Feasibility and accuracy of the screening for diabetic retinopathy using a fundus camera and an artificial intelligence pre-evaluation application.

AIMS: Periodical screening for diabetic retinopathy (DR) by an ophthalmologist is expensive and demanding. Automated DR image evaluation with Artificial Intelligence tools may represent a clinical and cost-effective alternative for the detection of retinopathy. We aimed to evaluate the accuracy and reliability of a machine learning algorithm. METHODS: This was an observational diagnostic precision study that compared human grader classification with that of DAIRET®, an algorithm nested in an electronic medical record powered by Retmarker SA. Retinal images were taken from 637 consecutive patients attending a routine annual diabetic visit between June 2021 and February 2023. They were manually graded by an ophthalmologist following the International Clinical Diabetic Retinopathy Severity Scale and the results were compared with those of the AI responses. The main outcome measures were screening performance, such as sensitivity and specificity and diagnostic accuracy by 95% confidence intervals. RESULTS: The rate of cases classified as ungradable was 1.2%, a figure consistent with the literature. DAIRET® sensitivity in the detection of cases of referable DR (moderate and above, "sight-threatening" forms of retinopathy) was equal to 1 (100%). The specificity, that is the true negative rate of absence of DR, was 80 ± 0.04. CONCLUSIONS: DAIRET® achieved excellent sensitivity for referable retinopathy compared with that of human graders. This is undoubtedly the key finding of the study and translates into the certainty that no patient in need of the ophthalmologist is misdiagnosed as negative. It also had sufficient specificity to represent a cost-effective alternative to manual grade alone.

A prediction model to assess the risk of egfr loss in patients with type 2 diabetes and preserved kidney function: The amd annals initiative.

OBJECTIVE: To develop and validate a model for predicting 5-year eGFR-loss in type 2 diabetes mellitus (T2DM) patients with preserved renal function at baseline. RESEARCH DESIGN AND METHODS: A cohort of 504.532 T2DM outpatients participating to the Medical Associations of Diabetologists (AMD) Annals Initiative was splitted into the Learning and Validation cohorts, in which the predictive model was respectively developed and validated. A multivariate Cox proportional hazard regression model including all baseline characteristics was performed to identify predictors of eGFR-loss. A weight derived from regression coefficients was assigned to each variable and the overall sum of weights determined the 0 to 8-risk score. RESULTS: A set of demographic, clinical and laboratory parameters entered the final model. The eGFR-loss score showed a good performance in the Validation cohort. Increasing score values progressively identified a higher risk of GFR loss: a score ≥ 8 was associated with a HR of 13.48 (12.96-14.01) in the Learning and a HR of 13.45 (12.93-13.99) in the Validation cohort. The 5 years-probability of developing the study outcome was 55.9% higher in subjects with a score ≥ 8. CONCLUSIONS: In the large AMD Annals Initiative cohort, we developed and validated an eGFR-loss prediction model to identify T2DM patients at risk of developing clinically meaningful renal complications within a 5-years time frame.

Healthcare resource use, direct and indirect costs of hypoglycemia in type 1 and type 2 diabetes, and nationwide projections. Results of the HYPOS-1 study.

BACKGROUND AND AIMS: To obtain an accurate picture of the total costs of hypoglycemia, including the indirect costs and comparing the differences between type 1 (T1DM) and type 2 diabetes mellitus (T2DM). METHODS AND RESULTS: HYPOS-1 was a multicenter, retrospective cohort study which analyzed the data of 2229 consecutive patients seen at 18 diabetes clinics. Data on healthcare resource use and indirect costs by diabetes type were collected via a questionnaire. The domains of inpatient admission and hospital stay, work days lost, and third-party assistance were also explored. Resource utilization was reported as estimated incidence rates (IRs) of hypoglycemic episodes per 100 person-years and estimated costs as IRs per person-years. For every 100 patients with T1DM, 9 emergency room (ER) visits and 6 emergency medical service calls for hypoglycemia were required per year; for every 100 patients with T2DM, 3 ER visits and 1 inpatient admission were required, with over 3 nights spent in hospital. Hypoglycemia led to 58 work days per 100 person-years lost by the patient or a family member in T1DM versus 19 in T2DM. The costs in T1DM totaled €90.99 per person-year and €62.04 in T2DM. Direct and indirect costs making up the total differed by type of diabetes (60% indirect costs in T1DM versus 43% in T2DM). The total cost associated with hypoglycemia in Italy is estimated to be €107 million per year. CONCLUSIONS: Indirect costs meaningfully contribute to the total costs associated with hypoglycemia. As compared with T1DM, T2DM requires fewer ER visits and incurs lower indirect costs but more frequent hospital use.

