The evolution of technology in the management of type 1 diabetes

Type 1 diabetes (T1D) is a chronic autoimmune disease characterised by progressive destruction of the pancreatic beta cells causing insulin deficiency and resultant hyperglycaemia.¹ T1D accounts for approximately 5-10 per cent of all diabetes and there are between 20,000 to 30,000 people currently living in Ireland with this disease.^2,3

There is currently no cure for T1D and the mainstay of treatment is physiological insulin replacement via multiple daily injections (MDI) of insulin or continuous subcutaneous insulin infusion (CSII) therapy. The aim of care and management is to support the person with diabetes to live a long and healthy life by achieving glycaemic targets, effectively managing cardiovascular risk factors, minimising hypoglycaemia, and maintaining quality of life.

People with T1D must learn to self-adjust their insulin doses in order to achieve glycaemic targets and minimise the risk of microvascular and macrovascular complications. The landmark Diabetes Control and Complications Trial (DCCT) demonstrated that intensive insulin therapy reduced the risk of development and progression of diabetic retinopathy, neuropathy, and nephropathy compared with conventional treatment in people with T1D.⁴

Data from the Epidemiology of Diabetes Interventions and Complications (EDIC) follow-up observational study confirmed sustained benefits of early intensive glycaemic control for diabetes related microvascular complications and a lower incidence of cardiovascular disease.⁵ The intervention group in the DCCT maintained a mean haemoglobin (Hb)A1c of approximately 7 per cent (53mmol/mol), and this study forms the basis for current glycaemic target recommendations for most people with T1D.⁴

One of the greatest challenges in the management of T1D is balancing tight glycaemic control with the risk of hypoglycaemia, and indeed there was a two- to three-fold increase in the rate of severe hypoglycaemia in the intervention group in DCCT.⁴ Reported rates of severe hypoglycaemia in unselected cohorts of people with T1D vary from 0.3-3 events per year, whilst non-severe hypoglycaemic events tend to occur twice per week.⁶

Structured education courses (eg, DAFNE) have been shown to reduce the incidence of severe hypoglycaemia and newer diabetes technologies can also help with many aspects of daily diabetes management including hypoglycaemia prevention.⁷

CGM: The evidence

Continuous glucose monitoring (CGM) devices have proven to be effective at improving glycaemic control whilst also reducing hypoglycaemia.^8,9 These devices consist of three components. A sensor containing a glucose oxidase electrode is inserted under the skin and reacts with glucose in the interstitial fluid to generate hydrogen peroxide and an electron signal.¹⁰

A transmitting device sends the electric signal to a receiving device (eg, an insulin pump, a mobile phone, a handheld receiver) which displays the glucose reading, the direction the glucose is travelling, and retrospective glucose data over the preceding number of hours.

These devices have multiple benefits, most obviously, the instant availability of 24 hours of sensor glucose data without any additional fingerstick glucose measurements required of the person with diabetes. In addition, trend arrows are useful for the person with diabetes when deciding on how much quick acting insulin to deliver. Alerts on sensors can warn the person with diabetes about impending hypo-or hyperglycaemia, allowing the person time to take action and prevent the event.

In randomised controlled trials, CGM has been associated with a HbA1c reduction of approximately 0.5 per cent compared with self-monitoring of blood glucose.^8,11 In 2021, the American Diabetes Association and European Association for the Study of Diabetes published a joint consensus statement recommending CGM for all people with T1D.¹² This was mirrored in the updated National Institute for Health and Care Excellence (NICE) guidelines published in 2022, and in the recently published guidelines for the management of T1D in Ireland.^3,13

As CGM is now the standard of care in T1D, there are numerous agreed-upon CGM metrics which provide guidance for clinicians and individuals with diabetes in using, interpreting, and reporting CGM data in clinical practice. The most commonly used metric to guide diabetes management is time in target range (sensor glucose between 3.9-10mmol/l), and the international consensus recommendation is that most people with type 1 and type 2 diabetes should aim to spend >70 per cent of their time with their sensor glucose values in the target range.¹⁴

Pump therapy

The first use of CSII with an insulin pump was described in 1978.¹⁵ However, rapid uptake of pump therapy did not occur until publication of the DCCT, as individuals on CSII had better glycaemic outcomes compared to those treated with MDI.⁴ Insulin pumps are small computerised devices that continuously deliver rapid acting insulin to the person with diabetes via a subcutaneous cannula.

