Skip to content

Proliferative Diabetic Retinopathy

Proliferative diabetic retinopathy (PDR) is a severe and potentially blinding complication of both type 1 and type 2 diabetes mellitus.[1]It represents the advanced stage of diabetic retinopathy (DR), a disease that affects multiple vascular and neural cell types within the retina.[1] PDR is widely recognized as the leading cause of new cases of blindness among working-aged adults.[1] This advanced form of DR is specifically characterized by neovascularization, the abnormal growth of new blood vessels on the retina.[1] It is strongly associated with a longer duration of diabetes and poor glycemic control.[2]

The progression to PDR begins with damage to the retinal vasculature, initially manifesting in early non-proliferative diabetic retinopathy (NPDR) stages with features such as microaneurysms, lipid and protein deposits, and cotton wool spots.[1] In PDR, the retina attempts to compensate for inadequate blood supply by growing new, fragile blood vessels. These new vessels are prone to bleeding, which can lead to vitreous hemorrhage, and can also cause tractional retinal detachment, both of which severely impair vision. Genetic factors play a significant role in an individual’s susceptibility to PDR, with studies estimating its heritability to be as high as 52%.[3] Genome-wide association studies (GWAS) have been conducted across various populations to identify specific genetic variants associated with PDR. For instance, the NADPH Oxidase 4 (NOX4) gene, particularly the SNP rs3913535 , has been suggested to be associated with severe DR, including PDR, in type 2 diabetes patients.[4] Other loci, such as WDR72, HLA-B, GAP43/RP11-326J18.1, and AL713866.1, have been identified in African populations.[2] In Japanese populations, RP1-90L14.1 has shown associations with DR and PDR.[5]The genetic landscape of PDR is complex, and there is considerable overlap in patient cohorts with different retinopathy subtypes, such as diabetic macular edema (DME), which complicates the search for specific contributing genes.[1]

PDR poses a critical threat to vision, often leading to irreversible blindness if not promptly managed. Complications such as vitreous hemorrhage and retinal detachment necessitate urgent medical intervention. Regular and thorough ophthalmological examinations are essential for diagnosing DR and classifying its severity, ranging from none to non-proliferative or proliferative.[5]While intensive glucose control is a cornerstone of diabetes management, some research indicates that it may not significantly affect the 5-year incidence of retinopathy rates.[6] This highlights the importance of understanding underlying biological mechanisms and genetic predispositions in addition to glycemic control for preventing progression to PDR.

As a leading cause of blindness in working-aged adults, PDR carries significant social and economic implications.[1] The loss of vision can severely impact an individual’s quality of life, independence, and ability to work, leading to substantial healthcare costs and lost productivity.[4] Understanding the genetic architecture of PDR, especially across diverse populations, is crucial for developing personalized risk assessment tools, early diagnostic methods, and targeted therapeutic strategies. Genetic research, including studies in diverse populations such as Africans, is vital to improve discovery for complex traits like PDR, acknowledging that genetic associations may vary across ethnic groups.[2]

Methodological and Statistical Constraints

Section titled “Methodological and Statistical Constraints”

Research into proliferative diabetic retinopathy is often constrained by study design and statistical limitations that can impact the robustness and interpretability of findings. Many studies contend with sample sizes that may be insufficient to detect genetic variants with modest effect sizes, leading to reduced statistical power and potentially missed associations.[3] Furthermore, imbalances in case-control sample sizes, particularly when control samples are limited, can adversely affect the accuracy of association analyses.[7]These statistical challenges contribute to a high rate of replication failures, where initial findings do not consistently reappear in independent cohorts, making it difficult to establish definitive genetic associations for proliferative diabetic retinopathy.[8] Another significant limitation arises from inconsistencies in study execution and data collection. The failure to consistently account for key covariates, such as duration of diabetes and glycemic control, can confound genetic analyses and obscure true associations.[3] Additionally, differences in diagnostic approaches among ophthalmologists across hospitals may introduce minor misclassification bias, affecting the reliability of case-control definitions.[7] Even technical variations, such as different genotyping laboratories or DNA extraction methods, can introduce systematic differences between cases and controls, necessitating careful statistical adjustment to avoid spurious findings.[9]

Phenotypic Heterogeneity and Measurement Challenges

Section titled “Phenotypic Heterogeneity and Measurement Challenges”

The complex nature of proliferative diabetic retinopathy (PDR) and its progression presents substantial challenges in phenotypic definition and ascertainment across studies. Variability in the classification of diabetic retinopathy (DR) and patient recruitment criteria can lead to discrepancies in study populations, hindering the comparability and replication of genetic findings.[6]Aggregating different stages of DR, such as combining proliferative DR with diabetic macular edema (DME) or non-proliferative DR (NPDR) with PDR into a broad “severe DR” or “any DR” category, can dilute specific genetic signals associated with PDR progression.[7]This lack of granular phenotyping may prevent the identification of variants specifically predisposing to the advanced stages of the disease.

Misclassification bias is also a concern when phenotypic ascertainment is incomplete or inconsistent. For instance, studies relying on limited-field fundus photography may miss subtle signs of DR, leading to an inaccurate classification of participants and potentially biasing results towards the null.[3] Similarly, the absence of a minimum duration of diabetes for control subjects means some individuals classified as controls could potentially develop DR later, further weakening the power to detect true genetic associations.[3] Furthermore, using a single measure of HbA1c, while practical, provides only a snapshot of glycemic control and may not accurately reflect long-term glycemic exposure, which is a critical risk factor for DR development.[3]

Ancestry, Generalizability, and Environmental Influences

Section titled “Ancestry, Generalizability, and Environmental Influences”

Generalizability of genetic findings for proliferative diabetic retinopathy is often limited by the ancestral composition of study cohorts and the interaction with environmental factors. Many identified genetic signals do not show consistent replication across different ethnic groups, suggesting that susceptibility loci may not be trans-ethnically shared.[6] This is further complicated by differences in minor allele frequencies and the magnitude of allelic effects across diverse populations, such as those observed between Chinese and Hispanic or African and African American cohorts.[6] Such population heterogeneity can lead to a lack of replication and complicates efforts to identify universally applicable genetic markers.

Beyond genetic background, environmental and genomic context differences, including varying levels of admixture in populations, can significantly confound genetic analyses.[10]These factors highlight the challenges in isolating specific genetic effects from broader population-specific influences. Crucially, the development of proliferative diabetic retinopathy is strongly influenced by non-genetic factors like duration of diabetes and glycemic control, hypertension, and dyslipidemia.[1] If these powerful environmental covariates are not adequately captured and modeled in genetic studies, their confounding effects can obscure underlying genetic predispositions and impact the interpretation of genetic associations.

