Library staff can assist you with:
Contact or visit your local CLIN Library to find out more about our full range of services and for assistance with your research project.
Have you tried our new CLIN Libraries website - our improved one-stop shop to search across all the Libraries’ and CIAP’s online resources?
Clin Exp Ophthalmol. 2025 Jun 9. doi: 10.1111/ceo.14560. Online ahead of print.
ABSTRACT
BACKGROUND: Artificial intelligence (AI) enhanced retinal screening could reduce the impact of diabetic retinopathy (DR), the leading cause of preventable blindness in Australia. This study assessed the performance and validity of a dual-modality, deep learning system for detection of vision-threatening diabetic retinopathy (vtDR) in a multi-ethnic community.
METHODS: Cross-sectional (algorithm-validation) study with the deep learning system assessing fundus photographs for gradability and severity of DR, and optical coherence tomography (OCT) scans for diabetic macular oedema (DMO). Internal validation of each algorithm was performed using a computer-randomised 80:20 split. External validation was by comparison to standard grading provided by two ophthalmologists in 748 prospectively recruited persons with diabetes (age ≥ 10) from hospital diabetes clinics and a general practice. Main outcome measures included sensitivity, specificity and the area under the receiver operating characteristic curve (AUC).
RESULTS: Internal validation revealed robust test characteristics. When compared to ophthalmologists, the system achieved an AUC of 0.92 (95% CI 0.90-0.94) for fundus photograph gradeability, 0.91 (95% CI 0.85-0.94) for the diagnosis of severe non-proliferative DR/proliferative DR and 0.90 (95% CI 0.87-0.96) for DMO detection from OCT scans. It demonstrated real-world applicability with an AUC of 0.94 (95% CI 0.91-0.97), sensitivity of 92.7% and specificity of 95.5% for detection of vtDR. Ungradable images occurred in 55 participants (7.4%).
CONCLUSIONS: The dual-modality, deep learning system can diagnose vtDR from fundus photographs and OCT scans with high levels of accuracy and specificity. This could support a novel model of care for DR screening in our community.
PMID:40491217 | DOI:10.1111/ceo.14560
Lancet Diabetes Endocrinol. 2025 May 26:S2213-8587(25)00093-2. doi: 10.1016/S2213-8587(25)00093-2. Online ahead of print.
NO ABSTRACT
PMID:40441172 | DOI:10.1016/S2213-8587(25)00093-2
Diabetes Res Clin Pract. 2025 Jul;225:112271. doi: 10.1016/j.diabres.2025.112271. Epub 2025 May 23.
NO ABSTRACT
PMID:40414307 | DOI:10.1016/j.diabres.2025.112271
Intern Med J. 2025 Apr 25. doi: 10.1111/imj.70078. Online ahead of print.
ABSTRACT
BACKGROUND: Aboriginal people have a high risk of type 2 diabetes (T2DM). Routine opportunistic testing using glycated hemoglobin (HbA1c) for diabetes in the emergency department (ED) offers an opportunity to detect undiagnosed diabetes and evaluate glycaemia for pre-existing diabetes.
AIMS: To evaluate the prevalence of pre-existing diabetes and assess random blood glucose (RBG) and HbA1c as screening tools for undiagnosed diabetes in Aboriginal people attending the ED.
METHODS: Demographic and RBG data were extracted for all Aboriginal adults presenting to a Sydney hospital ED over 6 months. Practitioners requested blood tests as per routine care, and in the final 3 months, HbA1c was automatically added to routine venous sampling. The primary outcome was change in diabetes diagnosis with the addition of HbA1c.
RESULTS: Overall, 1640 adult Aboriginal patients presented to the ED over 6 months (4.1% of all presentations), including 734 unique individuals tested during routine care. The prevalence of pre-existing T2DM was 12.0% (n = 88). Among those without known diabetes, 1.4% (n = 9) had glucose readings ≥11.1 mmol/L and 14.3% (n = 90) had glucose readings 7.0-11.0 mmol/L. For those without known diabetes with HbA1c measurement, there were 2.7% (n = 8) with HbA1c ≥6.5% and 4% (n = 12) with HbA1c 6.0%-6.4%. There was no overlap between those who had an HbA1c ≥6.5% and RBG ≥11.1 mmol/L.
