The varying health effects of eating fruit, as compared to drinking juice or smoothies, are often discussed. Curiously though little actual scientific research is directly cited showing the metabolic effects of eating fruit or drinking juice. This strange absence of data inspired me to search for myself to see what experimental results I could find in the literature. Somewhat surprisingly, I had to go back quite a way to find any research study describing such a comparison published in the Lancet on October 1st 1977 titled:

‘Depletion and disruption of dietary fibre. Effects on satiety, plasma-glucose, and serum-insulin’

These researchers from the Bristol Royal Infirmary and the Long Ashton Research Station were interested in testing the fibre hypothesis, proposed by Thomas L. Cleave and detailed in his 1974 book ‘The Saccharine Disease’. Cleaves’ proposition was that the consumption of refined, fibre-depleted, carbohydrates results in overnutrition and obesity and diabetes, due to reduced satiety and excessive insulin secretion.

To compared the presence or absence of fibre they compared whole apples, pureed apple, and apple juice with all the fibre removed. The apple purée, what would now be called a smoothie, represents a food in which the fibre is present but the physical structure of the food disrupted. Based in the south-west of England and having the expertise from the Long Ashton Research Station, a small government agricultural research institute specialising in apples, and other fruit production, the apple seems the natural choice of experimental fruit.

The researchers involved went to quite some length to make the study as well controlled as possible. All the apples used were Golden Delicious apples, harvested from a single plot of land. The apple juice was made by pressing fresh apples, removing any traces of fibre, and then pasteurised. Each meal contained 60 grams of carbohydrate, equal to 482 grams of whole apple, roughly three apples. The apple smoothie was blended just before consumption with 150 millilitres of water added. To make sure everything was equal the volunteers also drank 150 millilitres of water with the whole fruit and juice.

The apples, smoothie, and juice was fed to ten healthy volunteers (rather honestly reported as all being staff in the department) aged between 24 and 40, with a healthy body weight, and importantly with a good set of teeth. These volunteers fasted overnight and avoided any alcohol for 24 hours before each test meal.

Blood glucose responses

After eating the apples blood sugar increased, peaking at 30 minutes, and then quickly fell back to the fasting levels by 60 minutes. Blood sugar rose to the same level after drinking the smoothie or juice. What was more interesting is what happened after the blood sugar levels fall back down.

Instead of returning to normal levels after drinking the smoothie or juice the volunteers blood sugar dropped below fasting levels and remained lower for at least a couple of hours.

To see if drinking fast or slow influenced this effect the volunteers first drank their smoothie or juice fast as they wanted, and then in a second test drank gradually over the same time they had taken to eat their whole apples. Curiously, drinking slowly or quickly didn’t make any difference to their blood sugar response.

Insulin responses

Insulin also peaked around 30 minutes after eating the whole apples and then dropped back to normal levels. After drinking the apple juice slowly, the volunteers’ insulin peaked at almost double the level it reached after eating the apples whole. Curiously, insulin did not reach such a high level after drinking the apple juice quickly. The insulin response to drinking apple smoothie was somewhere in between, but still higher than eating the apples whole. As with the juice, when they drank the smoothie quickly their insulin did not increase as much.

Surprisingly, the fibre in whole apples did not slow down the rise in blood sugar after eating compared to juice, as if often claimed. More worryingly, the effect of juice or smoothie seemed to be to disrupt the normal mechanisms that keep blood sugar under control after the initial blood sugar rise after eating. The rebound fall in blood sugar in the second and third hours after drinking the smoothie or juice was in stark contrast to the steady levels seen after eating the apples.

The greater rises in insulin after drinking the apple juice or smoothie means that more insulin was needed to keep blood sugar the same. Although drinking quickly reduced the peak in insulin a bit it didn’t help prevent the rebound drop in blood sugar. This means that removing the fibre from the apple, or even just breaking up the physical structure of the fibre, contributes to disrupting the normal mechanisms that regulate blood sugar and insulin after drinking it, compared to eating the fruit whole.

If repeated regularly these inappropriately insulin responses and falls in blood sugar after drinking juice seem unlikely to be very good for you.  In real world situations apple juice or smoothies are likely to be drunk in greater amounts than would be eaten as whole apples, further exacerbating these effects.

Additional points of interest

One volunteer in the study had to be excluded from the analysis as “…soon after her apple meal, she passed several watery stools containing obvious particles of fruit.” Almost half a kilo of apple was obviously too much for this poor woman. Another volunteer, after drinking the apple juice, showed blood sugar levels below 2.0 mmol/l (36 mg/dl), a very low level! These unusual findings are rarely reported in research papers now but do go to show how individual people can be in their responses.

Reference

Lancet. 1977 Oct 1;2(8040):679-82.
Depletion and disruption of dietary fibre. Effects on satiety, plasma-glucose, and serum-insulin.
Haber GB, Heaton KW, Murphy D, Burroughs LF.

Abstract:

“Ten normal subjects ingested test meals based on apples, each containing 60 g available carbohydrate. Fibre-free juice could be consumed eleven times faster than intact apples and four times faster than fibre-disrupted purée. Satiety was assessed numerically. With the rate of ingestion equalised, juice was significantly less satisfying than purée, and purée than apples. Plasma-glucose rose to similar levels after all three meals. However, there was a striking rebound fall after juice, and to a lesser extent after purée, which was not seen after apples. Serum-insulin rose to higher levels after juice and purée than after apples. The removal of fibre from food, and also its physical disruption, can result in faster and easier ingestion, decreased satiety, and disturbed glucose homoeostasis which is probably due to inappropriate insulin release. These effects favour overnutrition and, if often repeated, might lead to diabetes mellitus.”