Association between Free Thyroxine Levels and Diabetic Retinopathy in Euthyroid Patients with Type 2 Diabetes Mellitus

Thyroid Hormone Signaling in Retinal Development and Function: Implications for Diabetic Retinopathy and Age-Related Macular Degeneration

Thyroid Hormones (THs) play a central role in the development, cell growth, differentiation, and metabolic homeostasis of neurosensory systems, including the retina. The coordinated activity of various components of TH signaling, such as TH receptors (THRs) and the TH processing enzymes deiodinases 2 and 3 (DIO2, DIO3), is required for proper retinal maturation and function of the adult photoreceptors, Müller glial cells, and pigmented epithelial cells. Alterations of TH homeostasis, as observed both in frank or subclinical thyroid disorders, have been associated with sight-threatening diseases leading to irreversible vision loss i.e., diabetic retinopathy (DR), and age-related macular degeneration (AMD). Although observational studies do not allow causal inference, emerging data from preclinical models suggest a possible correlation between TH signaling imbalance and the development of retina disease. In this review, we analyze the most important features of TH signaling relevant to retinal development and function and its possible implication in DR and AMD etiology. A better understanding of TH pathways in these pathological settings might help identify novel targets and therapeutic strategies for the prevention and management of retinal disease.

Association between HDL cholesterol with diabetic retinopathy in diabetic patients: a cross-sectional retrospective study

OBJECTIVE

Diabetic patients are often comorbid with dyslipidemia, however, the relationship between high-density lipoprotein cholesterol(HDL-C) and diabetic retinopathy (DR) in the adult diabetic population remains to be fully elucidated.The aim of this study is to evaluate the associations between HDL-C and DR in the United States adults with diabetes.

METHODS

A total of 1708 participants from the National Health and Nutrition Examination Survey (NHANES) 2005-2008 were enrolled in the present study. Fundus images of all study subjects were captured and evaluated using a digital camera and an ophthalmic digital imaging system, and the diagnosis of DR was made by the severity scale of the Early Treatment Diabetic Retinopathy Study (ETDRS).Roche Diagnostics were used to measure serum HDL-C concentration. The relationship of DR with HDL-C was investigated using multivariable logistic regression. The potential non-line correlation was explored with smooth curve fitting approach.

RESULTS

The fully-adjusted model showed that HDL-C positively correlated with DR(OR:1.69, 95%CI: 1.25-2.31).However, an inverted U-shaped association between them was observed by applying the smooth curve fitted method. The inflection point of HDL-C(1.99mmol/l) was calculated by utilizing the two-piecewise logistic regression model. In the subgroup analysis, the inverted U-shaped nonlinear correlation between HDL-C and DR was also found in female, Non-Hispanic White, and lower age groups.

CONCLUSION

Our study revealed an inverted U-shaped positive relationship between HDL-C and DR.The findings may provide us with a more comprehensive understanding of the association between HDL-C and DR.

Association of HbA1c Variability with Vibrating Perception Threshold in Middle-Aged and Elderly Patients with Type 2 Diabetes Mellitus: A Retrospective Cohort Study

Purpose

To investigate the relationship between HbA1c variability and vibrating perception threshold (VPT) in middle-aged and elderly patients with type 2 diabetes mellitus (T2DM).

Patients and Methods

A total of 367 middle-aged and elderly patients with T2DM were enrolled. All patients were categorized into the control and vibration sensation deficiency group (VSD) based on VPT. Clinical data were collected. The coefficient of variation of glycated hemoglobin A1c (HbA1c-CV) and the mean glycated hemoglobin A1c(HbA1c-Mean) are considered as indexes of HbA1c variability. Multivariate logistic regression analysis, the generalized linear model and ROC curve correlation analysis were used to analyze the correlation of various factors and VPT.

Results

The multivariate logistic regression analysis revealed that age, systolic blood pressure (SBP), and HbA1c-CV were identified as risk factors for vibration sensation deficiency in middle-aged and elderly patients with T2DM, while estimated glomerular filtration rate (eGFR), triiodothyronine (T3), and alanine aminotransferase (ALT) were considered as protective factors. In the unadjusted generalized linear model, a significant association was observed between HbA1c-CV and VPT values. After adjusting for age, diabetic duration, SBP, homeostatic model assessment for beta-cell function (HOMA-β), ALT, eGFR, T3, 24-hour urinary protein excretion levels, and HbA1c-Mean, HbA1c-CV remained significantly correlated with VPT values on both sides. (left side, B=2.560, 95% CI 1.298~3.823; P<0.001; right side, B=2.608, 95% CI 1.498~3.718, P<0.001). The area under the curve (AUC) for HbA1c-CV and VSD prevalence was 0.723, with a sensitivity of 79.85%, specificity of 56.22%.

