Anemia and Diabetic Kidney Disease Had Joint Effect on Diabetic Retinopathy Among Patients With Type 2 Diabetes

Prediction of proliferative diabetic retinopathy using machine learning in Latino and non-Hispanic black cohorts with routine blood and urine testing

PURPOSE

The objective was to predict proliferative diabetic retinopathy (PDR) in non-Hispanic Black (NHB) and Latino (LA) patients by applying machine learning algorithms to routinely collected blood and urine laboratory results.

METHODS

Electronic medical records of 1124 type 2 diabetes patients treated at the Bronxcare Hospital eye clinic between January and December 2019 were analysed. Data collected included demographic information (ethnicity, age and sex), blood (fasting glucose, haemoglobin A1C [HbA1c] high-density lipoprotein [HDL], low-density lipoprotein [LDL], serum creatinine and estimated glomerular filtration rate [eGFR]) and urine (albumin-to-creatinine ratio [ACR]) test results and the outcome measure of retinopathy status. The efficacy of different machine learning models was assessed and compared. SHapley Additive exPlanations (SHAP) analysis was employed to evaluate the contribution of each feature to the model's predictions.

RESULTS

The balanced random forest model surpassed other models in predicting PDR for both NHB and LA cohorts, achieving an AUC (area under the curve) of 83%. Regarding sex, the model exhibited remarkable performance for the female LA demographic, with an AUC of 87%. The SHAP analysis revealed that PDR-related factors influenced NHB and LA patients differently, with more pronounced disparity between sexes. Furthermore, the optimal cut-off values for these factors showed variations based on sex and ethnicity.

CONCLUSIONS

This study demonstrates the potential of machine learning in identifying individuals at higher risk for PDR by leveraging routine blood and urine test results. It allows clinicians to prioritise at-risk individuals for timely evaluations. Furthermore, the findings emphasise the importance of accounting for both ethnicity and sex when analysing risk factors for PDR in type 2 diabetes individuals.

Mediation effect of antithrombin III between chronic renal insufficiency and chronic coronary artery disease in T2DM patients

PURPOSE

The study aimed to investigate the potential effect of Antithrombin III (ATIII) between chronic renal insufficiency and chronic coronary artery disease (chronic CAD) in type 2 diabetes mellitus (T2DM) patients.

METHODS

T2DM patients hospitalized in ZhongDa Hospital from 2013 to 2018 were enrolled. Relationships between renal function, ATIII, and chronic CAD risk were explored using multivariate regression models. Multiplicative and additive interactions were investigated between ATIII and renal function for CAD risk, and the role of ATIII was determined by bootstrap mediation analysis in patients with chronic renal dysfunction.

RESULTS

A total of 4197 patients were included in the study, with a chronic CAD prevalence of 23.02%. Low ATIII level was statistically associated with chronic renal insufficiency and elevated CAD risk even after adjustments (P < 0.05). A positive correlation between renal function and ATIII was demonstrated, and each 1 SD increase in renal function, ATIII increased by 2.947% (2.406-3.488%, P < 0.001) and 0.969% (0.297-1.642%, P < 0.001) in crude and adjusted models respectively. Patients with decreased renal function and ATIII were at the highest chronic CAD risk (OR = 1.51, 95%CI:1.15-1.98, P < 0.05), while no multiplicative and additive interaction effects were significant. Bootstrap mediation analysis estimated that ATIII mediated approximately 4.27% of the effect of chronic renal insufficiency on chronic CAD risk.

CONCLUSION

ATIII may serve as a mediator between chronic renal insufficiency and chronic CAD, providing mechanistic clues for renal-heart association and new insight into clinical therapies.

The L-shape relationship between hemoglobin, albumin, lymphocyte, platelet score and the risk of diabetic retinopathy in the US population

Background

The primary aim of this study was to investigate the correlation between diabetic retinopathy (DR) and the HALP score (hemoglobin, albumin, lymphocyte, and platelet) in individuals with diabetes within the United States population.

