Latest Diabetes Tech News

May 12, 2023

Diabetes is a complex condition that requires constant monitoring and care. Fortunately, technology continues to evolve to make life easier for people with diabetes. In this blog post, we share some exciting updates on the latest developments in diabetes technology that can positively impact both patients and their caregivers.

Tandem Control IQ will work with Dexcom G7 and Libre this fall (in US)
Compare reimbursement of pumps and sensors with your neighboring countries
New study: Omnipod pods do not appear to be damaged by CT scan
Average TIR of 75% in Polish children and adolescents on a closed-loop
A fully closed-loop system with AndroidAPS

Continue reading below.

1. Tandem Control IQ will work with Dexcom G7 and Libre this fall (in US)

"I think that we've carefully evaluated that Abbott sensor. And I think we feel very confident that it will perform very well with Control IQ. We plan on using the sensor data as soon as we initiate the sensor session to power the system and to actually inform Control IQ. So, we have confidence that the system is going to work well." - John Sheridan (CEO Tandem) interview with Close Concerns

Tandem's latest quarterly results (3-5-2023) mentioned that in America, both the Dexcom G7 and the Libresensor were going to work with Tandem Control IQ by the fall of 2023.

In America, integration will first be offered with Dexcom G7, then with the FreeStyle Libre2, and shortly thereafter with the FreeStyle Libre3.
After that, integration with Dexcom G7 will also be available in Europe, and integration with Libre3 will be offered immediately.

Please note that this requires a software update, which will only be possible for insulin pumps that are still under warranty.

However, we know that many people carry their pumps beyond their 4-year warranty. Thus, they will have to continue with the Dexcom G6.

The question, of course, is whether Control IQ's results will be the same with these new sensors.

In any case, Tandem has no intention of waiting for official studies or evaluations.

Based on the results they have seen, they are confident that the results will be just as good.

2. Compare reimbursement of pumps and sensors with your neighbors

"Cost-effectiveness studies in both T1D and T2D have shown that pump use is cost-effective in several countries, leading to improvements in quality-adjusted life-years. [...] There is a need for updated guidelines for the use of CSII and HCL in individuals living with T2D based on the emerging evidence, with identification of and recommendations for the people who would benefit the most, which would eventually form a basis for reimbursement and health policies." - dr. Johan Jendle Diabetes, Obesity and Metabolism 2023

Many individuals who require sensors and insulin pumps often feel disadvantaged when they are unable to obtain proper reimbursement for these vital medical devices.

In response to their concerns, a common counterargument is that they should appreciate what they have since the situation in other countries is even more challenging.

But what about reimbursement for sensors and pumps in other countries?

That remains difficult to understand.

Dr. Jendle of Sweden has made an attempt to map this out in his article "Clinical and cost-effectiveness to enable access to technology and meet the needs of payers."

Indeed, in this article you will find this figure where you can clearly see in which European countries which technology is partially or fully reimbursed:

isCGM = FreeStyle Libre
rtCGM = alle andere sensoren
CSII = insulinepomp
SAP = sensorgestuurde pomp

For us, of course, it stands out that there are a lot of errors in the supposed reimbursement in Belgium and the Netherlands:

Indeed, in both Belgium and the Netherlands,

there is only partial reimbursement for rtCGM for people with type 1 and type 2 diabetes,
and also only partial reimbursement for SAP for people with type 1 diabetes.

This shows, on the one hand, how difficult it is to get a view of reimbursement in the different European countries,

and on the other hand that there are very big differences between the different countries.

People in Germany, Italy, Austria and Ireland currently seem to have the most access to insulin pumps and sensors.

Such data may help to convince the healthcare facilities of the underlying states to offer CGM (and insulin pumps) reimbursement for anyone with diabetes (on an intensive insulin regimen).

3. New study: Omnipod pods do not appear to be damaged by CT scan

“The probability of an adverse event being caused by exposing these devices to CT irradiation is extremely low, and it is greatly outweighed by the clinical benefit of a medically indicated CT examination.” - FDA Guidance Interference between CT and Electronic Medical Devices

Currently, the instructions for sensors and insulin pumps state that these devices should not be used under an RX or CT.

