Do Peaks and Valleys Matter?

Patient Expert

[Science of Diabetes]

One of the most common questions asked by people with diabetes is this: "What are nondiabetic blood glucose (BG) levels?" I discussed that in my recent blog "Normal Blood Glucose," which you can read on this site. But just in case you don't want to slog through that verbiage, the most important graph in that online lecture is here.

Another question that comes up over and over again is this: "Is it better for me to eat a food that causes a high peak for a very short time or a food that causes a small peak for a longer time?"

Different people give different answers to that question. Some argue that the only thing that matters is the A1c, so if the area under two curves one very high but coming down rapidly and the other one lower but lasting a long time was the same, it wouldn't matter.

However, a fascinating talk given last fall by endocrinologist Irl Hirsch suggests that there's more to consider than just the average BG level. As he says, "We need other metrics than the A1c."

If you have high-speed Internet access, you can view the entire presentation yourself here. I'll try to summarize what he said.

Most of us are familiar with the results of the Diabetes Control and Complications Trial (the DCCT). This landmark trial showed that BG levels do matter and are clearly related to the occurrence and progression of complications.

The A1c measures average BG levels, and the A1c was clearly related to the incidence of complications in this trial. Hence most clinicians and patients since then have focused on the A1c as a measure of good BG control. If your A1c is good, you're under good control and shouldn't develop significant complications.

But Hirsch asks, "Did we misinterpret the DCCT?" Sometimes we forget that because low BGs reduce the A1c, if you have a lot of very high BG levels and also a lot of very low BG levels, you can have the same A1c as someone who maintained good control 24 hours a day. Hirsch argues that the results of the DCCT suggest that fluctuation in BG levels going very high and then very low** can be as damaging as having high BG levels all the time**.

He supports this by comparing the results of the type 1 patients in the DCCT receiving "standard treatment," meaning one or two shots a day, and the type 1 patients in the trial receiving "intensive treatment," meaning three or more shots a day, adjusting the dosage according to the results of frequent BG testing and trying to keep their premeal BG levels between 70 and 120 and their postmeal BG levels under 180 mg/dL.

It turns out that the patients receiving the intensive treatment had fewer complications even when they had the same A1c as those in the standard-treatment group.

You can see the results here.

In the patients who had an A1c of 9, for example, there was a 50% reduction in the risk of retinopathy in those receiving the intensive treatment.

Next Hirsch asks why this should be so. He suggests and uses published research studies to provide good evidence for this that it's because high BG levels produce oxygen free radicals, or reactive oxygen species (ROS), and these ROS cause increases in the flux through 4 important metabolic pathways: the polyol, protein kinase C, hexosamine, and advanced glycation endproduct, or AGE, pathways.

Unless you're a biochemist, you don't need to understand what all these pathways do. You just need to know that high BG levels do more than just glycate proteins, producing AGEs and increasing the A1c: they also accelerate the flux through several different pathways. And these changes can be harmful.

You can see a chart of these pathways here.

These effects are not new. They've been known for some time. But most people have continued to focus on the A1c alone.

Hirsch then cites further studies that provide evidence that high BG levels cause more oxidative damage than normal BG levels, but alternating high and low BG levelscause even more damage than the high BG levels alone. Hence he suggests that glucose fluctuation may be involved in the development of oxidative stress and vascular injury.

This oxidative stress is not related to the A1c, fasting BG levels, fasting insulin levels, or mean BG levels. One way to measure it is to measure the urinary concentration of something called 8-iso-prostaglandin F-alpha, which Hirsch calls the "hemoglobin A1c of oxidative stress." Obviously, this test is not common today, but Hirsch feels that it will become common in the future.

One way to measure the glycemic variability that Hirsch and other scientists feel is so important is to measure something called MAGE (mean amplitude of glycemic excursions), but this is best done in a research setting, or perhaps with a continuous BG monitor. He says MAGE and oxidative stress are strongly correlated, and MAGE is more strongly correlated with oxidative stress than postprandial BG levels, which have been the topic of much discussion in recent years, because the postprandial levels only measure the highs, not the lows, and MAGE shows both.

An easier way to estimate glycemic variability is to calculate the standard deviation of your BG level readings (assuming you take a fair number of readings every day).

Oxidative stress can cause inflammation, which is also thought to cause vascular complications. And Hirsch showed an experiment in which nondiabetics saw an increase in one marker of inflammation when their BG levels were raised to only 180 mg/dL.

Hirsch points out that some of the experiments he cites were done in cell systems (in vitro) rather than in humans, and hence they're not conclusive. However, along with the DCCT results, they are very suggestive.

He concludes that all this "suggests that different therapeutic strategies now in use should be evaluated for their potential to minimize glycemic excursion, as well as their ability to lower A1."

This all makes intuitive sense to me. In general, the body doesn't like rapid changes, just as we wouldn't be very happy if the temperature in our homes kept going up to 100 and then down to 40, even though that meant the average temperature was 70 degrees.

And that's why, in a healthy person, the body has such exquisite controls to keep the BG levels in a very narrow range. When we lose those controls and keep bouncing up and down, is it any wonder that some damage can result?

If it turns out that it's glucose fluctuations that are causing complications, it's another reason that going on the American Diabetes Association's recommended low-fat, high-carbohydrate diet is probably a bad idea. With high levels of carbohydrate in your diet and no fat to slow them down, itâ�™s inevitable that your BG levels are going to bounce around, even when you "cover" the carbs with insulin.

If you want to read more about this, you can read an editorial by Hirsch and Michael Brownlee here and a research report here.