Methods

This cross-sectional investigation was based on the National Health and Nutrition Examination Survey (NHANES) database from 2003-2018. The following module calculated the HALP score: HALP score = [lymphocytes (/L) × hemoglobin (g/L) × albumin (g/L)]/platelets (/L). By performing the receiver operating characteristic (ROC) analysis, the optimal cutoff value of HALP was ascertained. Restricted cubic splines (RCS), multivariable logistic regression analysis, sensitivity analysis, and subgroup analysis were conducted to evaluate the effect of the HALP score on DR patients. Finally, the decision curve analysis (DCA) and clinical impact curve (CIC) were conducted to estimate the predictive power and clinical utility of the HALP score with clinical indicators.

Results

According to the cutoff value (42.9) determined by the ROC curve, the participants were stratified into a lower HALP group (HALPlow) and a higher HALP group (HALPhigh). An L-shaped relationship between HALP score and DR risk was presented in the RCS model (P for nonlinearity <0.001). The DR risk sharply decreased with the increase of HALP, and the decline reached a plateau when HALP was more than 42.9. After fully adjustment, the multivariate logistic regression analysis found that HALPlow was an independent risk factor for DR (OR = 1.363, 95% CI: 1.111-1.671, P < 0.001). Besides, sensitivity analysis showed consistent results. Furthermore, the combination of HALP score and clinical indicators demonstrated predictive power and clinical utility, as shown by the ROC curve, DCA, and CIC.

Conclusion

The HALP score has an L-shaped correlation with the risk of DR, and thus, the HALP score may contribute to the timely intervention of diabetes patients.

Associated factors of diabetic retinopathy in type 1 and 2 diabetes in Limpopo province in South Africa

Background

Diabetic retinopathy (DR) is the major cause of vision impairment or blindness in individuals who have diabetes. It has accounted for 2.6% of all cases of blindness, and 1.9% of all cases of vision impairments globally. There is a lack of data on the prevalence of diabetic retinopathy and its associated factors amongst diabetic rural populations. Hence, the current study aimed to determine factors associated with diabetic retinopathy (DR) among diabetes mellitus (DM) patients undergoing diabetic therapy.

Methods

The study was cross-sectional in design and the participants were selected using convenient sampling. STATA version 15 software was used for data analysis. Chi-square was used to compare proportions. Logistic regression was used to determine the relationship between DR and associated risk factors.

Results

The prevalence of DR was 35.3%, of which 32% were mild and 3.4% were moderate non-proliferative DR (NPDR). Females were more unemployed than males (32.1% versus 16.8%, p=0.0058). Males were found to drink alcohol (21.8% versus 1.9%, p<0.001) and smoke cigarettes (4% versus 0.3%, p=0.0034) more than females. Being aged ≥ 55 years (OR: 2.7, 95% CI: 1.6-4.4), with matric qualification (OR: 0.6; 95% CI: 0.4-1.0); employed (OR: 1.4, 95% CI: 1.2-1.6); having high systolic blood pressure (OR=1.4, 95%CI=1.1-1.7) were the independent determinants of DR.

Conclusions

The prevalence of diabetic retinopathy was 34%. DR was determined by high systolic blood pressure, old age, and employment. Although not statistically significant, gender, hyperglycemic state, poor glycemic control, smoking, and increased body mass index (BMI) were associated with increased risk of developing DR.

Association Between Anemia Status and the Risk of Different Types of Heart Failure: A RCSCD-TCM Study in China

We investigated the association between anemia status and the risk of heart failure (HF) in patients with coronary heart disease (CHD) based on a multi-center, large-sample and retrospective cross-sectional study including 89,207 patients. Heart failure was categorized as HF with reduced ejection fraction (HFrEF), HF with preserved ejection fraction (HFpEF), and HF with mid-range ejection fraction (HFmrEF). In multi-adjusted models, compared with patients without anemia, mild anemia (odds ratio [OR] 1.71; 95% confidence interval [CI] 1.53-1.91; P < .001), moderate anemia (OR 3.68; 95% CI, 3.25-4.17; P < .001), and severe anemia (OR 8.02; 95% CI, 6.50-9.88; P < .001) were associated with the risk of HF among CHD patients. Men aged <65 years were more likely to develop HF. In subgroup analyses, the multi-adjusted ORs and 95% CI of HFpEF, HFrEF, and HFmrEF related to anemia were 3.24 (95% CI 1.43-7.33), 2.22 (95% CI 1.28-3.84), and 2.55 (95% CI 2.24-2.89), respectively. These findings suggest that anemia might be associated with increased risk of different types of HF, especially HFpEF.