Conclusion

The risk of developing VSD increases proportionally with higher HbA1c-CV levels in middle-aged and elderly patients with T2DM. Reaching and maintaining blood glucose stability is essential to the mitigation of diabetes peripheral neuropathy occurrence.

Local modulation of thyroid hormone signaling in the retina affects the development of diabetic retinopathy

Thyroid hormone (TH) dyshomeostasis is associated with poor prognosis in acute and prolonged illness, but its role in diabetic retinopathy (DR) has never been investigated. Here, we characterized the TH system in the retinas of db/db mice and highlighted regulatory processes in MIO-M1 cells. In the db/db retinas, typical functional traits and molecular signatures of DR were paralleled by a tissue-restricted reduction of TH levels. A local condition of low T3 (LT3S) was also demonstrated, which was likely to be induced by deiodinase 3 (DIO3) upregulation, and by decreased expression of DIO2 and of TH receptors. Concurrently, T3-responsive genes, including mitochondrial markers and microRNAs (miR-133-3p, 338-3p and 29c-3p), were downregulated. In MIO-M1 cells, a feedback regulatory circuit was evidenced whereby miR-133-3p triggered the post-transcriptional repression of DIO3 in a T3-dependent manner, while high glucose (HG) led to DIO3 upregulation through a nuclear factor erythroid 2-related factor 2-hypoxia-inducible factor-1 pathway. Finally, an in vitro simulated condition of early LT3S and hyperglycemia correlated with reduced markers of both mitochondrial function and stress response, which was reverted by T3 replacement. Together, the data suggest that, in the early phases of DR, a DIO3-driven LT3S may be protective against retinal stress, while, in the chronic phase, it not only fails to limit HG-induced damage, but also increases cell vulnerability likely due to persistent mitochondrial dysfunction.

Association of thyroid stimulating hormone and time in range with risk of diabetic retinopathy in euthyroid type 2 diabetes

AIMS

Diabetic retinopathy (DR) can occur even in well-controlled type 2 diabetes, suggesting residual risks of DR in this population. In particular, we investigated the combined effect of thyroid function and glycaemic control assessed by an emerging metric, time in range (TIR) with DR.

MATERIALS AND METHODS

In this cross-sectional study, a total of 2740 euthyroid patients with type 2 diabetes were included. Thyroid indicators, including thyroid-stimulating hormone (TSH), free triiodothyronine, free thyroxine, thyroid peroxidase antibody and thyroglobulin antibody, were measured. TIR was measured using continuous glucose monitoring data.

RESULTS

Overall, the multivariable-adjusted odds ratios (ORs) for DR across ascending tertiles of TSH were 1.00 (reference), 1.06 (95% confidence interval [CI] 0.85-1.32), and 1.48 (95% CI 1.19-1.85). Even in well-controlled participants who achieved a TIR target of >70% (n = 1449), the prevalence of DR was 23.8%, which was significantly related to TSH (OR = 1.54, 95% CI 1.12-2.12, highest vs. lowest TSH tertile). Participants were then classified into 6 groups by the joint categories of TIR (>70%, ≤70%) and TSH (tertiles), and the multivariable-adjusted ORs for DR were highest in TIR ≤70% and the highest TSH tertile group (OR = 1.96, 95% CI 1.41-2.71) when compared with the TIR >70% and the lowest TSH tertile group.

CONCLUSIONS

In type 2 diabetic patients with well-controlled glycaemic status, higher TSH within the normal range was associated with an increased risk of DR. The combination of suboptimal TSH and TIR further increased the risk of DR.

The effect of Central and peripheral thyroid resistance indices on diabetic retinopathy: a study of hospitalized euthyroid patients with T2DM in China

OBJECTIVE

This study aims to explore the correlation between central and peripheral thyroid resistance indices and diabetic retinopathy(DR) in patients with type 2 diabetes mellitus (T2DM), so as to provide a clinical basis for the prevention and treatment of diabetic retinopathy.