Sex differences in food choices, adherence to dietary recommendations and plasma lipid profile in type 2 diabetes - The TOSCA.IT study.

BACKGROUND AND AIMS: Diabetic women have a more adverse plasma lipid profile than men. Sex differences in dietary habits may play a role, but are little investigated. The study evaluates the quality of diet, adherence to the nutritional recommendations of the Diabetes and Nutrition Study Group and their relation with plasma lipid in men and women with diabetes. METHODS AND RESULTS: We studied 2573 people, aged 50-75, enrolled in the TOSCA.IT study (clinicaltrials.gov; NCT00700856). Plasma lipids were measured centrally. Diet was assessed with a semi-quantitative food frequency questionnaire. Women had a more adverse plasma lipid profile than men. Women consumed significantly more legumes, vegetables, fruits, eggs, milk, vegetable oils, and added sugar, whereas men consumed more starchy foods, soft drinks and alcoholic beverages. This stands for a higher proportion (%) of energy intake from saturated fat and added sugar (12.0 ± 2.4 vs 11.5 ± 2.5 and 3.4 ± 3.2 vs 2.3 ± 3.2, P < 0.04), and a higher intake of fiber (11.2 ± 2.8 vs 10.4 ± 2.6 g/1000 Kcal/day) in women. Adherence to the recommendations for saturated fat and fiber consumption was associated with significantly lower LDL-cholesterol regardless of sex. Adherence to the recommendations for added sugars was associated with significantly lower triglycerides and higher HDL-cholesterol in men and women. CONCLUSIONS: Men and women with diabetes show significant differences in adherence to nutritional recommendations, but sex differences in plasma lipid profile are unlikely to be explained by nutritional factors. Adherence to the nutritional recommendations is associated with a better plasma lipid profile regardless of sex, thus reinforcing the importance of substituting saturated for unsaturated fat sources, increasing fiber and reducing added sugar intake.

A systematic review of acute pancreatitis as an adverse event of type 2 diabetes drugs: from hard facts to a balanced position.

The question whether antidiabetes drugs can cause acute pancreatitis dates back to the 1970s. Recently, old concerns have re-emerged following claims that use of incretins, a new class of drugs for type 2 diabetes, might increase the relative risk of acute pancreatitis up to 30-fold. Given that diabetes is per se a potent risk factor for acute pancreatitis and that drug-related acute pancreatitis is rare and difficult to diagnose, we searched the medical databases for information linking acute pancreatitis and type 2 diabetes drugs. Among the biguanides, both phenformin and metformin (the latter in patients with renal insufficiency) have been cited in case reports as a potential cause of acute pancreatitis. Sulphonylureas, as both entire class and single compound (glibenclamide), have also been found in cohort studies to increase its risk. No direct link was found between pancreatic damage and therapy with metaglinide, acarbose, pramlintide or SGLT-2 inhibitors. In animal models, thiazolinediones have demonstrated proprieties to attenuate pancreatic damage, opening perspectives for their use in treating acute pancreatitis in humans. Several case reports and the US Food and Drug Administration pharmacovigilance database indicate an association between acute pancreatitis and incretins, dipeptidyl peptidase-4 (DPP-4) inhibitors, and GLP-1 receptor agonists. To date, however, a clear-cut odds ratio for this association has been reported in only one of eight pharmacoepidemiological studies. Finally, none of the intervention trials investigating these compounds, including two large randomized controlled trials with cardiovascular endpoints, confirmed the purportedly increased risk of acute pancreatitis with incretin use.

Comparison of direct costs of type 2 diabetes care: different care models with different outcomes.

