August 02, 2018
Researchers have identified three "glucotypes," or differences in glucose spike patterns, in a study of 57 adults both with and without diabetes who wore a continuous glucose monitor (CGM).
The study, led by graduate student Heather Hall, from Stanford University, California, was published online July 24 in PLOS Biology.
"Importantly," Hall and colleagues write, "we found that even individuals considered normoglycemic by standard measures exhibit high glucose variability using CGM, with glucose levels reaching prediabetic and diabetic ranges 15% and 2% of the time, respectively."
"We saw that some folks who think they're healthy actually are misregulating glucose — sometimes at the same severity of people with diabetes — and they have no idea," said senior author Michael P. Snyder, PhD, professor and chair of genetics, also at Stanford University. "We think that's quite novel and it does mean that more people need to be on the alert than probably otherwise realize," he told Medscape Medical News.
Whether these glucose spikes are important clinically or not still remains to be determined in much larger trials, experts say.
In the second part of this study, the researchers showed that a breakfast of cereal with raisins and milk caused a large spike in glucose in almost all participants.
Snyder noted that CGM devices are "very powerful" for identifying foods that may cause blood glucose level spikes.
However, "it's a smallish study," he conceded, and they plan to continue the research in more people.
The study is "certainly hypothesis generating," Anne Peters, MD, from the University of Southern California, Los Angeles, told Medscape Medical News.
"It doesn't prove anything, but boy do I see it clinically," said Peters, who writes commentaries for Medscape Diabetes & Endocrinology. "We've never had this [CGM] tool before," she added. "It's just staggering what we're learning [from it]."
"I have these patients who have [severe] cardiovascular disease," she noted, "and if they have all these high spikes...my treatment is, I'll get their A1Cs down to 5.6%, [and] I'll do whatever I can to reduce the variability in these spikes, in the hope that I'll help slow progression of coronary disease."
Peters worries that those spikes increase the risk for cardiovascular disease or even cancer, "but I don't know for sure," she said. "None of us know, because we don't have the outcome [data from large trials]."
That is, you would need to have a trial with 2000 to 5000 people and divide them into these glucotype subsets, treat some to lower their postprandial glucose and not treat others, and then follow them to see if there is a benefit.
"Because if these postprandial spikes don't hurt you, then we don't have to worry," Peters noted. "Before I change everybody's lifestyle, we need to know that this has some clinical merit."
"But it's hard to believe that these glucose spikes don't matter," she stressed.
CGM Identifies Distinct Glycemic Patterns: Some 'Spikier' Than Others
The current way that diabetes or prediabetes is diagnosed does not take into consideration "nuances of glycemic patterns," Hall and colleagues write, but now a CGM device (which is available without a prescription in Europe) provides a tool to show how blood glucose fluctuates over time, a concept clinically known as glycemic variability.
Researchers aimed to characterize glycemic patterns in healthy individuals in the community, and then see how they responded to three standardized meals.
They recruited 32 women and 25 men aged 25 to 76 years (median age 51 years) who did not have a diagnosis of diabetes.
However, a further screening test showed that although 38 did not have diabetes, 14 had prediabetes and five had diabetes.
The study participants wore a Dexcom G4 CGM device for 3 to 4 weeks and were instructed to calibrate their monitors once or twice daily.
Researchers identified three patterns of low, moderate, or severe variability in glucose readings.
Overall, about half of the 57 participants had moderate variability (29 patients), but many had severe variability (22), and a few had low variability (six).
In the nondiabetic group, most participants had moderate variability (23), and fewer had severe (nine) or low variability (six).
In the prediabetic group, most had severe variability (10) and the rest had moderate variability (four).
And in the diabetes group, two participants had moderate variability and three had severe variability.
Cereal for Breakfast?
Next, 30 of the patients — three with diabetes, seven with prediabetes, and 20 without diabetes — were each given three different breakfasts with the same number of calories twice.
The breakfasts were a protein bar (moderate fat, moderate protein), a peanut butter sandwich (high fat, high protein), and cornflakes with added raisins and milk (high sugar, low fiber).
Most individuals (80% overall, and 80% of those without diabetes) had a high spike in glucose after eating the breakfast with cornflakes, raisins, and milk.
"You shouldn't eat all those carbs for breakfast," Peters said. "They just proved what I totally believe. Anybody who looks at [CGM] tracings will tell you the same thing. Again, we don't know metabolically what this means risk-wise, and you don't want to go crazy just treating a number, but these patterns are really striking."
'Are These Spikes Clinically Meaningful?'
Eric Topol, MD, executive vice president and professor of molecular medicine, Scripps Research Institute, La Jolla, California, and editor-in-chief of Medscape, is also intrigued by this research.
"Are these spikes clinically meaningful? We don't know yet," he tweeted.
In a large study of 5000 patients without diabetes (Diabetologia. 2018;61:101-107), having a high 30-minute glucose level in an oral glucose tolerance test was linked to an increased risk of diabetes, he noted. Researchers have also reported that hyperglycemia disrupted the intestinal barrier in a study in mice (Science. 2018;359:1376-1383), and other scientists have reported on a mechanistic link between hyperglycemia and cancer in another animal study (Nature. 2018;559:637-641).
According to Hall and colleagues, the findings in the current study suggest that traditional glycemic classification schemes "are overly simplistic."
Recent evidence, they note, "suggests that glycemic variability, more than fasting glycemia or HbA1c, predicts development of cardiovascular disease, possibly via oxidative damage causing endothelial dysfunction."
"As CGM becomes more accessible and affordable with newer technologies on the horizon, using this metric of glycemic variability may enable earlier and possibly expanded detection of individuals at risk for type 2 diabetes and cardiovascular disease," they write.
If this theory is validated in long-term studies, "use of CGM to define glucotypes might become an important tool for clinicians, and even individuals outside the medical system, to stratify their risk and to adopt interventions shown to prevent diabetes and cardiovascular disease."
"Like a Fitbit, why shouldn't we have a Glucosebit?" Peters wonders.
The authors have no relevant financial disclosures. Peters serves on advisory boards for Abbott, Becton Dickinson, Bigfoot Biomedical, Boehringer Ingelheim, Eli Lilly, Lexicon, Livongo, Medscape Diabetes & Endocrinology, Merck, Novo Nordisk, Omada Health, sanofi aventis, and Science 37. She has received research support from AstraZeneca, Dexcom, and MannKind, and serves on a speaker bureau for Novo Nordisk. Topol is editor-in-chief, Medscape, and serves as a director, officer, partner, employee, advisor, or consultant for Dexcom, Edico Genome, Illumina, Molecular Stethoscope, MyoKardia, Quest Diagnostics, and Walgreen. He has received research grants from the National Institutes of Health and Qualcomm Foundation.