番茄社区

Opinion Makers is an exclusive 番茄社区 video series, presenting leaders from all areas of medicine, offering their views on current topics in clinical care, research, and policy.

In this video, , of the University of California San Diego, talks about blinding of readouts from continuous glucose monitors. Why do patients do better at controlling their blood glucose when they can see values on the CGM? The answer has important consequences for managing patients who are at risk of hypoglycemia.

Today I'd like to talk about blinded versus unblinded or real time continuous glucose monitoring. Just to give you a quick background is that years ago when CGM first came on to the market it was all blinded. Patients would wear a device for 3 days, then bring the device back to the caregiver. They would look at the trends and patterns and try to make heads or tails.

But since then there has been a development of real-time unblinded CGM that gives the user -- the person with diabetes -- basically a fresh glucose value every 5 minutes with trend arrows, which way they're going, which way they're coming from and alerts that can alert them before they get extremely high or dangerously low.

Now, when you look at the literature on unblinded CGM, it's a very consistent group of papers that show basically the same thing. Improved quality of life but very importantly, less A1C, a reduced A1C and at the same time reduced hypoglycemia.

So that's exactly what you want: to try to get patients to goal without increasing the incidence of hypoglycemia. If you look at the literature on blinded CGM, trials using these devices where the patients have no idea what the value is showing, it's very limited and many studies are negative showing no improvement.

Now, I have to tell you, in a clinical setting, it's fair to say that blinded CGM makes perfect sense, when you're testing a medication, insulin oral agent versus placebo or some other competitor. That's fine. And if certain patients with type 2 diabetes maybe on orals, maybe with pre-diabetes where you want to pick up a pattern or a trend and they're not at risk for hypoglycemia, okay, that's fine too.

And of course if you're going to use unblinded CGM in type 2's and those situations -- I'm not here to say you slap it on and they're going to get better and everything is going to be hunky-dory, it's going to come with education. You're going to have to let patients know what to look for, what to pay attention to, to try to get them motivated into looking at their numbers and making some positive behavior changes.

Now, a lot of people will say, what about the Hawthorne effect? If you give it to patients and they can see the value, they're going to change their behavior. Their eating, their exercise, maybe other types of behavior and you're going to see an improvement of their diabetes not related to the drug, either insulin or oral agent. Well, I say that's a good thing. We want the Hawthorne effect.

So instead of adding on more and more medications to our patients who are not doing well, their A1C is not at goal and not really having enough time to get into their lifestyle of why they're not doing well, then you put this device on them, they see it, they see the numbers, they get motivated, they make positive changes, their A1C gets better while you've got not only the diagnosis of why their diabetes was not in a good control but you also got the therapy.

So I think the Hawthorne effect is a good thing, not a negative thing. I think in the long term CGM is going to really improve the lives of people with diabetes. There's no question, it has already reduced the incidence of severe hypoglycemia and potentially death from low blood sugar. And I can tell you, in 2016, it's the standard of care for people with type 1 diabetes. And I can -- also in the future when these devices get small, accurate, easy to apply and less expensive, they're going to be beneficial for all stages of type 2 diabetes, from pre- to oral agents to combination therapy to those on multiple daily injections. Thank you.

Comment