This is very inconvenient because because of this:

all sensors and pumps should theoretically be removed if the patient were to go under RX or CT,
the removed sensors and pumps (or part of them) often cannot be used afterwards,
causes some people to postpone necessary tests,
and temporary discontinuation of the sensor and/or insulin pump can lead to temporarily poorer glycemic control.

So the question is whether it will hurt so much to have the sensor and/or insulin pump go under the RX or CT with us?

Several studies happened in the past that could not show any significant effect of such X-rays:

Dexcom G6: Thomas et al. JDST 2021
Free Pro: Takatsu et al. Radiol Phys technol 2019

Dr. Dong of Texas has now investigated this for Omnipod pods as well (Dong et al. JDST 2023).

160 Omnipod Eros pods filled with water were examined for the effect of different doses of X-rays, corresponding to a typical CT scan.
The radiation had no effect on the functioning of the pods.

This kind of data will hopefully lead to the possibility of lifting the ban on sensors and insulin pumps under an RX and/or CT.

In practice, sensors and insulin pumps are already sometimes left on, as also indicated in this guideline for the mylife Loop system

It is recommended to turn off the auto mode, and to check the accuracy of the sensor with a finger prick after the CT scan.
Sometimes it is also recommended to disconnect the insulin pump if possible and/or shield the device with a lead apron.
Other guidelines are more conservative and still recommend removing the sensor and pump when going under a CT.

4. Average TIR of 75% in Polish children and adolescents on a closed-loop

"This population-based study shows that the HCL technology is superior to other treatment modalities in CGM-derived parameters and should be considered as a treatment of choice in all CwD fulfilling the indication criteria." - dr. Zdenek Sumnik Horm Res Paediatr 2023

We know that closed-loop systems systematically give better glycemic results than treatment with insulin pens or a manual pump.

Until now, however, there were no large real-world studies comparing the different treatments.

We usually used results from different studies to arrive at a comparison, e.g.

closed-loop systems: mean TIR 71-75% (Bergenstal et al. JAMA 2016, Brown et al, NEJM 2019, Carlson et al, DTT 2021)
insulin pens with CGM: mean TIR 57% (Beck et al. JAMA 2017, Bergenstal et al. NEJM 2015)
insulin pump with CGM: mean TIR 51% (Beck et al. JAMA 2017)
insulin pen without CGM: mean TIR 43% (Beck et al. JAMA 2017)

In Poland, a complete national diabetes registry was now analyzed for the first time.

This allowed the glycemic results of different treatments to be compared in 1 population.

This registry includes 90% of all children and adolescents with diabetes in Poland
and 52 out of 56 diabetes centers participate in it.
Every year all parameters are forwarded anonymously.

For this study, 2021 data were analyzed from all children and adolescents <19 years of age who had type 1 diabetes for >1 year.

A total of 3251 children and adolescents were included and divided according to the therapy they were using:

67.3% used multiple daily injections (MDI) and 32.7% used an insulin pump.
55% of children on an insulin pump used it with a closed-loop system (HCL).
By 2021, all Polish centers had equal access to 2 types of closed-loop systems, namely Tandem Control IQ and MiniMed 780G. A few used AndroidAPS.
Furthermore, a distinction was made between intermittent scanning CGM (isCGM), real-time CGM (rtCGM) and intermittent (<70% of the time) or no CGM (noCGM).

It then looked at the HbA1c, time in range (TIR) and glucose risk index (GRI).

As expected, the best results are obtained with a closed-loop system, with an average TIR of 75%, significantly better than the other groups.

It is notable, however, that the other groups do better than previous estimates, and glycaemia results in the children in Poland are generally very good!

This study shows that closed-loop technology is superior to other forms of insulin delivery and should be the preferred treatment in all children who meet the indication criteria of closed-loop systems.

It should be kept in mind, however, that an association was noted between the use of insulin pumps and an increase in BMI, which could potentially have a bad influence on cardiovascular risk factors.

In addition to this Polish study, a similar analysis appeared recently at a large American center.

Again, the use of CGM and closed-loop systems was shown to be associated with better glycemic control.
Unfortunately, only the HbA1c was discussed here and not the TIR.

5. A fully closed-loop system with AndroidAPS

"Our pilot study showed that FCL might be a realistic mode of treatment for people with T1D." - dr. Petruzelkova Pancreas4ALL Randomized Pilot Study

The concept of a completely closed-loop system, where you no longer have to enter carbs or meals, is getting closer and closer.