Unexplained Genetic Architecture and Knowledge Gaps

Section titled “Unexplained Genetic Architecture and Knowledge Gaps”

Despite advancements in genetic research, a significant portion of the genetic architecture underlying proliferative diabetic retinopathy remains unexplained, contributing to a substantial knowledge gap. Heritability estimates for proliferative diabetic retinopathy can be as high as 52%, yet many identified genetic variants account for only a small fraction of this inherited risk, indicating considerable missing heritability.[1]This suggests that numerous genetic factors, possibly with small individual effects or rare variants, are yet to be discovered. The inconsistency of findings from candidate gene studies and the lack of robust replication for many proposed polymorphisms further underscore an incomplete understanding of the disease’s genetic basis.[8]The modest effect sizes typically associated with genetic variants in complex diseases like proliferative diabetic retinopathy necessitate increasingly larger and more diverse international collaborations to achieve the statistical power required for novel locus discovery.[3] Current research efforts, even in large-scale genome-wide association studies, have not always yielded genome-wide significant findings, pointing to the need for more comprehensive approaches.[3]Future studies incorporating advanced techniques like whole-genome sequencing and highly refined phenotyping, including optical coherence tomography for detailed retinal assessment, are anticipated to reveal a role for rare variants and more specific genetic predispositions to disease progression.[7]

The genetic predisposition to proliferative diabetic retinopathy (PDR) is influenced by a range of single nucleotide polymorphisms (SNPs) affecting genes involved in diverse cellular functions, from non-coding RNA regulation to immune responses and metabolic balance. These variants can alter gene activity or protein function, contributing to the pathological neovascularization and inflammation characteristic of PDR.

Several variants associated with proliferative diabetic retinopathy (PDR) are found within or near non-coding RNA genes, highlighting their emerging importance in disease pathogenesis. The variantrs149201869 is located in LINC00607, a long intergenic non-coding RNA (FN1-DT). LncRNAs like LINC00607 play crucial regulatory roles in gene expression, influencing processes such as cell proliferation, apoptosis, and angiogenesis, all of which are critical in the development and progression of PDR. Similarly, rs73347124 is associated with LINC02672 (a Y_RNA), and rs74161190 is linked to MIR378C (another Y_RNA). YRNAs are a class of non-coding RNAs known to be involved in RNA processing and stress response, and their dysregulation can contribute to cellular dysfunction observed in diabetic complications. Alterations in these regulatory non-coding RNAs could affect the stability or translation of mRNAs involved in vascular health and inflammatory responses within the retina, thereby increasing susceptibility to the severe neovascularization characteristic of PDR.[5]Genetic studies have consistently demonstrated the significant impact of various genetic factors on the risk of developing diabetic retinopathy, including both protein-coding and non-coding regions.[3] Other notable variants impact genes central to cellular homeostasis and stress responses, pathways frequently disrupted in diabetes. The variant rs138683663 is associated with _SEC11C, a gene encoding a subunit of the signal peptidase complex, which is crucial for the proper processing and trafficking of proteins into the endoplasmic reticulum. Disruptions in this fundamental process can lead to protein misfolding and endoplasmic reticulum stress, contributing to cellular damage in the retina under chronic diabetic conditions. Similarly, variants rs115523882 and rs200295620 are linked to GOLIM4 (EGFEM1P), a gene involved in Golgi apparatus function, responsible for modifying, sorting, and packaging proteins for secretion or delivery to other organelles. Impaired Golgi function can lead to accumulation of misprocessed proteins and cellular dysfunction, exacerbating the pathology of PDR.[1] Additionally, rs7604016 is located within COMMD1, a gene implicated in copper metabolism and the regulation of the NF-κB signaling pathway, a key mediator of inflammation and cell survival. Dysregulation of COMMD1 could therefore influence the inflammatory processes and cellular resilience in the retinal microvasculature, impacting PDR progression.[11] The genetic landscape of PDR also includes variants affecting immune modulation and cellular redox balance, pathways intrinsically linked to chronic inflammation and oxidative stress in diabetes. The variant rs73228199 is associated with CD96, a cell surface receptor expressed on immune cells that modulates immune responses and cell adhesion. Alterations in CD96 function could impact the inflammatory milieu within the retina, influencing immune cell infiltration and the aberrant angiogenesis characteristic of PDR. Another significant variant, rs137949823 , is located near NNT (RNU6-381P), which codes for NADPH transhydrogenase. NNTplays a critical role in maintaining the cellular redox state by regulating NADPH levels, a coenzyme essential for antioxidant defenses.[4] Dysfunctional NNTcould lead to increased oxidative stress, a major contributor to microvascular damage in diabetic retinopathy. Furthermore,rs71354195 is linked to ZFP82, a zinc finger protein known to act as a transcription factor, regulating the expression of various genes involved in cell growth and differentiation. Changes in ZFP82 activity could alter gene programs crucial for retinal cell health and repair. Lastly, rs1414474 is associated with C1orf94, a gene whose precise function is still being elucidated but may contribute to cell signaling or inflammatory processes.[2] Collectively, these variants underscore the multifactorial genetic basis of PDR, involving complex interactions across immune, metabolic, and regulatory pathways.

RS IDGeneRelated Traits
rs149201869 FN1-DT - LINC00607proliferative diabetic retinopathy
rs138683663 SEC11C - GRPproliferative diabetic retinopathy
rs73347124 Y_RNA - LINC02672proliferative diabetic retinopathy
rs115523882
rs200295620
GOLIM4 - EGFEM1Pproliferative diabetic retinopathy
rs74161190 Y_RNA - MIR378Cproliferative diabetic retinopathy
rs137949823 NNT - RNU6-381Pproliferative diabetic retinopathy
rs73228199 CD96proliferative diabetic retinopathy
rs7604016 COMMD1proliferative diabetic retinopathy
rs71354195 ZFP82proliferative diabetic retinopathy
rs1414474 C1orf94proliferative diabetic retinopathy

Defining Proliferative Diabetic Retinopathy

Section titled “Defining Proliferative Diabetic Retinopathy”