The relationship between diabetic retinopathy and diabetic nephropathy in type 2 diabetes

Context

Diabetic retinopathy (DR) and diabetic nephropathy (DN), are major microvascular complications of diabetes. DR is an important predictor of DN, but the relationship between the severity of DR and the pathological severity of diabetic glomerulopathy remains unclear.

Objective

To investigate the relationship between severity of diabetic retinopathy (DR) and histological changes and clinical indicators of diabetic nephropathy (DN) in patients with type 2 diabetes mellitus (T2DM).

Methods

Patients with T2DM (n=272) who underwent a renal biopsy were eligible. Severity of DR was classified as non-diabetic retinopathy, non-proliferative retinopathy, and proliferative retinopathy (PDR). Relationship between DN and DR and the diagnostic efficacy of DR for DN were explored.

Results

DN had a higher prevalence of DR (86.4%) and DR was more severe. The sensitivity and specificity of DR in DN were 86.4% and 78.8%, while PDR was 26.4% and 98.5%, respectively. In DN patients, the severity of glomerular lesions (p=0.001) and prevalence of KW nodules (p<0.001) significantly increased with increasing severity of DR. The presence of KW nodules, lower hemoglobin levels, and younger age were independent risk factors associated with more severe DR in patients with DN.

Conclusion

DR was a good predictor of DN. In DN patients, the severity of DR was associated with glomerular injury, and presence of KW nodules, lower hemoglobin levels and younger age were independent risk factors associated with more severe DR.

Trial registration

ClinicalTrails.gov, NCT03865914.

Deep learning-based fundus image analysis for cardiovascular disease: a review

It is well established that the retina provides insights beyond the eye. Through observation of retinal microvascular changes, studies have shown that the retina contains information related to cardiovascular disease. Despite the tremendous efforts toward reducing the effects of cardiovascular diseases, they remain a global challenge and a significant public health concern. Conventionally, predicting the risk of cardiovascular disease involves the assessment of preclinical features, risk factors, or biomarkers. However, they are associated with cost implications, and tests to acquire predictive parameters are invasive. Artificial intelligence systems, particularly deep learning (DL) methods applied to fundus images have been generating significant interest as an adjunct assessment tool with the potential of enhancing efforts to prevent cardiovascular disease mortality. Risk factors such as age, gender, smoking status, hypertension, and diabetes can be predicted from fundus images using DL applications with comparable performance to human beings. A clinical change to incorporate DL systems for the analysis of fundus images as an equally good test over more expensive and invasive procedures may require conducting prospective clinical trials to mitigate all the possible ethical challenges and medicolegal implications. This review presents current evidence regarding the use of DL applications on fundus images to predict cardiovascular disease.

Research progress on the pathogenesis of diabetic retinopathy

Diabetic retinopathy is one of the most common and serious microvascular complications of diabetes mellitus. There are many factors leading to diabetic retinopathy, and its pathogenesis is still unclear. At present, there are still no effective measures for the early treatment of diabetic retinopathy, and the treatment options available when diabetes progresses to advanced stages are very limited, and the treatment results are often unsatisfactory. Detailed studies on the molecular mechanisms of diabetic retinopathy pathogenesis and the development of new therapeutic agents are of great importance. This review describes the potential pathogenesis of diabetic retinopathy for experimental studies and clinical practice.

Risk Factors for Diabetic Retinopathy in Chinese Patients with Different Diabetes Duration: Association of C-Peptide and BUN/Cr Ratio with Type 2 Diabetic Retinopathy

Background and Aim

Controlling the risk factors was the most effective strategy to prevent diabetic retinopathy (DR). This study aimed to recognize the risk factors of DR, and explores whether the effect of those factors is modified by diabetes mellitus (DM) duration.

Methods

A total of 1058 DM patients with information about DR assessment were included. DR was measured by a complete ophthalmic examination and was classified as having one or more distinct microaneurysms in the eyes. Data from the lab and clinical factors were gathered. Multivariate logistic analysis was used to examine the risk factors, and the best-fitting model was selected by a backward stepwise based on A1C.