METHODS

This study retrospectively analyzed 1249 euthyroid patients with T2DM hospitalized in the Department of Endocrinology, Peking University International Hospital from January 2017 to June 2022, including 852 males and 397 females, with an average age of 54.73 ± 13.40 years. According to the degree of DR, the patients were divided into three groups including the no diabetic retinopathy (NDR) group, non-proliferative diabetic retinopathy (NPDR) group and proliferative diabetic retinopathy (PDR) group.

RESULTS

Free thymidine (FT4), thyroid stimulating hormone (TSH), thyroid feedback quantile index (TFQI), thyrotropin-T4 resistance index (TT4RI), thyroid stimulating hormone index (TSHI) and free triiodothyronine/free thyroxine (FT3/FT4) levels among the three groups were significantly different, with the NDR group having lowest TSH, TFQI, TT4QI, TSHI and the highest in the PDR group (all p < 0.05). Logistic regression showed that after adjusting for age, body mass index (BMI), sex, diabetes duration, blood pressure, blood lipid, HbA1c, lower level of FT4 was an independent risk factor for DR, high level of TSH, TFQI, TSHI and TT4RI were independent risk factors for DR. Central and peripheral thyroid sensitivity indices have predictive value for DR, the overall predictive accuracy of FT3/FT4 was 0.61 (95%CI 0.57, 0.65), the overall predictive accuracy of TFQI was 0.66(95%CI 0.63, 0.70), the overall predictive accuracy of TSHI was 0.66(95%CI 0.62, 0.68), the overall predictive accuracy of TT4RI was 0.63 (95%CI 0.59, 0.66).

CONCLUSION

The reduction of central and peripheral thyroid hormone sensitivity is an independent risk factor for DR. These results can help predict the risk of the occurrence and development of DR, which may provide a clinical basis for the prevention and treatment of DR in T2DM patients.

Association between thyroid function and diabetes peripheral neuropathy in euthyroid type 2 diabetes mellitus patients

Previous studies disclosed that a high thyroid stimulating hormone level is an independent risk factor for diabetes peripheral neuropathy (DPN) in subclinical hypothyroidism (SCH) patients with type 2 diabetes mellitus (T2DM). However, whether thyroid metabolism has an effect on DPN in euthyroid T2DM patients remains unknown. The aim of this study was to identify the association between thyroid function and DPN in euthyroid T2DM patients. A set of 580 euthyroid T2DM patients was enrolled in the current study and stratified into DPN and Non-DPN groups. Mann-Whitney U test was performed to analyze the continuous variables of biochemical and thyroid metabolism indicators, and the Chi-square test was used to compare the categorical variables. Spearman correlation analysis was performed to analyze the relationship between clinical indicators and free thyroxine (FT4). By using the logistic regression analysis, the prevalence of DPN in different thyroid function indicators were evaluated. T2DM patients with DPN had obviously lower levels of aspartate aminotransferase (AST), alpha-hydroxybutyric dehydrogenase (α-HBDH), superoxide dismutase (SOD), calcium (Ca), creatinine (Cr), uric acid (UA), retinol binding protein (RBP), total protein (TP), albumin (ALB), alanine aminotransferase (ALT) and FT4 than the T2DM patients without DPN (P < 0.05). FT4 was associated with TP, prealbumin (PA), ALB, SOD, anion gap (AG), Ca, chlorine (Cl), UA, RBP, apoprotein A (Apo A), apoprotein B (Apo B), apoprotein E (Apo E), and total cholesterol (TC). According to the FT4 quartile, participants were sequentially divided into four groups to compare the prevalence of DPN between each group. The data suggested that the prevalence of DPN in these four groups was 53.79%, 53.28%, 54.97%, 38.10%, respectively. Moreover, compared with quartile 4, patients in quartile 1, 2, 3 all had a significantly higher risk of DPN (P = 0.007, P = 0.011, P = 0.004). The level of FT4 was negatively correlated with the prevalence of DPN in euthyroid T2DM patients.

Association Between Sensitivity to Thyroid Hormone Indices and Diabetic Retinopathy in Euthyroid Patients with Type 2 Diabetes Mellitus

OBJECTIVE

Normal thyroid hormone (TH) levels and their relation to microvascular complications in patients with type 2 diabetes mellitus (T2DM) have been studied. However, the relationship between TH sensitivity and diabetic retinopathy (DR) remains unclear. Thus, this study aimed to investigate the relationship between TH sensitivity and the risk of DR in euthyroid T2DM patients.

