There’s been a lot of activity on the topic of closed-loop insulin pumps, with much speculation about why they may or may not work, what challenges they face, and other areas of interest. I’ve been reading press releases from manufacturers, which seem to contradict what’s found in published studies from independent researchers. So, I thought I’d write up my findings. Turns out, it’s a bit of a long haul, so I broke them up into three separate articles.
In this first of three articles, “Why Controlling Glucose is so Tricky,” I examine what the natural body has to contend with to try to regulate glucose levels, and then measure that performance by reviewing glucose patterns of non-diabetics to explain why even nature struggles with this difficult task, largely because of how shockingly complex and delicate the metabolic system is. As the data will illustrate, even the natural human body struggles with it too.
This material will lay the groundwork for the next article in the series, “Closed-loop Insulin Pumps: Are We There Yet?” There, we’ll look at how closed-loop systems are performing in the field, where they work well, and what barriers remain for improvement.
In the last article in the series, we’ll talk about what new tech is needed (some of which is in development) that can help overcome these barriers.
Obesity in T1D is lower than the general public because life expectancy is lower, generally, by about 12 years (with a lot of variability, mostly governed by economic conditions). Because T1Ds die much sooner due to weight-related morbidities, their percentages skew more in their favor. If T1Ds lived longer, the ratio would be more on par with the general population.
I will be publishing a separate article on T1D lifespan separately–there’s a lot of fascinating data there.
So you are saying that T1Ds weight is generally the same as the rest of the population? I don’t believe that we die 12 years earlier than comparable life expectancy in nonT1Ds. My grandmother with T1D died at age 72 in 1960 after years of one shot of insulin a day. Having used insulin pumps for the last 15 years and at age 83, I am not ready to follow my grandmother… yet. There are quite a few really older people on this site. Are we “special”?
As I mentioned before, the data about T1D death rates (and causes) is complex due to the vast diversity of T1D demographics. It’s hard to nail down a specific figure (like 12 years), but that’s the figure most demographers cite when forced to come up with a single number. The variability rises dramatically when other factors are considered. My later article on this will get into the weeds on it.
The thing to keep in mind about all this stuff is that anecdotal evidence cannot be applied to broad populations. Yes, there are many overweight T1s and T2s that live into their golden years, and many studies seem to be homing on genetics playing a far more significant role than merely one’s A1c levels. Some people’s bodies are just better at dealing with these things better than others. But I personally wouldn’t want to rely solely on my good genes while whistling into the wind.
The author says in the article: “ T1Ds’ singular aim of lowering their A1c levels has resulted in their taking too much insulin, raising the rate of obesity.” I find that statement interesting…not sure if it is an accurate assessment of obesity in T1 or not.
Insulin is anabolic hormone that plays a role in inhibiting protein catabolism, stimulating lipogenesis, and slowing basal metabolism [4, 5, 9], resulting in increased fat accumulation [9, 10].
This was always known, but it wasn’t until the first formal trial involving tight glucose management in the DCCT was the data more formally quantified. The same paper above cites this finding:
Intensity of insulin treatment influences weight gain as shown in the Diabetes Control and Complications Trial (DCCT), where patients on intensive insulin therapy gained an average of 4.6 kg over 5 years, which is significantly more than patients in the study’s conventional arm [13].
Sadly, this known causality between insulin and body weight has an evil twin: Omitting insulin (OI) as a method to lose weight. In the literature review article, “A review of risk factors associated with insulin omission for weight loss in type 1 diabetes”, the authors begin the paper with the astonishing statistic that “60% of people with type 1 diabetes (T1D) admit to misusing insulin … for the purpose of weight loss, often referred to as diabulimia.” The condition is particularly acute with women, especially young girls.
So, yes, the simplistic aim of lowering A1c levels has resulted in the rise of obesity and other weight-management concerns, which oddly results in trading one set of risk factors for another. I provide more detail in the article, “HbA1c Tests and T1D: The Good, The Bad and the Ugly.”.
T1D 54 yrs here. Know quite a few T1Ds and none are obese, not even close. My endo insists I be on a statin so fluvastatin is one of 3 pills I take daily.
All stats I’ve read about T1Ds is that heart and stroke are the number 1 killers of diabetics, thus keeping at a low range LDL cholesterol is key through statin use. Weight will cause an extra burden especially if obesity is at hand. It’s tough but makes sense to me.
Know quite a few T1Ds and none are obese, not even close.
I know! Me too! It’s crazy, right?
It’s a good thing there are larger institutions and medical groups that collect data from around the whole country to get a more accurate, bird’s-eye view of the state of the T1D universe. As one might expect, the rate of obesity–or most anything else about T1D–is not evenly distributed throughout the population. We all live in in our corners of the world, where it is so easy to think that everywhere else is just like where we are.
Fortunately, we can read these peer-reviewed publications and get those realistic stats on what’s going on out there.
@argv Read your interesting paper. One major issue I see with a device to mimic the endocrine pancreatic functions is the lack of the internal negative feedback systems inherent in health islets.
I can’t find the paper that I read regarding Alpha and Beta cells having insulin and glucagon receptors, respectfully. These receptors do not respond to exogenous insulin or glucagon but only to the hormones within the islets themselves.
A fully automated insulin pump is going to be problematic in giving anything like good blood glucose management. I don’t think state of the art tech is up to the job and maybe not in the foreseeable future. It is a very complex problem when there may be varying levels of pancreatic endocrine disfunction.
Correct–and I will be addressing this in the third article in this series, which addresses what automated systems won’t be able to do. Next generation pump manufacturers are purporting to be working on supporting glucagon delivery, hoping to mimic alpha cell function, but there are lots of studies showing that this isn’t as easy as it appears. I will provide citations, but you land on the crux of the problem: The body’s natural method of delivering glucagon relies on the proximity between the alpha and beta cells and the cross-signaling between them. When insulin is released, there’s a signaling to alpha cells that cannot happen through exogenous insulin delivery, and current research shows that artificially delivering glucagon through subcutaneous tissue (through pumps) is rife with errors in dosage, absorption variability, and timeliness of efficacy (from 20m to 2h).
Again, that’s the third article in the series. The second article, which should be coming in the next few days, will be only addressing the current state of automated systems today and their performance metrics.
