lymphocytes) was closely related to the development of DME (4).

White blood cells (WBC) and their subtypes have long been used as biomarkers (5-7) and in the field of ophthalmology, the neutrophil-to-lymphocyte ratio (NLR) is a potential marker of neovascular glaucoma (8), which is closely related to the progression of DR (9) and can predict the effectiveness of anti-VEGF therapy for DME (10). However, there is no report on the correlation between leukocytes and their classification and DME.

The purpose of this study is to determine the relevant factors of the occurrence and development of DME through the detection of cheap WBCs and their classification, and apply them to the future clinical work to minimize the visual loss caused by DME. We present the following article in accordance with the STARD reporting checklist (available at https://apm.amegroups.com/article/view/10.21037/apm-22-962/rc).

Methods

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by ethics committee of Nantong First People’s Hospital (No. 2022KT108). Individual consent for this retrospective analysis was waived.

Subjects

DME patients who were treated in the hospital’s ophthalmology department and type 2 diabetes mellitus (T2DM) patients hospitalized in the endocrinology department from January 2019 to December 2020 were enrolled in this retrospective study. Demographic data, ophthalmic examination results, and laboratory measurements were retrieved from the medical records.

All T2DM patients were diagnosed according to the diagnostic criteria in the Chinese Diabetes Society Guidelines for the Prevention and Treatment of Type 2 Diabetes in China (2013) (11). Because all the DME patients were treatment-naive and the number of patients with severe/atrophic diabetic macular degeneration was small, we divided the DR patients according to the Early Treatment Diabetic Retinopathy Study (ETDRS) classification system and the optical coherence tomography (OCT)-based grading of diabetic maculopathy proposed by the European School for Advanced Studies in 2019 (12) into a No-DME group (n=33) and DME group (n=81); the DME group was further subdivided into early DME group (n=33) and late DME group (n=48).

The inclusion criteria are as follows: (I) age ≥18 years; (II) diagnosed with T2DM; and (III) T2DM with DR.

The exclusion criteria were as follows: (I) with cancer, endstage renal failure requiring dialysis, disabling stroke, coronary heart disease, tuberculosis, liver disease, blood disease, and/or rheumatoid diseases; (II) clinical evidence of any acute or chronic inflammation or infection; (III) systemic or topical use of non-steroidal anti-inflammatory drugs and/or steroids; (IV) history of prior ophthalmic surgery or intraocular inflammation or ischemia due to conditions other than DR; (V) having received laser therapy or intravitreal injections of anti-vascular growth factor drugs in both eyes due to DR or other conditions; and (VI) refractive interstitial clouding that affected fundus examination.

Examinations

Ocular examination: All patients underwent a detailed baseline ophthalmic examination including intraocular pressure measurement (Goldman applanation tonometry), best-corrected visual acuity (BCVA) measurement, slit-lamp biomicroscopy, and mydriatic fundus examination. Patients with T2DM and DR were selected according to the ETDRS criteria.

OCT: Spectral domain OCT (SD-OCT) was performed for all DR patients, with 6-mm OCT scans centered on the fovea (12 scans, with an interval of 15°). The following items were recorded: (I) central subfoveal thickness (CST) and macular volume (MV); the normal CST range within 1 mm of the fovea is 225–315 µm, and the normal MV range is 0.17–0.26 mm; (II) intraretinal cysts (IRCs), which are circular micro-reflective spaces within the neurosensory retina and may be located in the outer nuclear layer, inner nuclear layer, or ganglion cell layer. IRCs are graded according to height (13): mild: IRC ≤380 µm; and severe: IRC >380 µm; (III) external limiting membrane and ellipsoid zone, which are the first and second highly reflective bands on the outermost layer of OCT and can be graded according to their recognizability in the fovea: grade 0: intact, grade 1: partial destruction, grade 2: total destruction; and (IV) disorganization of the inner retinal layers (Dril), which refers to poorly defined borders between the foveal ganglion cell–inner plexiform layer complex, inner nuclear layer, and the outer plexiform layer. Dril is scored as: grade 0: without Dril, grade 1: with Dril. These values and morphological manifestations were measured and calculated in a 1-mm diameter circle centered on the fovea. If macular edema was present in both eyes, data were recorded for the one with the more severe condition and clearer OCT images. OCT images were graded by two experienced ophthalmologists in a double-blind setting. If they disagreed on the grading results, a fundus specialist made the final decision.