Proliferative diabetic retinopathy (PDR) represents the advanced and most severe stage of diabetic retinopathy (DR), a chronic, progressive, and potentially sight-threatening complication of diabetes mellitus.[12] It is precisely characterized by the presence of neovascularization, which involves the growth of abnormal new blood vessels on the surface of the retina or optic disc.[12] This pathological neovascularization is a critical diagnostic criterion, distinguishing PDR from earlier, less severe forms of DR.[10] PDR is recognized as the leading cause of new cases of blindness in working-aged adults and contributes significantly to vision loss among diabetic patients.[12]The development of PDR signifies extensive damage to the retinal microvasculature, progressing beyond the microaneurysms, hemorrhages, and exudates characteristic of non-proliferative diabetic retinopathy (NPDR).[12] The clinical significance of PDR lies in its high risk for severe vision impairment due to complications such as vitreous hemorrhage, tractional retinal detachment, and neovascular glaucoma, all stemming from these fragile new vessels.[10] Crucially, once PDR develops, regression to a state of no DR is generally not possible, underscoring its irreversible and advanced nature.[9]

The classification of diabetic retinopathy, including PDR, follows standardized systems crucial for diagnosis, prognosis, and research. The International Clinical Diabetic Retinopathy Disease Severity Scales provide a widely accepted framework, categorizing DR into stages such as none, non-proliferative diabetic retinopathy (NPDR), and proliferative diabetic retinopathy (PDR).[6] Within this framework, PDR specifically corresponds to the most advanced stages, often denoted as level R4 in some grading systems, which signifies the presence of neovascularization and/or retinal detachment.[10]The Early Treatment Diabetic Retinopathy Study (ETDRS) also provides a detailed severity scale, with PDR typically defined by an ETDRS score of 60 or higher in research contexts, indicating a high level of disease activity.[3]For some clinical and research purposes, a broader category of “severe diabetic retinopathy” is employed, which may encompass both severe background retinopathy (often level R3) and PDR (level R4).[9]This conceptual grouping acknowledges the rapid progression from severe NPDR to PDR, with severe background retinopathy often considered a precursor that will advance to PDR within a relatively short timeframe.[9]Conversely, controls in severe DR studies are stringently defined, often as individuals with type 2 diabetes whose screening records consistently show no DR or only mild background retinopathy (levels R0 or R1), and who have no history of laser photocoagulation treatment.[9]This categorical approach helps to delineate distinct disease states for effective management and genetic studies.

Diagnostic Criteria and Clinical Measurement

Section titled “Diagnostic Criteria and Clinical Measurement”

The diagnosis of proliferative diabetic retinopathy relies primarily on clinical examination, specifically a thorough fundus examination performed by a board-certified ophthalmologist.[6] The definitive diagnostic criterion for PDR is the direct observation of neovascularization in any retinal field or the presence of retinal detachment, which are hallmarks of the proliferative stage.[10]In addition to direct visualization, a history of laser photocoagulation treatment, a common intervention for PDR, is also a strong indicator of prior or current severe disease.[9]These clinical findings are then interpreted according to established severity scales, such as the International Clinical Diabetic Retinopathy Disease Severity Scales or the detailed ETDRS grading system.[3] In research settings, particularly for genome-wide association studies, precise operational definitions and measurement criteria are critical for distinguishing cases from controls. PDR cases are consistently defined by the presence of neovascularization and/or retinal detachment.[10]To enhance the power and precision of these studies, “super-controls” or “extreme phenotype” controls are often selected; these are individuals with type 2 diabetes who have high glycemic levels (e.g., fasting blood glucose of at least 169 mg/dL or HbA1c > 7.5%) and a long duration of diabetes (e.g., at least 10 years) but show no signs of diabetic retinopathy.[10] This rigorous control selection minimizes the risk of misclassifying individuals who might later develop PDR, ensuring a clearer genetic signal.[10] While duration of diabetes and HbA1c are known epidemiological risk factors for DR progression, their utility as direct diagnostic cut-offs for PDR is primarily in defining research cohorts rather than individual clinical diagnosis.[6]

Ocular Manifestations and Diagnostic Assessment

Section titled “Ocular Manifestations and Diagnostic Assessment”

Proliferative diabetic retinopathy (PDR), a late stage of diabetic retinopathy (DR), is primarily characterized by the development of neovascularization within the retina.[1] This new vessel formation, which can occur in any retinal field, is a definitive clinical sign for diagnosing PDR.[2] In severe cases, PDR can also present with retinal detachment.[2]The diagnosis and classification of DR severity, including PDR, are typically performed through a comprehensive fundus examination by a board-certified ophthalmologist, adhering to criteria such as the International Clinical Diabetic Retinopathy Disease Severity Scale.[5]

PDR represents a progression from earlier stages of DR, such as non-proliferative diabetic retinopathy (NPDR).[1] While NPDR is characterized by microaneurysms, lipid deposits, and cotton wool spots, only a subset of individuals with NPDR will advance to PDR; for instance, one study observed that 5.3% of Mexican Americans with type 2 diabetes and NPDR progressed to PDR.[6]The duration of diabetes and glycated hemoglobin (HbA1c) levels are considered significant determinants of DR progression, though the impact of intensive glucose control on the 5-year incidence of retinopathy rates has shown variability across studies.[6]Clinical and laboratory data, including HbA1c and duration of diabetes, are routinely measured and crucial for understanding disease evolution.[6]

Clinical Phenotypes and Diagnostic Stratification

Section titled “Clinical Phenotypes and Diagnostic Stratification”

PDR exhibits considerable phenotypic diversity, and patients may concurrently develop diabetic macular edema (DME), which involves fluid buildup in and beneath the macula, affecting central vision.[1]There is a recognized overlap in patient cohorts presenting with different retinopathy subtypes, complicating the study of underlying biological mechanisms.[1]When distinguishing PDR cases from controls in research, particularly in genetic studies, differences in clinical characteristics like gender, age, duration of diabetes, and HbA1c are often observed, though body mass index (BMI) may not differ significantly.[4]To enhance diagnostic precision and power in genetic analyses, “super-controls”—individuals with a long duration of type 2 diabetes (e.g., at least 10 years) and high glycemia (e.g., fasting blood glucose ≥ 169 mg/dL or HbA1c > 7.5%) but no signs of DR—are sometimes used to minimize misclassification bias.[2] Genetic factors also play a substantial role, with PDR demonstrating a heritability of 52%.[3]

Causes of Proliferative Diabetic Retinopathy

Section titled “Causes of Proliferative Diabetic Retinopathy”