Results

In the current study, 274 (25.9%) patients developed DR. In the entire subjects, baseline age, the level of C-peptide, and urinary creatinine were all presented as protective effects of DR, whose odds ratios (ORs) and 95% confidence intervals (CIs) were 0.79 (0.62, 0.99), 0.75 (0.61, 0.91), and 0.70 (0.52, 0.93), respectively. Conversely, systolic pressure (SBP), urinary albumin, and BUN/Cr ratio were the important risk factors for DR with ORs (95% CIs) 1.21 (1.01, 1.46), 1.55 (1.30, 1.84), and 1.33 (1.11, 1.59), respectively. In stratification analysis, females with higher SBP would be more likely to develop DR in the short-duration group, while C-peptide and urinary creatinine showed protective effects in the long-duration group. BUN/Cr ratio all presented as a risk factor, with ORs 1.38 (p = 0.041) and 1.33 (p = 0.014) in short- and long-duration groups, respectively.

Conclusion

Although renal functions presented a significant association with DR in all DM patients, the risk factors of DR varied widely in different disease-duration subjects. Target strategies to prevent DR should be put forward individually, considering the patient's DM duration. Improving the BUN/Cr ratio may be beneficial to delaying DR.

Diabetic Retinopathy - a Common Disease

Diabetic retinopathy (DR) is one of the most common complications of diabetes mellitus and one of the leading causes of visual impairment in working age individuals in the western world. The treatment of DR depends on its severity, so it is of great importance to detect patients as early as possible, in order to initiate early treatment and preserve vision. Despite currently insufficient screening participation, patients with diabetes already visit ophthalmological practices and clinics above average. Their medical care, including DR diagnostics and treatment has been making up an increasing proportion of ophthalmic activity for years. Since the prevalence of diabetes is increasing dramatically worldwide and a further increase is also predicted for Germany, the challenge for ophthalmologists is likely to grow considerably. As the same time, the diagnostic possibilities for differentiating DR and the therapeutic measures, especially with IVOM therapy, are becoming more and more complex, which increases the time burden in everyday clinical practice. The hope to avoid healthcare deficits and to further improve screening rates and visual acuity prognosis in patients with DR is based, among other things, on camera-assisted screening supported by artificial intelligence. Better diabetes management to reduce the prevalence of DR, as well as longer-acting drugs to treat DR, could also improve the care and help reduce the burden on ophthalmology practices.

Establishment and validation of a prognostic nomogram for long-term low vision after diabetic vitrectomy

Purpose

We aimed to evaluate the risk factors and develop a prognostic nomogram of long-term low vision after diabetic vitrectomy.

Methods

This retrospective study included 186 patients (250 eyes) that underwent primary vitrectomy for proliferative diabetic retinopathy with a minimum follow-up period of one year. Patients were assigned to the training cohort (200 eyes) or validation cohort (50 eyes) at a 4:1 ratio randomly. Based on a cutoff value of 0.3 in best-corrected visual acuity (BCVA) measurement, the training cohort was separated into groups with or without low vision. Univariate and multivariate logistic regression analyses were performed on preoperative systemic and ocular characteristics to develop a risk prediction model and nomogram. The calibration curve and the area under the receiver operating characteristic curves (AUC) were used to evaluate the calibration and discrimination of the model. The nomogram was internally validated using the bootstrapping method, and it was further verified in an external cohort.

Results

Four independent risk factors were selected by stepwise forward regression, including tractional retinal detachment (β=1.443, OR=4.235, P<0.001), symptom duration ≥6 months (β=0.954, OR=2.595, P=0.004), preoperative BCVA measurement (β=0.540, OR=1.716, P=0.033), and hypertension (β=0.645, OR=1.905, P=0.044). AUC values of 0.764 (95% CI: 0.699-0.829) in the training cohort and 0.755 (95% CI: 0.619-0.891) in the validation cohort indicated the good predictive ability of the model.

Conclusion

The prognostic nomogram established in this study is useful for predicting long-term low vision after diabetic vitrectomy.

The Association Between Lipid Profile and Subfoveal Choroidal Thickness in Chinese Patients with Proliferative Diabetic Retinopathy Secondary to Type 2 Diabetes

Introduction

Extensive studies have studied the factors associated with subfoveal choroidal thickness (SFCT). However, studies of the association between lipid profile and SFCT in patients with proliferative diabetic retinopathy (PDR) in type 2 diabetes remain limited. Thus, we aimed to investigate the relationship between lipid profile and SFCT in patients with PDR.