Collection of clinical data: Serum specimens were collected from the patients after a 12-h fast and before the administration of insulin and other drugs. Complete blood count including WBC, neutrophils, monocytes, lymphocytes, and platelets was performed with a hematology analyzer (Mindray, China), and an automatic biochemical analyzer (HITACHI, Japan) was used to measure the levels of glycosylated hemoglobin (HbA1c), fasting blood glucose (FBG), blood urea nitrogen (BUN), and creatinine. The NLR, monocyte-to-lymphocyte ratio (MLR), and platelet-to-lymphocyte ratio (PLR) were calculated by simple division of relevant blood parameters. NLR was calculated as neutrophil count (×10⁹/L)/lymphocyte count (×10⁹/L), MLR monocyte count (×10⁹/L)/lymphocyte count (×10⁹/L), and PLR as platelet count (×10⁹/L)/lymphocyte count (×10⁹/L). Age, sex, duration of type 2 diabetes, insulin use, blood pressure (BP), smoking history, whether there are other complications of diabetes, history of macrovascular disease, history of hypertension and duration of hypertension were recorded for all patients.

Statistical analysis

Dataset cleaning and statistical analysis were performed using R 4.0.5. Continuous variables are presented as the mean ± standard deviation, and categorical variables are expressed as number (percentage). Two-sided permutation test was applied for the comparison of continuous variables among groups. Chi-square test was used to compare the categorical variables between groups. P<0.05 was considered statistically significant.

In order to obtain the optimal model for disease prediction, we first considered the complete model (Model 1) with the main variables including age, sex, duration of diabetes, BP, blood glucose, HbA1c, WBCs and subtypes, platelets, platelet distribution width (PDW), creatinine, and BUN. Only neutrophil percentage (%neutrophils) and PLR were significant in Model 1. Therefore, stepwise regression analysis was performed, in which the Akaike information criterion (AIC) was applied. Model 2 was obtained through the smallest AIC information statistic. Because Model 2 was found to be with collinearity, it was further optimized to remove collinearity, which yielded an optimal Model 3.

Results

No-DME group versus DME group

The proportion of female patients showed a significant difference between the two groups (33.3% vs. 55.6%, P=0.03). The baseline data in two groups are shown in Table 1. Age, duration of T2DM, insulin use, systolic BP (SBP), diastolic BP (DBP), FBG, HbA1c, creatinine, BUN, WBC count, neutrophils, monocytes, lymphocytes, platelets, PDW, smoking history, whether there are other complications of diabetes, history of macrovascular disease, history of hypertension and duration of hypertension showed no significant differences between the No-DME and DME groups. In contrast, the DME group had significantly higher %neutrophils (63.27%±7.85% vs. 57.37%±9.52%, P=0.001), significantly lower %monocytes (6.88%±1.83% vs. 7.63%±1.77%, P=0.047) and significantly lower %lymphocytes (27.26%±6.59% vs. 30.35%±9.51%, P=0.032) than the No-DME group.

In order to obtain the optimal model of disease-related factors, we first considered the complete model (Model 1) (Table 2) with the main variables including age, sex, duration of diabetes, BP, FBG, HbA1c, WBCs and subtypes, platelets, PDW, creatinine, and BUN. Only %neutrophils and PLR were significant in Model 1. Model 2 was obtained through the smallest AIC information statistic (Table 3), but was found to be with collinearity and thus unable to predict diseases accurately, so it was further optimized to remove the collinearity, which yielded an optimal Model 3 (Table 4). With other predictors being the same, female sex had a significantly increased risk of DME (b=1.273, P=0.015) (Figure 1), %neutrophils was significantly associated with increased risk of DME (b=0.152, P=0.0006), and %lymphocytes was significantly associated with a reduced risk of DME (b=−0.027, P=0.179). The accuracy of the optimal model 3 can reach 82.46%. In 81 DME patients, the diagnostic accuracy of DME was 96.3% after using this model.

Figure 1 Sex distribution of No-DME and DME groups. DME, diabetic macular edema.

