Following the update of the recommended clinical targets for continuous glucose monitoring, guidelines author Simon Heller reflects on the benefits of the technology for people with diabetes and their clinicians, but explains why these may never apply to a large proportion of people.
A huge step forward
June of this year saw the update of previously published recommendations for continuous glucose monitoring (CGM), by a panel orchestrated by the ATTD Congress.
The inclusion of Simon Heller (University of Sheffield, UK) on this panel stemmed from his previous work with the International Hypoglycaemia Study Group, which led to the widespread adoption of the 3.9 and 3.0 mmol/L (70 and 54 mg/dL) blood glucose thresholds to define hypoglycemia.
These thresholds are included in the new, streamlined CGM guidelines, with people advised to spend no more than 4% of the day (approximately 1 hour) below the higher threshold and no more than 1% (about 15 minutes) below the lower threshold, which Heller describes as “quite easy to remember and quite meaningful.”
The guidelines’ overall focus is on time in range: the amount of time per day people spend with their blood glucose levels within the range of 3.9–10.0 mmol/L (70–180 mg/dL), the target for which is 70%.
Heller considers this a little ambitious for most people, suggesting that 50% is a more realistic target, but stresses: “I do think that time in range is a huge step forward in many ways and that’s particularly because it’s so much more informative to people with diabetes, compared to HbA1c [glycated hemoglobin].”
In the absence of CGM, people with diabetes and their clinicians are largely reliant on HbA1c, a measure of longer-term glucose control that can conceal large short-term fluctuations and can be expressed as a percentage or as mmol/mol, the latter being confusingly similar to the mmol/L units widely used to express blood glucose levels.
“Time in range is a far more meaningful concept,” says Heller, because it gives people with diabetes both a clear glucose range to aim for and a target time to remain within it. “It is just fantastic, and I think people with diabetes really get it.”
He believes that CGM “has changed practice hugely,” saying: “The days when people had to write the results in a book and [could] then fake them all have gone, because you either don’t bring your meter, which in itself is telling, or you acknowledge that.”
It can even be useful as a short-term diagnostic tool, as was shown in the DIACCOR study, in which just 1 week of CGM use allowed people with diabetes and their clinicians to identify glucose management issues that would not be apparent purely from HbA1c levels.
“The best one is giving insulin after you eat, instead of before, because around 50% of [people with type 1 diabetes] give their quick-acting insulin after they eat, because they can count their carbs,” says Heller.
“It makes total sense and it reflects the ludicrous impracticality of doctors saying, you should take your insulin half an hour before, which of course you should, all things being equal.”
But even short-term CGM data allow people to clearly see what happens when they delay their meal-time bolus, and also the effect that snacking without taking insulin boluses has on their blood glucose. “If you just see it through the afternoon going up and up and up and you just – hopefully, in no way punitive – just ask them what they think is happening,” says Heller.
He also highlights the ability to spot risky glucose management behaviors, such as taking an insulin bolus to correct blood glucose levels immediately before going to bed – “a really dodgy thing to do,” because of the risk for nocturnal hypoglycemia.
“Now, of course, it is all very well identifying that, but it is much more difficult to change behavior,” Heller concedes, but he suggests that short-term CGM use “will take you some way” towards this goal.
However, he stresses: “I guess the reason I would like to see CGM used most is: it just improves quality of life – certainly for those who are engaged.”
Two standards of care
Yet uptake of CGM remains low, even in healthcare systems where funding for the devices is available (including for the cheaper “flash” glucose monitoring devices), so the majority of people with type 1 diabetes continue to rely on self-monitored blood glucose, and therefore on the less intuitive and informative HbA1c targets.
“Then there is the type 2 population,” says Heller, “the vast 90% of the people [with diabetes] in the world, very few of them who will be using CGM for a long, long, long time and it is a meaningless concept for them. For them, HbA1c presumably is the measurement which we will continue to bash them over the head with.”
He asks: “So what do you do then? You’ve really got two standards haven’t you?”
Engagement reaps rewards
But even where access to CGM devices is straightforward, will they provide all the benefits seen in randomized trials? Heller cautions against such assumptions, pointing out that real-world rates of hypoglycemia have remained stubbornly unchanged over recent years, despite the introduction of the new technologies and insulins that reduced its incidence in clinical trials.
Likewise, “there is evidence in the US that glucose levels are rising, despite all these fancy new treatments,” says Heller.
“Why is that? Well it is because – as I have to keep saying again and again and again – the outcomes depend upon the person with the diabetes, doing what is extremely complicated. Most are very complicated in type 1, but even type 2 is so demanding of people and until we get that right, none of the most fancy insulin technology is making… well it will make a difference, but its difference is going to be so much less than people expect. It is going to be disappointing.”
