The effects of eating whole fruit or drinking fruit juice on blood glucose and insulin appears to be of some interest to people. As my last post on a study comparing apples, smoothies, and juice garnered some attention, I thought I would follow this up by looking a later study by the same researchers: “The role of dietary fiber in satiety, glucose, and insulin: studies with fruit and fruit juice“. This study expanded on the previous research on apples by comparing the effects of both oranges and grapes on blood glucose and insulin as either whole fruit or juice.

Oranges

South African Navel oranges were used for both the fruit and the juice. As the oranges had a lower amounts of sugar than the apples, the total sugar in each test meal was reduced to 50 grams, to avoid over-stuffing their volunteers with fruit. Even so, it took over 600 grams of peeled oranges to reach 50 gram dose of sugar. The equivalent 600 ml of orange juice was produced by hand using an orange reamer (some task!) and the fibre removed using a pectinase enzyme. As in the previous study, both test meals were fed to ten volunteers, on separate occasions, and after being fasted overnight.

Glucose

There wasn’t any difference in the spike in blood sugar after eating oranges or drinking the orange juice. While blood sugar dipped below fasting levels for both fruit and juice, it returned to normal rapidly after eating the oranges but the reactive hypoglycemia persisted to the end of the experiment after drinking juice.

Insulin

Insulin increased significantly more after drinking orange juice than after eating the oranges. The total insulin produced, calculated as the area under the curve, was also quite a bit larger after drinking the juice. Both the blood glucose spike and the insulin response were quite similar to what we saw from the apples in the previous post. It takes a lot more insulin to manage the sugar in orange juice than the same amount of sugar in oranges.

However, just to prove that nothing is totally consistent in nutrition, we now turn to grapes.

Grapes

Spanish black Napoleon grapes were used for both eating and pressing into juice. Each test meal contained 60 grams of sugar, which was equivalent to 339 grams of grapes or 323 ml of grape juice. The volunteers were required to drink the juice at the same speed they had eaten the fruit, which took an average of 19 minutes.

Glucose

As you can see, both the whole grapes and grape juice produced very similar blood glucose responses, with juice producing a slightly faster rise in glucose. Both reduced blood glucose below fasting levels by a similar amount.

Insulin

Paradoxically, and unlike with apples and oranges, the whole grapes tended to produce a higher insulin response than the grape juice. Althought insulin was only significantly higher at four time points, the overall area under the curve was greater for grapes than from grape juice.

Diluting grape juice

The authors of this study wondered whether the high osmolality of the grape juice was slowing its passage through the stomach and, therefore, reducing the insulin response. They repeated the tests with grape juice watered down half-and-half with water.

Diluting the grape juice didn’t significantly reduce the blood glucose peak compared to regular juice and, rather than improving the reactive hypoglycemia, it actually made it a bit worse. Why this should be is unclear. Diluting the juice had no effect on insulin levels compared to the undiluted juice.

This means the osmolality of the juice probably wasn’t the cause of the odd insulin responses.

Perhaps diluting fruit juice isn’t all that beneficial after all.

This follow-up study confirms that effects for apples and oranges are similar. That is, there is not much difference in blood glucose spikes between fruit and juice, but there is a greater increase in insulin production and prolonged reactive hypoglycemia after drinking juice.

The results for the grapes just goes to show how variable results can be in nutritional science when comparing between different foods. The authors of the study could not put forward any definite answer for why grapes behaved differently, other than suggesting that their lower fibre content and relative lack of internal structure may have played a role.

In general, it seems that eating some fruit is probably metabolically healthier than drinking fruit juice, although a little caution may be in order before generalising this to every type of fruit.

Reference:

Bolton RP, Heaton KW, Burroughs LF. (1981) The role of dietary fiber in satiety, glucose, and insulin: studies with fruit and fruit juice. American Journal of Clinical Nutrition.34(2):211-7.