METHODS

This retrospective study analyzed 422 T2DM patients and calculated their sensitivity to TH indices. Multivariable logistic regression, generalized additive model, and subgroup analysis were performed to examine the association between sensitivity to TH indices and DR risk.

RESULTS

After adjusting for covariates, the binary logistic regression model showed no statistically significant association between the sensitivity of TH indices and the risk of DR in euthyroid T2DM patients. However, a non-linear relationship was found between sensitivity to TH indices (thyroid-stimulating hormone index, thyroid feedback quantile index [TFQI]) and the risk of DR in the crude model; TFQI and DR in the adjusted model. The inflection point of the TFQI was 0.23. The effect size (odds ratio) on the left and right of the inflection point were 3.19 (95% confidence interval [CI]: 1.24 to 8.17 p=0.02) and 0.11 (95% CI: 0.01, 0.93 p=0.04), respectively. Moreover, this relationship was maintained by men stratified by sex. In euthyroid patients with T2DM, an approximate inverted U-shaped relationship and a threshold effect were demonstrated between TH index sensitivity and DR risk with sex differences. This study provided an in-depth understanding of the relationship between thyroid function and DR, which has important clinical implications for risk stratification and individual prediction.

Monosodium Glutamate-Induced Mouse Model With Unique Diabetic Retinal Neuropathy Features and Artificial Intelligence Techniques for Quantitative Evaluation

Objective

To establish an artificial intelligence-based method to quantitatively evaluate subtle pathological changes in retinal nerve cells and synapses in monosodium glutamate (MSG) mice and provide an effective animal model and technique for quantitative evaluation of retinal neurocytopathies.

Methods

ICR mice were subcutaneously injected with MSG to establish a model of metabolic syndrome. We then established a mouse model of type 1 diabetes, type 2 diabetes, and KKAy mouse model as control. The HE sections of the retina were visualized using an optical microscope. AI technology was used for quantitative evaluation of the retinal lesions in each group of rats. The surface area custom parameters of the retinal nerve fiber layer (RNFL), inner plexiform layer (IPL), inner nuclear layer (INL), and outer plexiform layer (OPL) were defined as SR, SIPL, SINL, and SOPL, respectively. Their heights were defined as HR, HIPL, HINL, and HOPL, and the number of ganglion cells was defined as A. Then, the attention-augmented fully convolutional Unet network was used to segment the retinal HE images, and AI technology to identify retinal neurocytopathies quantitatively.

Results

The attention-augmented fully convolutional Unet network increased PA and IOU parameters for INL, OPL, RNFL, and ganglion cells and was superior in recognizing fine structures. A quantitative AI identification of the height of each layer of the retina showed that the heights of the IPL and INL of the MSG model were significantly less than those of the control groups; the retinas of the other diabetic models did not exhibit this pathological feature. The RNFLs of type 2 diabetes were thinner, and the characteristics of retinopathy were not obvious in the other animal models. The pathological changes seen on HE images were consistent with the results of the quantitative AI evaluation. Immunohistochemistry results showed that NMDAR2A, GluR2, and NRG1 were significantly downregulated in the retina of MSG mice.

Conclusions

The MSG retinopathy model is closely associated with neurotransmitter abnormalities and exhibits important characteristics of retinal neurodegeneration, making it suitable for studying retinal neurocytopathies. The AI recognition technology for retinal images established in the present study can be used for the quantitative and objective evaluation of drug efficacy.

Assessment of causal direction between thyroid function and cardiometabolic health: a Mendelian randomization study

BACKGROUND

Growing evidence have demonstrated that thyroid hormones have been involved in the processes of cardiovascular metabolism. However, the causal relationship of thyroid function and cardiometabolic health remains partly unknown.

METHODS

The Mendelian randomization (MR) was used to test genetic, potentially causal relationships between instrumental variables and cardiometabolic traits. Genetic variants of free thyroxine (FT4) and thyrotropin (TSH) levels within the reference range were used as instrumental variables. Data for genetic associations with cardiometabolic diseases were acquired from the genome-wide association studies of the FinnGen, CARDIoGRAM and CARDIoGRAMplusC4D, CHARGE, and MEGASTROKE. This study was conducted using summary statistic data from large, previously described cohorts. Association between thyroid function and essential hypertension (EHTN), secondary hypertension (SHTN), hyperlipidemia (HPL), type 2 diabetes mellitus (T2DM), ischemic heart disease (IHD), myocardial infarction (MI), heart failure (HF), pulmonary heart disease (PHD), stroke, and non-rheumatic valve disease (NRVD) were examined.

