Food and diet

Low blood sugar 'affects food cravings'

“'Good' sugar is the secret to a slim figure,” according to The Daily Telegraph . The newspaper says that a new study has found that when our blood sugar levels drop we lose our ability to control desire and feel an increased urge to eat.

During the study researchers used scans to detect brain activity following a drop in glucose, which is the blood sugar that our cells use as a source of energy. They then compared their results to the participants’ desire to eat different foods and recorded how this related to their blood sugar levels. They found that small drops in blood sugar activated the region of the brain that produces a desire to eat, while adequate levels of blood sugar activated the region of the brain that controls impulses. Activation of this regulatory part of the brain by higher levels of blood sugar was found not to occur in obese individuals.

While these are intriguing results, the study was small, only involving 14 participants. This means the results should be interpreted cautiously, as smaller sample sizes are prone to being influenced by chance.

Where did the story come from?

The study was carried out by researchers from Yale University School of Medicine and the University of Southern California Keck School of Medicine. It was funded by the US National Institutes of Health.

The study was published in the peer-reviewed Journal of Clinical Investigation .

The study was covered accurately by the media. However, no news outlets reported on the small sample size, which is a major limitation of the research. Both the Daily Mail and The Daily Telegraph reported that the results mean that maintaining glucose levels is the “secret to staying slim”, an interpretation that is not supported by this small, short-term study.

What kind of research was this?

This was a small human experiment that exposed participants to images of food and non-food, and measured how exposure to these images related to their desire for food and their brain activity under varying blood sugar conditions. The researchers aimed to detect whether the participants’ desire to eat when presented with external cues would differ according to their blood sugar levels.

The small number of participants involved in the study (14 in total) means the results should be interpreted cautiously, especially as the participants were further divided into smaller subgroups based on weight (five obese versus nine non-obese).

What did the research involve?

The researchers recruited 14 healthy participants - nine male and five female. They had an average age of 30 years and an average BMI of 25.6. Five of the participants were obese and nine were not obese.

The participants were given a lunch prepared by the researchers and then examined using a function magnetic resonance imaging (fMRI) brain scan. During the scan the researchers controlled the participants’ blood sugar by giving them varying levels of glucose and insulin intravenously. The researchers held insulin levels constant, and varied the glucose levels. Glucose levels were initially held at normal levels (euglycaemia), and then slowly dropped to low blood sugar levels (mild hypoglycaemia). This was done over the course of two hours.

During the euglycaemia and mild hypoglycaemia phases, researchers showed the participants images of high-calorie food, low-calorie food and non-food images. After each image was shown, the researchers asked the participants to rate how much they liked the item shown in the image, on a scale of 1 to 9 (higher score meant they liked it more). The researchers then asked the participants to rate how much they wanted the item shown, again on a scale of 1 to 9. The high-calorie images included pictures of cake, ice cream, lasagne, crisps and steak. The low-calorie images included pictures of fruits, vegetables and tofu.

In addition to the behavioural ratings described above, the researchers measured the participants’ brain activity when they were looking at each image. An fMRI is able to measure brain activity in real-time by detecting which brain cells are using oxygen. To activate, brain cells need both oxygen and glucose from the blood.

The researchers recorded how much the participants reported liking and wanting each item, and the areas of the brain that were activated by seeing each of the images. They then compared which brain regions were active during the normal sugar (euglycaemic) phase versus the low sugar (hypoglycaemic) phase. They also assessed whether glucose levels influenced the ability of the food pictures to affect both brain activity and the feeling of desire for food. This was assessed using the rating scale.

What were the basic results?

During the normal glucose level (euglycaemia) phase, the non-obese participants showed more activity in two areas of the brain than during the hypoglycaemia phase. These areas of the brain, the prefrontal cortex (PFC) and the anterior cingulated cortex (ACC), were significantly more active regardless of the type of image presented. These areas of the brain are responsible for controlling impulses. The difference in activation did not occur in obese participants.
During the mild hypoglycaemia, compared with the euglycaemia phase, the researchers found:

  • Hunger ratings were significantly greater, with an average of 5.7 points during the hypoglycaemic phase versus an average of 4.5 points during the euglycaemic phase. Hunger ratings were similar in both the obese and non-obese participants.
  • In both obese and non-obese participants, two areas of the brain called the insula and striatum were significantly more active when presented with both high- and low-calorie food images. These areas of the brain are responsible for promoting feelings of desire and craving.
  • During hypoglycaemia wanting ratings were significantly higher (p=0.006) in response to high-calorie foods, but liking ratings were similar between the two phases.
  • There was no difference in brain activation in response to viewing low-calorie foods.

How did the researchers interpret the results?

The researchers concluded that small drops in glucose levels set in motion “adaptive mechanisms” that specifically increase the desire for high-energy and glucose-rich foods. That is, in response to blood sugar levels decreasing, the participants’ brains responded in ways that would increase desire to eat foods that would provide them with high levels of necessary sugars. They say that this activation occurred differently in obese people from in non-obese people.

The researchers say that, further to this, they were able to identify an interaction between blood glucose levels and external cues (the sight of food) that results in a drive to eat. They say that during the normal glucose phase, the activity in the PFC area of the brain (which controls impulses) decreased the desire for food in non-obese people. During the low glucose phase, however, a different region of the brain was activated in response to the sight of sugary foods. The activation of this region led participants to feel a desire for these foods.


This was a small human study that aimed to determine which areas of the brain were activated by the sight of food under different blood sugar levels. The use of both self-reported and brain imaging measurements provides information not only on physiological brain activity, but also on how this activity translates into consciously felt desires.

The researchers found that different areas of the brain are activated depending on the level of glucose available. When sufficient levels are present in the bloodstream, brain regions that control impulses seem to be activated. When low levels are present, brain regions that trigger desire and reward are more activated. The researchers say the level of activation of these regions differs depending on the weight of the individual.

When considering the implications of this research, it should be noted that the study was conducted under conditions that allowed the researchers to hold insulin levels constant artificially while manipulating glucose levels. This is not a state in which a person would find themselves naturally, as both insulin and glucose levels vary constantly. This feature of the study makes it difficult to generalise the results to a real world setting, particularly as, in everyday life, blood insulin levels would be expected to drop once sugar levels were too low.

This study has produced some interesting results but, ultimately, studies of this size are useful for generating theories rather than proving them. The sample size here (14 people) was very small and the results should be interpreted cautiously. Also, any comparisons between the obese and non-obese participants (five and nine people, respectively) are likely to be influenced by chance. Any further research attempts to confirm these results should involve more participants.

NHS Attribution