Historically in Ireland, uptake of pump therapy has been far lower than that of our European counterparts. However, recent data indicates that the number of pump users in Ireland is increasing, and approximately 15 per cent of people with T1D in Ireland are using insulin pump therapy.¹⁶

The main advantages of traditional pump therapy were the ability to vary basal insulin delivery and the ease of administering frequent boluses throughout the day. Although the evidence-base for pump therapy suggests their use is associated with reduced hypoglycaemia and improved quality of life, pump use only modestly improves glycaemic control.^17-19

Over the last 10 years or so, insulin pump technology has significantly progressed and most pump users are now using hybrid closed loop (HCL) therapy. HCL therapy (also called automated insulin delivery systems or the artificial pancreas) is the most advanced form of insulin delivery available. It involves the augmentation of pump therapy with integrated real-time CGM and a control algorithm which automatically adjusts basal insulin delivery every few minutes based on sensor glucose values.

Although there are differences between the various algorithms contained within the systems, typically, if sensor glucose is dropping or predicted to drop below the glucose target, the system will reduce basal insulin delivery. If sensor glucose is rising or predicted to rise above the glucose target, the system will increase insulin delivery.

Most systems will deliver auto-corrections to manage hyperglycaemia and will halt basal insulin delivery if sensor glucose is predicted to enter the hypoglycaemic range. HCL systems do not fully automate diabetes management and users must continue to deliver boluses for their meals. All of the systems have an over-ride feature which will make the algorithm less aggressive and this is commonly enabled by the person with diabetes to manage insulin delivery around exercise/activity.  There is now ample randomised control trial data, as well as real world evidence, demonstrating the benefits of HCL therapy for people with T1D.

The ADAPT study was a multi-centre randomised controlled trial, which compared advanced HCL therapy with the Medtronic 780G system with intermittently scanned CGM and MDI in 82 adults with T1D who had HbA1c levels above 8.0 per cent.²⁰

In this study, the mean HbA1c of the participants in the intervention group decreased from 9.0 per cent at baseline to 7.3 per cent after three months of HCL.²⁰ This improvement in HbA1c was sustained at six months. Participants in the control group had minimal improvement in their mean HbA1c levels which reduced from 9.1 per cent at baseline to 8.9 per cent at six months.²⁰

Studies comparing HCL therapy with sensor augmented pump therapy in young children and older adults with T1D have also yielded positive results, with HCL use being associated with higher time in range compared to sensor augmented pump therapy.^21,22

In 2021, the National Health Service in England announced funding for a real world pilot of HCL systems for people with T1D who had suboptimal HbA1c levels (>8.5 per cent) despite the use of CGM and traditional pump therapy. Hundreds of individuals from adult diabetes centres in England were invited to participate in this pilot and offered any of the commercially available closed loop systems at the time.

Glycaemic and psychosocial outcomes from 520 people with diabetes who were included in the pilot were assessed after six months of HCL therapy.²³ The use of HCL was associated with significant improvement in glycaemic metrics, with a mean reduction in HbA1c of 1.7 per cent and a mean increase in time in range of 27.6 per cent.²³ Use of HCL was also associated with a reduction in diabetes distress and an improvement in awareness of hypoglycaemia.²³

In the past, people with very high HbA1c levels, on MDI, would never have been considered suitable candidates for pump therapy due to concerns around adherence with glucose monitoring, performing set changes, engaging with the clinical team, etc.