BACKGROUNDS AND AIMS: To compare direct costs of four different care models and health outcomes in adults with type 2 diabetes. METHODS AND RESULTS: We used multiple independent data sources to identify 25,570 adults with type 2 diabetes residing in Turin, Italy, as of 1 July 2003. Data extracted from administrative data databases were used to create four care models ranging in organization from highly structured care (integrated primary and specialist care) to progressively less structured care (unstructured care). Regression analyses, adjusted for main confounders, were applied to examine the differences between the models in direct costs, mortality, and diabetes-related hospitalizations rates over a 4-year period. In patients managed according to the unstructured care model (i.e., usual care by a primary care provider and without strict guidelines adherence), excess of all-cause mortality was 84% and 4-year direct cost was 8% higher than in those managed according to the highly structured care model. Cost ratio analysis revealed that the major cost driver in the unstructured care model was hospital admissions, which were 31% higher than the rate calculated for the more structured care models. In contrast, spending on prescription medications and specialist consultations was higher in the highly structured care model. CONCLUSION: A diabetes care model that integrates primary and specialty care, together with practices that adhere to guideline recommendations, was associated with a reduction in all-cause mortality and hospitalizations, as compared with less structured models, without increasing direct health costs.

Improving quality of care in people with Type 2 diabetes through the Associazione Medici Diabetologi-annals initiative: a long-term cost-effectiveness analysis.

AIMS: The Associazione Medici Diabetologi-annals initiative is a physician-led quality-of-care improvement scheme that has been shown to improve HbA1c concentration, blood pressure, lipid profiles and BMI in enrolled people with Type 2 diabetes. The present analysis investigated the long-term cost-effectiveness of enrolling people with Type 2 diabetes in the Associazione Medici Diabetologi-annals initiative compared with conventional management. METHODS: Long-term projections of clinical outcomes and direct costs (in 2010 Euros) were made using a published and validated model of Type 2 diabetes in people with Type 2 diabetes who were either enrolled in the Associazione Medici Diabetologi-annals initiative or who were receiving conventional management. Treatment effects were based on mean changes from baseline seen at 5 years after enrolment in the scheme. Costs and clinical outcomes were discounted at 3% per annum. RESULTS: The Associazione Medici Diabetologi-annals initiative was associated with improvements in mean discounted life expectancy and quality-adjusted life expectancy of 0.55 years (95% CI 0.54-0.57) years and 0.48 quality-adjusted life years (95% CI 0.46-0.49), respectively, compared with conventional management. Whilst treatment costs were higher in the Associazione Medici Diabetologi-annals arm, this was offset by savings as a result of the reduced incidence and treatment of diabetes-related complications. The Associazione Medici Diabetologi-annals initiative was found to be cost-saving over patient lifetimes compared with conventional management [€ 37,289 (95% CI 37,205-37,372) vs € 41,075 (95% CI 40,956-41,155)]. CONCLUSIONS: Long-term projections indicate that the physician-led Associazione Medici Diabetologi-annals initiative represents a cost-saving method of improving long-term clinical outcomes compared with conventional management of people with Type 2 diabetes in Italy.

The SUBITO-DE study: sexual dysfunction in newly diagnosed type 2 diabetes male patients.

INTRODUCTION: No data on the prevalence of erectile dysfunction (ED) in subjects with newly diagnosed Type 2 diabetes mellitus (T2DM) are currently available. AIM: The aim of the present study was to estimate the prevalence of ED and its associated causes in a sample of male patients with recently diagnosed DM (<24 months) attending a diabetes care center. METHODS: The study comprised two phases: a cross-sectional analysis and a longitudinal reassessment of the data collected during the first phase. During the first phase, 1503 subjects (mean age 58.7±8.9 yr) from 27 centers were interviewed: 666 (43.3%) reported experiencing ED, 499 of which (mean age 58.8±8.8 yr) agreed to participate in the study (final enrolment rate, 33.3%). Concurrent morbidities were hypertension (55.3%), dyslipidemia (39.5%), and coronary heart disease (7.8%); chronic complications were neuropathy (8.9%), nephropathy (12.6%) and retinopathy (7.6%) in about one third of the sample at enrolment. RESULTS: Overall, about 20% of the patients reported having used ED drugs, but more than 50% had abandoned therapy because of the drug's ineffectiveness or high cost. The prevalence of hypogonadism was 46.9% (total testosterone level, 3.5 ng/ml). Some 20% of patients reported symptoms suggestive of depression. CONCLUSION: The present study provides data showing a high prevalence of ED, hypogonadism and depressive symptoms among male patients with newly diagnosed T2DM. Further analysis of the data will elucidate the specific determinants of such conditions and their longitudinal significance.

The role of the care model in modifying prognosis in diabetes.

