I’ve read in dozens of articles across the internet that reactive hypoglycemia is one of the first signs of diabetes. According to many prominent researchers (including Achim Starke, MD and Jean-Frederic Brun), it isn’t. In a recent email to Dr. Brun, I asked him why reactive hypoglycemia is often linked to prediabetes. His response was that in “medical studies emphasis is always put on pathologic situations.” Most of the time, says Dr. Brun, prediabetic hypoglycemia occurs late after the meal (more than 4 hours) while reactive hypoglycemia (hypoglycemia at 2-3 hours) is “…on the opposite (lowered risk of diabetes).”
In other words, if you have true reactive hypoglycemia (with symptoms appearing at 2-3 hours), the statistics say you are more likely to have a lower risk of diabetes. This article will help explain why even my own doctor got it wrong.
How can so many physicians get it wrong? As a mathematician who teaches statistics classes at the college level, I have a pretty good idea. Giving a patient a diagnosis of reactive hypoglycemia (or prediabetes) involves looking at a few numbers, analyzing those numbers, and coming up with a diagnosis for the patient. Unfortunately, physicians frequently misread health statistics, and do not know the probability that someone has a particular disease given the results from a screening test. That’s according to a report by the journal Psychological Science in the Public Interest published in US News.
One example given in the above article is a woman given a positive result on a mammogram. One hundred and sixty gynecologists were asked “What are the chances that the woman has breast cancer?”; only 20% got the right answer (which was that the woman had a 10% chance). Another 20% said she had a 1% chance, and an incredible 60% got it wrong and said she had an 81 or 90 percent chance of having cancer!
In order to illustrate why it isn’t easy to give a diagnosis for any particular disease, here is a graph of two patients who have just had a Glucose Tolerance Test (GTT). The patient with the red line is overweight (with most of her fat around the hips and thighs). The patient with the blue line is normal weight.
Which patient is at risk from diabetes? The overweight red patient? The normal weight blue patient? Or both?
Here are a couple of pieces of information to help you make up your mind:
1. Normal blood glucose values for a 75-gram oral glucose test are (from the National Institutes of Health):
- Fasting: 60 -100 mg/dL
- 1 hour: less than 200 mg/dL
- 2 hours: less than 140 mg/dL.
(Between 140 – 200 mg/dL is considered impaired glucose tolerance or prediabetes. This group is at increased risk for developing diabetes. Greater than 200 mg/dL is a sign of diabetes mellitus)
2. Prediabetic patients typically see hyperglycemia (>180 mg/dL) and hypoglycemia (>60 mg/dL) at the same time,
The correct answer is, you really can’t answer the question…at least, not from this test. The blue patient certainly looks like they could be prediabetic (and suffering from what doctors call glucose intolerance), but we don’t know anything about this patient’s other risk factors. The red patient looks borderline, but the GTT cannot be used to accurately diagnose reactive hypoglycemia (to find out why, see my article on tests for reactive hypoglycemia). Additionally, the red patient has excess body fat around her hips, which actually protects from diabetes! Here’s a quote from researcher Dr. JF Brun from page 10 of his report on Postprandial Reactive Hypoglycemia:
“…reactive hypoglycemia is frequently found in women with moderate lower body overweight…This situation seems to be associated with
a lower incidence of diabetes.”
In essence, if you’ve received a GTT result that looks like the red patient and if your doctor has told you that you are at risk for diabetes, ask for a Hyperglucidic Breakfast Test, especially if you don’t have other risk factors for diabetes. A Hyperglucidic Breakfast Test is the only test that will be able to tell you if you have a high sensitivity to insulin (and therefore a lower statistical risk of diabetes) or a low sensitivity to insulin (and an increased statistical risk for diabetes).
Blood glucose levels alone tell you practically nothing about your possible prediabetic status!
*update, Sept 20: despite having pretty severe reactive hypoglycemia, my endocrinologist confirmed that not only I am not prediabetic, but there is no evidence to link reactive hypoglycemia to diabetes. I am actually at “low risk for diabetes” (in my doctor’s words).