Materials and Methods

This was a retrospective cross-sectional study. The included participants were inpatients who underwent vitrectomy for PDR with type 2 diabetes and contralateral eyes of PDR patients meeting the criteria. Multivariate linear regression analysis was used to determine the independent association between lipid profile and SFCT.

Results

A total of 131 participants with PDR were enrolled in the final analysis. The average age of the participants was 55.76 ± 9.88 years, and the average SFCT was 276.10 ± 92.92 μm. Multivariate linear regression model results showed that in the fully adjusted model, total cholesterol, high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) had a negative association with SFCT (β = -16.51, 95% CI: -29.57, -3.46; P = 0.0148; β = -42.65, 95% CI: -82.60, -2.70; P = 0.0390; β = -17.89, 95% CI: -33.24, -2.54; P = 0.0245, respectively), while triglyceride was not significantly associated with SFCT (β = 5.23, 95% CI: -18.57, 29.02; P = 0.6678). Furthermore, the results of stratified analysis showed that except for triglyceride, the trends of total cholesterol, HDL-C, LDL-C, and SFCT were consistent among different stratifications in participants.

Conclusion

The cholesterol profile had a significant negative association with SFCT in Chinese PDR patients, but triglyceride was not significantly associated with SFCT. This suggests that these systemic imbalances contribute to choroidal changes, and often coexist in diabetic patients.

Development and validation of a diabetic retinopathy risk prediction model for middle-aged patients with type 2 diabetes mellitus

Objectives

The study aims to establish a predictive nomogram of diabetic retinopathy(DR) for the middle-aged population with type 2 diabetes mellitus (T2DM).

Methods

This retrospective study screened 931 patients with T2DM between 30 and 59 years of age from the 2011-2018 National Health and Nutrition Examination Survey database. The development group comprised 704 participants from the 2011-2016 survey, and the validation group included 227 participants from the 2017-2018 survey. The least absolute shrinkage and selection operator regression model was used to determine the best predictive variables. The logistic regression analysis built three models: the full model, the multiple fractional polynomial (MFP) model, and the stepwise (stepAIC) selected model. Then we decided optimal model based on the receiver operating characteristic curve (ROC). ROC, calibration curve, Hosmer-Lemeshow test, and decision curve analysis (DCA) were used to validate and assess the model. An online dynamic nomogram prediction tool was also constructed.

Results

The MFP model was selected to be the final model, including gender, the use of insulin, duration of diabetes, urinary albumin-to-creatinine ratio, and serum phosphorus. The AUC was 0.709 in the development set and 0.704 in the validation set. According to the ROC, calibration curves, and Hosmer-Lemeshow test, the nomogram demonstrated good coherence. The nomogram was clinically helpful, according to DCA.

Conclusion

This study established and validated a predictive model for DR in the mid-life T2DM population, which can assist clinicians quickly determining who is prone to develop DR.

Relationship between renal function and prognosis of Chinese proliferative diabetic retinopathy patients undergoing the first vitrectomy: protocol for a prospective cohort study

INTRODUCTION

China has the largest number of adults with diabetes aged 20-79 years (116.4 million) in 2019. Due to the socioeconomic condition or a lack of awareness of diabetic complications, many adults with diabetes have proliferative diabetic retinopathy (PDR) or renal function impairment at their first visit to the clinic for a sudden loss of vision, and pars plana vitrectomy (PPV) is required for their treatment. Risk factors for the outcomes and complications of PPV surgery in PDR patients have been widely explored in many epidemiological studies and clinical trials. However, few prospective studies have analysed the association between renal function and surgical outcomes in PDR.

METHODS AND ANALYSIS

This is a single-centre, prospective cohort study of PDR patients with type 2 diabetes mellitus who have definite indications for PPV surgery with or without renal function impairment. We will consecutively enrol PDR patients who meet the inclusion and exclusion criteria from November 2020 to December 2023. Each participant will be followed up for at least 6 months after surgery. Clinical data from medical records and vitreous fluid will be collected.Demographic characteristics and study outcomes will be summarised using descriptive statistics. The variation will be described and evaluated using the χ² test or Kruskal-Wallis test. Generalise additive mixed models will be used to explore the association between the renal profile and surgical outcomes including BCVA, and retinal and choroidal microvasculature/microstructure. Multivariate ordinal regression analysis will be used to detect the independent association between renal profile and BCVA changes, and smooth curve fitting will be employed to briefly present the tendency.