Early DME group versus late DME group

According to the OCT morphology of macular edema, we divided the DME group into an early DME group (n=33) and a late DME group (n=48). The proportion of females was 54.5% in the early DME group and 56.3% in the late DME group (P=0.88). Table 5 shows there were no statistically significant differences in the basic variables of these two groups. We also established the correlation model of the influencing factors of the two groups, and no significant predictor was found.

Discussion

DR refers to a series of lesions that occur in the ocular microvasculature of diabetic patients. DME occurs at any stage of DR and is also an important cause of vision loss in diabetic patients. A possible pathogenic mechanism may be early activation of retinal capillary endothelial cells that upregulates cell adhesion molecules, promoting WBC-endothelial cell adhesion and triggering the release of pro-inflammatory and vascular permeability factors, which together disrupt the BRB, causing increased vascular permeability, and ultimately leading to the development of DME (14). Intravitreal anti-VEGF drugs have replaced macular lasers as the first-line treatment option for DME in the past decade. In the Diabetic Retinopathy Clinical Research (15), approximately 50% of DME patients who received monthly intravitreal anti-VEGF injections still had retinal edema 1 year later, which progressed to irreversible visual impairment. According to the OCT-based grading of DME proposed by the European School for Advanced Studies in 2019 (12), it has four stages: early, advanced, chronic and atrophic maculopathy. Most OCT features associated with poor outcomes occur in the stages 2–4. Therefore, early detection and timely symptomatic treatment are essential to prevent irreversible vision loss. Unfortunately, the socioeconomic burden of diabetic complications such as DME is rising, and the detection of inflammatory factors in intraocular fluid and blood is not a practical strategy in early diagnosis and during follow-up due to its high cost and complex maneuver. Early detection of risk factors related to the occurrence of DME, close follow-up and strict blood glucose control should be carried out to delay or prevent its progression to the stage of irreversible macular structure and function.

High neutrophil and low lymphocyte counts in serum are closely associated with the macrovascular and microvascular complications of diabetes (3). WBCs and subtypes are readily and inexpensively measured in routine blood tests and are considered to be novel inflammatory biomarkers that reflect both adaptive immune responses (mediated by lymphocytes) and innate immune responses (mediated by neutrophils and monocytes).

When the BRB is still intact in early DR, low levels of inflammation (parainflammation) maintain homeostasis and restore function. However, long-term stimulation of the retina by chronic inflammation can lead to a maladapted immune system and dysregulated parainflammation, resulting in circulating immune cells (including neutrophils and monocytes) and serum proteins infiltrating the retina, accelerating the development of DR (2). Obasanmi et al. (16) investigated the relationship between changes in circulating WBCs and the progression of DR and concluded that DR is a consequence rather than a cause of systemic immune changes in T1DM. Activation of inflammatory and immune pathways is increasingly recognized as key mediators in the development and progression of DR (17).

Intravitreal injection of sustained-release steroids can effectively control DME (e.g., reducing macular thickness and improving visual function) (18). Immune cells (e.g., macrophages, neutrophils, and lymphocytes) have been detected in the epiretinal membrane of patients with diffuse DME (19), suggesting inflammation and immune dysregulation play key roles in the pathogenesis of DME. We speculate that DME is associated with increased innate cellular immunity (especially neutrophils) and decreased adaptive cellular immunity (especially lymphocytes).

In our current study, we divided the research subjects into No-DME and DME groups according to a structural measure of the macular area (OCT) in DR patients. The percentages of neutrophils, lymphocytes, and monocytes significantly differed between these two groups. The percentage stability of each WBC subtype is superior to pure neutrophil and lymphocyte counts as it is not easily affected by various physiological and pathological conditions and can better reflect the inflammatory state and immune response. Through statistical analysis, we found the optimal model of DME related risk factors, with an accuracy rate of 82.42% When the model was applied in 81 DME patients, diagnostic accuracy reached 96.3%.

Ilhan et al. (20) found the NLR was higher in their DME group than in two other control groups and concluded that NLR was a highly sensitive and specific diagnostic indicator of DME. Hu et al. (10) followed up 91 DME patients treated with intravitreal injection of anti-VEGF drugs and found that a higher NLR before treatment was associated with worse BCVA prognosis. However, in our present study, we did not find such a correlation between NLR and DME in either the No-DME or DME group; rather, we found in our model that %neutrophils and %lymphocytes correlated with the risk of DME. Such differences may be explained by differences in study design, patient characteristics (e.g., duration of diabetes), and study populations (e.g., ethnicity, region, and economic level).