Organizational factors in diabetes care can influence long- and medium-term outcomes, affecting the prognosis to the same extent as new therapies. A growing body of evidence supports the hypothesis that diabetes team consultation can favorably impact on hospital utilization, the costliest item in diabetes management, as well as on hospitalization rates, inpatient hospital length of stay, and re-admission rates. Moreover, the model of diabetes care has been reported to influence guidelines adherence, an additional factor linked to the variability in the quality of diabetes care. The strongest predictor and effect modifier of the quality of diabetes care is specialist referral. Compared to patients seen in primary care or other settings, those visiting a diabetes center are more likely to be monitored according to guidelines, regardless of the severity-of-disease effect, and to receive structured education, as well as more aggressive treatment when needed. Finally, at least eight published studies suggest that when continuity of care is shared with diabetes clinics, all-cause mortality and major cardiovascular events are both reduced. The sharing of care pathways between primary care providers and diabetes teams is likely to be the best and most affordable solution in the complex management of this chronic condition.

Factors associated with a rapid normalization of HbA1c in newly diagnosed type 2 diabetes patients seen in a specialist setting.

The time to achieve good metabolic control after diagnosis is essential for type 2 diabetes patients because it can influence long-term prognosis. This study aimed to elucidate the predictive role of several clinical and organization factors in normalizing metabolism within 6 months. A multi-centered, retrospective, observational study on 960 patients, with diabetes duration of 12 months or less, consecutively seen in 123 Italian clinics, was undertaken. Information about clinic's organization, along with data abstracted from medical records at enrollment (first visit) and after 6 months (follow-up visit), was collected. At 6 months, HbA1c dropped by -3.1 ± 2.2 points in those who achieved HbA1c <7 % (responders), whereas in non-responders (HbA1c ≥7 %), the mean reduction was -1.8 ± 1.9. The intervention markedly reduced lipids, blood pressure, BMI, and waist circumference, especially in responders. The presence of a diabetes team correlated with a likelihood of HbA1c normalization (OR 1.94, 1.17-3.22). By contrast, indicators of advanced disease such as previous retinopathy (0.53, 0.29-0.98), use of secretagogues (0.40, 0.25-0.64), high levels of HbA1c at first visit and related insulin use emerged as adverse factors. Early detection of diabetes, along with human resources and organization, was found to play a crucial role in rapidly attaining good metabolic control.

Addition of either pioglitazone or a sulfonylurea in type 2 diabetic patients inadequately controlled with metformin alone: impact on cardiovascular events. A randomized controlled trial.

BACKGROUND AND AIMS: Metformin is the first-line therapy in type 2 diabetes. In patients inadequately controlled with metformin, the addition of a sulfonylurea or pioglitazone are equally plausible options to improve glycemic control. However, these drugs have profound differences in their mechanism of action, side effects, and impact on cardiovascular risk factors. A formal comparison of these two therapies in terms of cardiovascular morbidity and mortality is lacking. The TOSCA.IT study was designed to explore the effects of adding pioglitazone or a sulfonylurea on cardiovascular events in type 2 diabetic patients inadequately controlled with metformin. METHODS: Multicentre, randomized, open label, parallel group trial of 48 month duration. Type 2 diabetic subjects, 50-75 years, BMI 20-45 Kg/m(2), on secondary failure to metformin monotherapy will be randomized to add-on a sulfonylurea or pioglitazone. The primary efficacy outcome is a composite endpoint of all-cause mortality, nonfatal myocardial infarction, nonfatal stroke, and unplanned coronary revascularization. Principal secondary outcome is a composite ischemic endpoint of sudden death, fatal and non-fatal myocardial infarction and stroke, endovascular or surgical intervention on the coronary, leg or carotid arteries, major amputations. Side effects, quality of life and economic costs will also be evaluated. Efficacy, safety, tolerability, and study conduct will be monitored by an independent Data Safety Monitoring Board. End points will be adjudicated by an independent external committee. CONCLUSIONS: TOSCA.IT is the first on-going study investigating the head-to-head comparison of adding a sulfonylurea or pioglitazone to existing metformin treatment in terms of hard cardiovascular outcomes. REGISTRATION: Clinicaltrials.gov ID NCT00700856.

Global cardiovascular risk management in different Italian regions: an analysis of the Evaluation of Final Feasible Effect of Control Training and Ultra Sensitisation (EFFECTUS) educational program.