CONCLUSIONS: There was a high prevalence of pre-existing diabetes among Aboriginal adults attending the ED. New diabetes diagnosis in the ED based on RBG or HbA1c was not common. Confirmatory testing for diabetes should be recommended to the general practitioners of patients in whom elevated HbA1c or RBG are identified.
PMID:40276855 | DOI:10.1111/imj.70078
Contemp Clin Trials. 2025 Jun;153:107915. doi: 10.1016/j.cct.2025.107915. Epub 2025 Apr 15.
ABSTRACT
BACKGROUND: 'Food is medicine' programs such as Produce Prescription (PRx) aim to integrate food-based nutrition programs into healthcare, for the prevention, management and treatment of diet-related diseases, typically for those experiencing food insecurity. However, the impact of PRx on health indicators in Australia has never been tested in a randomised trial.
OBJECTIVES: To determine the effect of PRx on blood glucose control and other health indicators in adults with type 2 diabetes experiencing hyperglycaemia and food insecurity and/or financial hardship in Australia.
METHODS: Using a parallel design randomised controlled trial, n = 224 participants will be randomised (1:1) to PRx or usual care. Over 26 weeks, the intervention group will receive a weekly delivery of fruits, vegetables, wholegrains and nuts, and up to 3 sessions with an accredited dietitian. Controls will receive usual care. The primary outcome is change in mean HbA1c over 26 weeks, comparing the intervention and control group. Secondary outcomes include between-group differences at 26 weeks in change in blood pressure, body weight, blood lipids, food and nutrition insecurity, person-reported outcome measures, medication use, and diet quality. Implementation outcomes assessed will include feasibility, acceptability, scalability and cost effectiveness.
DISCUSSION: This Australia-first PRx trial will provide novel and rigorous data for an intervention that may be feasible to improve health and health equity as part of the Australian healthcare system. We anticipate PRx will lead to a clinically meaningful reduction in HbA1c, contribute to improved health equity and long-term health benefits for adults with type 2 diabetes and food insecurity.
PMID:40239800 | DOI:10.1016/j.cct.2025.107915
This study sought to determine whether the association between attaining specific glycated hemoglobin (HbA1c) targets (<7.0% (<53 mmol/mol) and ≥8.0% (≥64 mmol/mol) compared with 7.0%–7.9%) over time and risk of incident coronary artery disease (CAD) was dependent on haptoglobin (Hp) phenotype in the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) study.
Prospectively collected HbA1c data from the ADVANCE biomarker case-cohort study, updated at 6 months and every 12 months thereafter over a median of 5.0 (IQR 4.5–5.3) years, were analyzed in relation to incident CAD in the Hp2-2 (n=1323) and non-Hp2-2 (n=2069) phenotypes separately using weighted multivariable-adjusted Cox regression models. Additional a priori stratifications by sex, race, previous cardiovascular disease (CVD), and type 2 diabetes duration were performed.
Mean HbA1c was similar in each phenotype group throughout the study. Compared with HbA1c of 7.0%–7.9%, HbA1c <7.0% was not associated with CAD risk for any phenotype group or subgroup. HbA1c ≥8.0% compared with 7.0%–7.9% over time was associated with higher CAD risk for the Hp2-2 phenotype only (HR 1.53, 95% CI 1.01 to 2.32; no significant association in the non-Hp2-2 type: 1.26, 0.89 to 1.77, p-interaction=0.71); this was pronounced when those with previous CVD at baseline were excluded (Hp2-2: 2.80, 1.41 to 5.53, p-interaction=0.03). Compared with HbA1c of <8.0%, having HbA1c ≥8.0% was associated with a 59% higher CAD risk among participants with the Hp2-2 phenotype (1.59, 1.12 to 2.26) and a 39% higher CAD risk among participants without the Hp2-2 phenotype (1.39, 1.03 to 1.88, p-interaction=0.97).