Proliferative diabetic retinopathy (PDR) is a complex microvascular complication of diabetes mellitus, characterized by new vessel formation in the retina. Its development is influenced by a combination of genetic predispositions, prolonged metabolic dysregulation, and various environmental factors. The multifactorial etiology of PDR, while postulated, remains an area of ongoing research.[5]

Proliferative diabetic retinopathy exhibits a significant heritability, estimated at 52%.[13] indicating a strong genetic component influencing susceptibility. Genome-wide association studies (GWAS) across diverse populations, including those from Latvia, Africa, China, Japan, and Taiwan, have been instrumental in identifying numerous genetic variants associated with the condition.[5] However, the genetic effects are often modest, and consistent replication of findings across studies has been challenging, necessitating large sample sizes and multiethnic analyses to improve discovery for complex traits.[3]Specific genes and single nucleotide polymorphisms (SNPs) have been implicated in PDR pathogenesis. For instance, variants in theNADPH Oxidase 4 (NOX4) gene, such as rs3913535 , have been associated with severe diabetic retinopathy.[4] Other candidate genes include ICAM1, PPARGC1A, and MTHFR.[4] as well as GRB2.[2] PLXDC2, and ARHGAP22, the latter two being involved in endothelial cell angiogenesis and capillary permeability.[14] Genetic variants related to inflammation, ciliopathy, free-iron radicals, and lipid metabolism, such as those in COMMD6-UCHL3, LRP2-BBS5, ARL4C-SH3BP4, TBC1D4, CFH, CFB, VEGF, TGFB, and IFNG genes, have also been suggested to play roles in retinal cell damage and the development of PDR.[3]

Metabolic Dysregulation and Environmental Modulators

Section titled “Metabolic Dysregulation and Environmental Modulators”

The duration of diabetes and the level of glycemic control, typically measured by HbA1c, are recognized as the two most significant epidemiological risk factors for the development and progression of diabetic retinopathy.[3]Poor long-term glycemic control leads to chronic damage to the retinal vasculature, setting the stage for the microaneurysms, hemorrhages, and ultimately, neovascularization characteristic of PDR. However, these factors alone do not entirely account for an individual’s susceptibility, as evidenced by studies showing that glycated hemoglobin and diabetes duration explained only a fraction of retinopathy risk.[6]While the researchs highlights the primary role of diabetes management, environmental factors also contribute to the broad context in which PDR develops. The genetic landscape of PDR can vary across different populations, suggesting potential influences from diverse environmental, lifestyle, and socioeconomic factors, although specific details regarding these are not extensively elaborated in the context.[4]Interestingly, some studies indicate that intensive glucose control might not always affect the 5-year incidence of retinopathy rates, suggesting a threshold effect or the involvement of additional, yet-to-be-fully-understood mechanisms in disease progression.[6]

The development of PDR is not solely dictated by individual genetic variants or environmental exposures but by their intricate interplay. Genetic predispositions can modify an individual’s response to metabolic stressors, influencing the likelihood and severity of retinopathy. Advanced analytical approaches, such as liability threshold modeling, are employed in genetic studies to account for strong covariates like the duration of diabetes and glycemic control, thereby enhancing the statistical power to identify genuine genetic associations and better understand these gene-environment interactions.[3]Beyond the direct genetic and metabolic factors, other elements contribute to the complex trajectory of PDR. Age is a frequently considered covariate in studies of diabetic retinopathy, reflecting its role as a general risk factor for various chronic conditions, including microvascular complications.[4]While specific details on developmental and epigenetic factors like early life influences, DNA methylation, or histone modifications directly causing PDR are not provided in the context, these mechanisms are increasingly recognized for their roles in modulating gene expression and disease susceptibility in response to environmental cues, potentially contributing to the multifactorial nature of PDR.

Pathophysiology of Retinal Damage and Neovascularization

Section titled “Pathophysiology of Retinal Damage and Neovascularization”

Proliferative diabetic retinopathy (PDR) is a severe microvascular complication of diabetes, posing a significant threat to vision and representing a leading cause of blindness among working-aged adults.[1], [5], [9]The disease progresses from earlier stages, known as non-proliferative diabetic retinopathy (NPDR), which are characterized by the formation of retinal microaneurysms, lipid and protein deposits, and cotton wool spots, all indicative of initial damage to the retinal vasculature.[1]Persistent or poorly controlled hyperglycemia, a primary factor in diabetes, is a major determinant influencing the progression of diabetic retinopathy by disrupting cellular homeostasis within the delicate retinal environment.[2], [6] The hallmark of PDR is the development of aberrant retinal neovascularization, where new, fragile blood vessels proliferate from the retina into the vitreous humor.[1], [10]This pathological process is significantly driven by oxidative stress, an imbalance resulting from excessive production of reactive oxygen species (ROS) that overwhelms the retina’s antioxidant defenses.[15] Specifically, NADPH Oxidase 4 (NOX4) is a key enzyme involved in generating these damaging ROS, and its activity contributes to aberrant retinal neovascularization.[15], [16] Studies have shown that deletion of NOX4 can reduce oxidative stress and associated injury, highlighting its critical role in the pathogenesis of PDR.[17]

Molecular Signaling and Cellular Dysregulation

Section titled “Molecular Signaling and Cellular Dysregulation”

Beyond oxidative stress, the progression of proliferative diabetic retinopathy involves intricate molecular signaling pathways and widespread cellular dysregulation. A central mediator of retinal neovascularization is Vascular Endothelial Growth Factor (VEGF), which is significantly upregulated under conditions of high glucose and actively promotes the growth of new blood vessels.[15] This VEGF upregulation can be mediated through signaling cascades such as the JAK/STAT pathway, a critical cellular response mechanism implicated in various diabetic microvascular complications.[18] Anti-VEGFtherapies are a common treatment strategy for PDR, underscoring the protein’s central role.[19] Furthermore, PDR is characterized by a breakdown of the blood-retinal barrier and chronic inflammatory processes. Inhibition of reactive oxygen species has been shown to ameliorate the integrity of the blood-retinal barrier, emphasizing its vulnerability to oxidative damage.[15] The inflammatory milieu in PDR patients is evidenced by elevated levels of cytokines associated with Th2 and Th17 cells in the vitreous fluid, alongside altered expression of cell adhesion molecules like L-selectin on lymphocytes, which promotes increased adhesion to the endothelium and exacerbates vascular damage.[20], [21] The Wnt/β-cateninsignaling pathway also plays a role, with its activation being promoted by oxidative stress, and its inhibition by agents such as fenofibrate showing beneficial effects on diabetic retinopathy.[16], [22]