In Heller’s clinical experience there is a wide spectrum of engagement among people with diabetes, stretching from the ultra-engaged people who get the best out of technology to the totally disengaged for whom the focus is simply about getting them a consistent dose of insulin.
“If you said to me: what is the breakdown? This is kind of rule-of-thumb, but nobody has really disagreed with me. I think 10% to 20% of people ‘get’ diabetes, I am talking type 1, but I suspect not many for type 2. They would do DAFNE self-management, even without any healthcare professionals because they are really fired up and they learn to do this themselves.”
These people are representative of the engaged subgroup who participate in clinical trials, in which the benefits of CGM are very clear, and who tend to do extremely well on diabetes devices in general, says Heller – sometimes to the point of not appreciating that less engaged people may not derive the same benefits they do. “They don’t understand how tough it is for 50% in the middle,” he says.
Technology plus high-quality diabetes education will make some difference in this largest group, he believes. “We are not going to achieve perfect glucose control in them, however, whatever the technology can say, but I think we can make a difference.”
Inappropriate patient targeting?
At the other end of the scale are the 20% to 25% of people “who hate diabetes” and do not want to engage with it. These people are “better off taking two shots of insulin a day,” says Heller, “and giving them a [CGM] device is a complete and utter waste of money.”
He cites an economic study from the USA, which concluded that if CGM were given to everyone, “you probably lose 30% who just put it on their shelf and never use it,” and highlighted the cost implications.
And “that’s the worry,” he says, that healthcare providers may hand out CGM devices to inappropriate subgroups. “They are giving them to people with high A1cs, when actually the people who really need them are the people with glucose values in range, because they would use it really well.”
Heller says: “I hope in the next year, people will begin to realize that depriving people who are reaching these targets with [self-monitored] blood glucose testing is unfair.
“It raises the ludicrous situation, as one of my patients said, ‘so in order to get one, I have got to let my glucose values go up and watch my glucose control deteriorate, is that what you’re telling me?’ “And I said: ‘That is exactly what I am telling you, or you have got to do eight tests a day’ and do I think that is absurd? I totally do, but at the moment that is the rule [in the UK].”
Educational needs
Two studies presented at this year’s EASD conference showed that use of CGM had the greatest impact on glycemic control, irrespective of people’s means of insulin delivery, which does “undoubtedly show that just handing out CGM can make a difference,” says Heller, “and that’s not to be dismissed.”
But as a clinical trial population represents an engaged subgroup, likely to get good results from diabetes devices, Heller warns: “I think healthcare professionals and patients are socialized into believing technology will solve their problems, but what it does is probably demand more of them.”
This can have unexpected benefits, he concedes, by more fully engaging people with their diabetes, leading to improved glucose control: “Just to wear a pump safely, you have to be engaged.” And this then is an important contributor to the effectiveness of insulin pumps. But he disagrees with people who dismiss the need for education, citing safety foremost, but also that “in fact it is very disempowering to engage people without [giving them] the skills.”
Giving people with diabetes these skills is made harder by the patchy availability of training for healthcare providers, which Heller attributes partly to the competencies-based education provided by colleges for medical specialists. “It is about physical examination. Self-management of diabetes doesn’t even come into it.”
In Heller’s own region, specialists working with people with diabetes have to sit in on a week-long DAFNE course and “they learn more in that week than sitting in clinic for a year,” he says.
“That should be mandatory, but when I and others have suggested it, the college committee, which is dominated by endocrinologists, just don’t understand what we are talking about.”
Gauging the impact
Despite the effort put into the CGM guidelines, it is hard for the authors to know the impact they are having, especially beyond the somewhat rarified bubble of diabetes technology specialists, where awareness is an obvious problem: “If you don’t read the guidelines, how do you learn about them?” asks Heller.
“Nobody is going to read the Diabetes Care paper,” he admits, although he hopes that their recommendations will filter through to national diabetes guidelines, which are more likely to be read.
The effects of CMG uptake and harmonized targets on real-world glycemic control also remain to be seen, and judging from the experience with hypoglycemia rates may be very hard to prove or refute, despite the increasing availability of large, high-quality datasets.
Nevertheless, Heller anticipates future guideline updates. He notes that they are currently based on consensus, and the thresholds and time in range recommendations remain arbitrary and may need to be revised as long-term data become available.
“There is always a danger that healthcare professionals, particularly doctors, think they know what they are talking about!”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2019 Springer Healthcare part of the Springer Nature group