The present ADVANCE analysis suggests that not having HbA1c ≥8.0%, rather than achieving HbA1c <7.0%, was found to be particularly important for CAD prevention among people with type 2 diabetes and the common Hp2-2 phenotype. While the subgroup analyses were likely underpowered, their inclusion is hypothesis generating and can be used in future meta-analyses to improve power and generalizability.
The effects of different glucose metabolic states and diabetes-controlled status on asymptomatic carotid atherosclerosis has not been well investigated. Herein, we aimed to investigate the association of different diabetes status with asymptomatic carotid plaques and carotid intima–media thickness (CIMT).
4752 participants aged over 40 years, free of stroke or myocardial infarction, from the China National Stroke Screen Survey programme were enrolled. Carotid plaque and CIMT were assessed using duplex ultrasonography. Logistic regression analysis was used to assess the relationship between diabetes status and the presence of asymptomatic carotid plaques or abnormal CIMT.
A total of 1977 (41.6%) subjects had carotid plaques and 804 (16.9%) had abnormal CIMT. In multivariate analyses, compared with normoglycemia, individuals with pre-diabetes showed significantly higher odds of asymptomatic carotid plaques (OR 1.22, 95% CI 1.03 to 1.45) and patients with diabetes also had higher odds (OR 1.66, 95% CI 1.41 to 1.92). Diabetes (OR 1.79, 95% CI 1.50 to 2.14) was associated with vulnerable plaques, while pre-diabetes was not (OR 1.17, 95% CI 0.96 to 1.43). Poorly controlled diabetes (HbA1c ≥7.5%) had higher odds of carotid plaques (OR 1.93, 95% CI 1.53 to 2.44), especially of vulnerable plaques (OR 2.03, 95% CI 1.55 to 2.67). No significant association was found between diabetes and abnormal CIMT.
Pre-diabetes and diabetes, especially poorly controlled diabetes, were associated with increased odds of asymptomatic carotid plaques. Implementing the most effective strategies to achieve optimal glycemic control is crucial for the prevention and management of atherosclerosis.
Pre-diabetes comprises diverse subphenotypes linked to varying complications, type 2 diabetes, and mortality outcomes. This study aimed to explore these outcomes across different pre-diabetes subphenotypes.
The dataset included adults without type 2 diabetes with baseline HbA1c and fasting plasma glucose (FPG) measurements from Siriraj Hospital, Bangkok, Thailand. The participants were classified into six subphenotypes via the k-means clustering method on the basis of age, body mass index, FPG, HbA1c, high-density lipoprotein cholesterol and alanine aminotransferase levels. The incidences of type 2 diabetes, long-term vascular complications and mortality were compared among subphenotypes over a median follow-up of 8.8 years, employing Kaplan-Meier curves and Cox regression analysis adjusted for sex, statin use and hypertension status.
Among the 4915 participants (mean age 60.1±10.1 years; 54.6% female), six clusters emerged: cluster 1, low risk (n=650; 13.2%); cluster 2, mild dysglycemia elderly (n=791; 16.1%); cluster 3, severe dysglycemia obese (n=1127; 22.9%); cluster 4, mild dysglycemia obese (n=963; 19.7%); cluster 5, severe dysmetabolic obese (n=337; 6.9%); and cluster 6, severe dysglycemia elderly (n=1042; 21.2%). Clusters were classified into diabetes risk subgroups: low risk (clusters 1 and 4) and high risk (clusters 3 and 5). Cluster 6 exhibited the highest risk, with significantly increased incidences of macrovascular complications (adjusted HR 2.22, 1.51–3.27) and type 2 diabetes (1.73, 1.42–2.12). In contrast, cluster 4 demonstrated the lowest risk, with significantly decreased incidences of new chronic kidney disease (0.65, 0.44–0.96), microvascular complications (0.62, 0.43–0.89) and mortality (0.25, 0.10–0.63).
Our pre-diabetes phenotyping approach effectively provides valuable insights into the risk of type 2 diabetes, vascular complications and mortality in individuals with pre-diabetes. Those with high-risk phenotypes should be prioritized for type 2 diabetes and cardiovascular interventions to mitigate risks.