Genetic Predisposition and Regulatory Mechanisms

Section titled “Genetic Predisposition and Regulatory Mechanisms”

Genetic factors significantly contribute to an individual’s susceptibility to developing proliferative diabetic retinopathy, with studies estimating the heritability of PDR to be around 52%.[3], [13] Genome-wide association studies (GWAS) have been instrumental in identifying specific genetic variants and loci potentially linked to this severe condition. For example, the NADPH Oxidase 4 (NOX4) gene has been identified as a candidate gene associated with severe diabetic retinopathy in type 2 diabetes, reinforcing its role in the oxidative stress mechanisms underlying the disease.[9] Beyond NOX4, research has explored other genes and their polymorphisms for associations with DR and PDR. These include variations in genes such as Vascular Endothelial Growth Factor (VEGF), Transforming Growth Factor Beta (TGF-β), and Interferon Gamma (IFN-γ), all of which are critical components of angiogenic and immune regulatory pathways.[23] Furthermore, genes like WDR72 have been implicated in the proliferative angiogenesis of the microvasculature, while specific HLA-B alleles, such as HLA-B*07:02 and HLA-B*45:01, have shown associations with PDR in certain populations, highlighting the complex genetic architecture and potential ethnic variability in disease susceptibility.[10] The multifactorial etiology of DR also points to the involvement of regulatory elements, including long intergenic non-coding RNAs (lncRNAs), which can modulate gene expression patterns crucial for retinal health.[5]

Proliferative diabetic retinopathy is not merely a localized ocular disease but rather an integral part of broader systemic diabetic complications, reflecting widespread metabolic and physiological disruptions. It is widely recognized that the presence of retinopathy serves as a predictor for other diabetic complications, suggesting common underlying pathophysiological mechanisms across various microvascular beds.[24]There is also a significant overlap in patient populations experiencing different retinopathy subtypes, such as diabetic macular edema (DME), which indicates shared pathways of retinal damage and complicates the search for highly specific genetic determinants.[1]The systemic nature of diabetes means that factors such as overall glycemic exposure and blood pressure exert considerable influence on the trajectory of DR, affecting both its progression and potential remission.[2] This highlights the critical role of systemic metabolic control in mitigating the risk and severity of ocular complications. Furthermore, key biomolecules and genetic components, like HLA-B and GAP43, are expressed in both the retina and the pancreas, suggesting potential systemic interactions or shared molecular signals that contribute to the manifestation of diabetic complications in multiple organs.[10] The association of gene polymorphisms, such as those in the RAGE(Receptor for Advanced Glycation Endproducts) gene, with both diabetic retinopathy and nephropathy further exemplifies how systemic factors and genetic predispositions contribute to multi-organ damage in individuals with diabetes.[25]

Proliferative diabetic retinopathy (PDR) is fundamentally characterized by aberrant vascular growth, driven by a complex interplay of signaling pathways that promote angiogenesis and disrupt vascular integrity. A central player in this process is Vascular Endothelial Growth Factor (VEGF), whose expression is upregulated under pathological conditions, leading to the formation of fragile new blood vessels that are prone to leakage and hemorrhage.[19] This VEGF-mediated neovascularization is further exacerbated by oxidative stress, as inhibition of reactive oxygen species (ROS) can downregulate VEGF expression and improve blood-retinal barrier function.[15] Other growth factors such as Fibroblast Growth Factor 5 (FGF5) also contribute to choroidal neovascularization, with its expression being implicated in the progression of retinal pathologies.[26] The NADPH Oxidase 4 (NOX4) enzyme is a significant source of ROS in the retina, and its overexpression promotes both oxidative stress and aberrant retinal neovascularization, partly by activating pathways such as the Wnt pathway and enhancing insulin-stimulated angiogenesis.[15] Downstream of these growth factors and oxidative stress, intracellular signaling cascades like the Akt pathway, specifically enhanced Akt-3 stimulation, play a crucial role in cell survival and proliferation of vascular cells, further contributing to the pathological vascular remodeling observed in PDR.[27] Moreover, the JAK/STAT signaling pathway, which can be inhibited to suppress VEGFupregulation under high glucose conditions, integrates signals from various cytokines and growth factors to regulate cell proliferation and inflammation, highlighting its significance in the disease’s progression.[18] Neuropeptide Y (NPY) has also been identified as a novel mechanism for ischemic angiogenesis, indicating a neurovascular component to the proliferative process.[28]

Chronic inflammation and dysregulated immune responses are critical drivers in the pathogenesis of PDR, contributing to vascular damage and the proliferative cascade. Elevated levels of various cytokines, particularly those associated with Th2 and Th17 cells, are found in the vitreous fluid of PDR patients, indicating an active inflammatory milieu.[20]This inflammatory environment includes key mediators like Tumor Necrosis Factor (TNF), where systemic soluble TNF receptors are associated with the severity of diabetic retinopathy, suggesting a role for TNF-alpha signaling in disease progression.[29] The complement system, an integral part of innate immunity, is also implicated, with polymorphisms in genes such as CFH and CFBassociated with retinopathy, indicating that dysregulation of complement activation contributes to retinal injury.[30] Furthermore, cellular adhesion molecules play a crucial role in mediating inflammatory cell infiltration and vascular damage. Altered L-selectin expression in lymphocytes and increased adhesion to the endothelium in PDR patients signify enhanced leukocyte recruitment to the retinal vasculature.[21] Similarly, variants in ICAM1(Intercellular Adhesion Molecule 1) are associated with different phenotypes of diabetic retinopathy, further linking leukocyte adhesion and endothelial dysfunction to the disease.[31] These integrated inflammatory signals contribute to the breakdown of the blood-retinal barrier and create a permissive environment for neovascularization.