RESULTS

Genetically predicted FT4 levels were associated with SHTN (odds ratio = 0.48; 95% CI = 0.04−0.82,P = 0.027), HPL (odds ratio = 0.67; 95% CI = 0.18−0.88,P = 0.023), T2DM (odds ratio = 0.80; 95% CI = 0.42−0.86,P = 0.005), IHD (odds ratio = 0.85; 95% CI = 0.49−0.98,P = 0.039), NRVD (odds ratio = 0.75; 95% CI = 0.27−0.97,P = 0.039). Additionally, genetically predicted TSH levels were associated with HF (odds ratio = 0.82; 95% CI = 0.68−0.99,P = 0.042), PHD (odds ratio = 0.75; 95% CI = 0.32−0.82,P = 0.006), stroke (odds ratio = 0.95; 95% CI = 0.81−0.97,P = 0.007). However, genetically predicted thyroid function traits were not associated with EHTN and MI.

CONCLUSIONS

Our study suggests FT4 and TSH are associated with cardiometabolic diseases, underscoring the importance of the pituitary-thyroid-cardiac axis in cardiometabolic health susceptibility.

Serum Antithyroglobulin Antibody Levels Are Associated with Diabetic Retinopathy among Euthyroid Type 2 Diabetes Patients: A Hospital-Based, Retrospective Study

BACKGROUND

Previous studies have revealed that the variation of thyroid indicators may be associated with the risk of diabetic retinopathy (DR) among euthyroid type 2 diabetes (T2D) patients. But the specific conclusions are currently inconsistent.

METHODS

This is a hospital-based retrospective survey. We recruited 1,145 euthyroid T2D patients and checked the thyroid function and fundus photographs. The modified Airlie House classification system was used to categorize the stages of DR. The association between thyroid indicators and different stages of DR was analyzed.

RESULTS

We divided free triiodothyronine (FT3) into tertiles and found that the prevalence of mild nonproliferative DR (NPDR) was significantly higher in T2, compared with T1 (32.0% vs. 25.2%, p < 0.05). When FT3 was within the level of T2, FT3 could be an independent risk factor for mild NPDR (OR 1.426, 95% CI (1.031, 1.971), p < 0.05). In addition, the prevalence of severe NPDR and proliferative DR (PDR) was significantly higher in thyroglobulin antibody (TgAb) positive group (8.8% vs. 4.1%, p < 0.05) and vice versa (33.3% vs. 18.4%, p < 0.05). TgAb positivity was also an independent risk factor for severe NPDR and PDR (OR 2.212, 95% CI (1.244, 3.934), p < 0.05).

CONCLUSIONS

We hardly observed a significant change in DR risk with the elevation or reduction of serum TSH or thyroid hormone within the reference interval. Although the slightly elevated FT3 may be associated to mild NPDR, the extensibility of this result remains to be seen. For T2D patients with euthyroid function, there may be a significant correlation between serum TgAb positivity and severe NPDR and PDR.

Association of Thyroid Hormone Levels with Microvascular Complications in Euthyroid Type 2 Diabetes Mellitus Patients

PURPOSE

To examine the prevalence of different microvascular complications and investigate the association between thyroid hormones (THs) and these complications in euthyroid patients with type 2 diabetes mellitus (T2DM).

METHODS

A total of 248 T2DM patients were analyzed retrospectively for the study. All patients received a detailed and standard assessment to identify diabetic peripheral neuropathy (DPN), diabetic nephropathy (DN), and diabetic retinopathy (DR). Multivariate logistic regression was carried out to analyze the association between THs and diabetic microvascular complications.

RESULTS

The study found the prevalence of any microangiopathy to be 72.18% (n = 179). At the same time, the prevalence of DPN was 54.84% (n=136), while that of DN was 31.85% (n=79). Likewise, the prevalence of DR was 35.48% (n=88). The odds ratios (ORs) for free triiodothyronine (FT3) developing any microangiopathy, DPN, DN and DR were 0.200, 0.361, 0.310, and 0.588 (P<0.05), respectively. Also, the ORs for free thyroxine (FT4) developing any microangiopathy, DPN, DN and DR were 0.643, 0.800, 0.702 and 0.726 (P<0.05), respectively. Lastly, the ORs for thyroid-stimulating hormone (TSH) developing DPN was 1.57 (95% CI: 1.148-2.137, P=0.005).