AndroidAPS, an open-source closed-loop algorithm, has made great progress in recent months, and more and more people are using it without entering their meals.

Tim Street did an interview with n=93 AndroidAPS users in January 2023, and it showed that

only 51/93 surveyed always gave a bolus at meals (their mean TIR was 86%),
15/93 never gave a meal bolus (their mean TIR was also 86%),
and the rest 27/93 of the user occasionally gave a meal bolus (their mean TIR was 83%).

It was noticed that there was no clear difference between the mean TIR in these 3 groups.

Note: only the TIR was surveyed here, which may not give a complete picture of glycaemic control.

Meanwhile, there are several ways to make it easier to handle meals when using AndroidAPS.

These include using

supermicroboluses (SMB) and unannounced meal function (UAM)
automations (in the AndroidAPS app, you can set a temporary insulin dosage based on your glycemia, trend arrow, insulin on board or location, among other things)
automated insulin management interface (AIMI)
experimental variants of AndroidAPS:
- autoISF (insulin sensitivity adapts to glycaemia or glycaemia trend)
- Boost (gives more insulin faster by using a dynamic insulin sensitivity)
- Eating Now (elements of Boost, only activated when you enter a treatment)
- or Tsunami (exponential Smoothing)

To get good results without entering carbohydrates, most people do use faster-acting insulin (eg Lyumjev) and a low-carb diet.

Also published this month was the Pancreas4ALL study, already briefly discussed at ISPAD 2022.

Here, in n=16 adolescents with type 1 diabetes on AndroidAPS, glycemic control for 3 days during admission was compared between

group 1 = entering grams of carbohydrates in the bolus calculator,
group 2 = entering grams of carbohydrates in the "add carbs" function,
and group 3 = entering no carbs.

A regular version of AndroidAPS was used, led by Milos Kozak, the lead developer of AndroidAPS:

Dana-RS pump with Fiasp, Dexcom G6 and AndroidAPS version 3.1.03 with SMB and UAM
Target value at 100 mg/dl (5.5 mmol/l)
People ate an average of 182, 155 and 165 grams of carbohydrates per day in the 1st, 2nd and 3rd groups, respectively

There were no significant differences in glycaemic control between groups

The TIR was 83%, 80% and 81% in the 1st, 2nd and 3rd groups, respectively.
Time between 54-70 mg/dl (3.0-2.8 mmol/l): 3.55%, 2.35% and 1.4% in the 1st, 2nd and 3rd groups, respectively
Time < 54 mg/dl (< 3 mmol/l): 1.05%, 0.00% and 0.00% in the 1st, 2nd and 3rd groups, respectively
Mean glycaemia was 120, 131 and 127 mg/dl (6.65, 7.3 and 7.05 mmol/l) in the 1st, 2nd and 3rd groups, respectively

Dr Petruzelkova concluded from this that it is possible to use AndroidAPS as a fully closed-loop system.

Of course, this was a very small study, and further research is needed before this can be recommended.

The evolution of diabetes technology never stands still, and open-source closed-loop systems play an important role in this progress.

In addition to the developments already mentioned, people using open-source closed-loop systems are constantly testing new strategies.

These innovations aim to further refine the algorithms and improve the functionality of the system.

One of the interesting developments is the use of microdoses of glucagon to treat hypoglycemias. It was noted here, for example, that the action of glucagon may be limited if alcohol is involved.
Another promising development is the integration of a pedometer into the algorithm. By monitoring activity, the system can take into account longer periods of inactivity or increased activity, for example, as well as the first steps of the day. Thanks to in-app automation capabilities, you can easily set a specific (temporary) insulin release based on this.
There are also those working on integrating a heart rate monitor.

The enthusiasm and dedication of developers to further improve open-source closed-loop algorithms are obvious.

They are working hard to optimize their performance and increase their functionality.

If you want to stay on top of the latest developments, we recommend following Tim Street of Diabettech.

He is an authority on diabetes technology and regularly shares updates on his own exciting experiments and discoveries in this field.

The future of diabetes technology is promising, and open-source closed-loop systems are also playing a crucial role in the drive to improve the quality of life for people with diabetes.

Stay tuned to find out what's possible in this rapidly evolving landscape of innovative solutions.