Oxidative Stress and Metabolic Dysregulation

Section titled “Oxidative Stress and Metabolic Dysregulation”

Metabolic abnormalities inherent in diabetes, particularly chronic hyperglycemia, lead to profound oxidative stress and dysregulation of various metabolic pathways, which are central to the development and progression of PDR. The overproduction of reactive oxygen species (ROS), largely driven by enzymes like NADPH Oxidase 4 (NOX4), is a key mechanism that damages retinal cells and vasculature.[15] This oxidative stress is intertwined with the activation of the Receptor for Advanced Glycation Endproducts (RAGE), a receptor whose polymorphisms are associated with diabetic retinopathy and which mediates cellular damage from advanced glycation endproducts formed under hyperglycemic conditions.[25] Mitochondrial defects are also recognized as fundamental drivers of degenerative retinal diseases, impacting cellular energy metabolism and contributing to increased ROS production.[32] The master metabolic regulator PPARGC1A(Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1-alpha) has genetic variants associated with different phenotypes of diabetic retinopathy, highlighting its role in mitochondrial biogenesis and metabolic adaptation in retinal cells.[31]Furthermore, the metabolism of homocysteine is a significant factor, with elevated levels serving as a risk factor for both nephropathy and retinopathy in type 2 diabetes, implicating folate metabolism and endothelial dysfunction in disease progression.[33] Variants in the CAPN2/CAPN8locus, associated with serum alpha-carotene concentrations, suggest a link between antioxidant status and retinal health, where deficiencies in protective metabolic pathways can exacerbate oxidative damage.[34]

Cellular Homeostasis and Regulatory Pathways

Section titled “Cellular Homeostasis and Regulatory Pathways”

Maintaining cellular homeostasis in the retina is critical, and its disruption through various regulatory mechanisms contributes to PDR. Telomerase components, such as the AAA-ATPase NVL2, are essential for holoenzyme assembly, suggesting that telomere maintenance and cellular senescence pathways could play a role in the longevity and function of retinal cells under diabetic stress.[35] Cellular trafficking and vesicle formation, regulated by proteins like EHD1 and EHD3, are important for the proper localization and function of signaling receptors and the secretion of growth factors, which can impact retinal cell communication and response to injury.[36]Beyond direct signaling, gene regulation mechanisms, including those influenced by long intergenic non-coding RNAs (lncRNAs), are implicated in the complex etiology of diabetic retinopathy, suggesting that epigenetic and transcriptional control play a role in disease susceptibility and progression.[5] The interplay of these regulatory mechanisms ensures the appropriate cellular response to stress; however, in diabetes, their dysregulation leads to a cascade of events that culminate in the pathological features of PDR, including impaired neurovascular repair.[37]The overall systems-level integration of these pathways, where metabolic stress feeds into inflammatory and angiogenic signaling through complex feedback loops and pathway crosstalk, ultimately dictates the emergent properties of the disease, providing multiple points for potential therapeutic intervention.

Risk Stratification and Prognostic Indicators

Section titled “Risk Stratification and Prognostic Indicators”

Proliferative diabetic retinopathy (PDR) is a severe, sight-threatening complication of diabetes, with significant heritability estimated at 52%.[3]Identifying individuals at high risk for PDR progression is crucial for personalized medicine and prevention strategies. Key prognostic indicators include the duration of diabetes and glycemic control (HbA1c levels), although recent studies show varying conclusions on the direct impact of intensive glucose control on retinopathy incidence.[6] Genetic factors also play a role, with studies identifying specific loci such as NOX4 (specifically rs3913535 ), WDR72, HLA-B, GAP43/RP11-326J18.1, and AL713866.1as potential susceptibility markers for severe diabetic retinopathy or PDR, even after accounting for traditional risk factors like age, gender, diabetes duration, and HbA1c.[2]The integration of genetic information with clinical covariates like duration of diabetes and glycemic control, potentially through liability threshold modeling, can enhance the power of risk assessment and improve the prediction of disease progression and treatment response.[3] While numerous candidate gene and genome-wide association studies (GWAS) have been conducted, the modest effect sizes of genetic variants and the need for large, well-phenotyped cohorts, along with consistent accounting for strong covariates, present challenges for replication and widespread clinical application.[3] However, continued research into these genetic markers holds promise for more precise risk stratification, enabling earlier intervention for those most likely to develop advanced PDR.

The diagnosis and ongoing management of PDR rely on standardized clinical assessments and monitoring strategies. PDR is characterized by the presence of abnormal new vessels (neovascularization) in the retina.[1]and is diagnosed through fundus examination by board-certified ophthalmologists, often classified using the International Clinical Diabetic Retinopathy Disease Severity Scale.[6] Regular retinal screening programs are vital for early detection, as their implementation has been associated with a lower proportion of patients requiring referral to ophthalmology.[38]Monitoring strategies involve routine screening where approximately 60% of diabetic patients may have non-proliferative diabetic retinopathy (NPDR) and about 20% present with active or regressed PDR.[4] Timely referral to specialists and appropriate treatment selection, such as photocoagulation, are critical to prevent vision loss, which PDR is a leading cause of among working-aged adults.[1]The overlap in patient cohorts with PDR and diabetic macular edema (DME) also necessitates comprehensive ocular assessment, as DME can occur independently or concurrently with PDR.[1]

Proliferative diabetic retinopathy is not merely an isolated ocular condition but often reflects broader systemic complications of diabetes. It is a significant complication of both type 1 and type 2 diabetes and can coexist with or predict the development of other diabetic complications.[24]Specifically, retinopathy, including its severe proliferative form, has been shown to predict incident renal dysfunction and diabetic nephropathy in patients with diabetes.[39] highlighting the shared microvascular pathology across different organ systems.

Beyond direct organ damage, systemic inflammatory markers and vascular factors are associated with PDR severity. For instance, elevated plasma levels of systemic soluble tumor necrosis factor receptors 1 and 2, and genetic variants related to P-selectin, have been linked to the severity of diabetic retinopathy in specific ethnic populations.[40]While a general clinical association between retinopathy and nephropathy is recognized, genetic studies have shown population-specific nuances, such as a lack of association of certain genetic variants for diabetic retinopathy with diabetic nephropathy in Taiwanese patients.[41] underscoring the complex interplay of genetic and environmental factors in multi-organ diabetic complications.

Frequently Asked Questions About Proliferative Diabetic Retinopathy

Section titled “Frequently Asked Questions About Proliferative Diabetic Retinopathy”

These questions address the most important and specific aspects of proliferative diabetic retinopathy based on current genetic research.


1. Why did my sibling with diabetes get eye problems but I haven’t yet?

Section titled “1. Why did my sibling with diabetes get eye problems but I haven’t yet?”

Even if you both have diabetes, your individual genetic makeup can make a big difference. Studies show that susceptibility to PDR can be as high as 52% inherited. This means specific genetic variations you carry, like those in genes such as NOX4 or WDR72, might make you more or less prone to developing severe eye complications compared to your sibling.