CONCLUSION

The study concludes that serum FT3 and FT4 levels are negatively associated with any microangiopathy, DPN, DN and DR in euthyroid patients with T2DM, independent of traditional risk factors. However, the TSH levels are positively associated with DPN. Future larger sample-size studies are needed to confirm the relationship between thyroid hormone levels and microvascular complications in euthyroid patients with T2DM.

The effect of coexisting autoimmune thyroiditis in children with Type 1 diabetes on optical coherence tomography results

OBJECTIVE

To assess the influence of thyroid hormones status and coexistence of autoimmune thyroiditis on optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) results in children with Type 1 diabetes (T1D).

RESEARCH DESIGN AND METHODS

In the prospective, observational study (n = 175) we analyzed the impact of thyroid hormones on OCT results and the differences between the matched groups of children with T1D (n = 84; age = 13.14 ± 3.6; diabetes duration = 5.99 ± 3.3 years) and the children with T1D and autoimmune thyroiditis (AT) (n = 20; age = 13.94 ± 3.6; diabetes duration = 6.7 ± 4 years). We analyzed the following parameters: fovea avascular zone (FAZ), foveal thickness (FT), parafoveal thickness (PFT), ganglion cell complex (GCC), loss volume (global-GLV, focal-FLV), capillary vessel density: superficial (whole-wsVD, foveal-fsVD, parafoveal-psVD), and deep (whole-wdVD, foveal-fdVD, parafoveal-pdVD. The differences between the groups were tested by the unpaired t-Student test, Mann-Whitney U test as appropriate, whereas p level .05 was recognized as significant.

RESULTS

We detected the significant correlations between thyroid-stimulating hormone (TSH) level and PFT (r = -0.14; p < .05), psVD (r = -0.18; p < .005). The level of free triiodothyronine (FT3) was correlated with psVD (r = -0.14; p < .05). We found significant correlation between free thyroxine (FT4) and fsVD (r = -0.17; p < .01). In the studied T1D and AT groups there were statistical differences in FT (p < .005), PFT (p < .03), GCC (p < .01), and GLV (p < .003). We did not observe any significant differences in the FAZ area between the groups.

CONCLUSIONS

In our patients the co-occurrence of T1D and AT worsens the status of retinal parameters. Further studies are necessary to observe these relations and their potential influence on the occurrence of diabetic retinopathy (DR).

Diabetic Retinopathy Environment-Wide Association Study (EWAS) in NHANES 2005-2008

Several circulating biomarkers are reported to be associated with diabetic retinopathy (DR). However, their relative contributions to DR compared to known risk factors, such as hyperglycaemia, hypertension, and hyperlipidaemia, remain unclear. In this data driven study, we used novel models to evaluate the associations of over 400 laboratory parameters with DR compared to the established risk factors. Methods: we performed an environment-wide association study (EWAS) of laboratory parameters available in National Health and Nutrition Examination Survey (NHANES) 2007–2008 in individuals with diabetes with DR as the outcome (test set). We employed independent variable (feature) selection approaches, including parallelised univariate regression modelling, Principal Component Analysis (PCA), penalised regression, and RandomForest™. These models were replicated in NHANES 2005–2006 (replication set). Our test and replication sets consisted of 1025 and 637 individuals with available DR status and laboratory data respectively. Glycohemoglobin (HbA1c) was the strongest risk factor for DR. Our PCA-based approach produced a model that incorporated 18 principal components (PCs) that had an Area under the Curve (AUC) 0.796 (95% CI 0.761–0.832), while penalised regression identified a 9-feature model with 78.51% accuracy and AUC 0.74 (95% CI 0.72–0.77). RandomForest™ identified a 31-feature model with 78.4% accuracy and AUC 0.71 (95% CI 0.65–0.77). On grouping the selected variables in our RandomForest™, hyperglycaemia alone achieved AUC 0.72 (95% CI 0.68–0.76). The AUC increased to 0.84 (95% CI 0.78–0.9) when the model also included hypertension, hypercholesterolemia, haematocrit, renal, and liver function tests.