Growing evidence suggests that sex is an important factor in DR (21,22). The average age of female patients in our study was ≈58 years, and the incidence of DME was significantly higher in females than in males, with statistical significance. In a previous study of 42 patients with severe DR, we also found that the incidence of DME in women was significantly higher than that in men (4). Estrogen exerts its protective effect on the retina by activating the PI3K/Akt signaling cascade, serving as antioxidant, and exerting a vasodilatory effect (23). A Korean study of postmenopausal women showed that those not receiving estrogen replacement therapy had a higher incidence of eye diseases, including anterior polar cataracts and other retinal diseases (24). Among T2DM patients with a diabetic history of more than 10 years, age over 60 years, and a relatively intermediate economic status in China, females had a higher prevalence of DR than males (25). Thus, postmenopausal estrogen decline is closely related to the occurrence of DME, although prospective studies are needed to further investigate the causal relationship between these two factors.

In our current study, WBCs and subtypes did not significantly differ between the early and late DME groups, which might be due to the small sample size; however, it is also possible that there are other unknown inflammatory markers during the development of DME.

Our study has some limitations. First, as a retrospective study, it only analyzed data related to WBCs and subtypes; second, the sample size was small, and additional prospective and multicenter studies are needed; third, the role of the imbalanced immune system in the occurrence and development of DME needs to be confirmed by more evidence from clinical studies and basic research; and fourth, the development of DME from an early stage to a more advanced stage involves a series of changes in morphology, and the influential factors and mechanism of action were not elucidated in the present study.

In conclusion, WBCs and subtypes are representative markers of inflammation and the immune system, and their associations with DR have been demonstrated; however, few have described their relationships with DME. As shown in our current study, with all other predictors being the same, %neutrophils was significantly associated with increased risk of DME and %lymphocytes was significantly associated with a reduced risk of DME. Females are more likely to develop DME. The model we established had a predictive accuracy of 82.46%. In addition, imbalance of the immune system plays a key role in the pathogenesis of DME, and it will become an important target for DME treatment in the future.

Acknowledgments

Funding: This study was supported by the Scientific Research Program of Nantong Health and Family Planning Commission (Nos. MB2020004 and MB2019005).

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://apm.amegroups.com/article/view/10.21037/apm-22-962/rc

Data Sharing Statement: Available at https://apm.amegroups.com/article/view/10.21037/apm-22-962/dss