2. I manage my blood sugar really well, but should I still worry about PDR?

Section titled “2. I manage my blood sugar really well, but should I still worry about PDR?”

Yes, you should still be aware. While excellent blood sugar control is crucial for managing diabetes, research suggests that even intensive glucose control might not completely prevent PDR in everyone. Genetic factors play a significant role in your individual susceptibility, meaning some people have a higher risk despite good control due to their inherited predispositions.

3. Can a genetic test tell me if I’m at high risk for losing my vision from diabetes?

Section titled “3. Can a genetic test tell me if I’m at high risk for losing my vision from diabetes?”

Potentially, yes. Understanding your genetic makeup is becoming increasingly important for personalized risk assessment. Researchers are working to identify specific genetic variations, like those in genes such as NOX4 or RP1-90L14.1, that could indicate a higher individual risk for severe diabetic eye complications like PDR, even before symptoms appear.

4. My family is from Africa; does that change my risk for this eye problem?

Section titled “4. My family is from Africa; does that change my risk for this eye problem?”

Yes, your ethnic background can influence your genetic risk. Research shows that specific genetic associations for PDR can vary across different populations. For example, certain genetic markers, like those found near WDR72 or HLA-B, have been identified in African populations, suggesting unique genetic predispositions that might affect your personal risk.

5. Why do some people with diabetes get severe eye damage, and others only mild issues?

Section titled “5. Why do some people with diabetes get severe eye damage, and others only mild issues?”

This difference often comes down to individual genetic susceptibility. Even with similar diabetes duration and blood sugar control, some people have a genetic predisposition that makes them more likely to develop severe complications like PDR. Studies estimate that genetic factors can account for up to 52% of this susceptibility, influencing how your retina responds to diabetes-related damage.

6. If I get PDR, will I still be able to work and live independently?

Section titled “6. If I get PDR, will I still be able to work and live independently?”

PDR poses a critical threat to vision, and if not managed promptly, can lead to irreversible blindness. This loss of vision can severely impact your quality of life, independence, and ability to work, making early detection and urgent medical intervention for complications like vitreous hemorrhage or retinal detachment extremely important.

7. Is it true that genetics are more important than how long I’ve had diabetes?

Section titled “7. Is it true that genetics are more important than how long I’ve had diabetes?”

Both are very important, but genetics can play a surprisingly large role. While a longer duration of diabetes and poor glycemic control are strongly linked to PDR, your genetic makeup also significantly influences your risk. For some individuals, genetic factors might even override the benefits of strict glucose control, making them more susceptible to PDR despite managing their diabetes for a shorter period.

8. If my doctor says I have early eye damage, can I stop it from getting worse?

Section titled “8. If my doctor says I have early eye damage, can I stop it from getting worse?”

Regular monitoring and understanding your individual risk are key. Early stages of diabetic retinopathy, like NPDR, can progress to PDR, characterized by fragile new blood vessels. While intensive glucose control is important, recognizing your genetic predispositions can help your doctor tailor prevention strategies, as genetics influence how your body responds and if the condition will worsen.

9. Why is finding answers for my eye problem so slow for researchers?

Section titled “9. Why is finding answers for my eye problem so slow for researchers?”

It can be challenging for researchers to pinpoint exact causes for specific eye problems because studies often combine different types of diabetic eye conditions. When they aggregate PDR with less severe forms or with diabetic macular edema, it can dilute the unique genetic signals specific to PDR. This complexity makes it harder to identify the precise genetic factors that lead to particular conditions.

10. Why do some people develop these fragile blood vessels in their eyes, but others don’t?

Section titled “10. Why do some people develop these fragile blood vessels in their eyes, but others don’t?”

It’s largely due to individual genetic susceptibility. In PDR, the retina tries to compensate for poor blood supply by growing new, fragile vessels, a process called neovascularization. Your unique genetic makeup, with a heritability of up to 52%, determines how prone your body is to this abnormal vessel growth, even when faced with similar diabetes-related stress.


This FAQ was automatically generated based on current genetic research and may be updated as new information becomes available.

Disclaimer: This information is for educational purposes only and should not be used as a substitute for professional medical advice. Always consult with a healthcare provider for personalized medical guidance.

[1] Graham PS, et al. “Genome-wide association studies for diabetic macular edema and proliferative diabetic retinopathy.”BMC Med Genet, vol. 19, 2018, p. 71.

[2] Liu Y, Wang M, Morris AD, Doney AS, Leese GP, Pearson ER & Palmer CN. (2013). Glycemic exposure and blood pressure influencing progression and remission of diabetic retinopathy: a longitudinal cohort study in GoDARTS.Diabetes Care, 36, 3979–3984.

[3] Pollack S, et al. “Multiethnic Genome-wide Association Study of Diabetic Retinopathy using Liability Threshold Modeling of Duration of Diabetes and Glycemic Control.”Diabetes, vol. 67, no. 12, 2018, pp. 2707-2717.

[4] Meng, D. et al. “NADPH oxidase 4 insulin-stimulated expression angiogenesis in vitro.” PLoS One, 2012.

[5] Awata T, et al. “A genome-wide association study for diabetic retinopathy in a Japanese population: potential association with a long intergenic non-coding RNA.”PLoS One, vol. 9, no. 11, 2014, p. e111715.

[6] Sheu WH, et al. “Genome-wide association study in a Chinese population with diabetic retinopathy.”Hum Mol Genet, vol. 22, no. 15, 2013, pp. 3165–3173.

[7] Imamura, M. et al. “Genome-Wide Association Studies Identify Two Novel Loci Conferring Susceptibility to Diabetic Retinopathy in Japanese Patients with Type 2 Diabetes.”Hum Mol Genet, vol. 30, no. 8, 2021, pp. 719-729.

[8] Hsieh, A. R. et al. “Lack of association of genetic variants for diabetic retinopathy in Taiwanese patients with diabetic nephropathy.”BMJ Open Diabetes Res Care, vol. 8, no. 1, 2020, e000965.

[9] Meng W, Butterworth J, Bradley DT, Hughes AE, Soler V, Calvas P & Malecaze F. (2018). A genome-wide association study suggests new evidence for an association of the NADPH Oxidase 4 (NOX4) gene with severe diabetic retinopathy in type 2 diabetes.Acta Ophthalmol, 96, e904-e910.

[10] Liu C, Chen G, Bentley AR, Doumatey A, Zhou J, Adeyemo A, et al. (2019). Genome-wide association study for proliferative diabetic retinopathy in Africans.NPJ Genom Med, 4, 20.