However, it is likely that this cohort will derive the most benefit from HCL systems. Indeed, a recent observational study investigating the impact of initiating closed loop therapy in high risk individuals with diabetes who had a mean HbA1c of 10.5 per cent at baseline, demonstrated substantial and sustained improvements in both glycaemic control and quality of life after 12 months of HCL therapy.²⁴

In 2023, NICE criteria for HCL eligibility was updated and now all children with T1D under 18 years of age are eligible for HCL therapy. Adults with T1D are eligible for HCL therapy if, despite the use of either CGM or open-loop pump therapy, they have a HbA1c above 7.5 per cent, are experiencing disabling hypoglycaemia, or trying for pregnancy.²⁵

In Ireland, there are no specific eligibility criteria for HCL in T1D. In theory, anyone with T1D could access HCL, however, significant geographical variation exists in terms of the number of adult diabetes services providing pump therapy.

Available technology 

There are currently three commercially available HCL systems in Ireland; the Medtronic 780G system with Smartguard algorithm, the Tandem T:slim X2 pump with Control-IQ algorithm, and the Ypsomed pump with mylife loop algorithm. Unfortunately, Omnipod 5, which is currently the only tubeless automated insulin delivery system, is not available in Ireland at present. There are several factors that need to be considered by the person with diabetes when choosing the right HCL for themselves.

There are differences in the licensing for the systems which may need to be taken into account for young people with diabetes or those planning pregnancy. Some people may have a preference for a particular pump or CGM device which may dictate their choice of HCL system. There are also subtle differences in the algorithms which may have an impact on which system the person with diabetes deems most suitable for their needs. Fortunately, glycaemic outcomes from each of the three systems are very similar.

Data from thousands of users of these systems have demonstrated that the mean time in range associated with these systems is 71.5-72.6 per cent, with minimal hypoglycaemia.^26,27,28 Glycaemic outcomes with HCL can be further improved by optimising algorithm settings, pre-meal bolusing, accurate carbohydrate counting, and appropriate management of hypoglycaemia, set failures, and activity.

With HCL therapy, as with MDI regimes, optimal bolusing behaviours is associated with higher time in range, ie, bolusing for all meals in advance of the meals. An ancillary study from a randomised control trial evaluating HCL in children with T1D demonstrated time in range of 80 per cent when timing of boluses were optimum.²⁹

However, time in range was still 59 per cent when the children had two or more suboptimal boluses per day. Furthermore, when children missed two or more boluses per day, they were still able to achieve a time in range of 62 per cent, indicating that the closed loop system in this study compensated remarkably well for missed meal boluses.²⁹

Historically, in many diabetes services, completion of a structured education course (eg, DAFNE) was a pre-requisite for commencement on insulin pump therapy. However, HCL therapy can somewhat alleviate hyperglycaemia resulting from an underestimation of carbohydrate intake by increasing basal insulin delivery and administering autocorrections. T

his was demonstrated in the FLEX study, which evaluated adolescents using the Medtronic 780G system, and compared accurate carbohydrate counting with simplified meal announcements, ie, inputting fixed carbohydrate amounts into the bolus advisor for each meal.³⁰ After 12 months of follow-up, there was no significant difference in HbA1c levels between the two groups, although time in range was higher in those who were accurately carbohydrate counting.³⁰

The importance of completing set changes every two to three days cannot be reiterated strongly enough to people using HCL systems. Any reduction in insulin absorption because of a problem at the insertion site or an issue with the infusion set can affect the algorithm as the system operates based on the assumption that the person is receiving the insulin being delivered by the pump.

The absence of long-acting basal insulin in pump users contributes to the rapid development of diabetic ketoacidosis in the event of an infusion set failure. With HCL therapy, it is very unusual to have glucose readings above 15mmol/l for over two hours. If this occurs, the person with diabetes should give the pump one opportunity to correct the hyperglycaemia, and if this does not result in an improvement in glucose levels, infusion site failure should be considered and a complete set change performed.

Conclusion

Over the last decade considerable progress has been made in diabetes care with advancements in automated insulin delivery systems. HCL technology may be useful in a wide array of patient populations including older adults, those with cystic fibrosis-related diabetes, or chronic kidney disease, and could also address unmet needs in inpatient diabetes care.

The technology continues to evolve and more advanced systems with more aggressive algorithms and no requirement for food boluses are in development. The future looks bright for T1D care, but increasing access and ensuring equitable access to these technologies is likely to challenge diabetes services going forward.