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://apm.amegroups.com/article/view/10.21037/apm-22-962/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by ethics committee of Nantong First People’s Hospital (No. 2022KT108). Individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bressler SB, Ayala AR, Bressler NM, et al. Persistent Macular Thickening After Ranibizumab Treatment for Diabetic Macular Edema With Vision Impairment. JAMA Ophthalmol 2016;134:278-85. [Crossref] [PubMed]
2. Xu H, Chen M. Diabetic retinopathy and dysregulated innate immunity. Vision Res 2017;139:39-46. [Crossref] [PubMed]
3. Cardoso CRL, Leite NC, Salles GF. Importance of hematological parameters for micro- and macrovascular outcomes in patients with type 2 diabetes: the Rio de Janeiro type 2 diabetes cohort study. Cardiovasc Diabetol 2021;20:133. [Crossref] [PubMed]
4. Zhu Y, Cai Q, Li P, et al. The relationship between peripheral blood inflammatory markers and diabetic macular edema in patients with severe diabetic retinopathy. Ann Palliat Med 2022;11:984-92. [Crossref] [PubMed]
5. Verdoia M, Schaffer A, Barbieri L, et al. Impact of diabetes on neutrophil-to-lymphocyte ratio and its relationship to coronary artery disease. Diabetes Metab 2015;41:304-11. [Crossref] [PubMed]
6. Mertoglu C, Gunay M. Neutrophil-Lymphocyte ratio and Platelet-Lymphocyte ratio as useful predictive markers of prediabetes and diabetes mellitus. Diabetes Metab Syndr 2017;11:S127-31. [Crossref] [PubMed]
7. Kuranari Y, Tamura R, Tsuda N, et al. Prognostic Significance of Preoperative Neutrophil-to-Lymphocyte Ratio in Patients With Meningiomas. Front Oncol 2020;10:592470. [Crossref] [PubMed]
8. Zhang A, Ning L, Han J, et al. Neutrophil-To-Lymphocyte Ratio as a Potential Biomarker of Neovascular Glaucoma. Ocul Immunol Inflamm 2021;29:417-24. [Crossref] [PubMed]
9. Ulu SM, Dogan M, Ahsen A, et al. Neutrophil-to-lymphocyte ratio as a quick and reliable predictive marker to diagnose the severity of diabetic retinopathy. Diabetes Technol Ther 2013;15:942-7. [Crossref] [PubMed]
10. Hu Y, Cheng Y, Xu X, et al. Pretreatment neutrophil-to-lymphocyte ratio predicts prognosis in patients with diabetic macular edema treated with ranibizumab. BMC Ophthalmol 2019;19:194. [Crossref] [PubMed]
11. The Chinese Diabetes Society. Chinese Diabetes Society Guidelines for the Prevention and Treatment of Type 2 Diabetes in China (2013). Chinese Journal of Diabetes 2014;6:447-98.
12. Panozzo G, Cicinelli MV, Augustin AJ, et al. An optical coherence tomography-based grading of diabetic maculopathy proposed by an international expert panel: The European School for Advanced Studies in Ophthalmology classification. Eur J Ophthalmol 2020;30:8-18. [Crossref] [PubMed]
13. Gerendas BS, Prager S, Deak G, et al. Predictive imaging biomarkers relevant for functional and anatomical outcomes during ranibizumab therapy of diabetic macular oedema. Br J Ophthalmol 2018;102:195-203. [Crossref] [PubMed]
14. Yang X, Scott HA, Monickaraj F, et al. Basement membrane stiffening promotes retinal endothelial activation associated with diabetes. FASEB J 2016;30:601-11. [Crossref] [PubMed]
15. Diabetic Retinopathy Clinical Research Network. Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology 2010;117:1064-77.e35. [Crossref] [PubMed]
16. Obasanmi G, Lois N, Armstrong D, et al. Circulating Leukocyte Alterations and the Development/Progression of Diabetic Retinopathy in Type 1 Diabetic Patients - A Pilot Study. Curr Eye Res 2020;45:1144-54. [Crossref] [PubMed]
17. Forrester JV, Kuffova L, Delibegovic M. The Role of Inflammation in Diabetic Retinopathy. Front Immunol 2020;11:583687. [Crossref] [PubMed]
18. Dugel PU, Bandello F, Loewenstein A. Dexamethasone intravitreal implant in the treatment of diabetic macular edema. Clin Ophthalmol 2015;9:1321-35. [Crossref] [PubMed]
19. Tamura K, Yokoyama T, Ebihara N, et al. Histopathologic analysis of the internal limiting membrane surgically peeled from eyes with diffuse diabetic macular edema. Jpn J Ophthalmol 2012;56:280-7. [Crossref] [PubMed]
20. Ilhan C, Citirik M, Uzel MM, et al. The usefulness of systemic inflammatory markers as diagnostic indicators of the pathogenesis of diabetic macular edema. Arq Bras Oftalmol 2020;83:299-304. [Crossref] [PubMed]
21. Bek T. Systemic risk factors contribute differently to the development of proliferative diabetic retinopathy and clinically significant macular oedema. Diabetologia 2020;63:2462-70. [Crossref] [PubMed]
22. Nakayama Y, Yamaguchi S, Shinzato Y, et al. Retrospective exploratory analyses on gender differences in determinants for incidence and progression of diabetic retinopathy in Japanese patients with type 2 diabetes mellitus. Endocr J 2021;68:655-69. [Crossref] [PubMed]
23. Nuzzi R, Scalabrin S, Becco A, et al. Gonadal Hormones and Retinal Disorders: A Review. Front Endocrinol (Lausanne) 2018;9:66. [Crossref] [PubMed]
24. Na KS, Jee DH, Han K, et al. The ocular benefits of estrogen replacement therapy: a population-based study in postmenopausal Korean women. PLoS One 2014;9:e106473. [Crossref] [PubMed]
25. Li M, Wang Y, Liu Z, et al. Females with Type 2 Diabetes Mellitus Are Prone to Diabetic Retinopathy: A Twelve-Province Cross-Sectional Study in China. J Diabetes Res 2020;2020:5814296. [Crossref] [PubMed]