[11] Ustinova M, et al. “Novel susceptibility loci identified in a genome-wide association study of type 2 diabetes complications in population of Latvia.” BMC Med Genomics, vol. 14, 2021, p. 18.

[12] Graham, P.S., et al. “Genome-wide association studies for diabetic macular edema and proliferative diabetic retinopathy.”BMC Medical Genetics, vol. 20, no. 1, 2019, p. 38.

[13] Hietala K, Forsblom C, Summanen P, Groop PH, FinnDiane Study Group. (2008). Heritability of proliferative diabetic retinopathy.Diabetes, 57(8), 2176–80.

[14] Huang, Y. C. et al. “Genome-wide association study of diabetic retinopathy in a Taiwanese population.”Ophthalmology, vol. 118, no. 4, 2011, pp. 642–8.

[15] Li J, Wang JJ, Yu Q, Chen K, Mahadev K & Zhang SX. (2010). Inhibition of reactive oxygen species by Lovastatin downregulates vascular endothelial growth factor expression and ameliorates blood-retinal barrier breakdown in db/db mice: role of NADPH oxidase 4. Diabetes, 59, 1528–1538.

[16] Li J, Wang JJ & Zhang X. (2014). NADPH Oxidase4-derived H2O2 promotes aberrant retinal neovascularization via activation of retinopathy.Ophthalmol Vis Sci, 55, E-Abstract 2955.

[17] Thallas-Bonke V, Jha JC, Gray SP et al. (2014). Nox-4 deletion reduces oxidative stress and injury by PKC-a-associated mech-anisms in diabetic nephropathy. Physiol Rep, 2, e12192.

[18] Marrero MB, Banes-Berceli AK, Stern DM, et al. (2006). Role of the JAK/STAT signaling pathway in diabetic nephropathy. Am J Physiol Renal Physiol, 290, F762–8.

[19] Jardeleza MS, Miller JW. (2009). Review of anti-VEGF therapy in proliferative diabetic retinopathy.Semin Ophthalmol, 24(2), 87–92.

[20] Takeuchi M, Sato T, Tanaka A, et al. (2015). Elevated levels of cytokines associated with Th2 and Th17 cells in vitreous fluid of proliferative diabetic retinopathy patients.PLoS One, 10, e0137358.

[21] MacKinnon JR, Knott RM, Forrester JV. (2004). Altered L-selectin expression in lymphocytes and increased adhesion to endothelium in patients with diabetic retinopathy.Br J Ophthalmol, 88, 1137–1141.

[22] Liu Q, Li J, Liu Z & Ma J. (2014). Salutary effect of fenofibrate on diabetic retinopathy via inhibiting oxidative stress-mediated Wnt pathway activation.Invest Ophthalmol Vis Sci, 55, E-Abstract 1027.

[23] Paine SK, Basu A, Mondal LK, et al. (2012). Association of vascular endothelial growth factor, transforming growth factor beta, and interferon gamma gene polymorphisms with proliferative diabetic retinopathy in patients with type 2 diabetes.Mol Vis, 18, 2749–2757.

[24] El-Asrar AMA, Al-Rubeaan KA, Al-Amro SA, et al. (2001). Retinopathy as a predictor of other diabetic complications.Int Ophthalmol, 24, 1–11.

[25] Lindholm E, Bakhtadze E, Sjögren M, Cilio CM, Agardh E, Groop L & Agardh CD. (2006). The -374 T/A polymorphism in the gene encoding RAGE is associated with diabetic nephropathy and retinopathy in type 1 diabetic patients.Diabetologia, 49, 2745–2755.

[26] Kitaoka, T. et al. “Expression of FGF5 in choroidal neovascular membranes associated with ARMD.” Curr Eye Res, 1997.

[27] Sandirasegarane, L. et al. “Enhanced stimulation of Akt-3/protein kinase B-gamma in human aortic smooth muscle cells.” Biochem Biophys Res Commun, 2003.

[28] Zukowska, Z. et al. “Neuropeptide Y: a novel mechanism for ischemic angiogenesis.” Trends Cardiovasc Med, 2003.

[29] Chen, Y.D. “Systemic soluble tumor necrosis factor receptors 1 and 2 are associated with severity of diabetic retinopathy in Hispanics.” Ophthalmology, 2012.

[30] Wang, J. et al. “Association of CFH and CFB gene polymorphisms with retinopathy in type 2 diabetic patients.” Mediators Inflamm, 2013.

[31] Sim~oes, M.J. et al. “Genetic variants in ICAM1, PPARGC1A and MTHFR are potentially associated with different phenotypes of diabetic retinopathy.” Ophthalmologica, 2014.

[32] Ferrington, D.A. et al. “Mitochondrial defects drive degenerative retinal diseases.” Trends Mol Med, 2020.

[33] Brazionis, L. et al. “Homocysteine and diabetic retinopathy.” Diabetes Care, 2008.

[34] D’Adamo, C.R. et al. “The CAPN2/CAPN8 locus on chromosome 1q is associated with variation in serum alpha-carotene concentrations.” J Nutrigenet Nutrigenomics, 2016.

[35] Her, J. et al. “The AAA-ATPase NVL2 is a telomerase component essential for holoenzyme assembly.” Biochem Biophys Res Commun, 2012.

[36] Galperin, E. et al. “EHD3: a protein that resides in recycling tubular and vesicular membrane structures.” 2002.

[37] Ebenezer, G. J. et al. “Impaired neurovascular repair in subjects with diabetes following experimental intracutaneous axotomy.” Brain, 2011.

[38] Leese, G. P., et al. “Implementation of national diabetes retinal screening programme is associated with a lower proportion of patients referred to ophthalmology.” Diabetic Medicine, vol. 22, 2005, pp. 1112–1115.

[39] Benitez-Aguirre, P. Z., et al. “Retinal vascular geometry predicts incident renal dysfunction in young people with type 1 diabetes.” Diabetes Care, vol. 35, 2012, pp. 599–604.

[40] Kuo, J. Z., et al. “Systemic soluble tumor necrosis factor receptors 1 and 2 are associated with severity of diabetic retinopathy in Hispanics.”Ophthalmology, vol. 119, 2012, pp. 1041–1046.

[41] Hsieh, A. R., et al. “Lack of association of genetic variants for diabetic retinopathy in Taiwanese patients with diabetic nephropathy.”BMJ Open Diabetes Research & Care, vol. 7, 2019, e000889.