BACKGROUND AND AIM: The Final Evaluation Feasible Effect of Ultra Control Training and Sensitization (EFFECTUS) is an educational program, aimed at improving global CV risk stratification and management in Italy. The present study evaluates differences on clinical approach to global CV risk among physicians involved in the EFFECTUS program and stratified in three geographical macro-areas (North, Center, South) of our Country. METHODS AND RESULTS: Physicians were asked to submit data already available in their medical records, covering the first 10 adult outpatients, consecutively seen in the month of May 2006. Overall, 1.078 physicians (27% females, aged 50 ± 7 years) collected data of 9.904 outpatients (46.5% females, aged 67 ± 9 years), among which 3.219 (32.5%) were residents in Northern, 3.652 (36.9%) in Central and 3.033 (30.6%) in Southern Italy. A significantly higher prevalence of major CV risk factors, including obesity, physical inactivity, hypertension and diabetes, was recorded in Southern than in other areas. Accordingly, Southern physicians more frequently prescribed antihypertensive, glucose and lipid lowering agents than other physicians, who paid significantly more attention to life-style changes in their clinical practice. CONCLUSIONS: This analysis of the EFFECTUS study demonstrates a high prevalence of CV risk factors in Italy, particularly in Southern areas, and indicates some important discrepancies in the clinical management of global CV risk among physcians working in different Italian regions.

Impact of physicians' age on the clinical management of global cardiovascular risk: analysis of the results of the Evaluation of Final Feasible Effect of Control Training and Ultra Sensitisation Educational Programme.

AIM: To evaluate the potential impact of physicians' age on global cardiovascular (CV) risk management in the population of the Evaluation of Final Feasible Effect of Ultra Control Training and Sensitisation (EFFECTUS) study. METHODS: Involved physicians were stratified into three age groups (≤ 45, 46-55 and > 55 years), and asked to provide clinical data covering the first 10 adult outpatients, consecutively seen in May 2006. RESULTS: Overall 1078 physicians, among whom 219 (20%) were aged ≤ 45, 658 (61%) between 46 and 55, and 201 (19%) > 55 years, collected data of 9904 outpatients (46.5% female patients, aged 67 ± 9 years), who were distributed into three corresponding groups: 2010 (20%), 6111 (62%) and 1783 (18%), respectively. A higher prevalence of myocardial infarction and stroke was recorded by younger physicians rather than those aged > 46 years. Older physicians frequently recommended life-style changes, whereas a higher number of antihypertensive, antiplatelet, glucose and lipid-lowering prescriptions was prescribed by physicians aged ≤ 45 years. CONCLUSIONS: This analysis of the EFFECTUS study indicates a higher prevalence of vascular diseases among outpatients who were followed by younger physicians, who prescribed a higher number of CV drugs than older physicians. These older physicians have more attitude for prescribing favourable life-style changes than younger physicians.

Persistent platelet activation in patients with type 2 diabetes treated with low doses of aspirin.

BACKGROUND: The percentage of diabetic patients who do not benefit from the protective effect of aspirin is larger than in other populations at cardiovascular risk. OBJECTIVE: We compared the ability of aspirin to suppress TxA2 and platelet activation in vivo, in type-2 diabetics vs. high-risk non-diabetic patients. METHODS: Urinary 11-dehydro-TXB2, plasma sCD40 L, and sP-selectin were measured, together with indices of low-grade inflammation, glycemic control, and lipid profile, in 82 patients with type-2 diabetes and 39 without diabetes, treated with low doses of aspirin. RESULTS: Urinary 11-dehydro-TxB2, plasma sCD40L and sP-selectin were significantly higher in diabetics than in controls: [38.9 (27.8-63.3) vs. 28.5 (22.5-43.9) ng mmol(-1) of creatinine, P = 0.02], [1.06 (0.42-3.06) vs. 0.35 (0.22-0.95) ng mL(-1); P = 0.0001], [37.0 (16.8-85.6) vs. 20.0 (11.2-35.6) ng mL(-1), P = 0.0001], respectively. The proportion of individuals with diabetes increased across quartiles of 11-dehydro-TxB2, sCD40L, and sP-selectin, with the highest quartiles of 11-dehydro-TxB2, sCD40L and sP-selectin, including 66%, 93.3%, and 93.3% of individuals with diabetes. Markers of platelet activation positively correlated with indices of glycemic control but not with markers of low-grade inflammation. CONCLUSIONS: Platelet dysfunction associated with insufficient glycemic control, may mediate persistent platelet activation under aspirin treatment.