medwireNews: Day 2 of the Diabetes UK 2017 Professional Conference saw talks on diabetic retinopathy and glucose variability, and explored the link between diabetes and mental health further with insights into how type 1 diabetes-related eating disorders can be identified and managed.
Diabetic retinopathy: screening, prevention and new technologies
Graham Leese from Dundee University began the retinopathy session with a discussion on referable retinopathy rates.
“The magic figure” that we should be aiming for is a rate of 2%, he said. However, he underlined that this figure “has been drummed up with no real good evidence behind it,” with most current screening programs having a rate of 3–4%.
Risk factors – such as type and duration of diabetes, glycated hemoglobin (HbA1c) levels, and blood pressure – can predict the risk for retinopathy, and “we need to spread the net widely” to further refine the risk factors, he said.
“What age should screening for diabetic retinopathy begin?,” asked Rebecca Thomas (Swansea University) in her presentation on Should retinal screening begin at 12 years of age? She presented data showing that the incidence of retinopathy increases with advancing age and increasing diabetes duration. The recommended age to start screening for retinopathy among patients with type 1 diabetes varies, she said, with the American Diabetes Association advocating screening for those aged 10 years or older, whereas Scandinavian guidelines recommend screening from the onset of puberty. In the UK, the Royal College of Ophthalmology recommends screening in those aged 12 years and older.
Based on results from a Diabetic Eye Screening Wales study of 3391 children and adolescents with type 1 diabetes showing that the prevalence of background diabetic retinopathy increased from around 10% among children younger than 13 years old to 70% among those aged between 13 and 18 years old, Thomas concluded that UK guidelines “seem appropriate.”
“Beginning screening beyond 15 years would probably be too late,” she said.
Adnan Tufail, from Moorfields Eye Hospital in London, talked about the potential of machine learning to aid screening for diabetic retinopathy. He outlined that the current “rigorous system” for screening in England is labor- and cost-intensive, and said that it may be “problematic to continue delivering this service.”
He described a study of 20,258 patients of multiple ethnicities (41.4% White, 34.7% Asian, and 19.5% Black) undergoing screening in central London with multiple camera types. The study found that the Retmarker and EyeArt systems achieved “acceptable sensitivity” and “sufficient specificity” for referable retinopathy when compared with human graders.
These two systems “have the potential to reduce cost in developed world healthcare economies and to aid delivery of diabetic retinopathy screening in developing or remote healthcare settings,” he concluded.
To round up the retinopathy session, David Preiss (University of Oxford, UK) gave an overview of data on the relationship between lipid-lowering or antihypertensive therapy and retinopathy. Among patients with no retinopathy, blood pressure (BP) lowering reduces the development of retinopathy, he said, while in those with non-proliferative retinopathy, BP reduction leads to retinopathy regression. However, he observed that BP lowering “has not been shown to prevent proliferative retinopathy and macular edema.” There is “no convincing evidence yet that statins have any meaningful effect on diabetic retinopathy”, he added.
And he concluded that the UK “is ideally placed for trials of interventions to prevent the development/progression of diabetic retinopathy.”
Managing eating disorders in patients with diabetes
The session Dissolving the water: the rising tide of type 1 diabetes related eating disorders revisited the concept of eating disorders in patients with diabetes, exploring the associations further and offering some practical guidance for healthcare providers who suspect their patients may be experiencing both conditions.
Janet Treasure OBE, a psychiatrist from Kings College London who specializes in the treatment of eating disorders, explained that “recognizing eating disorders in the context of diabetes can be very difficult” given that some symptoms of eating disorders could be mistaken for diabetic symptoms.
In her presentation, Jacqueline Allan (Birkbeck University of London) gave some examples from eating disorder screening tools that could also be indicative of type 1 diabetes, such as “Do you feel controlled by food?” and “Can you always stop eating when you want to?”
As a patient with type 1 diabetes herself, Allan pointed out that “2am fridge binges” are sometimes necessary to overcome hypoglycemia, and she is not always able to stop eating when full due to her insulin levels after a bolus.
Treasure also explained that diabetologists and eating disorder specialists are sometimes “talking at cross purposes,” as diabetic physicians can sometimes encourage perfectionism in diabetes management in terms of regular blood testing and optimal timing of insulin and meals, whereas “we in eating disorder clinics see perfectionism and such high standards as part of the problem.”
Khalida Ismail (Kings College London) spoke about the “disappointing” results of a systematic review finding that only one randomized controlled trial has included people with type 1 diabetes and an eating disorder. The trial found no change in glycemic control, but some improvement in eating disorder behaviors, with an intervention consisting of psycho-education and family work.
“This leaves us in the position of a blank page” in terms of how to manage these patients, she said.
“I think this is really exciting. It’s an opportunity for all the different stakeholders in the care of people with type 1 diabetes and eating disorders to get together and think about new innovations and new ways of working,” she added.
“Rather than imposing a medical model, we can really think about what patients want, what their experiences have been, and what’s gone wrong in the past.”
So “what can we do whilst we’re waiting for the evidence to emerge?” Ismail asked. She suggested that medical strategies could include graded exposure to insulin, a dietetic focus on carbohydrates rather than calories, and prescribing antidepressants for impulse control. On the other side of the coin, she recommended psychological strategies such as addressing misconceptions surrounding insulin and weight change, family work, and developing non-diabetes goals could be beneficial from a psychological perspective.
In the short term, goals could include bringing blood glucose down safely without excessive weight gain, and reducing psychological distress. And in the long term, it is important that patients develop a “good enough” relationship with diabetes, concluded Ismail.
Allan ended the session on a positive note, saying that research on managing type 1 diabetes-related eating disorders has “been a long time coming,” but “the tide is turning.”
Is glucose variability important?
In the afternoon session on glucose variability, Nick Oliver from Imperial College London asked whether changes in glucose levels are important for people with diabetes. “This is a difficult question that I think remains, in part, unanswered,” he said.
Data suggest that there is something “over and above HbA1c” that has an impact on microvascular and macrovascular complications in people with diabetes, and it is “biologically plausible” that variations in glucose levels have physiological effects, he said. However, although anecdotal evidence suggests that glycemic variability is important, Oliver pointed out that there is very little longitudinal data on the impact of glucose variability, and there is no gold standard measure.
Pratik Choudhary (King’s College London) noted that type 1 diabetes is an inherently variable condition, and patients experience day-to-day variability in glucose levels that make therapeutic adjustment difficult.
“Glycemic variability and mean glucose are so tightly interlinked that designing a study that decides whether variability on its own is important is very challenging,” he said.
Torben Biester (Diabetes Centre for Children and Adolescents, Hannover, Germany) provided some practical recommendations for managing the challenge of glucose variability during exercise. “Extra carbs should be avoided,” he said, and suggested that the calculated insulin dose should be halved prior to sport.
Looking to the future, Biester suggested that closed-loop systems could help reduce glucose variability during physical activity. In the Physi-Dream study involving 20 children and adolescents, using closed-loop control gave rise to a reduction in glucose variability compared with standard insulin pump use during and after sport, he said.
Throughout his talk, Biester underlined the need to educate patients with type 2 diabetes before sport.
“We should tell patients that they should not be afraid of errors,” but should rather learn from them, he concluded.
medwireNews is an independent medical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2017