ETHICS AND DISSEMINATION

The trial has received ethical approval from the West China Hospital of Sichuan University. Results of this trial will be disseminated through publication in peer-reviewed journals and presentations at local and international meetings.

TRIAL REGISTRATION NUMBER

ChiCTR2000039698.

An online diabetic retinopathy screening tool for patients with type 2 diabetes

AIM

To develop a useful diabetic retinopathy (DR) screening tool for patients with type 2 diabetes mellitus (T2DM).

METHODS

A DR prediction model based on the Logistic regression algorithm was established on the development dataset containing 778 samples (randomly assigned to the training dataset and the internal validation dataset at a ratio of 7:3). The generalization capability of the model was assessed using an external validation dataset containing 128 samples. The DR risk calculator was developed through WeChat Developer Tools using JavaScript, which was embedded in the WeChat Mini Program.

RESULTS

The model revealed risk factors (duration of diabetes, diabetic nephropathy, and creatinine level) and protective factors (annual DR screening and hyperlipidemia) for DR. In the internal and external validation, the recall ratios of the model were 0.92 and 0.89, respectively, and the area under the curve values were 0.82 and 0.70, respectively.

CONCLUSION

The DR screening tool integrates education, risk prediction, and medical advice function, which could help clinicians in conducting DR risk assessments and providing recommendations for ophthalmic referral to increase the DR screening rate among patients with T2DM.

Factors protecting against diabetic retinopathy in a geriatric Indian cohort

Purpose:

Diabetic retinopathy (DR) is a potentially sight-threatening complication of diabetes mellitus. The majority of cases are in older adults. This study aims to evaluate modifiable and nonmodifiable protective factors against DR in a geriatric Indian population.

Methods:

This retrospective observational study uses data from a multitiered ophthalmology network to evaluate several demographic and clinical variables against diabetic retinopathy and visual acuity.

Results:

Our data show that high myopia, the female sex, and no cataract surgery are associated with lower prevalence of DR (OR = 0.21, 0.65, and 0.76, respectively; P < 0.001). We also found that among those with DR, people categorized as payers, retirees, and those living in urban or metropolitan areas have better visual acuity (OR = 0.65, 0.65, 0.83, and 0.73, respectively; P < 0.001). Among those with DR, females, presence of cataracts, and no cataract surgery had lower associations with sight-threatening DR (STDR) (OR = 0.68, 0.37, and 0.76, respectively; P < 0.001). Prevalence of DR decreased in older age groups while controlling for DM duration.

Conclusion:

It is probable that high myopia, the female sex, and better glycemic control are protective against DR and STDR in our study cohort of adults over 60 years of age. It is possible that occupations involving manual labor, delayed cataract surgery, and living past the age of 70 are also protective against DR.

Distribution of diabetic retinopathy in diabetes mellitus patients and its association rules with other eye diseases

The study aims to explore the distribution characteristics and influencing factors of diabetic retinopathy (DR) in diabetes mellitus (DM) patients and association rules of eye diseases in these patients. Data were obtained from 1284 DM patients at Henan Provincial People's Hospital. Association rules were employed to calculate the probability of the common occurrence of eye-related diseases in DM patients. A web visualization network diagram was used to display the association rules of the eye-related diseases in DM patients. DR prevalence in people aged < 40 years (≥ 58.5%) was higher than that in those aged 50-60 years (≤ 43.7%). Patients with DM in rural areas were more likely to have DR than those in urban areas (56.2% vs. 35.6%, P < 0.001). DR prevalence in Pingdingshan City (68.4%) was significantly higher than in other cities. The prevalence of DR in patients who had DM for ≥ 5 years was higher than in other groups (P < 0.001). About 33.07% of DM patients had both diabetic maculopathy and DR, and 36.02% had both diabetic maculopathy and cataracts. The number of strong rules in patients ≥ 60 years old was more than those in people under 60 in age, and those in rural areas had more strong rules than those in urban areas. DM patients with one or more eye diseases are at higher risks of other eye diseases than general DM patients. These association rules are affected by factors such as age, region, disease duration, and DR severity.